Anti-rollover detection method of crane

Abstract

The invention provides an anti-rollover detection method for a crane. The anti-rollover detection method comprises the following steps: step 1, acquiring operation data, equipment data and state data of the crane; step 2, performing fuzzy evaluation on the operation data, the equipment data and the state data through a fuzzy hierarchy algorithm, and obtaining weighted fuzzy evaluation values corresponding to the operation data, the equipment data and the state data based on the weights corresponding to the operation data, the equipment data and the state data, and outputting the safety evaluation level of the crane. According to the method, various conditions in the operation process of the hoisting equipment are integrated, the obtained heterogeneous data are efficiently and accurately processed and analyzed, the efficiency of anti-rollover real-time detection of the hoisting equipment is improved, the operation and maintenance cost of the hoisting equipment is saved, and the operation safety of the hoisting equipment is improved.

Description

Technical Field

[0001] The invention relates to the technical field of special equipment safety assessment, and in particular to an anti-rollover detection method for a crane. Background Art

[0002] Lifting equipment is crucial in the fields of industry, construction and port logistics, especially in ports, where they play a central role in cargo loading and unloading and ship unloading. Through automation, digitalization and intelligent technologies, port lifting machinery has improved the safety and reliability of operations and reduced safety accidents. However, due to the complexity of the operating environment and the variability of load conditions, safety accidents of lifting equipment occur frequently, so it is particularly important to monitor the safety of these equipment.

[0003] To prevent accidents, port cranes are usually equipped with anti-tilt devices based on the principle of mechanical balance. With the increasing complexity of logistics operations and the improvement of automation levels, these devices face new challenges and require higher levels of intelligence and automation to adapt to complex working conditions, while also focusing on cost-effectiveness and ease of maintenance. Summary of the invention

[0004] The present invention provides an anti-rollover detection method for a crane, which is used to solve the defects of the prior art that the anti-rollover device of the crane has insufficient automation, is difficult to cope with complex working conditions and operation requirements, is difficult to use previous response experience to perform efficient and accurate safety assessment and automatic control, and cannot operate stably for a long time.

[0005] The present invention provides a method for detecting an anti-tilt of a crane, comprising the following steps: Step 1, obtaining real-time data of the crane; Step 2, based on the real-time data and its mapping relationship with the operation data, equipment data and status data, the safety evaluation level of the crane is calculated through a fuzzy hierarchy algorithm.

[0006] The present invention provides a method for detecting an anti-tilt of a crane, further comprising step 3: Based on the operation data, the equipment data and the status data, as well as the corresponding safety assessment level, the DNN deep neural network is used to predict the crane's safety wind protection level.

[0007] Among them, in step 1, the operation data are illegal operation data or operation error data; the equipment data are structural parameters and operation status; and the status data are wind environment data and crane position height.

[0008] Wherein, the step 2 further comprises: Step 21, acquiring the real-time data of the crane, and obtaining a corresponding mapping relationship between the real-time data and the operation data, the equipment data, and the status data of the crane; Step 22, based on the real-time data and the corresponding mapping, using a fuzzy hierarchy algorithm, calculating the fuzzy relationship and corresponding weights between the operation data, the device data and the status data; Step 23, based on the fuzzy relationship and the corresponding weight, obtain the membership of the operation data, the device data and the status data corresponding to different safety evaluation levels, and obtain a preprocessed data set consisting of the membership of the operation data, the device data and the status data to the safety evaluation level.

[0009] Wherein, the step 22 further includes: Step 221, acquiring the real-time data, and a corresponding mapping relationship between the real-time data and the operation data, the equipment data, and the status data of the crane; Step 222, based on the real-time data and the corresponding mapping relationship, using a fuzzy evaluation matrix to reflect the fuzzy relationship between the operation data, the device data and the status data; Step 223, based on the real-time data and the corresponding mapping relationship, using a hierarchical analysis algorithm, calculate the corresponding weights of the operation data, the device data and the status data.

[0010] Wherein, the step 3 further comprises: Step 31, inputting the preprocessed data set into the DNN deep neural network; Step 32, based on the pre-processed data, using the activation function to perform calculations to predict the wind safety level score; Step 33, outputting a windproof safety assessment result according to the windproof safety grade score.

[0011] Wherein, step 32 further includes: Step 321, based on the preprocessed data, using a cross-validation method, determining the input weight of the activation function to obtain a current activation function; Step 322: Based on the current activation function, the DNN deep neural network is used to extract features related to wind safety and output the wind safety level.

[0012] The present invention also provides a crane anti-rollover detection device, which includes: a signal acquisition device, used to obtain the operation data, the equipment data and the status data of the crane; a data analysis device, used to calculate the operation data, the equipment data and the status data using a fuzzy hierarchical algorithm and a DNN deep neural network, and output the windproof safety assessment result under the current state of the crane; an output device, used to obtain a warning signal, safety level or control instruction based on the windproof safety assessment result.

[0013] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any anti-rollover detection method for a crane provided by the present invention.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the anti-rollover detection methods for a crane of the present invention when executing the computer program.

[0015] The present invention provides a method and device for detecting the anti-rollover of a crane. By comprehensively considering various influencing factors in the operating equipment, a machine learning method is used to process multi-source heterogeneous data that affects the safety of the equipment operation, and a real-time monitoring model for the operation status of the lifting equipment is constructed. The method and device of the present invention comprehensively consider the data of the lifting equipment itself, the human factors and environmental factors that affect the operation of the lifting equipment, and use advanced deep neural network algorithms to achieve efficient and accurate real-time detection of the safety status of the lifting equipment. The method of the present invention achieves an improvement in the efficiency of real-time detection of anti-rollover of the crane, saves the operation and maintenance costs of the lifting equipment, and improves the safety of the operation of the lifting equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 It is a schematic diagram of the flow of the crane anti-tipping detection device provided by the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] Crane anti-tilt detection is used to evaluate the stability of cranes during operation to prevent cranes from tipping over due to loss of balance. This detection is essential to ensure the safety of crane operations, especially in environments such as ports, construction sites and industrial facilities, where the stability of cranes is directly related to personnel safety and equipment integrity.

[0020] The present invention realizes real-time monitoring of the operating status of lifting equipment by deeply integrating machine learning technology, greatly improving the safety of the work site. The method and device can comprehensively capture the key information that affects the safe operation of lifting equipment by analyzing the equipment's own data, human operation factors, and environmental factors. Using the deep neural network algorithm, the present invention can efficiently and accurately process these multi-source heterogeneous data, build an accurate real-time monitoring model for the operating status of lifting equipment, and provide operators with immediate safety feedback.

[0021] In specific implementation, the anti-rollover detection device of the present invention adopts advanced sensor technology, which can monitor the movement of the lifting machinery and wind environment parameters such as wind speed and wind direction in real time. These data are crucial for evaluating the stability of the lifting machinery in complex environments. By integrating accurate algorithm models, the device can calculate the wind load that the lifting machinery will face, which is extremely critical for preventing rollover accidents caused by excessive wind. In addition, the device also has a threshold correction function, which can dynamically adjust the safety threshold according to real-time data to adapt to changing environmental conditions.

[0022] Combining these data, the device of the present invention can provide a comprehensive evaluation of the wind resistance performance of lifting equipment, thereby providing scientific and objective data support for the safety management of lifting machinery. This comprehensive evaluation not only includes the status of the equipment itself, but also covers human operation and environmental factors, making safety management more comprehensive and accurate. Through this real-time monitoring and evaluation, the present invention can significantly enhance operational safety, reduce potential risks, and provide a strong guarantee for the safe operation of lifting machinery.

[0023] Figure 1 FIG. 1 is a flow chart of the anti-tilt detection method for a crane provided by the present invention. Figure 1As shown, it includes: step 1, obtaining the real-time data of the crane; step 2, based on the real-time data and its mapping relationship with operation data, equipment data and status data, calculating the safety evaluation level of the crane through a fuzzy hierarchical algorithm.

[0024] The anti-tilt detection method for the crane further comprises step 3: Based on the operation data, the equipment data and the status data, as well as the corresponding safety assessment level, the DNN deep neural network is used to predict the crane's safety wind protection level.

[0025] The DNN deep neural network includes an input layer, a hidden layer and an output layer. The input layer is used for preprocessed data, including the operation data, the device data and the status data; the hidden layer contains a nonlinear function for learning and simulating the function mapping relationship of the complex relationship between the operation data, the device data and the status data; the output layer outputs the windproof safety assessment result and calculates the windproof safety level.

[0026] Among them, in step 1, the operation data are illegal operation data or operation error data; the equipment data are structural parameters and operation status; and the status data are wind environment data and crane position height.

[0027] Among them, the equipment data are structural parameters and operating status. The structural parameters include the structural strength of the equipment. The operating status includes the equipment lifting weight, the equipment posture angle, acceleration data and angular velocity data during the operation of the equipment. The status data are wind environment data and crane position height. The wind environment data includes wind speed and wind direction. The crane position height is the height of the crane estimated based on the air pressure.

[0028] The operation data, equipment data and status data of the crane are obtained through a signal acquisition device. The operation data include illegal operations, operational errors, and professional skills of personnel, which are usually monitored and collected through the terminal equipment of the crane; the equipment data include equipment structural strength, lifting weight, boom angle, and crane posture (yaw angle, pitch angle and roll angle), acceleration during movement, angular velocity, and height data, which are usually collected through a high-precision sensor array. Specifically, the high-precision sensor array includes a waterproof protective cover, a ten-axis attitude angle sensor, a wind detection device, and an air pressure sensor.

[0029] Wherein, the step 2 further comprises: Step 21, acquiring the real-time data of the crane, and obtaining a corresponding mapping relationship between the real-time data and the operation data, the equipment data, and the status data of the crane; Step 22, based on the real-time data and the corresponding mapping, using a fuzzy hierarchy algorithm, calculating the fuzzy relationship and corresponding weights between the operation data, the device data and the status data; Step 23, based on the fuzzy relationship and the corresponding weight, obtain the membership of the operation data, the device data and the status data corresponding to different safety evaluation levels, and obtain a preprocessed data set consisting of the membership of the operation data, the device data and the status data to the safety evaluation level.

[0030] Among them, the real-time data refers to the various indicator data affecting the safe operation of the lifting equipment obtained through the collection equipment, and the corresponding mapping relationship refers to the corresponding mapping relationship between the real-time data and the three pre-set data categories of the operation data, the equipment data and the status data automatically generated during data acquisition.

[0031] The fuzzy hierarchy algorithm is a comprehensive evaluation method based on fuzzy mathematics and hierarchical analysis. This method can be used to construct safety evaluation indicators for complex operating environments and changing load conditions during the operation of lifting equipment. This method can handle the uncertainty and fuzziness of evaluation indicators and improve the accuracy of safety evaluation. Through the fuzzy hierarchy algorithm, the fuzzy relationship and corresponding weights between the operation data, the equipment data and the status data are calculated.

[0032] Among them, the fuzzy relationship and the corresponding weight are used to comprehensively evaluate factors such as the operation data, the equipment data and the status data to obtain the membership corresponding to the corresponding data, that is, the attribution relationship of each data relative to the evaluation index, and a more accurate safety assessment is performed based on the corresponding limit value.

[0033] The pre-processed data set includes the degree of membership of the operation data, the equipment data and the status data with the safety evaluation level.

[0034] Wherein, the step 22 further includes: Step 221, acquiring the real-time data, and obtaining a corresponding mapping relationship between the real-time data and the operation data, the equipment data, and the status data of the crane; Step 222, based on the real-time data and the corresponding mapping relationship, using a fuzzy evaluation matrix to reflect the fuzzy relationship between the operation data, the device data and the status data; Step 223, based on the real-time data and the corresponding mapping relationship, using a hierarchical analysis algorithm, calculate the corresponding weights of the operation data, the device data and the status data.

[0035] The fuzzy evaluation matrix is ​​used to construct an evaluation index system and weight distribution, handle the uncertainty and fuzziness between evaluation indicators, and then quantitatively describe the safety status of the crane. Specifically, the fuzzy matrix is ​​constructed: , The fuzzy matrix is ​​used to calculate the product of each row of elements in the judgment matrix, calculate the cubic root of the product, and normalize the square root.

[0036] Based on the operation data, the device data and the status data, a hierarchy analysis algorithm is used to calculate corresponding weights of the operation data, the device data and the status data.

[0037] The hierarchical analysis algorithm is used to calculate the weights of the operation data, the equipment data and the status data, reflecting their importance in the evaluation process.

[0038] Among them, based on the fuzzy relationship, the membership of the operation data, the equipment data and the status data corresponding to different safety evaluation levels is obtained through expert scoring; the different safety evaluation levels refer to the results of the product of the corresponding weights determined by the fuzzy evaluation matrix and the hierarchical analysis algorithm, with 0.6, 0.7, 0.8 and 0.9 being taken as the limit values, and the closer to 1, the safer the system.

[0039] Step 31, inputting the preprocessed data set into the DNN deep neural network; Step 32, based on the pre-processed data, using the activation function to perform calculations to predict the wind safety level score; Step 33: output a windproof safety assessment result based on the windproof safety grade score.

[0040] The preprocessed data set is formed by collecting multiple groups of operation data, equipment data, and status data, and is expressed in the form of a two-dimensional matrix, wherein each column represents a data category and each row is a group of corresponding data.

[0041] The activation function is a custom nonlinear function, which can better simulate the nonlinear relationship between wind load and crane structure. This relationship is often beyond the processing range of traditional activation functions. By designing a custom activation function and increasing the curvature of the function at the extreme value, the adaptability of the model to complex dynamic environments can be improved, thereby improving the accuracy of wind safety level prediction.

[0042] Based on the windproof safety grade score, a windproof safety assessment result is output.

[0043] The safety level score is a score of 0-100 output by the DNN deep neural network. The wind safety is finally evaluated according to the rules of less than 60 (very poor), 60-70 (poor), 70-80 (average), 80-90 (good), and greater than 90 (very good). According to the safety level score, the corresponding wind safety assessment results are output in a total of 5 types, including: no risk, low risk (no manual intervention or confirmation required), high risk (manual attention or prevention of overturning required), warning risk (immediate measures required), and faulty (overturned).

[0044] Wherein, step 32 further includes: Step 321, based on the preprocessed data, using a cross-validation method, determining the input weight of the activation function to obtain a current activation function; Step 322: Based on the current activation function, using the DNN deep neural network, extract features related to wind safety and output a wind safety grade score.

[0045] Wherein, the activation function is the Sigmoid function: , Preferably, during the model training process, the input weights of the Sigmoid function can be adjusted for multiple rounds of parameters based on the preprocessed data and the cross-validation method, and finally the input weights that can ensure the safety of the crane in the most dangerous situation can be determined.

[0046] In one embodiment, the input weight is 1.6, that is, the Sigmoid function is: .

[0047] Among them, the DNN deep neural network automatically adjusts the weights of parameters such as the convolution kernel through the back propagation algorithm and the gradient descent algorithm to minimize the loss function and extract features related to wind safety. The introduction of the cross-validation method reduces the risk of overfitting during the training of the deep neural network. Through the hierarchical representation of learning data through a multi-layer structure, each layer may capture features of different levels of data, thereby constructing a feature representation space for the operation data, the equipment data, and the status data, and then extracting features related to wind safety.

[0048] Among them, according to the characteristics related to wind safety, the wind safety level score is output.

[0049] The present invention also provides a crane anti-rollover detection device, characterized in that the system includes: a signal acquisition device, used to obtain the operation data, the equipment data and the status data of the crane; a data analysis device, used to calculate the operation data, the equipment data and the status data using a fuzzy hierarchical algorithm and a DNN deep neural network, and output the windproof safety assessment result under the current state of the crane; an output device, used to obtain a warning signal, safety level or control instruction based on the windproof safety assessment result.

[0050] The signal acquisition device is used to obtain the operation data, equipment data and status data of the crane. The operation data includes illegal operations, operational errors, and professional skills of personnel, which are usually monitored and collected through the terminal equipment of the crane; the equipment data includes equipment structural strength, lifting weight, boom angle, and crane posture (yaw angle, pitch angle and roll angle), acceleration during movement, angular velocity, and height data, which are usually collected through a high-precision sensor array. Specifically, the high-precision sensor array includes a waterproof protective cover, a ten-axis attitude angle sensor, a wind detection device, and an air pressure sensor.

[0051] Wherein, the data analysis device is used to calculate the operation data, the equipment data and the status data using a fuzzy hierarchical algorithm and a DNN deep neural network, and output the wind safety assessment result under the current state of the crane. The data analysis device sends the collected data to the edge computing device in real time through the serial port for analysis and calculation, so as to realize the assessment and analysis of the safety status of the crane. Through multiple rounds of training of the model, based on the input data and cross-validation, the input weights of the hidden layer activation function are further optimized, and the hidden layer is used to extract features related to wind safety. Finally, the wind safety level score is determined based on the features related to wind safety extracted by the DNN deep neural network.

[0052] Wherein, the output device is used to obtain the warning signal, safety level or control instruction according to the wind safety assessment result. The output device is a terminal receiving module responsible for receiving the warning signal or control instruction transmitted by the signal analysis module, and presenting it to the operator in an intuitive and easy-to-understand manner. In the practice of wind safety assessment of cranes, the output device can be connected to the control system of the port crane to realize automated control operation. This intelligent control system can automatically adjust the operating status of the crane according to the real-time monitored wind speed and wind direction data to ensure the safe and stable operation of the crane in a strong wind environment. Through this automated control, delays and errors in human operations can be reduced, and the response speed can be improved, so that measures can be taken quickly in the most dangerous situation to maximize the safety of the crane.

[0053] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any one of the crane anti-rollover detection methods provided by the above method.

[0054] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the crane anti-rollover detection methods provided by the above-mentioned methods when executing the computer program.

[0055] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0056] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0057] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting an anti-tilt of a crane, characterized in that: include: Step 1, obtaining real-time data of the crane; Step 2, based on the real-time data and its mapping relationship with the operation data, equipment data and status data, the safety evaluation level of the crane is calculated through a fuzzy hierarchy algorithm.

2. The method according to claim 1, characterized in that Also includes step 3: Based on the operation data, the equipment data and the status data, as well as the corresponding safety assessment level, the DNN deep neural network is used to predict the crane's safety wind protection level.

3. The method according to claim 1, characterized in that In step 1, the operation data are illegal operation data or operation error data; the equipment data are structural parameters and operation status; and the status data are wind environment data and crane position height.

4. The method according to claim 1, characterized in that: The step 2 also includes: Step 21, acquiring the real-time data of the crane, and obtaining a corresponding mapping relationship between the real-time data and the operation data, the equipment data, and the status data of the crane; Step 22, based on the real-time data and the corresponding mapping, using a fuzzy hierarchy algorithm, calculating the fuzzy relationship and corresponding weights between the operation data, the device data and the status data; Step 23, based on the fuzzy relationship and the corresponding weight, obtain the membership of the operation data, the device data and the status data corresponding to different safety evaluation levels, and obtain a preprocessed data set consisting of the membership of the operation data, the device data and the status data to the safety evaluation level.

5. The method according to claim 4, characterized in that The step 22 further comprises: Step 221, acquiring the real-time data, and obtaining a corresponding mapping relationship between the real-time data and the operation data, the equipment data, and the status data of the crane; Step 222, based on the real-time data and the corresponding mapping relationship, using a fuzzy evaluation matrix to reflect the fuzzy relationship between the operation data, the device data and the status data; Step 223, based on the real-time data and the corresponding mapping relationship, using a hierarchical analysis algorithm, calculate the corresponding weights of the operation data, the device data and the status data.

6. The method according to claim 4, characterized in that The step 3 also includes: Step 31, inputting the preprocessed data set into the DNN deep neural network; Step 32, based on the pre-processed data, using the activation function to perform calculations to predict the wind safety level score; Step 33, outputting a windproof safety assessment result according to the windproof safety grade score.

7. The method according to claim 6, characterized in that Step 32 also includes: Step 321, based on the preprocessed data, using a cross-validation method, determining the input weight of the activation function to obtain a current activation function; Step 322: Based on the current activation function, use the DNN deep neural network to extract features related to wind safety and output a wind safety level score.

8. A crane anti-tilt detection device, characterized in that: The system comprises: A signal acquisition device, used to acquire the operation data, the equipment data and the status data of the crane; A data analysis device, used to process the operation data, the equipment data and the status data by using a fuzzy hierarchical algorithm and a DNN deep neural network, and output the wind safety assessment result of the crane in the current state; The output device is used to obtain a warning signal, a safety level or a control instruction according to the wind safety assessment result.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the anti-rollover detection method for a crane as claimed in any one of claims 1 to 7 is implemented.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the anti-rollover detection method for the crane according to any one of claims 1 to 7 is implemented.

Images