Special equipment digital dynamic management system and method

Abstract

A kind of special equipment digital dynamic management method and system, including the following steps: collecting and processing the data of each special equipment, generating digital twin model, and using C4.5 decision tree model for intelligent analysis and decision, optimization operation strategy;Using the optimized digital twin model to simulate operation, verify the feasibility of actual operation, provide simulation data to operation management reference;Real-time monitoring the running condition of special equipment, analyze and decide in combination with real-time data and simulation data;Early warning to potential dangerous situation, and record and check all operation data;Store all running data, exchange information with third party platform, update equipment model and operation parameter.Data centralization and real-time update, improve management efficiency, early warning potential dangerous situation and record all operation data, reduce the occurrence of safety accidents, improve data utilization efficiency, solve data island problem, carry out all-round digital dynamic management to special equipment.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of special equipment operation management, and more particularly relates to a special equipment digital dynamic management system and method. BACKGROUND

[0002] Special equipment is of various types and in large quantity, which brings great difficulty to the safety management of special equipment. At present, the traditional special equipment management mode mainly relies on manual management, and has problems such as scattered data, untimely information update, and low work efficiency. At the same time, the safety awareness of the special equipment using units is not strong, and there are phenomena such as illegal purchase, use and modification of special equipment, which leads to frequent safety accidents, and once an accident occurs, it will bring great loss and threat to personal safety and property.

[0003] At present, the field of special equipment in China is not closely combined with digital technology, and the digital level is still low. Although some achievements have been made in intelligent monitoring, damage identification, risk early warning and accident prediction in recent years, and a number of risk monitoring platforms have also been developed, the data island phenomenon is still serious. In addition, most of the special equipment digital management systems only consider the collection of data resources, and lack sufficient consideration of the utilization of the collected resources, and there are still few systems that can truly realize the dynamic operation management, risk control and data conversion and utilization of special equipment. Therefore, in order to solve the above problems, it is urgent to build a digital dynamic special equipment management system, and truly realize the dynamic management of special equipment basic information, real-time update of supervision information, real-time early warning of safety hazards, and real-time analysis and statistics of data, so as to improve the safety performance and production efficiency of special equipment.

[0004] Prior art document 1 (CN104656459B) discloses a crane monitoring and management simulation system, which is divided into crane monitoring and management and crane simulation. The crane monitoring and management is composed of a local monitoring and management system and a remote monitoring and management system, which realizes the functions of state monitoring of the crane, video monitoring of the operation area, fault detection of the crane and management of the crane.

[0005] The deficiency of prior art document 1 is that the crane monitoring and management simulation system in the prior art mainly obtains the safety action parameter range by pre-action simulation, and then controls the actual operation process of the crane. Under the condition of simultaneous operation of a large number of special equipment, large task quantity and tight construction period, it is difficult to play the timeliness and dynamic management characteristics. SUMMARY

[0006] In order to solve the deficiencies in the prior art, the present application provides a special equipment digital dynamic management system and method.

[0007] The present application adopts the following technical solutions.

[0008] The first aspect of the present application provides a special equipment digital dynamic management method, comprising the following steps:

[0009] Collect and process various data of the special equipment, generate a digital twin model, and use a C4.5 decision tree model for intelligent analysis and decision-making to optimize the operation strategy;

[0010] Use the optimized digital twin model to perform simulation operation, verify the feasibility of the actual operation, and provide simulation data for operation management reference;

[0011] Real-time monitoring of the operation of the special equipment, combined with real-time data and simulation data for analysis and decision-making; warning of potential dangerous situations, and recording and checking all operation data;

[0012] Store all operation data, exchange information with third-party platforms, update equipment models and operation parameters, and conduct comprehensive digital dynamic management of the special equipment.

[0013] Preferably, the digital twin model is constructed specifically comprising the following steps:

[0014] Install three-dimensional space sensors and video sensors to monitor the action trajectory and work behavior of the equipment;

[0015] Verify the accuracy of the monitoring data through the video sensor and calculate the deviation value;

[0016] Check whether the deviation is within a reasonable range to generate a digital twin model.

[0017] Preferably, the C4.5 decision tree model is constructed specifically comprising the following steps:

[0018] According to the stratified sampling strategy, the operation data set of the special equipment is divided into a training set and a validation set, which are used to construct and evaluate the performance of the decision tree model;

[0019] Calculate the information gain ratio, select the optimal partition attribute, and use a recursive method to construct the C4.5 decision tree until all leaf nodes contain samples of the same class;

[0020] Starting from the leaf node, perform post-pruning processing on each non-leaf node from bottom to top, so that the model's performance on new data is consistent with its performance on training data;

[0021] Continue to traverse other non-leaf nodes of the decision tree upwards, repeatedly perform post-pruning processing operations until the root node, obtain the pruned decision tree model, and determine the running state of the special equipment, including start, emergency stop, safety, and danger.

[0022] Preferably, the formula for calculating the information gain ratio and the information entropy is as follows:

[0023] Entropy(D) = -∑pi log2 pi i = 1 npilog2pi

[0024] wherein:

[0025] pi is the proportion of the i-th class of samples in the data set, and n is the number of classes of the data set;

[0026] The formula of information gain is as follows:

[0027] Gain(A) = Entropy(D) ∑v∈values (A)|Dv||D|×Entropy(Dv)

[0028] wherein:

[0029] valuesA represents all possible values of attribute A, Dv represents the sample subset of attribute A taking value v, Dv represents the size of the sample subset, and D represents the total number of samples of the data set D;

[0030] The formula of intrinsic value of attribute A is as follows:

[0031] IV(A) = ∑v∈values(A)|Dv||D|log2|Dv||D|

[0032] The formula of information gain ratio of attribute A is as follows:

[0033] GainRatio(A) = Gain (A) IV(A)

[0034] wherein:

[0035] Gain(A) represents the information gain of attribute A; and IV(A) represents the intrinsic value of attribute A;

[0036] The feature with the maximum information gain ratio is selected as the division feature of the current node.

[0037] Preferably, the post-pruning process specifically comprises the following steps:

[0038] Temporary pruning: the current non-leaf node is changed into a leaf node, the class label of the node is kept unchanged, and all the child nodes of the node are pruned;

[0039] Evaluation on the validation set: the performance of the decision tree after temporary pruning is evaluated using the validation set, and the accuracy is calculated;

[0040] Determination of whether to prune: the accuracy before and after temporary pruning is compared, if the accuracy does not change significantly, the pruning operation is performed, otherwise the node is kept without pruning.

[0041] Preferably, the simulation operation using the optimized digital twin model specifically includes the following steps:

[0042] Confirm the type of special equipment and the motion equation of each measuring point;

[0043] Simulate the operation of the special equipment, including forward, backward, turning, rotating, lifting, stretching, retracting, and load simulation, and trigger the simulation effect through the front-end button;

[0044] Prioritize the operation of the special equipment, check whether there is motion interference between the two special equipment, and determine the action range when working together;

[0045] According to the height of the boom and the working amplitude, set the high-voltage line warning special effect and the circular ring warning special effect, and check whether the simulation process of the special equipment interferes with the warning line.

[0046] Preferably, the running condition of the special equipment is monitored in real time, and the real-time data and simulation data are analyzed and decided, specifically including the following steps:

[0047] According to the working range of the special equipment obtained by the simulation module, set more than 0 and less than or equal to 2 three-dimensional space sensors at each key measuring point of the special equipment, and work according to the preset action instruction;

[0048] Compare the analysis data in the real measuring point and the special equipment simulation module, and record the change value and change trend of each key point;

[0049] Substitute the measurement value of each key point into the C4.5 decision tree model to judge the running state of the special equipment and make action decision.

[0050] Preferably, the potential dangerous situation is warned, and all operation data is recorded and checked, including alarming for emergency stop, giving the dangerous measuring point in the BIM model, outputting and reporting the identification information.

[0051] Preferably, all running data is stored, information exchange is carried out with a third party platform, equipment model and operation parameters are updated, and the special equipment is managed in a full range of digital dynamic management, including saving each special equipment running information as incremental information, connecting with the third party platform for adding special equipment model information, adding corresponding mechanical equation and motion equation.

[0052] The second aspect of the application provides a special equipment digital dynamic management system, which comprises a digital modeling module, a special equipment simulation module, a running management module, a safety warning module and an information storage module.

[0053] The digital modeling module is used for collecting and processing data, generating a digital twin model, and optimizing operation strategies through intelligent analysis and decision-making using a C4.5 decision tree model.

[0054] The special equipment simulation module is used for operation simulation of the optimized digital twin model, verification of actual operation, and provision of simulation data to the operation management module.

[0055] The operation management module is used for real-time monitoring of the operation of the special equipment, analysis and decision-making in combination with real-time data and simulation data.

[0056] The safety warning module is used for early warning of potential dangerous situations, and recording and checking of all operation data.

[0057] The information storage module is used for saving all operation data, information exchange with a third-party platform, updating of equipment models and operation parameters, and comprehensive digital dynamic management of the special equipment.

[0058] Compared with the prior art, the present application has at least the following beneficial effects:

[0059] (1) Comprehensive digital management is realized: through digital modeling and information storage, data is centralized and updated in real time, management efficiency is improved, the boundary conditions of the substation construction site are simulated digitally, and the objective environment of the construction operation is born in the digital information layer;

[0060] (2) Safety performance is improved: potential dangerous situations are early warned in real time, and all operation data are recorded and checked, thereby reducing the occurrence of safety accidents, three-dimensional dynamic modeling of commonly used special construction machinery is performed, and safety verification of the construction operation scheme is completed;

[0061] (3) Operation decision-making is optimized: through intelligent analysis and decision-making optimization, real-time data of the special equipment are combined to realize rapid analysis and decision-making of the special equipment, operation strategies of the special equipment are optimized, and production efficiency is improved;

[0062] (4) Management efficiency is improved: through real-time monitoring and simulation operation verification of the special equipment, efficient use of data is realized, and the data island problem is solved;

[0063] (5) Real-time data exchange and updating: through information exchange between the information storage and a third-party platform, real-time updating of equipment models and operation parameters is ensured, and dynamic management of basic information and real-time updating of supervision information are realized. BRIEF DESCRIPTION OF DRAWINGS

[0064] Figure 1 is a special equipment digital dynamic management flowchart provided according to an embodiment of the present application;

[0065] Figure 2 is a special equipment digital dynamic management system architecture provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0066] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. The embodiments described in the present application are only a part of the embodiments of the present application, but not all the embodiments. Based on the spirit of the present application, other embodiments obtained by those skilled in the art without any creative work are within the protection scope of the present application.

[0067] Taking two types of automobile cranes (50T, 130T) in the construction site of a substation reconstruction project as an example, the specific implementation of the special equipment digital management system is as follows:

[0068] As shown in Figure 1 , embodiment 1 of the present application provides a special equipment digital dynamic management method, comprising the following steps:

[0069] Collect and process various data of the special equipment, generate a digital twin model, and use a C4.5 decision tree model for intelligent analysis and decision-making to optimize the operation strategy;

[0070] The construction of the digital twin model specifically includes the following steps:

[0071] High-precision three-dimensional space sensors are set for the mechanical arm and motion nodes of the crane, and the position and action of the crane are remotely collected, stored, checked, and optimized. The special mechanical operation process is mapped to the digital twin virtual operation site through an artificial intelligence algorithm, and the construction process is recorded in all directions:

[0072] S11 uses high-precision three-dimensional space sensors to monitor the action trajectory and operation behavior of the equipment. The chassis part includes factors such as the frame, the supporting leg, the axle, the tire, and the cab. The upper part includes factors such as the control room, the crane arm, the hook, the rope, the rotating mechanism, the lifting mechanism, and the amplitude changing mechanism.

[0073] S12 verifies the accuracy of the monitoring data and calculates the deviation value through the video sensor;

[0074] S13 checks whether the deviation is within a reasonable range and generates a digital twin model.

[0075] A C4.5 decision tree model is constructed for real-time analysis and decision-making of the equipment working data:

[0076] S21 divides the training set and the validation set:

[0077] According to the stratified sampling strategy, the historical operation data set of the crane is divided into a training set and a validation set, which are used to build and evaluate the performance of the decision tree model.

[0078] S22 Select the optimal partition attribute:

[0079] The formula for calculating information entropy is:

[0080] Entropy(D)= i=1npilog2pi

[0081] Where pi is the proportion of the i-th class of samples (chassis, loading) in the data set, and n is the number of classes in the data set.

[0082] The formula for calculating information gain is:

[0083] Gain(A)=Entropy(D) ∑v∈values (A)|Dv||D|×Entropy(Dv)

[0084] Where values (A) represents all possible values of attribute A, such as the range of rotation and extension of the boom, Dv represents the sample subset of attribute A with value v, |Dv| represents the size of the sample subset, and |D| represents the total number of samples in the data set D.

[0085] The formula for calculating the intrinsic value of attribute A is:

[0086] IV(A)= ∑v∈values(A)|Dv||D|log2|Dv||D|

[0087] The information gain ratio formula for attribute A is:

[0088] GainRatio(A)=Gain (A)IV(A)

[0089] Select the feature with the maximum information gain ratio as the partition feature of the current node.

[0090] Use a recursive method to build a C4.5 decision tree until all leaf nodes contain samples of the same class.

[0091] S23 Post-pruning processing:

[0092] Starting from the leaf nodes, prune each non-leaf node from bottom to top. For each non-leaf node, perform pruning operations according to steps a, b, and c as follows.

[0093] a. Temporary pruning: change the current non-leaf node to a leaf node, but keep the node class label unchanged, that is, cut off all its child nodes.

[0094] b. Evaluation on the validation set: evaluate the performance of the temporarily pruned decision tree on the validation set, and calculate the accuracy.

[0095] c. Determine whether to prune: compare the accuracy before and after temporary pruning. If the accuracy decreases slightly or does not change significantly, perform pruning operation, otherwise keep the node without pruning.

[0096] Continue to traverse other non-leaf nodes of the decision tree upwards, and repeat steps a, b, and c until the root node is reached.

[0097] Get the pruned decision tree model.

[0098] Using the optimized digital twin model to perform simulation operation, verify the feasibility of actual operation, and provide simulation data for operation management reference;

[0099] Preferably, the simulation operation using the optimized digital twin model comprises the following steps:

[0100] Select the digital twin model of the crane, restore the motion trajectory, working radius, working efficiency and mechanical influence on the object being constructed, and analyze the construction safety under different boundary conditions:

[0101] S31 confirms the crane model (50T, 130T) and the respective measurement point motion equation;

[0102] S32 simulates the forward, backward, turning, rotating, lifting, telescoping, retracting and load bearing of the two cranes, and triggers the model motion through the front button:

[0103] (1) Forward / backward / turning simulation, direct input, plus / minus modification, "crane movement" button, plus / minus button to trigger simulation effect;

[0104] (2) Hoist arm rotation simulation, direct input, plus / minus modification, "hoist arm rotation" button, plus / minus button to trigger simulation effect, parameters as follows: 50t: 0~360°, 1.6r / min; 130t: 0~360°, 1.5r / min;

[0105] (3) Hoist arm lifting simulation, direct input, plus / minus modification, "hoist arm elevation" button, plus / minus button to trigger simulation effect, "hoist arm lifting / falling back" button to trigger the effect of hoist arm lifting to the maximum value / falling back, parameters as follows: 50t: -1°~80°, 45s; 130t: -0.5°~81.5°, 55s;

[0106] (4) Crane boom stretching simulation, direct input, add or subtract modification, "crane boom length" button, add or subtract button trigger simulation effect, "crane boom extension / retraction" button triggers the crane boom to extend to the maximum value / retract effect, parameters as follows: 50t: 5-section arm, basic arm 11.6m full arm 45.5m, each arm length about 8.5m, 105s; 130t: 7-section arm, basic arm 13.6m full arm 73m, each arm length about 9.9m, 660s;

[0107] (5) Hook retraction simulation, direct input, add or subtract modification, "hook retraction" button, add or subtract button trigger simulation effect, "hook down / put up" button triggers the hook to the ground / put up effect, parameters as follows: the lifting / dropping interval of the hook / rope refers to the maximum lifting height of each section of the arm; 50t: maximum lifting height 45m; 130t: maximum lifting height 75m;

[0108] (6) "steel structure lifting" button, "steel structure release" button simulates lifting / release components, and the lifting performance is derived from the corresponding technical documents.

[0109] S33 prioritizes each movement link of the crane, checks whether there is motion interference between the two cranes, and determines the action range of each node when the two cranes work together;

[0110] S34 sets high-voltage line warning special effects and circular ring warning special effects, and checks whether the special equipment simulation process interferes with the warning line. According to the boom height, set the high-voltage line warning special effect (for example, 220kV Dongda 2585 line #34~#34-1, 2586 line #32~#32-1 segment high-voltage line, high limit rope protection line is set along the west side of the high-voltage line, the distance from the vertical projection of the high-voltage line to the outside and west is 15m, the line height is 20m, and the construction ensures the safe operation of the high-voltage line); The boom height calculation method is as follows: refer to the maximum lifting height of each section of the arm: 50t: maximum lifting height 45m; 130t: maximum lifting height 75m; According to the working amplitude, set the circular ring warning special effect, parameters as follows: 50t: 3~38m; 130t: 3~62m.

[0111] Real-time monitoring of the operation of special equipment, analysis and decision-making combined with real-time data and simulation data; warning of potential dangerous situations, and recording and checking all operation data;

[0112] Preferably, real-time monitoring of the operation of special equipment, analysis and decision-making combined with real-time data and simulation data, specifically including the following steps:

[0113] The working range of the special equipment obtained by the simulation module is combined, 1-2 high-precision three-dimensional space sensors are arranged at each key measuring point of the crane, and the change value and change trend of each key point are found in time by comparing the real measuring point and the analysis data in the simulation module of the special equipment according to preset action instructions.

[0114] The measurement values of the key nodes are substituted into the C4.5 decision tree model to judge the start, emergency stop, safety and dangerous conditions of the crane operation, and the action decision is made autonomously.

[0115] Preferably, the potential dangerous condition is warned, and all operation data are recorded and checked, including alarming and prompting for the emergency stop condition, giving the dangerous measuring point in the BIM model, outputting and reporting the identification information.

[0116] All running data are stored, information exchange with a third-party platform is performed, the equipment model and operation parameters are updated, and the special equipment is managed in a full-digital dynamic manner.

[0117] Preferably, all running data are stored, information exchange with a third-party platform is performed, the equipment model and operation parameters are updated, and the special equipment is managed in a full-digital dynamic manner, including saving each special equipment running information as incremental information, connecting with the third-party platform, and being used for adding special equipment model information, adding corresponding mechanical equations and motion equations.

[0118] As shown in Figure 2 Embodiment 1 of the present application provides a special equipment digital dynamic management system:

[0119] including a digital modeling module, a special equipment simulation module, a running management module, a safety warning module and an information storage module;

[0120] The digital modeling module is used for collecting and processing data, generating a digital twin model, and using a C4.5 decision tree model to intelligently analyze and decide to optimize the operation strategy;

[0121] The special equipment simulation module is used for operating simulation on the optimized digital twin model, verifying the actual operation, and providing simulation data to the running management module;

[0122] The running management module is used for real-time monitoring of the running condition of the special equipment, and analyzing and deciding in combination with real-time data and simulation data;

[0123] The safety warning module warns the potential dangerous condition, and records and checks all operation data;

[0124] The information storage module is used for saving all operation data, exchanging information with a third party platform, updating equipment models and operation parameters, and performing all-round digital dynamic management on special equipment.

[0125] The beneficial effects of the present application include at least:

[0126] (1) Realize all-round digital management: through digital modeling and information storage, ensure data centralization and real-time update, improve management efficiency, realize digital simulation of substation engineering site construction boundary conditions, and duplicate the objective environment of construction operation at the digital information level;

[0127] (2) Improve safety performance: through safety early warning, real-time early warning of potential dangerous situations and recording and checking of all operation data, reduce the occurrence of safety accidents, three-dimensional dynamic modeling of commonly used special construction machinery and simulation of the whole construction process, safety checking and verification of construction operation scheme;

[0128] (3) Optimize operation decision: through intelligent analysis and decision optimization, realize rapid analysis and decision of special equipment action combined with real-time data of special equipment, realize optimization of special equipment operation strategy, and improve production efficiency;

[0129] (4) Improve management efficiency: through real-time monitoring of operation management and simulation operation verification of special equipment, realize efficient use of data, and solve the problem of data island;

[0130] (5) Real-time data exchange and update: through information exchange between information storage and a third party platform, ensure real-time update of equipment models and operation parameters, realize dynamic management of basic information and real-time update of supervision information.

[0131] The present disclosure can be a system, a method, and / or a computer program product. The computer program product can include a computer readable storage medium having computer readable program instructions loaded thereon for causing a processor to implement various aspects of the present disclosure.

[0132] Finally, it should be noted that the above examples are only used to illustrate the technical solutions of the present application and not to limit it. Although the present application has been described in detail with reference to the above examples, those skilled in the art should understand that the specific embodiments of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered within the protection scope of the claims of the present application.

Claims

1. A method for digital dynamic management of special equipment, characterized in that, The method comprises the following steps: Collect and process various data of the special equipment, generate a digital twin model, and use a C4.5 decision tree model for intelligent analysis and decision-making to optimize the operation strategy; The construction of the digital twin model specifically comprises the following steps: Install three-dimensional space sensors and video sensors to monitor the action trajectory and operation behavior of the equipment; install three-dimensional space sensors on the mechanical arm and motion nodes of the crane, collect, store, correct and optimize remote data sets of the crane, map the special mechanical operation process to the digital twin virtual operation site through an artificial intelligence algorithm, and realize all-around recording of the construction process; verify the accuracy of the monitoring data through the video sensors and calculate the deviation value; check whether the deviation is within a reasonable range to generate the digital twin model; Use the optimized digital twin model for simulation operation to verify the feasibility of the actual operation and provide simulation data for operation management reference; Real-time monitor the operation of the special equipment, analyze and make decisions in combination with real-time data and simulation data, give early warning for potential dangerous situations, and record and check all operation data; the simulation operation using the optimized digital twin model specifically comprises the following steps: Confirm the type of the special equipment and the motion equation of each measuring point; simulate the operation of the special equipment, including forward, backward, turning, rotating, lifting, stretching, retracting, and load bearing simulation, and trigger the simulation effect through the front-end button; prioritize the operation of the special equipment, check whether there is motion interference between the two special equipments, and determine the action range during joint work; According to the height of the boom and the working range, set up high-voltage line warning special effects and circular ring warning special effects, and check whether the simulation process of the special equipment interferes with the warning line; in combination with the working range of the special equipment obtained by the simulation module, set up more than 0 and less than or equal to 2 three-dimensional space sensors at each key measuring point of the special equipment, and work according to the preset action instruction; Compare the analysis data in the real measuring points and the simulation module of the special equipment, record the change value and trend of each key point; put the measurement value of each key point into the C4.5 decision tree model to judge the running state of the special equipment and make action decisions; Store all operation data, exchange information with a third-party platform, update the equipment model and operation parameters, and conduct all-around digital dynamic management of the special equipment.

2. The special equipment digital dynamic management method according to claim 1, wherein: The construction of the C4.5 decision tree model specifically comprises the following steps: According to a stratified sampling strategy, divide the operation data set of the special equipment into a training set and a validation set to construct and evaluate the performance of the decision tree model; Calculate the information gain ratio, select the optimal partition attribute, and use a recursive method to construct the C4.5 decision tree until all leaf nodes contain samples of the same class; From the leaf nodes, perform post-pruning processing on each non-leaf node from bottom to top, so that the performance of the model on new data is consistent with the performance on the training data; Continue to traverse the other non-leaf nodes of the decision tree upwards, repeatedly perform the post-pruning processing operation until the root node, obtain the pruned decision tree model, and judge the running state of the special equipment, including starting, emergency stopping, safety, and danger.

3. The special equipment digital dynamic management method according to claim 2, wherein: The formula for calculating the information gain ratio and the information entropy is as follows: In the formula: is the proportion of the class sample in the data set, is the proportion of the class sample in the data set, is the number of classes of the data set; The formula for calculating the information gain is as follows: In the formula: representing attributes all possible values, representing attributes taking values a subset of samples, representing the size of a subset of samples, representing the total number of samples of a dataset ; Attributes The formula for the intrinsic value of the attribute is as follows: Attributes The information gain ratio formula is as follows: In the formula: information gain of the attribute information gain of the attribute information gain of the attribute information gain of the attribute Select the feature with the maximum information gain ratio as the division feature of the current node.

4. The special equipment digital dynamic management method according to claim 2, wherein: The post-pruning processing specifically includes the following steps: Temporary pruning: change the current non-leaf node to a leaf node, keep the node class label unchanged, and cut off all its child nodes; Evaluation on the validation set: evaluate the performance of the pruned decision tree using the validation set, and calculate the accuracy rate; Determine whether to prune: compare the accuracy rates before and after temporary pruning. If the accuracy rate does not change significantly, perform pruning operation, otherwise, keep the node without pruning.

5. The special equipment digital dynamic management method according to claim 1, wherein: The potential dangerous situation is warned, and all operation data are recorded and checked, including alarming for emergency stopping, giving the dangerous measuring point in the BIM model, outputting and reporting the identification information.

6. The special equipment digital dynamic management method according to claim 1, wherein: All running data are stored, information is exchanged with a third-party platform, equipment models and operation parameters are updated, and the special equipment is managed in a full range of digital dynamic management, including saving each special equipment running information as incremental information, connecting with the third-party platform for adding special equipment model information, adding corresponding mechanical equations and motion equations.

7. A special equipment digital dynamic management system, comprising: A digital modeling module, a special equipment simulation module, a running management module, a safety warning module, and an information storage module; The digital modeling module is used for collecting and processing data, generating a digital twin model, and using a C4.5 decision tree model to optimize operation strategies through intelligent analysis and decision-making; The digital twin model is constructed by the following steps: Install three-dimensional space sensors and video sensors to monitor the action trajectory and operation behavior of the equipment; install three-dimensional space sensors on the mechanical arm and motion nodes of the crane, collect, store, check, and optimize remote data sets of the crane, map the special mechanical operation process to the digital twin virtual operation site through artificial intelligence algorithms, and realize full-range recording of the construction process; verify the accuracy of the monitoring data through the video sensors and calculate the deviation value; check whether the deviation is within a reasonable range, and generate a digital twin model; The special equipment simulation module is used for operation simulation of the optimized digital twin model, verification of actual operation, and provision of simulation data to the running management module; The running management module is used for real-time monitoring of the running state of the special equipment, analysis and decision-making based on real-time data and simulation data. ​ The safety warning module warns potential dangerous situations, records and checks all operation data; the simulation operation using the optimized digital twin model specifically includes the following steps: Confirm the type of special equipment and the motion equation of each measuring point; simulate the operation of special equipment, including forward, backward, turning, rotating, lifting, stretching, retracting, and load bearing simulation, trigger the simulation effect through the front-end button; prioritize the operation of special equipment, check if there is motion interference between the two special equipment, and determine the action range when working together; According to the height of the boom and the working amplitude, set the high-voltage line warning special effect and the circular ring warning special effect, check if the special equipment simulation process interferes with the warning line; combined with the working range of the special equipment obtained by the simulation module, set more than 0 and less than or equal to 2 three-dimensional space sensors at each key measuring point of the special equipment, and work according to the preset action instruction; Compare the analysis data in the real measuring point and the special equipment simulation module, record the change value and trend of each key point; substitute the measurement value of each key point into the C4.5 decision tree model to judge the running state of the special equipment and make action decisions; The information storage module is used to save all operation data and exchange information with third-party platforms, update equipment models and operation parameters, and conduct comprehensive digital dynamic management of special equipment.

Images