Real-time operation safety monitoring and early warning system for tower crane

Abstract

The invention discloses a tower crane real-time operation safety monitoring and early warning system, which comprises a load real-time estimation module, and the load real-time estimation module is used for estimating the real-time operation safety of a tower crane by deploying a tension sensor in a steel wire rope traction path of the tower crane in combination with a lifting hook structure form, pulley multiplying power and gravitational acceleration. The actual suspended load mass value is inversed in real time; and the dynamic torque fusion calculation module is used for constructing a torque real-time dynamic estimation and risk grading model based on a multi-source data fusion algorithm so as to carry out risk identification on the operation of the tower crane. The problems that a traditional tower crane depends on artificial experience to estimate the hoisting weight or a simple electronic scale to weigh, precision is poor, response is slow, and misjudgment is prone to being caused are solved, intelligent identification of the operation risk of the tower crane is achieved, and safe operation monitoring is carried out in the construction process of the tower crane.

Description

Technical Field

[0001] This invention relates to safety monitoring technology, specifically to a real-time operation safety monitoring and early warning system for tower cranes. Background Technology

[0002] Tower cranes are crucial equipment in high-rise building construction, and their safe operation directly impacts construction efficiency and personnel safety. To ensure the stability and reliability of tower crane operations, various technologies are widely used in the industry to monitor and control their operational status, primarily including the following key technologies:

[0003] (1) Limiting devices: Tower cranes are usually equipped with lifting limit, trolley limit, slewing limit and other devices to control the tower crane to stop automatically when it approaches the maximum working range to prevent danger.

[0004] (2) Torque limiter: This is a common protective device on tower cranes. It can calculate the current working torque based on parameters such as lifting weight and radius. Once the preset value is exceeded, it will issue an alarm or automatically stop the operation.

[0005] (3) Load detection device: The actual weight on the hook is determined by the weighing sensor installed on the hoisting mechanism, thereby detecting whether there is an overload.

[0006] (4) Display device: Some tower cranes are equipped with simple display instruments that can display some operating parameters to help operators keep track of the equipment status.

[0007] In current tower crane operations, a common safety monitoring method is to use torque limiters in conjunction with weight sensors for basic overload protection. This system installs weight sensors at the boom and hook positions of the tower crane to measure the weight of the lifted object, and simultaneously uses angle sensors to measure the boom's tilt angle. This data is transmitted to the controller to calculate the tower crane's current operating torque. When the system determines that the torque or weight exceeds a preset safety range, it alerts the operator via an alarm, and some systems can automatically stop operation to prevent danger. This safety monitoring method has the following main drawbacks:

[0008] (1) The data is not comprehensive enough: it can only detect some key parameters, such as the load and angle, and cannot monitor multiple operating data at the same time (such as the torque change trend, load change rate, etc.), so the overall operating status is not comprehensive enough.

[0009] (2) Inadequate information display: The display interface is usually a simple numerical value or alarm light, lacking a graphical interface. Operators have difficulty intuitively understanding the current status of the tower crane, which affects the accuracy of judgment.

[0010] (3) Lack of intelligent analysis capabilities: It only provides basic threshold alarms and cannot analyze and predict based on historical data or operating trends, thus failing to achieve higher-level safety early warning.

[0011] (4) Insufficient response: Some systems have low acquisition frequency and large processing delay, resulting in the monitoring data not being updated in real time, which poses a security risk.

[0012] There is an existing safety monitoring method for tower cranes. This method uses the following modules to collect and preliminarily process operational data and perform simple judgments:

[0013] (1) Install sensors: Install weight sensors, angle sensors and torque sensors on the tower crane to collect basic data such as weight, angle and torque of the tower crane during the hoisting process.

[0014] (2) Simple data processing: The collected data will first undergo simple processing, such as removing obviously abnormal values ​​to make the data more stable.

[0015] (3) Set alarm thresholds: The system sets some fixed safety ranges, such as the maximum allowable lifting weight or torque. When the real-time data exceeds these ranges, the system will alarm to remind the operator to pay attention.

[0016] (4) Record data: This data can also be saved for later viewing or analysis.

[0017] This safety monitoring method for tower cranes has the following main drawbacks:

[0018] (1) Single monitoring data: This technology mainly monitors a single parameter, such as lifting weight or torque, and lacks comprehensive monitoring of multiple operating data of tower cranes (such as boom angle, wind speed, etc.), making it impossible to fully understand the working status of tower cranes.

[0019] (2) Static threshold setting: The system relies on a fixed safety threshold to trigger an alarm, and cannot dynamically adjust the safety range according to different working environments or historical data, resulting in an inability to flexibly cope with complex operating situations.

[0020] (3) Limited data processing capabilities: The data processing process is relatively simple, and it can only remove obvious outliers. It lacks analysis and prediction of data trends and cannot provide higher-level intelligent decision support.

[0021] (4) Slow response speed: The system has a slow response speed, which may delay the feedback of monitoring results, especially during heavy loads or complex operations, and may not be able to give timely and accurate warnings. Summary of the Invention

[0022] The purpose of this invention is to overcome the shortcomings of the prior art and provide a real-time operation safety monitoring and early warning system for tower cranes to monitor the safe operation of tower cranes during construction.

[0023] To achieve the above objectives, the technical solution of the present invention is as follows:

[0024] A real-time operation safety monitoring and early warning system for tower cranes includes:

[0025] The real-time load estimation module uses tension sensors deployed in the wire rope traction path of the tower crane, combined with the hook structure, pulley ratio and gravitational acceleration, to calculate the actual load mass in real time.

[0026] The dynamic torque fusion calculation module is based on a multi-source data fusion algorithm to construct a real-time dynamic torque estimation and risk classification model to identify risks in the operation of tower cranes.

[0027] Optionally, the real-time load estimation module, by deploying tension sensors in the wire rope traction path of the tower crane and combining the hook structure, pulley ratio, and gravitational acceleration, can calculate the actual hoisted mass value in real time, including:

[0028] Modeling the tower crane lifting system as a pulley force system, we get: (1)

[0029] Where m is the hoisting mass, F T , where g is the traction force measured by the sensor, n is the acceleration due to gravity, and n is the pulley system transmission ratio.

[0030] By introducing a compensation coefficient α to dynamically correct equation (1), we obtain: (2)

[0031] Optionally, the real-time load estimation module, by deploying tension sensors in the wire rope traction path of the tower crane and combining the hook structure, pulley ratio, and gravitational acceleration, can calculate the actual hoisted mass value in real time, and further includes:

[0032] The traction force data measured by the sensor is processed using a moving weighted average method: (3)

[0033] Will Substituting into equation (2), we get: (4).

[0034] Optionally, the dynamic torque fusion calculation module, based on a multi-source data fusion algorithm, constructs a real-time dynamic torque estimation and risk classification model to identify risks in tower crane operation, including:

[0035] Establish a mathematical model for torque estimation: (5)

[0036] Where M is the load torque generated by the current working arm of the tower crane, m is the hoisted mass, g is the gravitational acceleration, and L is the horizontal distance from the hook to the center of the tower crane; (6)

[0037] Where L0 is the basic boom length, r is the pulley radius or telescopic section length, and θ is the boom pitch angle;

[0038] Substituting equation (6) into equation (5), we get: (7)

[0039] Introducing a multivariate fusion model: (8)

[0040] Among them, M t The current moment is the comprehensive torque index, dM / dt is the torque change rate, representing the sudden dynamic impact; ΔM is the deviation of the average torque; w1, w2, w3 are the fusion weights, satisfying w1+w2+w3=1;

[0041] Based on the calculated M t With the maximum safe torque M of the tower crane max The ratio is introduced into the risk grading function R: (9)

[0042] The risk status is then classified based on the value of R.

[0043] Optionally, the tower crane real-time operation safety monitoring and early warning system also includes an operation trend prediction module, used to predict the potential safety impact of the next lifting action.

[0044] Optionally, the operation trend prediction module uses an operation trend prediction model based on historical behavior data and sensor parameters to predict the potential safety impact of the next hoisting action.

[0045] Optionally, the operation trend prediction module predicts the potential safety impact of the next hoisting action based on an operation trend prediction model using historical behavior data and sensor parameters, including:

[0046] Construct an operation sequence dataset:

[0047] The key operating parameters of the tower crane are combined into a characteristic time series, defined as follows: (10)

[0048] Where, m t For the weight being lifted; L t The current amplitude; θ t v is the current pitch angle; t This represents the current speed of the trolley. The rate of change of pitch angle; The rate of change of amplitude;

[0049] Construct an operation history window W={X t−n ,...,X t};

[0050] Introducing a state prediction function:

[0051] Using an improved weighted recursive model, predict the trend of operational changes in the next few seconds: (11)

[0052] Where, α i For decreasing weight coefficients, satisfying ∑α i =1; Predicting the state at a future moment;

[0053] Substituting into formula (7), we obtain the future torque change: (12)

[0054] By comparing the current state with the predicted state, a trend risk indicator T is defined. r : (13)

[0055] If T r >δ indicates that the trend is heading in an unsafe direction, issuing an early warning:

[0056] T r <δ1: Stable;

[0057] δ1≤T r <δ2: Slight deviation;

[0058] T r ≥δ2: High trend risk, intervention required;

[0059] δ represents the trend risk threshold, δ1 represents the first trend risk threshold, and δ2 represents the second trend risk threshold. Specifically, the value of δ is typically set based on the statistical boundary between safe and unsafe conditions in historical data, derived through experimental regression, risk assessment models, or expert experience. The system supports dynamic adjustment of δ or adaptive setting based on operating conditions.

[0060] Optionally, the tower crane real-time operation safety monitoring and early warning system also includes an anomaly identification and intelligent alarm module, used to identify dangerous conditions in real time and trigger alarms.

[0061] Optionally, the anomaly identification and intelligent alarm module is used to identify dangerous states in real time and trigger alarms, including:

[0062] The key data collected by the sensor is represented as a multi-dimensional state vector: (14)

[0063] Among them, T t For temperature environment parameters, H t For humidity environmental parameters, M t This represents the currently calculated operating torque;

[0064] By using clustering learning or rule learning models, normal behavior data can be fitted to establish a feature distribution model and identify deviant behaviors.

[0065] For each time t, calculate the current state Z. t Abnormal rating: (15)

[0066] Where: z t,i Let μ be the state value of the i-th term; i σ i Let w be the mean and standard deviation of the i-th feature under normal conditions; i These are the feature weights.

[0067] When S a When (t)>λ, it is determined to be an abnormal state; λ is the abnormal scoring threshold.

[0068] Optionally, the tower crane real-time operation safety monitoring and early warning system also includes a safety decision and graphical interaction module, used to display the results of the real-time load estimation module, dynamic torque fusion calculation module, operation trend prediction module, and anomaly identification and intelligent alarm module in real time.

[0069] Compared with the prior art, the advantages of this invention are as follows:

[0070] This application addresses the problems of delayed data acquisition, high operational risks, and lack of abnormal early warning in existing tower crane operations. It proposes a tower crane operation safety monitoring system that integrates real-time multi-parameter acquisition and intelligent analysis, offering the following advantages:

[0071] (1) Realize the real-time acquisition of key operating data and solve the problem of missing data: By deploying a variety of high-precision sensors, it is possible to achieve comprehensive, continuous and real-time acquisition of key data during the operation of tower cranes, providing a real and reliable data foundation for subsequent analysis and judgment.

[0072] (2) Accurate operational status perception and judgment capability is constructed through algorithm model: An algorithm system with torque calculation model, status prediction model and anomaly scoring mechanism as the core is constructed, which can analyze the operational stability, overload trend and dynamic abnormal changes of tower crane in real time.

[0073] (3) Significantly improve operational safety and reduce accident risk: The system automatically judges whether the current operating status is close to the safety boundary, issues early warnings, assists operators in safety intervention, and greatly reduces the operational risk caused by human judgment errors.

[0074] (4) Improve operator response speed and judgment ability through human-computer interaction module: The graphical visual interface and intelligent suggestion module, combined with color coding, trend charts, danger warnings and other methods, enable operators to quickly understand the current operating status and make scientific decisions without complicated calculations.

[0075] (5) It has good system scalability and application adaptability: The algorithm and system structure of this application adopt a modular design, which is convenient to integrate with the existing tower crane control system or smart construction site platform. It can also adjust parameters and fine-tune algorithms according to different tower crane models or operation scenarios, and has good versatility and flexibility. Attached Figure Description

[0076] Figure 1 A schematic diagram of the composition of the tower crane real-time operation safety monitoring and early warning system provided in the embodiments of this application. Detailed Implementation

[0077] Example:

[0078] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0079] See Figure 1 As shown in the figure, the real-time operation safety monitoring and early warning system 100 for tower cranes provided in this application embodiment mainly includes:

[0080] The real-time load estimation module 110 utilizes tension sensors deployed along the wire rope traction path of the tower crane. By combining the hook structure, pulley ratio, and gravitational acceleration, it calculates the actual load mass in real time. During tower crane operation, the load mass is a core parameter for determining whether the operation is overloaded. Therefore, this system includes the real-time load estimation module 110. Based on this module, by deploying high-precision tension sensors along the wire rope traction path and combining the hook structure, pulley ratio, and gravitational acceleration, it accurately calculates the actual load mass in real time. This solves the problems of traditional tower cranes relying on manual experience to estimate the load or using simple electronic scales, which suffers from poor accuracy, slow response, and susceptibility to misjudgments.

[0081] The dynamic torque fusion calculation module 120, based on a multi-source data fusion algorithm, constructs a real-time dynamic torque estimation and risk classification model to identify risks in tower crane operation. During tower crane lifting operations, not only does the weight being lifted need real-time monitoring, but the torque fluctuations caused by changes in lifting distance (amplitude) and angle are also key factors determining its stability and safety. Therefore, this system includes the dynamic torque fusion calculation module 120, which, based on a multi-source data fusion algorithm, constructs a real-time dynamic torque estimation and risk classification model to intelligently identify operational risks of tower cranes.

[0082] In one specific embodiment, the real-time load estimation module, by deploying tension sensors in the wire rope traction path of the tower crane and combining the hook structure, pulley ratio, and gravitational acceleration, calculates the actual hoisted mass value in real time, including:

[0083] Modeling the tower crane lifting system as a pulley force system, we get: (1)

[0084] Where m is the hoisting mass, F T , where g is the traction force measured by the sensor, n is the acceleration due to gravity, and n is the pulley system transmission ratio.

[0085] By introducing a compensation coefficient α to dynamically correct equation (1), we obtain: (2)

[0086] Because real-time sensor data fluctuates greatly, this system also processes the raw traction force data measured by the sensors using a sliding weighted average method, as follows: (3)

[0087] Will Substituting into equation (2), we get: (4).

[0088] The compensation coefficient α is determined as follows:

[0089] (1) If the suspended object has a slight sway, the sensor force value will fluctuate instantaneously. This fluctuation is then measured by the moving average filter MA. k Smoothing correction;

[0090] (2) The system learns the error between the sensor response and the actual calibrated weight under different working conditions, and uses the least squares method to dynamically fit an empirical correction curve α=f(F T˙ ,θ) represents the rate of change of force, and θ is the boom elevation angle.

[0091] Therefore, this real-time load estimation module has the following technical advantages:

[0092] (1) Adaptable to various hoisting combination scenarios: Supports complex structures such as single-arm cranes, multi-ratio pulleys, and dynamic variable ratio cranes;

[0093] (2) High dynamic response: The filtering and compensation mechanism ensures that the load estimation is stable and reliable even during startup, braking or swaying;

[0094] (3) Supports offline calibration and online adaptation: Establishes empirical coefficient curves through historical data models to achieve self-learning compensation.

[0095] In one specific embodiment, the dynamic torque fusion calculation module, based on a multi-source data fusion algorithm, constructs a real-time dynamic torque estimation and risk classification model to identify risks in the operation of tower cranes, including:

[0096] Establish a mathematical model for torque estimation:

[0097] The slewing boom of a tower crane can be simplified as a single-sided cantilever beam, and the basic formula for calculating the moment is: (5)

[0098] Where M is the load torque generated by the current working arm of the tower crane, m is the hoisted mass, g is the gravitational acceleration, and L is the horizontal distance from the hook to the center of the tower crane; (6)

[0099] Where L0 is the basic boom length, r is the pulley radius or telescopic section length, and θ is the boom pitch angle;

[0100] Substituting equation (6) into equation (5), we get: (7)

[0101] Because the operating state of a tower crane is affected by multiple parameters, directly calculating the torque cannot fully reflect its dynamic changes. Therefore, a multivariate fusion model is introduced: (8)

[0102] Among them, M t dM / dt is the comprehensive torque index at the current moment, representing the torque change rate and sudden dynamic impact; ΔM is the deviation of the average torque; w1, w2, w3 are the fusion weights, satisfying w1+w2+w3=1.

[0103] Thus, by introducing a multivariate fusion model, the model has a high dynamic response capability and can detect sudden changes in the hoisting process in a timely manner. It is especially suitable for high-risk transients such as hoisting start-up, slewing braking, and sudden swaying.

[0104] Based on the calculated M t With the maximum safe torque M of the tower crane max The ratio is introduced into the risk grading function R: (9)

[0105] And risk status is classified according to the value of R, including:

[0106] R<70%: Green and safe;

[0107] 70%≤R<90%: Yellow alert;

[0108] 90%≤R<100%: Red, high risk;

[0109] R≥100%: Black sign prohibits lifting; the system will automatically prompt for braking.

[0110] The graphical interface provides real-time visualization of torque changes and current status, helping operators to monitor the stability of tower cranes.

[0111] In complex operating conditions, relying solely on real-time data to determine the current state is often inadequate. Even minor adjustments during tower crane operation can lead to a sudden increase in torque or overload risk. Therefore, in a preferred embodiment, the system further includes an operation trend prediction module 130. Based on historical behavior data and sensor parameters, this module uses an operation trend prediction model to predict the potential safety impacts of the next lifting action in advance, thereby enhancing the system's preventative safety capabilities. Specifically, this includes:

[0112] Construct an operation sequence dataset:

[0113] The key operating parameters of the tower crane are combined into a characteristic time series, defined as follows: (10)

[0114] Where, m t For the weight being lifted; L t The current amplitude; θ t v is the current pitch angle; t This represents the current speed of the trolley. The rate of change of pitch angle; The rate of change of amplitude;

[0115] All the key operating parameters mentioned above are collected in real time by the corresponding sensors, and an operating behavior history window W={X t−n ,...,X t};

[0116] Introducing a state prediction function:

[0117] Using an improved weighted recursive estimation (WRE) model, predict the trend of operational changes in the next few seconds: (11)

[0118] Where, α i For decreasing weight coefficients, satisfying ∑α i =1; Predicting the state at a future moment;

[0119] Substituting into formula (7), we obtain the future torque change: (12)

[0120] By comparing the current state with the predicted state, a trend risk indicator T is defined. r : (13)

[0121] If T r >δ indicates that the trend is heading in an unsafe direction, issuing an early warning:

[0122] T r <δ1: Stable;

[0123] δ1≤T r <δ2: Slight deviation;

[0124] T r ≥δ2: High trend risk, intervention required.

[0125] δ is the trend risk threshold, δ1 is the first trend risk threshold, and δ2 is the second trend risk threshold.

[0126] Thus, through the above operations, compared with the traditional static judgment method, it has the ability to "foresee the future", which greatly enhances the intelligence and proactive security of the system.

[0127] Although predictive models can anticipate trends, sudden abnormal states (such as sensor distortion, sudden load impact, and accidental operation) may still occur during the operation of tower cranes. Therefore, in a preferred embodiment, this system further integrates an anomaly detection and alarm module 140, which uses intelligent algorithms to identify dangerous states in real time and trigger a visual + audible and visual alarm mechanism. Specifically, this includes:

[0128] Definition and data modeling of abnormal behavior:

[0129] The key data collected by the sensor is represented as a multi-dimensional state vector: (14)

[0130] Among them, T t For temperature environment parameters, H t For humidity environmental parameters, M t This represents the currently calculated operating torque;

[0131] By using clustering learning or rule learning models, normal behavior data can be fitted to establish a feature distribution model and identify deviant behaviors.

[0132] Introducing a multidimensional anomaly scoring function, for each time t, the current state Z is calculated. t Abnormal rating: (15)

[0133] Where: z t,i Let μ be the state value of the i-th term; i σ i Let w be the mean and standard deviation of the i-th feature under normal conditions; i These are the feature weights.

[0134] When S a When (t)>λ, it is determined to be an abnormal state, where λ is the abnormal scoring threshold.

[0135] Anomaly classification and adaptive response mechanism: This system not only identifies whether an anomaly exists, but also further classifies the anomaly type. Common anomalies include:

[0136] Abnormal load: Sudden increase or decrease in load;

[0137] Abnormal torque: instantaneous exceedance or excessive fluctuation;

[0138] Sudden change in amplitude: moving too far in a short period of time;

[0139] Sensor malfunction: data jumps or static drift.

[0140] Each type of exception has a corresponding response strategy, such as:

[0141] Issue an audible and visual alarm;

[0142] Control system locking operation;

[0143] Report abnormal data to the management platform;

[0144] Displays error messages and suggested solutions.

[0145] Therefore, this anomaly detection and alarm module can combine "rules + model" dual recognition logic to improve detection accuracy; the multi-dimensional scoring function adapts to feature differences and is not affected by a single indicator; the response mechanism is linked with the user interface to improve on-site decision-making efficiency.

[0146] Since all key data and conclusions obtained from monitoring and algorithm analysis ultimately need to be transmitted to tower crane operators or site managers, in a preferred embodiment, this system also includes a safety decision-making and graphical interaction module 150, which, combined with a safety decision-making assistance algorithm, improves human-computer interaction efficiency and decision-making quality. A real-time graphical display interface, through an embedded terminal or host computer system, displays the following information in real time:

[0147] Current weight of hook (m) t ;

[0148] Operating radius L t ;

[0149] Crane boom angle θ t ;

[0150] Torque variation curve M(t);

[0151] Environmental data such as wind speed, temperature, and humidity;

[0152] Predict the trend chart for the next 5-10 seconds;

[0153] Current anomaly score S a (t) and anomaly type annotation;

[0154] Visualization methods include color-coded charts, radar charts, and trend line charts, making complex data "clear at a glance".

[0155] Dynamic alert algorithm for safe and risk zones:

[0156] The system maps load torque and operating amplitude data to a two-dimensional safe operation diagram, using the following conversion relationship:

[0157] Based on the allowable moment boundary B of the tower crane design, the operating state is mapped as follows:

[0158] The system provides real-time feedback on its operating status through color coding, dynamic jumping, vibration alerts, and other methods.

[0159] Decision support and suggestion mechanism:

[0160] When the system predicts an overload trend or the current state is approaching a risk boundary, it will automatically generate suggestions, such as:

[0161] "The current torque is approaching the safety limit. Please reduce the working range."

[0162] "The estimated lifting weight will exceed the red line in 6 seconds. It is recommended to retract the hook in advance."

[0163] "Crosswind interference exists. Please check for surrounding obstacles."

[0164] It is recommended to generate a rule tree-based and simple inference engine, with the following logic:

[0165] In summary, this application addresses the problems of delayed data acquisition, high operational risks, and lack of abnormal early warning in existing tower crane operations by proposing a tower crane operation safety monitoring system that integrates real-time multi-parameter acquisition and intelligent analysis, which has the following beneficial effects:

[0166] (1) Realize the real-time acquisition of key operating data and solve the problem of missing data: By deploying a variety of high-precision sensors, it is possible to achieve comprehensive, continuous and real-time acquisition of key data during the operation of tower cranes, providing a real and reliable data foundation for subsequent analysis and judgment.

[0167] (2) Accurate operational status perception and judgment capability is constructed through algorithm model: An algorithm system with torque calculation model, status prediction model and anomaly scoring mechanism as the core is constructed, which can analyze the operational stability, overload trend and dynamic abnormal changes of tower crane in real time.

[0168] (3) Significantly improve operational safety and reduce accident risk: The system automatically judges whether the current operating status is close to the safety boundary, issues early warnings, assists operators in safety intervention, and greatly reduces the operational risk caused by human judgment errors.

[0169] (4) Improve operator response speed and judgment ability through human-computer interaction module: The graphical visual interface and intelligent suggestion module, combined with color coding, trend charts, danger warnings and other methods, enable operators to quickly understand the current operating status and make scientific decisions without complicated calculations.

[0170] (5) It has good system scalability and application adaptability: The algorithm and system structure of this application adopt a modular design, which is convenient to integrate with the existing tower crane control system or smart construction site platform. It can also adjust parameters and fine-tune algorithms according to different tower crane models or operation scenarios, and has good versatility and flexibility.

[0171] The above embodiments are merely illustrative of the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made based on the essence of the content of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A real-time operation safety monitoring and early warning system for a tower crane, characterized in that, Comprise: The load real-time estimation module reverses the actual hoisting load mass value in real time by deploying a tension sensor in the steel wire rope traction path of the tower crane, combining the hook structure form, the pulley ratio and the gravitational acceleration, comprising: The dynamic torque fusion calculation module constructs a torque real-time dynamic estimation and risk classification model based on a multi-source data fusion algorithm to identify risks in the operation of the tower crane.

2. The real-time operation safety monitoring and early warning system of the tower crane according to claim 1, characterized in that, The load real-time estimation module reverses the actual hoisting load mass value in real time by deploying a tension sensor in the steel wire rope traction path of the tower crane, combining the hook structure form, the pulley ratio and the gravitational acceleration, comprising: Model the tower crane hoisting system as a pulley force system, and obtain: （1） where m is the load mass, F T is the measured traction force by the sensor, g is the acceleration of gravity, and n is the transmission ratio of the pulley block. Introduce a compensation coefficient α to dynamically correct formula (1), and obtain: （2）。 3. The real-time operation safety monitoring and early warning system of a tower crane according to claim 2, characterized in that, The load real-time estimation module reverses the actual hoisting load mass value in real time by deploying a tension sensor in the steel wire rope traction path of the tower crane, combining the hook structure form, the pulley ratio and the gravitational acceleration, further comprising: Process the traction force data measured by the sensor by sliding weighted average method: （3） Will Substituting into equation (2), we get: （4）。 4. The real-time operation safety monitoring and early warning system of tower crane according to claim 1, characterized in that, The dynamic torque fusion calculation module constructs a torque real-time dynamic estimation and risk classification model based on a multi-source data fusion algorithm to identify risks in the operation of the tower crane, comprising: Establish a torque estimation mathematical model: （5） Wherein M is the load torque generated by the current working arm of the tower crane, m is the hoisting load mass, g is the gravitational acceleration, and L is the horizontal distance from the hook to the center of the tower crane; （6） Wherein L0 is the base arm length, r is the trolley radius or telescopic segment length, and θ is the hoisting arm pitch angle; Substitute formula (6) into formula (5) to obtain: （7） Introduce a multi-variable fusion model: （8） wherein M t is the current moment of the integrated force moment indicator, dM / dt is the rate of change of the moment, representing a sudden dynamic impact; ΔM is the deviation of the average moment; w1, w2, w3 are fusion weights, satisfying w1+w2+w3=1; According to the calculated M t The maximum safe moment M max The ratio, which introduces the risk classification function R: （9） And classify the risk state according to the value of R.

5. The real-time operation safety monitoring and early warning system of tower crane according to claim 4, characterized in that, Further comprising an operation trend prediction module for predicting the safety impact of the next hoisting behavior.

6. The real-time operation safety monitoring and early warning system of a tower crane according to claim 5, characterized in that, The operation trend prediction module predicts the safety impact of the next hoisting behavior based on the operation trend prediction model of historical behavior data and sensor parameters.

7. The tower crane real-time operational safety monitoring and warning system, as claimed in claim 6, wherein, The operation trend prediction module predicts the safety impact of the next hoisting behavior based on the operation trend prediction model of historical behavior data and sensor parameters, comprising: Construct an operation sequence data set: Form a feature time sequence with key operation parameters of the tower crane, defined as follows: （10） where m t is the load mass; L t is the current amplitude; θ t is the current pitch angle; v t is the current trolley running speed; is the pitch angle change rate; is the amplitude change rate; constructing an operational behavior history window W = {X t−n ,...,X t} Introduce a state prediction function: Use an improved weighted recursive model to predict the operation trend in the future few seconds: （11） wherein α i are decreasing weight coefficients, satisfying∑α i =1. is the state prediction of the future time instant; Substitute formula (7) to obtain the future torque change: （12） Using the current state and the predicted state comparison, a trend risk indicator T is defined r : （13） If T r >δ, it indicates that the trend is going in an unsafe direction, and an early warning is issued: T r <δ1: stable; δ1≤ T r <δ2: slight deviation; T r ≥ δ2: Trend risk high, intervention needed; δ is the trend risk threshold value, δ1 is the first trend risk threshold value, and δ2 is the second trend risk threshold value.

8. The real-time operation safety monitoring and early warning system of a tower crane according to any one of claims 1-7, characterized in that, Further comprising an abnormality identification and intelligent alarm module for real-time identification of dangerous states and triggering of alarms.

9. The tower crane real-time operational safety monitoring and warning system, as claimed in claim 8, wherein, The abnormality identification and intelligent alarm module is used for real-time identification of dangerous states and triggering of alarms, comprising: Express the key data collected by the sensor as a multi-dimensional state vector: （14） where T t is a temperature environmental parameter, H t is a humidity environmental parameter, M t is the currently calculated operating torque; Fit the normal behavior data by a clustering learning or rule learning model to establish a feature distribution model and identify deviated behaviors; For each time instant t, compute the abnormality score of the current state Z t : （15） where: z t,i is the ith state value; μ i , σ i is the mean and standard deviation of the ith feature under normal conditions; w i is the feature weight; When S a (t) > λ, i.e. the abnormal state is determined; λ is an abnormal score threshold value.

10. The tower crane real-time operational safety monitoring and warning system, as claimed in claim 9, wherein, The security decision and graphical interaction module is used for real-time display of results of the load real-time estimation module, the dynamic moment fusion calculation module, the operation trend prediction module and the abnormality identification and intelligent alarm module.

Images