Monitoring and early warning method and system for container portal crane

Abstract

The invention discloses a monitoring and early warning method and system for a container portal crane, and relates to the technical field of crane monitoring and early warning. According to the monitoring and early warning method for the container portal crane, monitoring time sequence data of the portal crane are continuously obtained, risk extraction processing is conducted on the monitoring time sequence data of the portal crane, and a risk assessment index set of each time period of the portal crane is obtained; comprising a structure risk polymerization index, a crane behavior risk index and an operation environment interference polymerization index; and comprehensively analyzing the risk assessment index set of each time period of the portal crane to obtain a risk coupling judgment index of each time period of the portal crane. According to the invention, a preset early warning measure is taken for the portal crane based on the risk coupling judgment index of each time period; therefore, the capability of predicting and early warning potential safety hazards is remarkably improved, the safety of the crane in the operation process is effectively guaranteed, and misjudgment and missed judgment are avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of crane monitoring and early warning, and in particular to a monitoring and early warning method and system for a container gantry crane. Background Art

[0002] As one of the important equipment in modern ports and logistics industries, container gantry cranes are widely used in container loading, unloading, transportation and stacking operations. This type of crane has a strong load capacity and high operating accuracy. It is usually used in large-scale container storage and transportation sites such as docks and warehouses. With the development of container logistics industry, the application of gantry cranes is also increasing. Its working efficiency and safety are crucial to modern logistics systems. However, under high-intensity working environment, gantry cranes have some potential safety hazards during long-term operation, which affects the working efficiency of the crane and may also cause equipment damage, casualties or serious economic losses. In addition, the traditional gantry crane monitoring system mainly relies on manual inspections and regular equipment inspections. Its monitoring methods are relatively single, and it is difficult to grasp the operating status of the equipment in real time and comprehensively, and it is difficult to accurately reflect the overall operating risk of the equipment.

[0003] Prior art, such as the real-time monitoring system and method for the operation status of container gantry crane yard disclosed in the patent application with announcement number: CN114132842B, includes a stand-alone control system and a remote operation console in the central control room, the stand-alone control system is connected to the remote operation console in communication, including a stand-alone PLC, several network cameras, and a distributed processing platform; the stand-alone PLC controls the operation of the crane and sends the current operation status to the distributed processing platform; several network cameras are set at different positions of the crane, which can shoot the operation dynamics or working area of ​​the crane under different viewing angles in real time; the distributed processing platform includes a video stream access module, a core algorithm module, a PLC interaction module, an alarm information output module, and a WEB server. The present invention makes full use of the existing remote control camera, the distributed front-end computing processing unit, integrates the crane operation status and multi-view cameras, and the multi-functions are automatically switched according to the operation status, with a high degree of intelligence, which effectively improves the operation efficiency and safety protection level.

[0004] Based on the above solution, it is found that the limitations of the existing technology include at least the following problems: it is difficult for the existing technology to extract and analyze multiple dimensional data involved in the operation of the gantry crane, such as structural operation data, lifting behavior data and environmental interference data. There is a lack of comprehensive analysis and efficient processing of different types of data, which makes the risk judgment less accurate. For example, the containers under the spreader are not stacked in a standardized manner. At the same time, the load is eccentric and shakes during the lifting process, which can easily lead to the instability of the spreader or excessive structural stress. The existing technology often analyzes these factors independently and ignores their mutual influence. It is difficult to comprehensively judge the linkage effect of structural risks and operating environment risks on safety, and thus it is difficult to accurately judge the safety hazards of the crane during operation. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a monitoring and early warning method and system for a container gantry crane, which solves the problem that the prior art is difficult to conduct multi-dimensional comprehensive analysis and ignores its interaction, thus leading to inaccurate risk judgment.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a monitoring and early warning method for a container gantry crane, comprising the following steps: continuously acquiring monitoring time series data of the gantry crane, the monitoring time series data including structural operation time series data, hoisting behavior time series data, and operation video stream time series data; performing risk extraction processing on the monitoring time series data of the gantry crane respectively to obtain a risk assessment index set for each time period of the gantry crane, including a structural risk aggregation index, a hoisting behavior risk index, and an operating environment interference aggregation index; performing a comprehensive analysis on the risk assessment index set for each time period of the gantry crane to obtain a risk coupling determination index for each time period of the gantry crane; taking preset early warning measures for the gantry crane based on the risk coupling determination index for each time period.

[0007] Furthermore, the specific formula for calculating the risk coupling determination index of a gantry crane in a certain period of time is as follows: ; Among them, FpD is the risk coupling determination index of the gantry crane in a certain period of time, GxH is the structural risk aggregation index of the gantry crane in a certain period of time, α 1 is the structural risk adjustment coefficient stored in the database, DxF is the risk index of the gantry crane's hoisting behavior in a certain period of time, ZgR is the operating environment interference aggregation index of the gantry crane in a certain period of time, α 2 is the behavior environment aggregation adjustment coefficient stored in the database, α 3 Aggregate adjustment coefficients for structural behaviors stored in the database.

[0008] Furthermore, the structural operation timing data for each time period includes the main beam impact response index, the outrigger force distribution value, the main beam strain variation index, and the vibration coordination index. The specific steps for obtaining the structural risk aggregation index for each time period of the gantry crane are as follows: a comprehensive analysis is performed on the structural operation timing data of the gantry crane to obtain a set of structural operation risk assessment indexes for each time period of the gantry crane, including a structural abnormal response intensity index and a structural coordination stability index; and a comprehensive analysis is performed on the set of structural operation risk assessment indexes for each time period of the gantry crane to obtain the structural risk aggregation index for each time period of the gantry crane.

[0009] Furthermore, the specific steps for obtaining the structural operation risk assessment index set of the gantry crane in each time period are as follows: read the main beam impact response index and the main beam strain variation index of the gantry crane in each time period, and perform a comprehensive analysis to obtain the structural abnormal response intensity index of the gantry crane in each time period; read the outrigger force distribution value and the vibration coordination index of the gantry crane in each time period, and perform a comprehensive analysis to obtain the structural coordination stability index of the gantry crane in each time period.

[0010] Furthermore, the specific formula for calculating the structural risk aggregation index of a gantry crane in a certain period of time is as follows: ; Among them, GxH is the structural risk aggregation index of the gantry crane in a certain period of time, YcX is the structural abnormal response intensity index of the gantry crane in a certain period of time, β 1 is the abnormal response adjustment coefficient stored in the database, XtW is the structural coordination stability index of the gantry crane in a certain period of time, β 2 is the coordination stability adjustment coefficient stored in the database, β 3 is the interaction coupling adjustment coefficient stored in the database, β 4 is the coupling sensitivity adjustment coefficient stored in the database.

[0011] Furthermore, the hoisting behavior time series data includes the hoisting equipment jerk frequency value, load eccentricity value, load sway coordination index, hoisting equipment inertia energy accumulation index, control stability index, and braking inertia dragging distance value in each time period. The specific steps for obtaining the hoisting behavior risk index of the gantry crane in each time period are as follows: comprehensively analyze the hoisting behavior time series data of the gantry crane to obtain a set of hoisting risk assessment indexes for each time period of the gantry crane, including an operation consistency risk index and a load coupling disturbance risk index; comprehensively analyze the hoisting risk assessment index set for each time period of the gantry crane to obtain the hoisting behavior risk index of the gantry crane in each time period.

[0012] Furthermore, the operation video stream time series data includes operation video stream data for each time period, and the operation video stream data includes several frames of operation image data, and the operation image data is specifically the pixel value and two-dimensional coordinates of each pixel point in the operation image. The specific steps of obtaining the operation environment interference aggregation index of each time period of the gantry crane are as follows: the pixel value and two-dimensional coordinates of each pixel point in each frame of the operation image of the gantry crane in each time period are input into the pre-trained operation risk identification model for comprehensive analysis to obtain the operation risk assessment index set of the gantry crane in each time period, including the operation obstacle interference index, the non-operation target misentry index, and the container stacking interference index; the operation risk assessment index set of the gantry crane in each time period is comprehensively analyzed to obtain the operation environment interference aggregation index of the gantry crane in each time period.

[0013] Furthermore, the operation risk identification model is specifically a visual transformer network, and the visual transformer network includes an input layer, an image coding layer, a visual transformer encoder layer, a risk attention feature extraction layer, and a task output layer. The specific steps of obtaining the operation risk assessment index set for each time period of the gantry crane are as follows: in the input layer of the visual transformer network, the operation video stream time series data of the gantry crane is received and preprocessed; in the image coding layer of the visual transformer network, the preprocessed operation video stream time series data of the gantry crane is subjected to image block coding processing to obtain an image block embedding vector sequence of each frame of the operation image of the gantry crane in each time period; in the visual transformer encoder layer of the visual transformer network, each frame of each time period of the gantry crane is subjected to image block coding processing to obtain an image block embedding vector sequence of each frame of the operation image of the gantry crane in each time period; The image block embedding vector sequence of the operating image is subjected to multi-layer attention transformation encoding processing to obtain the image block global feature vector sequence of each frame of the operating image of the gantry crane in each time period; in the risk attention feature extraction layer of the visual transformer network, feature extraction is performed on the image block global feature vector sequence of each frame of the operating image of the gantry crane in each time period to obtain the risk area feature vector set of each frame of the operating image of the gantry crane in each time period; in the task output layer of the visual transformer network, branch prediction processing is performed on the risk area feature vector set of each frame of the operating image of the gantry crane in each time period to obtain the operating obstacle interference index, non-operating target misentry index and container stacking interference index of the gantry crane in each time period, that is, the operating risk assessment index set.

[0014] Furthermore, the specific steps of taking preset early warning measures for the gantry crane based on the risk coupling determination index of each time period are as follows: reading the risk coupling determination index of each time period of the gantry crane, and performing a comprehensive analysis to obtain the risk coupling determination index of the next time period of the gantry crane; comparing and analyzing the risk coupling determination index of the next time period of the gantry crane with the preset risk coupling determination index threshold interval; if the risk coupling determination index of the next time period of the gantry crane is lower than the lower limit of the preset risk coupling determination index threshold interval, it is marked as a low risk, and the first early warning measure is taken; If the risk coupling determination index of the gantry crane in the next time period is between the preset risk coupling determination index threshold range, it is marked as medium risk and the second warning measure is taken; if the risk coupling determination index of the gantry crane in the next time period is higher than the preset upper limit of the risk coupling determination index threshold range, it is marked as high risk and the third warning measure is taken.

[0015] A monitoring and early warning system for a container gantry crane comprises: a monitoring data acquisition module, used for continuously acquiring monitoring time series data of the gantry crane, wherein the monitoring time series data comprises structural operation time series data, hoisting operation behavior time series data and operation video stream time series data; a risk extraction and processing module, used for respectively performing risk extraction and processing on the monitoring time series data of the gantry crane to obtain a risk assessment index set for each time period of the gantry crane, including a structural risk aggregation index, a hoisting operation behavior risk index and an operation environment interference aggregation index; a comprehensive risk assessment module, used for performing comprehensive analysis on the risk assessment index set for each time period of the gantry crane to obtain a risk coupling determination index for each time period of the gantry crane; and a risk intelligent early warning module, used for taking preset early warning measures for the gantry crane based on the risk coupling determination index for each time period.

[0016] The present invention has the following beneficial effects: (1) The monitoring and early warning method of the container gantry crane realizes multi-dimensional data fusion and risk assessment by comprehensively analyzing the structural operation data, hoisting behavior data and operation video stream data of the gantry crane, and performs coupling analysis on multiple risk assessment index sets in each time period to obtain more accurate and comprehensive risk judgment, thereby ensuring the safety of the gantry crane during operation. For example, by combining structural risk and hoisting behavior risk, it can identify potential instability risks caused by load eccentricity and hoisting gear shaking, thereby significantly improving the prediction and early warning capabilities of safety hazards, thereby effectively ensuring the safety of the crane during operation and avoiding misjudgment and missed judgment.

[0017] (2) The monitoring and early warning method of the container gantry crane introduces a visual transformer network to perform deep learning analysis on the operation video stream data of the gantry crane, so that it can accurately identify potential interference factors in the operating environment. After processing the complex information in each frame of the video, through image block encoding, global semantic modeling and local attention mechanism, it deeply explores the problems such as interference from operating obstacles, misentry of non-operating targets and interference from container stacking, and can identify and evaluate the dynamic risks in the operating environment with higher accuracy, thereby improving the real-time monitoring capability of the operating environment, providing more accurate data support for early warning, and ensuring the timeliness and accuracy of early warning.

[0018] (3) The monitoring and early warning method of the container gantry crane automatically predicts the risk status of the next period by performing trend analysis and risk change rate adjustment on the risk coupling judgment index of the gantry crane in each period, and triggers early warning measures of different levels according to the prediction results, so as to ensure the safety of the operation, thereby improving the real-time and accuracy of the safety response. It can also adjust the early warning strategy in real time according to the actual operation situation, thereby improving the real-time and accuracy of the safety response, thereby avoiding the lag and false alarm rate of the early warning, and then ensuring that the gantry crane can provide accurate early warning in a complex and changeable operating environment, thereby improving the safety and operation efficiency during the operation.

[0019] (4) The monitoring and early warning system of the container gantry crane, through modular design, realizes real-time monitoring and intelligent early warning of multi-dimensional risks in the operation process of the container gantry crane, thereby improving the automation and intelligence level of the system. For example, the risk extraction and processing module conducts detailed risk extraction and analysis on various types of data to obtain the structural risk aggregation index, crane operation risk index and operating environment interference aggregation index for each time period. Through the collaborative analysis of multivariate data, a comprehensive assessment of various potential risks that may exist in the operation process can be made, thereby reducing the risk of human operational errors, thereby improving the system's autonomous decision-making and real-time response capabilities, and providing more accurate, real-time and intelligent guarantees for the safe operation of the container gantry crane.

[0020] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 The present invention is a flow chart of a monitoring and early warning method for a container gantry crane.

[0022] Figure 2 The present invention is a flowchart of the specific steps of obtaining the operating environment interference aggregation index of the gantry crane in each time period in a monitoring and early warning method of the container gantry crane.

[0023] Figure 3 It is a time sequence diagram of an operation risk assessment index set of a gantry crane in a monitoring and early warning method of a container gantry crane of the present invention.

[0024] Figure 4 The present invention is a block diagram of a monitoring and early warning system for a container gantry crane. DETAILED DESCRIPTION

[0025] See also Figure 1 The embodiment of the present invention provides a technical solution: a monitoring and early warning method for a container gantry crane, comprising the following steps: continuously acquiring monitoring time series data of a (container) gantry crane, the monitoring time series data including structural operation time series data, hoisting behavior time series data, and operation video stream time series data; performing risk extraction processing on the monitoring time series data of the gantry crane respectively to obtain a risk assessment index set for each time period of the gantry crane, including a structural risk aggregation index, a hoisting behavior risk index, and an operation environment interference aggregation index; performing a comprehensive analysis on the risk assessment index set for each time period of the gantry crane to obtain a risk coupling determination index for each time period of the gantry crane; taking preset early warning measures for the gantry crane based on the risk coupling determination index for each time period.

[0026] The specific formula for calculating the risk coupling determination index of a gantry crane in a certain period of time is as follows: ; Among them, FpD is the risk coupling determination index of the gantry crane in a certain period of time, GxH is the structural risk aggregation index of the gantry crane in a certain period of time, α 1 is the structural risk adjustment coefficient stored in the database, DxF is the risk index of the gantry crane's hoisting behavior in a certain period of time, ZgR is the operating environment interference aggregation index of the gantry crane in a certain period of time, α 2 is the behavior environment aggregation adjustment coefficient stored in the database, α 3 Aggregate adjustment coefficients for structural behaviors stored in the database.

[0027] It should be explained that the formula This item is used to adjust the mutual influence between the structural risk aggregation index and the crane operation behavior risk index, and to control the amplification or suppression of the risk coupling judgment index by the interaction between the structural risk aggregation index and the crane operation behavior risk index.

[0028] α 1 , α 2 , α 3It can be obtained through the following steps: using historical data, combined with the structural risk aggregation index, the crane operation behavior risk index, and the working environment interference aggregation index, statistical regression analysis is performed to quantify the specific impact of each factor on the risk coupling judgment index, thereby fitting the initial weight value; secondly, using the sensitivity analysis method, adjust the value range of each coefficient, observe its impact on the risk coupling judgment result, ensure the stability and rationality of the model, and based on the actual situation, correct and optimize the preliminary fitted coefficients to finally determine the coefficient value.

[0029] Specifically, the structural operation timing data for each period include the main beam impact response index, the outrigger force distribution value, the main beam strain variation index, and the vibration coordination index. The specific steps for obtaining the structural risk aggregation index for each period of the gantry crane are as follows: a comprehensive analysis is performed on the structural operation timing data of the gantry crane to obtain a set of structural operation risk assessment indexes for each period of the gantry crane, including a structural abnormal response intensity index and a structural coordination stability index; and a comprehensive analysis is performed on the set of structural operation risk assessment indexes for each period of the gantry crane to obtain the structural risk aggregation index for each period of the gantry crane.

[0030] Among them, the main beam impact response index is the structural response of the main beam caused by impact or load change. It can obtain the vibration acceleration value at each time point in the period through the acceleration sensor, and perform statistical analysis and mean processing respectively to obtain the maximum vibration acceleration value and the mean vibration acceleration, and perform ratio processing, that is, the maximum vibration acceleration value / mean vibration acceleration. The result is the main beam impact response index.

[0031] The outrigger force distribution value is the degree of imbalance of the pressure borne by each outrigger of the container gantry crane. That is, the pressure borne by each outrigger at each time point in the period can be obtained by a pressure sensor, and standard deviation processing is performed. Based on the standard deviation processing result, mean processing is performed, and the result obtained is the outrigger force distribution value.

[0032] The main beam strain variation index is a measure of the dispersion of strain values ​​of multiple strain measuring points of the main beam of the container gantry crane within the same monitoring period during the period. It can be achieved by arranging multiple resistance strain gauges on the surface of the main beam, and obtaining the strain value of each strain measuring point at each time point in the period, and performing mean and standard deviation processing respectively to obtain the strain mean and strain standard deviation at each time point, and performing ratio processing, that is, strain standard deviation / strain mean, and performing mean processing based on the ratio processing result, and the result is the main beam strain variation index.

[0033] The vibration synergy index is a measure of the similarity and degree of synergy between the vibration amplitudes of multiple key structural parts (such as the mid-span area of ​​the main beam, the connection between the main beam and the outrigger, the connection point between the bottom of the outrigger and the track, etc.). It can obtain the vibration amplitude of each key structural part at each time point through a vibration sensor, and calculate the correlation value of the vibration amplitudes of adjacent key structural parts at each time point based on the Pearson correlation coefficient, and perform mean processing. The result is the vibration synergy index.

[0034] The specific steps for obtaining the structural operation risk assessment index set of the gantry crane in each period are as follows: read the main beam impact response index and the main beam strain variation index of the gantry crane in each period, and conduct a comprehensive analysis (first perform normalization processing, and perform weighted processing based on the normalization processing result, and map the weighted processing result to between 0 and 1 based on the sigmoid function), and obtain the structural abnormal response intensity index of the gantry crane in each period; read the outrigger force distribution value and vibration coordination index of the gantry crane in each period, and conduct a comprehensive analysis (first perform normalization processing, and perform weighted processing based on the normalization processing result, and map the weighted processing result to between 0 and 1 based on the sigmoid function), and obtain the structural coordination stability index of the gantry crane in each period.

[0035] The specific formula for calculating the structural risk aggregation index of a gantry crane in a certain period of time is as follows: ; Among them, GxH is the structural risk aggregation index of the gantry crane in a certain period of time, YcX is the structural abnormal response intensity index of the gantry crane in a certain period of time, β 1 is the abnormal response adjustment coefficient stored in the database, XtW is the structural coordination stability index of the gantry crane in a certain period of time, β 2 is the coordination stability adjustment coefficient stored in the database, β 3 is the interaction coupling adjustment coefficient stored in the database, β 4 is the coupling sensitivity adjustment coefficient stored in the database.

[0036] It should be explained that the formula This item is used to adjust the risk linkage amplification effect that is prone to occur in the gantry crane structure under critical instability conditions when the structural abnormal response intensity index is high and the structural coordination stability index is low, so as to avoid the structural risk aggregation index being too high or too low.

[0037] β 1 , β 2 , β 3 , β 4It can be obtained through the following steps: Based on historical data, the initial impact weights of each variable (structural abnormal response intensity index, structural coordination stability index) on the structural risk aggregation index are determined through statistical regression analysis. Then, the range of coefficients is adjusted using the sensitivity analysis method to evaluate the stability and applicability of these parameters to the formula output. Next, the weights are further fitted through model optimization (such as machine learning algorithms or multi-objective optimization) to ensure that the formula can accurately reflect the risk status of the actual crane structure.

[0038] In this implementation scheme, the limitations of a single indicator are avoided through the fusion analysis of multiple structural data, and the key factors in the operation of the structure are accurately considered, so that complex structural risks can be effectively identified. Secondly, historical data is used for statistical regression analysis to quantify the impact of different structural risk factors, and the adjustment coefficient is adjusted through sensitivity analysis to ensure the real-time update and adaptability of the structural risk assessment formula, so that it can dynamically respond to the changes of the gantry crane in different working environments, ensure the real-time and accuracy of risk assessment, and introduce an interactive coupling adjustment coefficient to fine-tune the structural risk aggregation index, thereby effectively preventing the imbalance of the overall risk assessment caused by the abnormality of a single risk indicator and providing a more accurate risk judgment. Finally, through a more refined structural risk aggregation analysis, the potential risks of structural instability or excessive stress during crane operation can be identified, providing a more accurate decision-making basis for the early warning system, thereby effectively improving the overall safety assurance capability of the gantry crane and preventing accidents caused by inaccurate risk judgment.

[0039] Specifically, the time series data of lifting behavior include the sling jerk frequency value, load eccentricity value, load sway coordination index, sling inertia energy accumulation index, control stability index, and braking inertia drag distance value in each time period. The specific steps to obtain the lifting behavior risk index of the gantry crane in each time period are as follows: conduct a comprehensive analysis on the time series data of the gantry crane's lifting behavior to obtain a set of lifting risk assessment indexes for each time period of the gantry crane, including an operation consistency risk index and a load coupling disturbance risk index; conduct a comprehensive analysis (i.e., perform weighted processing) on ​​the set of lifting risk assessment indexes for each time period of the gantry crane to obtain the lifting behavior risk index of the gantry crane in each time period.

[0040] Among them, the sling jerk frequency value is the number of acceleration mutations that occur in the sling during the period, which is used to identify the operational stability during the lifting process. It can obtain the acceleration value of the sling at each time point in the period through the acceleration sensor, and compare it with the set acceleration threshold (the historical acceleration values ​​of the sling at several historical time points can be obtained, and the average value is processed, and the result is set as the acceleration threshold) for judgment and analysis. The time point above the acceleration threshold is marked as an acceleration mutation, and statistical analysis is performed. The result is the sling jerk frequency value.

[0041] The load eccentricity value is the degree of offset between the actual load-bearing mass center and the geometric center, reflecting whether there is asymmetric loading of the load. It can obtain the four-point force values ​​of the four hanging point wire ropes and the corresponding three-dimensional position coordinates through tension sensors, and obtain the three-dimensional position coordinates of the load center based on static moment equilibrium, and obtain the three-dimensional position coordinates of the geometric center of the load. Based on the Euclidean distance formula, the three-dimensional position coordinates of the load center and the three-dimensional position coordinates of the geometric center are analyzed and calculated, and the result is the load eccentricity value.

[0042] The load sway coordination index is the degree of synchronization between the front and rear swing angles and the left and right swing angles of the hoist during the swinging process. The pitch angle and roll angle at each time point in the period can be obtained through the inertial measurement unit, and the degree of time synchronization between the two is calculated based on the Pearson correlation, and the average processing is performed. The result is the load sway coordination index.

[0043] The inertial energy accumulation index of the sling is the total amount of inertial kinetic energy accumulated by the sling due to continuous movement during the period. It can collect the horizontal acceleration signal in real time through the sling inertial measurement unit, combined with the current load mass (it can obtain the total vertical tension borne by the sling through the tension sensor, and the load mass is analyzed based on Newton's second law). The square value of the acceleration per unit time in the period is calculated and integrated to obtain the accumulated inertial kinetic energy, which is used to identify whether there is a continuous accumulation of energy during the operation of the sling, and then evaluate the potential structural vibration and impact risks during the lifting process.

[0044] The control stability index is the degree of deviation between the running speeds of the trolley and the spreader during the period and the trolley speed reference value (by obtaining the trolley speed at several historical time points and performing average processing) and the spreader speed reference value (by obtaining the spreader speed at several historical time points and performing average processing). The trolley running speed and the spreader running speed at each time point in the period can be obtained by speed sensors, and average processing is performed to obtain the average trolley running speed and the average spreader running speed, and then ratio processing is performed with the trolley speed reference value and the spreader speed reference value, respectively. Weighted processing is performed based on the ratio processing results, and the result obtained is the control stability index.

[0045] The braking inertia drag distance value is the additional distance of inertia dragging after the stop command of the trolley and the spreader is issued during the period. The displacement change from the issuance of the command to the complete stop of the trolley or the spreader can be obtained through the displacement sensor, and the displacement change of the trolley and the spreader is weighted. The result is the braking inertia drag distance value.

[0046] The specific steps to obtain the lifting risk assessment index set of the gantry crane in each time period are as follows: read the sling jerk frequency value, control stability index, and braking inertia dragging distance value of the gantry crane in each time period, and conduct a comprehensive analysis (first perform standardization processing, and then perform weighted processing based on the standardization processing result) to obtain the operation consistency risk index (operation execution stability and dynamic response consistency) of the gantry crane in each time period; read the load eccentricity value, load sway coordination index, and sling inertia energy accumulation index of the gantry crane in each time period, and conduct a comprehensive analysis (first perform standardization processing, and then perform weighted processing based on the standardization processing result) to obtain the load coupling disturbance risk index (the degree to which the load state causes disturbance risk to the sling shaking behavior) of the gantry crane in each time period.

[0047] In this implementation scheme, through real-time monitoring of multiple key parameters such as the frequency of sling jerk movements, load eccentricity, load sway coordination, inertial energy accumulation, etc., the stability and safety of various behaviors in the lifting process can be comprehensively evaluated, thereby ensuring an accurate evaluation of the crane lifting behavior. Secondly, through the analysis of indicators such as the frequency of sling jerk movements and the control stability index, it is helpful to judge whether the crane operation is stable, whether there are abnormal conditions such as sudden acceleration changes, and avoid safety hazards such as structural instability or sling collision caused by inconsistent operation or unstable load during operation. Finally, the comprehensive analysis of the lifting behavior data can help the system make more detailed risk predictions during real-time monitoring. For example, when the sling moves jerkily, the load is unstable or sways, it can quickly identify and evaluate potential risks, trigger the early warning mechanism, and avoid accidents. This refined risk identification and early warning capability can effectively improve the safety and response speed of the entire crane operation process, and through weighted processing and dynamic evaluation of different risk indicators, it can adjust the risk index according to the actual operation status and optimize the early warning strategy.

[0048] Specifically, Figure 2As shown, the operation video stream time series data includes the operation video stream data of each time period, and the operation video stream data includes several frames of operation image data. The operation image data is specifically the pixel value and two-dimensional coordinates of each pixel point in the operation image. The specific steps of obtaining the operation environment interference aggregation index of each time period of the gantry crane are as follows: the pixel value and two-dimensional coordinates of each pixel point in each frame of the operation image of the gantry crane in each time period are input into the pre-trained operation risk identification model for comprehensive analysis to obtain the operation risk assessment index set of each time period of the gantry crane, including the operation obstacle interference index (identifying the interference of static objects in the descent path of the spreader), the non-operation target misentry index (identifying dynamic intrusion behaviors such as unplanned entry of personnel and vehicles in the lifting operation area), and the container stacking interference index (identifying the abnormal state of container stacking at the target drop point of the spreader); the operation risk assessment index set of each time period of the gantry crane is comprehensively analyzed to obtain the operation environment interference aggregation index of each time period of the gantry crane.

[0049] Among them, the specific formula for calculating the operating environment interference aggregation index of the gantry crane in a certain period of time is as follows: ; Among them, ZgR is the interference aggregation index of the operating environment of the gantry crane in a certain period of time, ZaW is the interference index of the operating obstacles of the gantry crane in a certain period of time, FmB is the non-operating target error index of the gantry crane in a certain period of time, μ 1 is the superposition adjustment coefficient stored in the database, DgR is the container stacking interference index of the gantry crane in a certain period of time, μ 2 It is the stacking interference adjustment coefficient stored in the database.

[0050] It needs to be explained that μ 1 , μ 2 It can be obtained through the following steps: Based on historical data, determine the initial impact weight of each variable (such as the coupling interaction between the operating obstacle interference index and the non-operating target misentry index, and the container stacking interference index) on the operating environment interference aggregation index through statistical regression analysis. Then, use the sensitivity analysis method to adjust the value range of the coefficient to evaluate the stability and applicability of these parameters to the formula output. Next, further fit the weights through model optimization (such as multi-objective optimization) to ensure that the formula can accurately reflect the interference status of the actual working environment.

[0051] The specific implementation example of calculating the operating environment interference aggregation index of a gantry crane in a certain period of time is as follows. The following data are available: including calculating the operating obstacle interference index, non-operating target misentry index, and container stacking interference index of the gantry crane in 5 periods of time, as shown in Table 1 and Figure 3 : Table 1 Example of time series data of gantry crane operation risk assessment index

[0052] The interaction adjustment coefficient μ stored in the database 1 Approximately: 0.317; The stacking interference adjustment coefficient μ stored in the database 2 Approximately: 0.573; Substituting the data in Table 1 and the above adjustment coefficient into the specific formula for calculating the working environment interference aggregation index of the gantry crane in a certain period of time, we get: The first period of the gantry crane working environment interference aggregation index = ln (1 + 0.317 × (0.103 + 1) × (0.042 + 1)) × √ (exp (0.537 × (1 + 0.121))) ≈ 0.427; The interference aggregation index of the working environment of the gantry crane in the second period = ln (1 + 0.317 × (0.254 + 1) × (0.126 + 1)) × √ (exp (0.537 × (1 + 0.194))) ≈ 0.521; The interference aggregation index of the working environment of the gantry crane in the third period = ln (1 + 0.317 × (0.346 + 1) × (0.413 + 1)) × √ (exp (0.537 × (1 + 0.253))) ≈ 0.676; The fourth period of the gantry crane working environment interference aggregation index = ln (1 + 0.317 × (0.268 + 1) × (0.234 + 1)) × √ (exp (0.537 × (1 + 0.401))) ≈ 0.601; The working environment interference aggregation index of the gantry crane in the fifth period = ln (1 + 0.317 × (0.153 + 1) × (0.092 + 1)) × √ (exp (0.537 × (1 + 0.687))) ≈ 0.475.

[0053] The operation risk identification model is specifically a visual transformer network (visual transformer). The visual transformer network includes an input layer, an image encoding layer, a visual transformer encoder layer, a risk attention feature extraction layer, and a task output layer. The specific steps for obtaining the operation risk assessment index set of the gantry crane in each time period are as follows: In the input layer of the visual transformer network, the operation video stream time series data of the gantry crane is received (the pixel value and two-dimensional coordinate of each pixel point in each frame of the operation image in each time period) and preprocessed; in the image encoding layer of the visual transformer network, the preprocessed operation video stream time series data of the gantry crane is subjected to image block encoding processing (each frame of the operation image is divided into several image blocks according to a fixed size, and then the pixel value of each image block is linearly mapped, and the two-dimensional image block is converted into a vector representation of a fixed dimension, that is, an image block embedding vector. At the same time, for each image block, its two-dimensional coordinate position in the original image is embedded, and the position information is embedded into the corresponding vector using a position encoding method), and an image block embedding vector sequence of each frame of the operation image of the gantry crane in each time period is obtained (several image block embedding vectors sequence); in the visual transformer encoder layer of the visual transformer network, the image block embedding vector sequence of each frame of the working image of the gantry crane in each period is subjected to multi-layer attention transformation encoding processing (each layer of the transformer encoder includes a multi-head self-attention mechanism submodule, a feedforward neural network submodule, and a residual connection and layer normalization module. In each encoder layer, the multi-head self-attention mechanism is used to establish the dependency relationship between the current image block and all other image blocks. The global attention calculation mechanism is used to enable each image block vector to integrate the context information in the entire image. Then, the feedforward neural network, that is, a two-layer fully connected neural network, is used to perform a nonlinear transformation on the sub-attention output result to enhance the feature expression ability. At the same time, residual connection and layer normalization operations are introduced after the output of each submodule to maintain feature stability and improve model training efficiency, so that the image block vector obtains deeper and richer global semantic information layer by layer. Each vector not only retains the position information of its original image block, but also integrates the context features from other areas in the entire image), and obtains the image block global feature vector sequence of each frame of the working image of the gantry crane in each period;In the risk attention feature extraction layer of the visual transformer network, feature extraction is performed on the global feature vector sequence of the image blocks of each frame of the operation image of the gantry crane in each period (the target detection network stored in the database is called to automatically identify the spreader appearing in the image and locate the bounding box, so as to obtain the rectangular area range of the spreader in the current frame image, and according to the position of the box in the image space, several subordinate sub-areas related to the spreader space are automatically generated, including: a descending path area composed of several rows of image blocks extending downward from the spreader frame as the center, a spreader landing area composed of image blocks aligned directly below the spreader frame, and an operation buffer area formed by image blocks within a certain range around the spreader frame. The above areas can be accurately calibrated in the image block sequence through simple pixel coordinate expansion and image block index mapping, and the image block vector set corresponding to the calibrated risk sub-area is extracted from the global feature sequence of the image block, and a local self-attention mechanism is constructed for each sub-area, such as generating three sets of vector representations of query, key and value for each sub-area image block set, and based on The soft maximization attention function calculates the semantic correlation weights between image blocks, and then performs weighted fusion on the original feature vectors within the region to obtain an enhanced image block sequence. The enhanced sequence is then further pooled, including weighted average or maximum response selection, to output a regional feature vector with the same dimension as the original image block as a high-dimensional semantic expression representing the potential risk structure, abnormal target or dynamic interference state of the sub-region. For example, the feature vector of the operating obstacle region contains feature dimensions such as regional surface continuity, texture mutation density, and grayscale gradient variance that characterize the existence of obstacles. The feature vector of the non-operating target misentry region contains structural activation features such as contour symmetry, target movement trajectory pattern, and height-width ratio deviation that reflect the intrusion behavior of dynamic targets. The feature vector of the container stacking interference region contains regional discriminant features such as edge line alignment, height gradient anomaly, and irregular corner point distribution that characterize the stability of the container stacking structure. The risk region feature vector set of each frame of the operating image of the gantry crane in each period is obtained.In the task output layer of the visual transformer network, the risk area feature vector set of each frame of the operating image of the gantry crane in each period is subjected to branch prediction processing (an independent task branch network structure is set for each type of operating risk, corresponding to the operating obstacle interference risk index, the non-operating target misentry risk index and the container stacking interference risk index, each task branch includes a set of fully connected neural network modules for risk discrimination, and the input is the feature vector of the corresponding area, and then through multi-layer linear transformation, non-linear activation function, such as ReLU or GELU and regularization processing, such as Dropout, the input features are deeply semantically mapped and pattern classified. Compressed, and finally a scalar value is output, and the scalar value is normalized to between 0 and 1 by the Sigmoid activation function, which is used to represent the probability of occurrence or confidence strength of the risk corresponding to the current area in the frame image. For example, the operating obstacle area feature vector is input to the obstacle After the risk discrimination network, if the output value is close to 1, it means that the system highly determines that there is a significant object interference risk in the area directly below the spreader. The output of the non-operating target misentry branch is close to 1, indicating that there is a clear human target or dynamic object in the current image in the spreader operation area; and when the container stacking interference branch outputs a higher value, it means that the container structure in the current spreader landing area has abnormal stacking characteristics such as tilt and misalignment. Three risk indexes are calculated for each frame of the image, and finally the three index values ​​of all frames are aggregated and analyzed within a monitoring period, including average, maximum, sliding weighted and other strategies, to obtain the operation obstacle interference index, non-operating target misentry index and container stacking interference index of the period, which constitute the operation risk assessment index set of the current operation cycle of the gantry crane). The operation obstacle interference index, non-operating target misentry index and container stacking interference index of each period of the gantry crane are obtained, that is, the operation risk assessment index set. ;

[0054] Among them, the input layer is used to receive the operation video stream image of the gantry crane and perform image standardization and pixel information preprocessing.

[0055] The image coding layer is used to receive the operating video stream images of the gantry crane and perform image standardization and pixel information preprocessing.

[0056] The encoder layer of the visual transformer is used to model the global semantic relationship between image patches through a multi-layer attention mechanism, generating a sequence of global feature vectors of image patches with context-awareness.

[0057] The risk attention feature extraction layer is used to delineate the risk area based on the position of the spreader, and perform local attention enhancement on the image blocks in each sub-area to extract the risk area feature vector.

[0058] The task output layer is used to extract the three regional feature vectors and input them into their respective risk discrimination networks to output the operation obstacle interference index, non-operation target misentry index and container stacking interference index.

[0059] And the pre-training process of the visual transformer network is as follows: Obtain a labeled image dataset, including a large amount of labeled image data with targets such as spreaders, containers, and work areas, preprocess the images and target location information, and divide the labeled image dataset into a sample training set and a sample verification set.

[0060] Initialize the visual transformer network, including the weights and bias parameters of the image block embedding coding layer, transformer encoder layer, and task output layer, using random initialization or loading existing model parameters.

[0061] Training is performed based on the sample training set, with the number of cycles set (e.g., 100 times), and in each training cycle, the forward propagation process is performed (the sample data is input into each layer of the visual transformer network, and then the three risk index prediction values ​​under the current batch are output), the loss function is calculated (the difference between the predicted output value and the true label is used to calculate the loss function, for example, the binary cross entropy loss function is used to calculate the prediction error of the operating obstacle interference index, or the mean square error function is used to measure the error size), and the back propagation and parameter update are performed (the gradient value of each neural network parameter is automatically calculated, and the weights and bias parameters in the network are updated based on the set optimization algorithm, such as Adam or SGD).

[0062] After each training cycle, an evaluation and analysis is performed based on the sample validation set. That is, for each frame of the validation image, the model will generate an image block embedding vector, a global feature sequence, and a risk area feature representation in turn, and output three risk prediction indexes: the interference index of operating obstacles, the index of misentry of non-operating targets, and the interference index of container stacking. Next, the predicted three risk indices are compared and analyzed with the true labels marked in the validation set samples, and the validation loss value of each type of risk output is calculated, such as the cross entropy loss or the mean square error. At the same time, multiple performance evaluation indicators are statistically analyzed, such as the average prediction accuracy (Accuracy), AUC (area under the curve), Precision, Recall, F1 score, etc., to comprehensively measure the performance of the model in various risk identification tasks. In addition, the distribution trend of the output values ​​of each risk index in the validation set, the frequency of abnormal predictions, and the degree of gradient fluctuation are monitored to determine whether the model has an overfitting tendency or insufficient training. An early stopping mechanism is also provided. If there is no obvious improvement in the validation indicators for several consecutive rounds, the subsequent training is automatically terminated, the optimal model parameters are retained, and the training process ends to obtain a trained network model.

[0063] In this implementation, by combining the operation video stream data with the visual transformer network for multi-dimensional risk identification, it can not only handle static obstacles in the lifting operation in each period, but also dynamically identify the misentry of non-operating targets and abnormal container stacking, so as to comprehensively monitor the crane operation environment, thereby improving the comprehensiveness of operation safety monitoring. Secondly, through the multi-layer attention mechanism of the visual transformer network, it can go deep into the feature extraction at the image block level and accurately identify the potential risks of the spreader's descent path, landing area and surrounding buffer zone. For example, when identifying interference from operating obstacles, the system can detect texture changes, surface continuity, The grayscale gradient and other features are analyzed to detect possible object interference, thereby effectively reducing the risk of false alarms and missed alarms. In addition, through the pre-training and adaptive optimization of the visual transformer network, the accuracy of the model in different operating environments can be continuously improved, and the misjudgment caused by overfitting or insufficient training can be effectively reduced, thereby improving the long-term stability and accuracy. Finally, by adopting the global feature and regional feature extraction method based on image blocks, the risk assessment results are not only highly accurate, but also have strong interpretability, so that the operator can intuitively understand the safety risk level of the current operating area, and further improve the efficiency and safety of operation management.

[0064] Specifically, the specific steps for taking preset early warning measures for gantry cranes based on the risk coupling determination index of each time period are as follows: read the risk coupling determination index of each time period of the gantry crane, and conduct a comprehensive analysis to obtain the risk coupling determination index of the next time period of the gantry crane; compare and analyze the risk coupling determination index of the next time period of the gantry crane with the preset risk coupling determination index threshold interval; if the risk coupling determination index of the next time period of the gantry crane is lower than the lower limit of the preset risk coupling determination index threshold interval, then (the next time period) is marked as low risk, and the first early warning measure is taken, specifically recording the current state as stable operation without triggering any active control intervention; if the risk coupling determination index of the next time period of the gantry crane is between the preset risk coupling determination index threshold interval, then (the next time period) It is marked as medium risk, and the second early warning measure is taken, specifically triggering a prompt early warning signal, such as a voice prompt, to remind the operator to pay attention to the operating status, and automatically start the frequency enhancement monitoring mode, speed up the data refresh frequency, track risk fluctuations in real time, and limit some risk actions, such as prohibiting rapid lifting and limiting the left and right extreme lateral movement of the hoist; if the risk coupling determination index of the next time period of the gantry crane is higher than the preset upper limit of the risk coupling determination index threshold range, then (the next time period) is marked as high risk, and the third early warning measure is taken, specifically triggering a red alarm signal and an audible and visual alarm, and simultaneously informing the operator of the gantry crane of the alarm information, and starting the automatic intervention mechanism, including: slowing down operation, suspending some actions, locking the operation authority of the high-risk area, and recording the time period as a high-risk historical sample for risk model adjustment and working condition reconstruction.

[0065] Among them, the specific steps of obtaining the risk coupling determination index of the next time period of the gantry crane are as follows: perform trend analysis on the risk coupling determination index of each time period of the gantry crane (such as the rate of change between two adjacent time periods, that is, the difference divided by the value of the previous time period, and then perform weighted processing), and obtain the risk coupling determination index change rate of several groups of prediction time periods of the gantry crane (such as 3 consecutive time periods as a group); and perform mean processing on the risk coupling determination index of each time period of the gantry crane (that is, perform mean processing on the risk coupling determination index of 3 consecutive time periods), and obtain the operation risk determination benchmark value of each group of prediction time period of the gantry crane, and perform risk trend adjustment processing in combination with the risk coupling determination index change rate, that is, according to the risk coupling determination index change rate (increase or decrease), adjust the benchmark value of the group up and down, and obtain the predicted risk coupling determination index of each group of prediction time period of the gantry crane, and use the corresponding group of prediction time period as the risk coupling determination index of the next time period.

[0066] In this implementation scheme, by performing trend analysis and risk change rate adjustment based on the risk coupling determination index of each time period, the risk status of the gantry crane in the next time period can be accurately predicted, and corresponding early warning measures can be taken according to the prediction results, so as to respond to changes in the operating environment in real time, avoid the lag and false alarm rate of the early warning, and ensure a more timely safety response. Secondly, based on the comparison between the risk coupling determination index and the preset threshold, the system can divide the risk status into low, medium and high risks, and trigger corresponding early warning measures, so as to refine the risk control strategy, help operators and managers take action in advance, reduce the occurrence of emergencies, thereby enhancing predictability and foresight. Finally, through the dynamic adjustment of the risk change rate, the risk management strategy can be intelligently optimized, risk fluctuations can be tracked in real time, and a quick response can be made according to the actual situation, further improving the safety and stability of gantry crane operations.

[0067] See also Figure 4 The embodiment of the present invention provides a technical solution: a monitoring and early warning system for a container gantry crane, comprising: a monitoring data acquisition module, used to continuously acquire monitoring time series data of the gantry crane, the monitoring time series data including structural operation time series data, hoisting behavior time series data, and operation video stream time series data; a risk extraction and processing module, used to perform risk extraction and processing on the monitoring time series data of the gantry crane respectively, to obtain a risk assessment index set for each time period of the gantry crane, including a structural risk aggregation index, a hoisting behavior risk index, and an operating environment interference aggregation index; a comprehensive risk assessment module, used to perform a comprehensive analysis on the risk assessment index set for each time period of the gantry crane, to obtain a risk coupling determination index for each time period of the gantry crane; a risk intelligent early warning module, used to take preset early warning measures for the gantry crane based on the risk coupling determination index for each time period.

[0068] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0069] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A monitoring and early warning method for a container gantry crane, characterized in that: The following steps are involved: Continuously acquiring monitoring time series data of the gantry crane, wherein the monitoring time series data includes structure operation time series data, crane operation behavior time series data, and operation video stream time series data; The monitoring time series data of the gantry crane are processed for risk extraction respectively to obtain the risk assessment index set of the gantry crane in each period, including the structural risk aggregation index, the crane operation behavior risk index, and the operating environment interference aggregation index; Comprehensively analyze the risk assessment index set of each period of the gantry crane to obtain the risk coupling determination index of each period of the gantry crane; Based on the risk coupling judgment index in each time period, preset early warning measures are taken for gantry cranes.

2. The container gantry crane monitoring and early warning method according to claim 1 is characterized in that: The specific formula for calculating the risk coupling determination index of a gantry crane in a certain period of time is as follows: ； Among them, FpD, GxH, DxF, and ZgR are the risk coupling judgment index, structural risk aggregation index, crane operation behavior risk index, and working environment interference aggregation index of the gantry crane in a certain period of time, respectively. α1, α2, and α3 are the structural risk adjustment coefficient, behavior environment aggregation adjustment coefficient, and structure behavior aggregation adjustment coefficient stored in the database, respectively.

3. The container gantry crane monitoring and early warning method according to claim 1 is characterized in that: The structural operation time series data for each period includes the main beam impact response index, the outrigger force distribution value, the main beam strain variation index, and the vibration coordination index. The specific steps for obtaining the structural risk aggregation index for each period of the gantry crane are as follows: The structural operation sequence data of the gantry crane are comprehensively analyzed to obtain the structural operation risk assessment index set of the gantry crane in each period, including the structural abnormal response intensity index and the structural coordination stability index; A comprehensive analysis is then conducted on the structural operation risk assessment index set of the gantry crane in each period to obtain the structural risk aggregation index of the gantry crane in each period.

4. The container gantry crane monitoring and early warning method according to claim 3 is characterized in that: The specific steps to obtain the structural operation risk assessment index set of the gantry crane in each period are as follows: Read the main beam impact response index and main beam strain variation index of the gantry crane in each period, and conduct a comprehensive analysis to obtain the structural abnormal response strength index of the gantry crane in each period; The force distribution value of the outriggers and the vibration coordination index of the gantry crane in each period are read, and a comprehensive analysis is performed to obtain the structural coordination stability index of the gantry crane in each period.

5. The monitoring and early warning method for container gantry crane according to claim 3 is characterized in that: The specific formula for calculating the structural risk aggregation index of a gantry crane in a certain period of time is as follows: ； Among them, GxH, YcX, and XtW are the structural risk aggregation index, structural abnormal response intensity index, and structural coordination stability index of the gantry crane in a certain period of time respectively, and β1, β2, β3, and β4 are the abnormal response adjustment coefficient, coordination stability adjustment coefficient, interactive coupling adjustment coefficient, and coupling sensitivity adjustment coefficient stored in the database respectively.

6. The container gantry crane monitoring and early warning method according to claim 1 is characterized in that: The hoisting behavior time series data includes the sling jerk frequency value, load eccentricity value, load sway coordination index, sling inertia energy accumulation index, control stability index, and braking inertia drag distance value in each period. The specific steps for obtaining the hoisting behavior risk index of the gantry crane in each period are as follows: The hoisting behavior time series data of the gantry crane are comprehensively analyzed to obtain the hoisting risk assessment index set of the gantry crane in each period, including the operation consistency risk index and the load coupling disturbance risk index; A comprehensive analysis is conducted on the lifting risk assessment index set of the gantry crane in each period to obtain the lifting behavior risk index of the gantry crane in each period.

7. The container gantry crane monitoring and early warning method according to claim 1 is characterized in that: The operation video stream time series data includes the operation video stream data of each time period, and the operation video stream data includes several frames of operation image data. The operation image data is specifically the pixel value and two-dimensional coordinates of each pixel point in the operation image. The specific steps of obtaining the operation environment interference aggregation index of each time period of the gantry crane are as follows: The pixel value and two-dimensional coordinates of each pixel point in each frame of the operation image of the gantry crane in each period are input into the pre-trained operation risk identification model for comprehensive analysis, and the operation risk assessment index set of the gantry crane in each period is obtained, including the operation obstacle interference index, the non-operation target misentry index, and the container stacking interference index; A comprehensive analysis is performed on the operation risk assessment index set of the gantry crane in each time period to obtain the operation environment interference aggregation index of the gantry crane in each time period.

8. The container gantry crane monitoring and early warning method according to claim 7 is characterized in that: The operation risk identification model is specifically a visual transformer network, which includes an input layer, an image encoding layer, a visual transformer encoder layer, a risk attention feature extraction layer, and a task output layer. The specific steps of obtaining the operation risk assessment index set for each period of the gantry crane are as follows: In the input layer of the visual transformer network, the time series data of the operation video stream of the gantry crane is received and preprocessed; In the image coding layer of the visual transformer network, the pre-processed gantry crane operation video stream time series data is subjected to image block coding to obtain an image block embedding vector sequence of each frame of the gantry crane operation image in each period; In the visual transformer encoder layer of the visual transformer network, a multi-layer attention transformation encoding process is performed on the image block embedding vector sequence of each frame of the operating image of the gantry crane in each period, so as to obtain a global feature vector sequence of the image block of each frame of the operating image of the gantry crane in each period; In the risk attention feature extraction layer of the visual transformer network, feature extraction is performed on the global feature vector sequence of the image block of each frame of the operating image of the gantry crane in each period, and a risk area feature vector set of each frame of the operating image of the gantry crane in each period is obtained; In the task output layer of the visual transformer network, the risk area feature vector set of each frame of the operating image of the gantry crane in each time period is processed by branch prediction to obtain the operating obstacle interference index, non-operating target misentry index, and container stacking interference index of the gantry crane in each time period, that is, the operating risk assessment index set.

9. The container gantry crane monitoring and early warning method according to claim 1, characterized in that: The specific steps for taking preset early warning measures for gantry cranes based on the risk coupling determination index in each period are as follows: Read the risk coupling determination index of each period of the gantry crane, and conduct a comprehensive analysis to obtain the risk coupling determination index of the gantry crane in the next period; Compare and analyze the risk coupling determination index of the gantry crane in the next period with the preset risk coupling determination index threshold range respectively; If the risk coupling determination index of the gantry crane in the next period is lower than the preset lower limit of the risk coupling determination index threshold range, it is marked as low risk and the first warning measure is taken; If the risk coupling determination index of the gantry crane in the next period is within the preset risk coupling determination index threshold range, it is marked as medium risk and the second early warning measure is taken; If the risk coupling determination index of the gantry crane in the next period is higher than the upper limit of the preset risk coupling determination index threshold range, it is marked as a high risk and the third warning measure is taken.

10. A monitoring and early warning system for a container gantry crane, using the monitoring and early warning method for a container gantry crane according to any one of claims 1 to 9, characterized in that: include: A monitoring data acquisition module is used to continuously acquire monitoring time series data of the gantry crane, wherein the monitoring time series data includes structure operation time series data, crane operation behavior time series data, and operation video stream time series data; The risk extraction and processing module is used to perform risk extraction and processing on the monitoring time series data of the gantry crane respectively, and obtain the risk assessment index set of each period of the gantry crane, including the structural risk aggregation index, the crane operation behavior risk index, and the working environment interference aggregation index; The comprehensive risk assessment module is used to comprehensively analyze the risk assessment index set of each period of the gantry crane to obtain the risk coupling determination index of each period of the gantry crane; The risk intelligent early warning module is used to take preset early warning measures for gantry cranes based on the risk coupling judgment index of each time period.

Images