Automatic lifting balance weight adjusting system of intelligent tower crane

Abstract

The invention relates to the field of tower cranes, in particular to an intelligent tower crane hoisting balance weight automatic adjusting system which comprises a data collection module, an intelligent control module, a balance weight adjusting module and an early warning module which are in communication connection, and the data collection module is used for obtaining tower crane data and environment data in real time through a sensor; the intelligent control module is used for receiving data of the sensor and performing intelligent analysis to obtain a decision instruction; the balance weight adjusting module is used for converting the instruction sent by the intelligent control module into an action signal and adjusting the movable balance weight; and the early warning module is used for carrying out early warning grading according to the abnormal condition detected by the system and sending out an alarm signal of a corresponding grade. According to the system, the safety of the crane can be improved, the working efficiency is improved, the operation difficulty and the labor cost are reduced, and the reliability and adaptability of the system are enhanced.

Description

Technical Field

[0001] The invention relates to the field of tower cranes, in particular to an automatic adjustment system for the hoisting balance weight of an intelligent tower crane. Background Art

[0002] With the continuous development of the construction industry, tower cranes play a vital role in construction. However, in the lifting process of traditional tower cranes, the adjustment of the counterweight often relies on manual experience, which is not only inefficient but also difficult to ensure accuracy. In a complex construction environment, factors such as lifting weight, amplitude, angle, wind speed and inclination will affect the tower force and stability of the tower crane. If the counterweight cannot be adjusted in time and accurately, the crane may become unstable or even cause a safety accident. Although some existing adjustment systems can monitor some parameters, they often lack real-time and dynamic performance. It is difficult for traditional systems to dynamically adjust the counterweight according to the changes of various factors in real time during the lifting process, and they cannot meet the needs of complex and changing working environments. In addition, the existing tower crane early warning system is usually relatively simple and cannot provide accurate graded early warnings according to different degrees of danger. When abnormal situations occur, it is impossible to provide clear danger level information to operators in time, and it is difficult to take targeted measures. Summary of the invention

[0003] To solve the above problems, the present invention provides an automatic adjustment system for the lifting balance counterweight of an intelligent tower crane. The system can improve the safety of the crane, improve work efficiency, reduce operating difficulty and labor costs, and enhance the reliability and adaptability of the system.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] An intelligent tower crane lifting balance counterweight automatic adjustment system includes: a communication-connected data collection module, an intelligent control module, a counterweight adjustment module and an early warning module, wherein:

[0006] The data collection module is used to obtain tower crane data and environmental data in real time through sensors;

[0007] The intelligent control module is used to receive sensor data and perform intelligent analysis to obtain decision instructions;

[0008] The counterweight adjustment module is used to convert the instructions sent by the intelligent control module into action signals to adjust the movable balance weight;

[0009] The early warning module is used to classify early warnings according to abnormal conditions detected by the system and issue alarm signals of corresponding levels.

[0010] Furthermore, the sensor comprises:

[0011] The lifting weight sensor is installed at the hook, the amplitude sensor is installed on the boom, the angle sensor is installed between the tower body and the boom, the wind speed sensor is installed on the top of the tower body and the boom, and the tilt sensor is installed on the tower body.

[0012] Furthermore, the receiving of sensor data and performing intelligent analysis to obtain decision instructions comprises the following steps:

[0013] A1: The data collected by the data collection module in real time is verified by cyclic redundancy check, and the received data is filtered and normalized;

[0014] A2: Based on the processed sensor data, in a distributed framework environment, intelligent analysis is performed through the counterweight movement model to comprehensively determine the movement direction and distance of the counterweight.

[0015] Furthermore, the intelligent analysis is performed by using a counterweight movement model; comprising the following steps:

[0016] A21: Perform multi-model fusion judgment based on the lifting weight, amplitude, angle, wind speed and tilt obtained by the sensor to obtain the corresponding judgment results;

[0017] A22: Conduct a comprehensive analysis based on the judgment results, integrate the results and make the final decision.

[0018] Furthermore, the multi-model fusion judgment is performed respectively according to the lifting weight, amplitude, angle, wind speed and tilt obtained by the sensor; comprising the following steps:

[0019] A211: By establishing a physical relationship model between weight, amplitude, angle, wind speed and inclination and the force and stability of the tower, the direction and size of the counterweight movement are calculated;

[0020] A212: Through machine learning algorithms, historical sensor data is trained to establish a prediction model of weight, amplitude, angle, wind speed and tilt and the direction and size of counterweight movement to predict the direction of counterweight movement.

[0021] Furthermore, the physical relationship model of weight, amplitude, angle, wind speed and inclination and the tower body stress and stability is established; comprising the following steps:

[0022] A2111: According to the principles of physics, a mathematical model is established for the stress conditions of tower cranes under different working conditions. Based on the stress model, the stability index of the tower crane is calculated through the overturning moment ratio;

[0023] A2112: Based on experimental data and theoretical analysis, a physical relationship model between weight, amplitude, angle, wind speed and inclination and the tower body stress and stability is established through multiple linear regression method;

[0024] A2113: Based on the real-time collected sensor data and the established physical relationship model, calculate the direction and size of the counterweight movement under the current working state.

[0025] Furthermore, the historical sensor data is trained by a machine learning algorithm to establish a prediction model of weight, amplitude, angle, wind speed and tilt and the direction and size of counterweight movement; comprising the following steps:

[0026] A2121: Through the multi-layer perceptron neural network structure, the lifting weight, amplitude, angle, wind speed and tilt are input, the counterweight movement direction and size are output, ReLU is used as the activation function of the hidden layer, and the mean square error is used as the loss function;

[0027] A2122: Divide the preprocessed data into 70% training set, 15% validation set and 15% test set, randomly initialize the weights and bias parameters of the neural network through normal distribution, and train the neural network through the training set;

[0028] A2123: Calculate the gradient of the loss function with respect to the model parameters through the back-propagation algorithm, update the model parameters through the Adam optimization algorithm, optimize the model performance by adjusting the hyperparameters of the neural network, evaluate the model performance under different hyperparameter combinations through the validation set data, and evaluate the trained model through the test set data;

[0029] A2124: Collects data from lifting weight, amplitude, angle, wind speed and tilt sensors in real time, and inputs the pre-processed real-time data into the trained neural network model to predict the optimal counterweight movement direction and size.

[0030] Furthermore, the comprehensive analysis is performed based on the judgment results, and the results are integrated, including: fusing the results of multiple models to obtain a fusion result through a weighted summation method, evaluating each model separately, and making a comprehensive judgment based on the two evaluation results.

[0031] Furthermore, the instructions issued by the intelligent control module are converted into action signals to adjust the movable balancing weight, including: receiving instructions in real time, parsing according to a preset instruction format, converting digital instruction information into pulse signals, converting speed information in the instructions into a motor speed control signal through pulse width modulation technology, realizing direction information by controlling the forward and reverse rotation of the motor, amplifying and driving the converted action signal through an amplifier, the actuator acting according to the received action signal to adjust the position of the movable balancing weight, monitoring the status of the actuator during the execution of the action, and sending feedback information back to the intelligent control module through the position sensor and the speed sensor, the intelligent control module performing closed-loop control and continuously adjusting the instructions.

[0032] Furthermore, the method of grading the warning according to the abnormal situation detected by the system and issuing an alarm signal of the corresponding level includes the following steps:

[0033] B1: Collect data in real time through sensors, analyze the collected sensor data, compare it with the preset normal range and warning classification standards, and determine the warning level based on the data analysis results;

[0034] B2: The abnormal situation that is close to but not reaching the safety limit for crane operation and safety is set as the first-level warning, the situation that is close to but not reaching the serious danger level for crane operation and safety is set as the second-level warning, and the situation that reaches the serious danger level for crane operation and safety is set as the third-level warning;

[0035] B3: When the warning level is determined, the control system immediately triggers the corresponding alarm signal. The first-level warning uses flashing yellow light and low-frequency buzzing sound, the second-level warning uses orange light and high-frequency buzzing sound, and the third-level warning uses red light, continuous high-frequency buzzing sound and vibration alarm. The alarm signal is transmitted to the crane operating room and related monitoring center.

[0036] The beneficial effects of the present invention are:

[0037] 1. By establishing precise physical models and machine learning prediction models, the direction and size of counterweight movement can be accurately calculated to ensure the balance and stability of the tower crane during the lifting process, greatly reducing the risk of overturning due to imbalance. Early warning classification can detect abnormal conditions or dangers in a timely manner, and issue corresponding alarm signals according to different levels, so that operators can quickly take countermeasures and effectively avoid accidents.

[0038] 2. The intelligent automatic adjustment system can quickly adjust the movable balancing weight in real time according to changes in factors such as lifting weight, amplitude, angle, wind speed and tilt, reducing the time and uncertainty of manual intervention and improving the efficiency of lifting operations. The multi-model fusion method combines the accuracy of the physical model and the flexibility of the machine learning model, and can more quickly derive the optimal counterweight adjustment plan, further shortening the adjustment time. The multi-model fusion and comprehensive judgment methods improve the reliability of the system. Even if a model has a problem, the comprehensive judgment mechanism can be adjusted through the results and evaluations of other models to ensure the stable operation of the system.

[0039] 3. Operators no longer need to rely on experience to make complex counterweight adjustment judgments. The system automatically provides accurate adjustment solutions, which reduces the difficulty of operation and reduces the dependence on professional experience. The automated adjustment system reduces the need for a large number of manual operations, thereby reducing labor costs. The system can adapt to different working scenarios and complex and changing environmental conditions, such as different wind speeds, lifting weights and tilt angles, and has strong adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a module schematic diagram of the automatic adjustment system of the lifting balance counterweight of the intelligent tower crane.

[0041] Figure 2 It is a schematic flow chart of the intelligent control module.

[0042] Figure 3 It is a schematic flow chart of the early warning module. DETAILED DESCRIPTION

[0043] See also Figure 1 , Figure 2 and Figure 3 As shown, the present invention relates to an automatic adjustment system for the lifting balance counterweight of an intelligent tower crane, comprising: a data collection module, an intelligent control module, a counterweight adjustment module and an early warning module connected in communication, wherein:

[0044] The data collection module is used to obtain tower crane data and environmental data in real time through sensors;

[0045] The intelligent control module is used to receive sensor data and perform intelligent analysis to obtain decision instructions;

[0046] The counterweight adjustment module is used to convert the instructions sent by the intelligent control module into action signals to adjust the movable balance weight;

[0047] The early warning module is used to classify early warnings according to abnormal conditions detected by the system and issue alarm signals of corresponding levels.

[0048] Furthermore, the sensor comprises:

[0049] The lifting weight sensor is installed at the hook, the amplitude sensor is installed on the boom, the angle sensor is installed between the tower body and the boom, the wind speed sensor is installed on the top of the tower body and the boom, and the tilt sensor is installed on the tower body.

[0050] Specifically, the lifting weight sensor installed at the hook is used to monitor the weight of the lifted object in real time and provide the system with key lifting load data; the amplitude sensor is installed on the boom to measure the amplitude of the boom, that is, the horizontal distance from the hook to the tower body, to help the system understand the working range and status of the crane; the angle sensor is installed between the tower body and the boom to monitor the angle of the boom so that the system can accurately calculate the working posture and torque of the crane; the wind speed sensor is installed on the top of the tower body and the boom to obtain the wind speed data in the environment, because the wind has an important influence on the stability of the tower crane, and the system can adjust the counterweight and operation strategy according to the wind speed; the tilt sensor is installed on the tower body to detect the tilt angle of the tower body to ensure that the crane remains vertical during operation and to detect potential unstable factors in time.

[0051] After the sensor is installed, the sensor is calibrated to ensure that the data output by the sensor is accurate and reliable. The calibration process requires the use of equipment such as standard weights and angle measuring tools, and the operation is carried out according to the sensor calibration method. Perform functional tests on the sensor to simulate different working conditions and check whether the sensor can accurately obtain data and transmit it to the control system. After the sensor is installed and the network is configured, perform data transmission tests, send simulated data or actually operate the crane to check whether the sensor data can be accurately and timely transmitted to the control system. The data acquisition module in the control system is responsible for receiving data from the sensor, and can collect data from various sensors in real time, and perform preliminary processing and sorting. The frequency of data collection is set according to the working characteristics of the crane and the requirements of the system to ensure that key data can be obtained in a timely manner, while avoiding excessive data causing the system to be overloaded. The collected data needs to be stored for subsequent analysis and regulation. Use the database management system MySQL to store data on a local server or a cloud server. Data storage takes into account the security and reliability of the data, and takes backup and recovery measures to prevent data loss.

[0052] Furthermore, the receiving of sensor data and performing intelligent analysis to obtain decision instructions comprises the following steps:

[0053] A1: The data collected by the data collection module in real time is verified by cyclic redundancy check, and the received data is filtered and normalized;

[0054] A2: Based on the processed sensor data, in a distributed framework environment, intelligent analysis is performed through the counterweight movement model to comprehensively determine the movement direction and distance of the counterweight.

[0055] Specifically, in step A1, data from each sensor is received in real time based on the communication connection with the sensor, and data acquisition card, communication protocol stack and other technical means are used to realize data reception. The received data is checked and verified to ensure the integrity and accuracy of the data. The received data is filtered through a digital filter to remove noise and interference. The data is normalized to unify the data of different sensors into the same scale range, which is helpful for subsequent data analysis and processing.

[0056] Furthermore, the intelligent analysis is performed by using a counterweight movement model; comprising the following steps:

[0057] A21: Perform multi-model fusion judgment based on the lifting weight, amplitude, angle, wind speed and tilt obtained by the sensor to obtain the corresponding judgment results;

[0058] A22: Conduct a comprehensive analysis based on the judgment results, integrate the results and make the final decision.

[0059] Specifically, in the distributed architecture environment, Spark master nodes and slave nodes are configured, the collected data is stored in HDFS, real-time data is continuously written to HDFS, and the written Spark application is submitted to the Spark cluster for execution. The running status of the Spark job is monitored, including task progress, resource usage, and performance indicators. Optimization is performed based on the monitoring results, including adjusting the parallelism, memory allocation, and data partitioning of the job to improve the efficiency and performance of the job.

[0060] Furthermore, the multi-model fusion judgment is performed respectively according to the lifting weight, amplitude, angle, wind speed and tilt obtained by the sensor; comprising the following steps:

[0061] A211: By establishing a physical relationship model between weight, amplitude, angle, wind speed and inclination and the force and stability of the tower, the direction and size of the counterweight movement are calculated;

[0062] A212: Through machine learning algorithms, historical sensor data is trained to establish a prediction model of weight, amplitude, angle, wind speed and tilt and the direction and size of counterweight movement to predict the direction of counterweight movement.

[0063] Furthermore, the physical relationship model of weight, amplitude, angle, wind speed and inclination and the tower body stress and stability is established; comprising the following steps:

[0064] A2111: According to the principles of physics, a mathematical model is established for the stress conditions of tower cranes under different working conditions. Based on the stress model, the stability index of the tower crane is calculated through the overturning moment ratio;

[0065] A2112: Based on experimental data and theoretical analysis, a physical relationship model between weight, amplitude, angle, wind speed and inclination and the tower body stress and stability is established through multiple linear regression method;

[0066] A2113: Based on the real-time collected sensor data and the established physical relationship model, calculate the direction and size of the counterweight movement under the current working state.

[0067] Specifically, the principle of physics and the mechanical structure model of the crane are used in combination with sensor data for analysis. By establishing a relationship model between factors such as weight, amplitude, angle, wind speed and inclination and the force and stability of the tower body, the direction and size of the counterweight movement are calculated. According to the geometric structure and mechanical principles of the crane, a mathematical model is established. For example, according to the principle of moment balance, the moment of the tower body when lifting heavy objects and the balancing moment of the counterweight on the tower body are calculated. Combined with sensor data, such as the weight of the lifting object provided by the lifting weight sensor, the boom amplitude provided by the amplitude sensor, the boom angle provided by the angle sensor, the wind speed provided by the wind speed sensor, and the tower inclination angle provided by the inclination sensor, etc., they are substituted into the mathematical model for calculation. By analyzing the calculation results, the direction and size of the counterweight movement are determined to achieve the balance and stability of the tower body. For example, if the lifting weight is large and the boom amplitude is large, and the wind speed is large, causing the tower body to tend to tilt to one side, it is calculated that the counterweight needs to be moved a certain distance in the opposite direction to increase the balancing moment and prevent the tower body from tilting.

[0068] More specifically, according to the principles of physics, the stress conditions of tower cranes under different working conditions are analyzed. The effects of gravity, lifting force, wind force, inertia force, etc. on the tower body, boom and balance arm are mainly considered, and a mathematical model is established to decompose each force into different coordinate axes to calculate the resultant force and moment on the tower body. Based on the force analysis model, the stability index of the tower crane is calculated, and the stability of the crane is evaluated using the stability coefficient or overturning moment ratio index. Considering the safety factor requirements under different working conditions, the stable boundary conditions of the crane are determined.

[0069] Through experimental data and theoretical analysis, a physical relationship model between factors such as weight, amplitude, angle, wind speed and tilt and the force and stability of the tower body is established, and modeling is performed using multiple linear regression or neural network. The physical relationship model should be able to accurately reflect the impact of counterweight movement on the force and stability of the tower body. Based on the real-time collected sensor data and the established relationship model, the direction and size of the counterweight movement under the current working state are calculated, and an optimization algorithm is used to find the optimal counterweight movement solution while meeting the stability requirements.

[0070] Furthermore, the historical sensor data is trained by a machine learning algorithm to establish a prediction model of weight, amplitude, angle, wind speed and tilt and the direction and size of counterweight movement; comprising the following steps:

[0071] A2121: Through the multi-layer perceptron neural network structure, the lifting weight, amplitude, angle, wind speed and tilt are input, the counterweight movement direction and size are output, ReLU is used as the activation function of the hidden layer, and the mean square error is used as the loss function;

[0072] A2122: Divide the preprocessed data into 70% training set, 15% validation set and 15% test set, randomly initialize the weights and bias parameters of the neural network through normal distribution, and train the neural network through the training set;

[0073] A2123: Calculate the gradient of the loss function with respect to the model parameters through the back-propagation algorithm, update the model parameters through the Adam optimization algorithm, optimize the model performance by adjusting the hyperparameters of the neural network, evaluate the model performance under different hyperparameter combinations through the validation set data, and evaluate the trained model through the test set data;

[0074] A2124: Collects data from lifting weight, amplitude, angle, wind speed and tilt sensors in real time, and inputs the pre-processed real-time data into the trained neural network model to predict the optimal counterweight movement direction and size.

[0075] Specifically, the preprocessed data set is divided into a training set, a validation set, and a test set. The training set is used to train the neural network model, the validation set is used to adjust the hyperparameters of the model, and the test set is used to evaluate the performance of the model. The neural network is trained using the training set data. The training process usually uses a back-propagation algorithm to calculate the gradient of the loss function to the model parameters, and then uses an optimization algorithm to update the model parameters to reduce the value of the loss function. The Adam optimization algorithm was chosen because it has the advantages of adaptive learning rate and fast convergence speed. During the training process, the model performance is optimized by adjusting the hyperparameters of the neural network. The hyperparameters include learning rate, number of hidden layers, number of neurons, batch size, etc. The trained model is evaluated with the test set data. The evaluation indicators include root mean square error (RMSE), mean absolute error (MAE), coefficient of determination (R 2 ) etc. If the model performance does not meet the requirements, readjust the model structure, hyperparameters or increase the amount of data for training.

[0076] During the operation of the tower crane, sensor data such as lifting weight, amplitude, angle, wind speed and tilt are collected in real time. The real-time collected data is subjected to the same preprocessing steps as the training data, including data cleaning and normalization. The preprocessed real-time data is input into the trained neural network model to predict the optimal counterweight movement direction and size.

[0077] Furthermore, the comprehensive analysis is performed based on the judgment results, and the results are integrated, including: fusing the results of multiple models to obtain a fusion result through a weighted summation method, evaluating each model separately, and making a comprehensive judgment based on the two evaluation results.

[0078] Specifically, set initial weights for the physical model and the machine learning model respectively, and make preliminary settings based on experience or prior knowledge. For example, set the weights of both models to 0.5, that is, give them equal importance. Use trial and error or adjust the weights based on some performance indicators. For example, by comparing the accuracy and stability of the fusion results under different weight combinations, gradually optimize the weight values. Or consider dynamically adjusting the weights according to different working scenarios or crane states. For example, in the case of high wind speeds, prefer the results of the physical model, thereby increasing the weight of the physical model. Since the results of the physical model and the machine learning model may be in different numerical ranges, the results need to be standardized. Use maximum and minimum standardization or Z-score standardization to map the results to the same numerical range.

[0079] According to the determined weights, the standardized results of the physical model and the machine learning model are weighted and summed: Assume that the result of the physical model is P, the result of the machine learning model is M, and the weight of the physical model is w p , the machine learning model weight is w m (wp +w m =1), then the fusion result F = w p ×P+w m ×M.

[0080] Evaluate the results of the physical model through evaluation indicators, such as calculating the root mean square error (RMSE) and mean absolute error (MAE) between the prediction results of the physical model and the actual counterweight movement. Analyze the stability and reliability of the physical model under different working conditions, such as its performance under different wind speeds and lifting weights. Also use evaluation indicators to evaluate the prediction results of the machine learning model, and calculate the error index between the prediction results and the actual values, as well as the performance indicators of the model such as accuracy and recall. Evaluate the generalization ability of the machine learning model, that is, the ability to predict unseen data. Compare the evaluation results of the physical model and the machine learning model. Analyze the advantages and disadvantages of the two models in different aspects, such as the physical model is more stable under certain specific working conditions, while the machine learning model is more accurate in other cases. Adjust the weight or adopt other fusion strategies according to the specific situation. Make a comprehensive judgment decision based on the fusion results and the evaluation results of the two models. If the fusion result is within the acceptable error range and the evaluation results of the two models are relatively good, it is considered that the system is operating normally, and the fusion result will continue to be used for counterweight adjustment. If the fusion result deviates greatly from the actual situation, or the evaluation result of a model shows that it has a problem, further measures need to be taken, such as readjusting the model parameters, checking the sensor data, etc.

[0081] Furthermore, the instructions issued by the intelligent control module are converted into action signals to adjust the movable balancing weight, including: receiving instructions in real time, parsing according to a preset instruction format, converting digital instruction information into pulse signals, converting speed information in the instructions into a motor speed control signal through pulse width modulation technology, realizing direction information by controlling the forward and reverse rotation of the motor, amplifying and driving the converted action signal through an amplifier, the actuator acting according to the received action signal to adjust the position of the movable balancing weight, monitoring the status of the actuator during the execution of the action, and sending feedback information back to the intelligent control module through the position sensor and the speed sensor, the intelligent control module performing closed-loop control and continuously adjusting the instructions.

[0082] Specifically, first define the format of the instruction issued by the intelligent control module. The instruction contains information such as the direction of the counterweight movement (such as forward, backward, left, right), the distance or angle of movement, and the speed of movement. For example, the instruction uses one byte of data to represent the direction, two bytes of data to represent the distance, and one byte of data to represent the speed. After receiving the instruction, the control unit of the actuator parses it. According to the definition of the instruction format, the received data is split into different fields, and the direction, distance, and speed information are extracted respectively. For example, the received data is "0x12 0x34 0x56". According to the definition of the instruction format, the direction may be parsed as forward, the distance is 0x3456 (set in millimeters), and the speed is medium speed. The parsed instruction information is converted into an action signal that the actuator can understand, and the digital instruction information is converted into a pulse signal. For the motor-driven actuator, the speed information in the instruction is converted into the motor speed control signal, using pulse width modulation (PWM) technology. The direction information is realized by controlling the forward and reverse rotation of the motor, and the distance information is realized by calculating the number of steps or time the motor needs to rotate. For example, if the speed in the instruction is medium speed, the corresponding PWM duty cycle is set to 50%. If the direction is forward, the motor is controlled to rotate forward. If the distance is 100 mm, the number of steps the motor needs to rotate is calculated according to the step angle and transmission ratio of the motor, and then the motor is controlled to rotate by sending the corresponding number of pulse signals. The converted action signal may need to be amplified and driven to meet the power requirements of the actuator, which is achieved by using a motor driver, amplifier or relay. For example, for a motor-driven actuator, the PWM signal is amplified by the motor driver to provide sufficient current and voltage to drive the motor to rotate.

[0083] The actuator acts according to the received action signal to adjust the position of the movable balancing weight. For example, the motor-driven actuator controls the rotation of the motor according to the direction and distance information in the instruction, and drives the balancing weight to move through the transmission device. During the execution of the action, it is necessary to monitor the status of the actuator and send feedback information back to the intelligent control module, which is achieved by using sensors (such as position sensors, speed sensors, and current sensors). For example, the position sensor monitors the position of the balancing weight in real time and feeds back the position information to the intelligent control module. The intelligent control module determines whether the action is executed correctly based on the feedback information and makes necessary adjustments. Based on the feedback information, the intelligent control module performs closed-loop control and continuously adjusts the instructions to ensure that the balancing weight can be accurately moved to the target position. For example, if the feedback information shows that the position of the balancing weight is still deviated from the target position, the intelligent control module generates instructions again to adjust the action of the actuator until the balancing weight reaches the target position.

[0084] Furthermore, the method of grading the warning according to the abnormal situation detected by the system and issuing an alarm signal of the corresponding level includes the following steps:

[0085] B1: Collect data in real time through sensors, analyze the collected sensor data, compare it with the preset normal range and warning classification standards, and determine the warning level based on the data analysis results;

[0086] B2: The abnormal situation that is close to but not reaching the safety limit for crane operation and safety is set as the first-level warning, the situation that is close to but not reaching the serious danger level for crane operation and safety is set as the second-level warning, and the situation that reaches the serious danger level for crane operation and safety is set as the third-level warning;

[0087] B3: When the warning level is determined, the control system immediately triggers the corresponding alarm signal. The first-level warning uses flashing yellow light and low-frequency buzzing sound, the second-level warning uses orange light and high-frequency buzzing sound, and the third-level warning uses red light, continuous high-frequency buzzing sound and vibration alarm. The alarm signal is transmitted to the crane operating room and related monitoring center.

[0088] Specifically, the first-level warning is defined as abnormal conditions that may have a minor impact on the operation and safety of the crane, such as slight changes in wind speed, small changes in tilt angles, and a certain proportion of the lifting weight approaching but not reaching the safety limit. The second-level warning is defined as a situation that has a certain impact on the operation and safety of the crane, such as wind speed reaching a certain intensity but not exceeding the dangerous level, large tilt angles but not reaching the serious dangerous level, and a certain higher proportion of the lifting weight reaching the safety limit. The third-level warning is defined as a situation that poses a serious threat to the operation and safety of the crane, such as strong winds, severe tilts, and lifting weight exceeding the safety limit. Continuously use the sensors installed on the tower crane (lifting weight sensor, amplitude sensor, angle sensor, wind speed sensor, tilt sensor, etc.) to collect data in real time, and transmit the collected data to the control system at a certain frequency for analysis and processing. Analyze the collected sensor data and compare it with the preset normal range and warning classification standards, such as calculating the current wind speed and comparing it with the wind speed threshold corresponding to different warning levels; calculating the tilt angle and comparing it with the tilt angle threshold of each level; judging whether the lifting weight reaches the limit ratio of different warning levels. The warning level is determined based on the results of data analysis. If the data is within the first-level warning range, it is determined as a first-level warning. Similarly, whether the data meets the conditions for second-level and third-level warnings is determined to determine the corresponding warning level.

[0089] When the warning level is determined, the control system immediately triggers the corresponding alarm signal. The alarm signal is transmitted to the crane operation room and related monitoring centers by wired or wireless means to ensure that operators and managers can receive the alarm information in time. After receiving the alarm signal, the operator should immediately take corresponding measures according to the warning level. For the first-level warning, you can pay close attention to the status of the crane and take preventive measures; for the second-level warning, suspend the operation, check and adjust the crane; for the third-level warning, immediately stop the crane operation and take emergency safety measures. Relevant managers should activate the corresponding emergency plan according to the alarm level, organize professionals to inspect and repair the crane, and ensure the safe operation of the crane.

[0090] The above implementation modes are merely descriptions of the preferred implementation modes of the present invention, and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary engineering and technical personnel in the field shall fall within the protection scope determined by the claims of the present invention.

Claims

1. An intelligent tower crane lifting balance weight automatic adjustment system, characterized in that: include: The data collection module, intelligent control module, counterweight adjustment module and early warning module are connected by communication, wherein: The data collection module is used to obtain tower crane data and environmental data in real time through sensors; The intelligent control module is used to receive sensor data and perform intelligent analysis to obtain decision instructions; The counterweight adjustment module is used to convert the instructions sent by the intelligent control module into action signals to adjust the movable balance weight; The early warning module is used to classify early warnings according to abnormal conditions detected by the system and issue alarm signals of corresponding levels.

2. The automatic adjustment system for lifting balance weight of intelligent tower crane according to claim 1 is characterized in that: The sensor comprises: The lifting weight sensor is installed at the hook, the amplitude sensor is installed on the boom, the angle sensor is installed between the tower body and the boom, the wind speed sensor is installed on the top of the tower body and the boom, and the tilt sensor is installed on the tower body.

3. The automatic adjustment system for lifting balance weight of intelligent tower crane according to claim 1 is characterized in that: The method of receiving sensor data and performing intelligent analysis to obtain decision instructions includes the following steps: A1: The data collected by the data collection module in real time is verified by cyclic redundancy check, and the received data is filtered and normalized; A2: Based on the processed sensor data, in a distributed framework environment, intelligent analysis is performed through the counterweight movement model to comprehensively determine the movement direction and distance of the counterweight.

4. The automatic adjustment system for lifting balance weight of intelligent tower crane according to claim 3 is characterized in that: The intelligent analysis is performed by using a counterweight movement model; comprising the following steps: A21: Perform multi-model fusion judgment based on the lifting weight, amplitude, angle, wind speed and tilt obtained by the sensor to obtain the corresponding judgment results; A22: Conduct a comprehensive analysis based on the judgment results, integrate the results and make the final decision.

5. The automatic adjustment system for lifting balance weight of intelligent tower crane according to claim 4 is characterized in that: The method of performing multi-model fusion judgment according to the lifting weight, amplitude, angle, wind speed and tilt obtained by the sensor comprises the following steps: A211: By establishing a physical relationship model between weight, amplitude, angle, wind speed and inclination and the force and stability of the tower, the direction and size of the counterweight movement are calculated; A212: Through machine learning algorithms, historical sensor data is trained to establish a prediction model of weight, amplitude, angle, wind speed and tilt and the direction and size of counterweight movement to predict the direction of counterweight movement.

6. The automatic adjustment system for lifting balance weight of intelligent tower crane according to claim 5 is characterized in that: The method of establishing a physical relationship model between weight, amplitude, angle, wind speed and inclination and the force and stability of the tower body comprises the following steps: A2111: According to the principles of physics, a mathematical model is established for the stress conditions of tower cranes under different working conditions. Based on the stress model, the stability index of the tower crane is calculated through the overturning moment ratio; A2112: Based on experimental data and theoretical analysis, a physical relationship model between weight, amplitude, angle, wind speed and inclination and the tower body stress and stability is established through multiple linear regression method; A2113: Based on the real-time collected sensor data and the established physical relationship model, calculate the direction and size of the counterweight movement under the current working state.

7. The automatic adjustment system for lifting balance weight of intelligent tower crane according to claim 5 is characterized in that: The method of training the historical sensor data by a machine learning algorithm to establish a prediction model of weight, amplitude, angle, wind speed and tilt and the direction and size of counterweight movement includes the following steps: A2121: Through the multi-layer perceptron neural network structure, the lifting weight, amplitude, angle, wind speed and tilt are input, the counterweight movement direction and size are output, ReLU is used as the activation function of the hidden layer, and the mean square error is used as the loss function; A2122: Divide the preprocessed data into 70% training set, 15% validation set and 15% test set, randomly initialize the weights and bias parameters of the neural network through normal distribution, and train the neural network through the training set; A2123: Calculate the gradient of the loss function with respect to the model parameters through the back-propagation algorithm, update the model parameters through the Adam optimization algorithm, optimize the model performance by adjusting the hyperparameters of the neural network, evaluate the model performance under different hyperparameter combinations through the validation set data, and evaluate the trained model through the test set data; A2124: Collects data from lifting weight, amplitude, angle, wind speed and tilt sensors in real time, and inputs the pre-processed real-time data into the trained neural network model to predict the optimal counterweight movement direction and size.

8. The automatic adjustment system for lifting balance weight of intelligent tower crane according to claim 4 is characterized in that: The comprehensive analysis is performed based on the judgment results, and the results are integrated, including: fusing the results of multiple models through a weighted summation method to obtain a fusion result, evaluating each model separately, and making a comprehensive judgment based on the two evaluation results.

9. The automatic adjustment system for lifting balance weight of intelligent tower crane according to claim 1 is characterized in that: The method converts the command issued by the intelligent control module into an action signal to adjust the movable balancing weight, including: receiving the command in real time, parsing according to a preset command format, converting the digital command information into a pulse signal, converting the speed information in the command into a motor speed control signal through a pulse width modulation technology, realizing the direction information by controlling the forward and reverse rotation of the motor, amplifying and driving the converted action signal through an amplifier, the actuator taking action according to the received action signal to adjust the position of the movable balancing weight, monitoring the state of the actuator during the execution of the action, and sending feedback information back to the intelligent control module through a position sensor and a speed sensor, and the intelligent control module performing closed-loop control and continuously adjusting the command.

10. The automatic adjustment system for lifting balance weight of intelligent tower crane according to claim 1, characterized in that: The method of grading the warning according to the abnormal conditions detected by the system and issuing the corresponding alarm signal comprises the following steps: B1: Collect data in real time through sensors, analyze the collected sensor data, compare it with the preset normal range and warning classification standards, and determine the warning level based on the data analysis results; B2: The abnormal situation that is close to but not reaching the safety limit for crane operation and safety is set as the first-level warning, the situation that is close to but not reaching the serious danger level for crane operation and safety is set as the second-level warning, and the situation that reaches the serious danger level for crane operation and safety is set as the third-level warning; B3: When the warning level is determined, the control system immediately triggers the corresponding alarm signal. The first-level warning uses flashing yellow light and low-frequency buzzing sound, the second-level warning uses orange light and high-frequency buzzing sound, and the third-level warning uses red light, continuous high-frequency buzzing sound and vibration alarm. The alarm signal is transmitted to the crane operating room and related monitoring center.

Images