Research And Application Of Intelligent Crane Safety Monitoring System

Tower cranes is a common tool of materials-transport, it has been more widely used for their ability to accelerate the construction schedule, reduce labor intensity, save manual labour, reduce project cost and other reasons, But it is also accidents-prone agency, significant personal casualties accidents occur in the process of installation and removal of the standard section seriously endanger the life and property of people. Therefore, it is extremely necessary to develop a tower crane intelligent safety monitoring system to make the tower crane fault-tolerant to avoid serious accidents because of human errors. The main work is completed as follows:Firstly, the overall plan of crane intelligent security monitoring system's basic monitoring function is determined on the basis of research and analysis on the operating characteristics of the crane:PLC (programmable logic controller) is used to complete the tasks of controlling crane rise and fall. Crane running parameters are collectted by various types of sensors, the collected-data are transmitted to the monitor core PLC, the processed-data are sent to the touch screen, and the data are compared with the thresholds, if the data over the thresholds, PLC will run the sound and light alarm to the complete crane safety monitoring task successfully.Secondly, the detailed detection plan of crane operating parameters and status is identified on the basis of research and analysis on the structure and working principle of each hardware portion of crane. Then, the selection of hardware of crane intelligent security monitoring system such as:controller PLC, switching power supply, analog expansion modules, relays, fuses, etc. Production of controlling box, I/O address assignment, line-connection work are completed. In addition, the automatic control program, the data-monitoring progratru alarm procedures> jacking animation program for monitoring system are written and configuration work of HMI is completed.In addition, in order to enhance the intelligent level of the monitoring system, the idea that use the binocular vision measurement technology to achieve crane anti-collision is proposed in this paper. The principle of vision binocular 3D measurement is figured out, a mathematical model for the measurement system is set up, all aspects of binocular vision measurement process are determined. However, only the deep study of the most important step of measurement system?mage feature extraction and matching is made in this paper due to my energy, time constraints and other objective factors. For the low real-time of SIFT algorithm, an improvement scheme that make high-dimensional feature description vector matching after dimension reduction by PCA algorithm is proposed, simulation experiment by MATLAB is made and the desired matching results is obtained.Finally, the monitoring system also add comprehensive alarm function for much of parameter besides single parameter alarm function. Idler wheel pressure of sleeve frame, top lift, tilt angle, boom deflection angle, wind speed are seen as input variables, the operating parameters of the crane with known security status are as samples. The SVM classifier make supervised-learning on the samples data, the effect of different kernel functions and parameters on SVM classifier is discussed, classifier parameters are selected after optimization by mixed method of grid search and cross-validation to establish perfect classification model, the model is used to test the running parameters of crane with unknown security status to judge the secure state comprehensively. Finally, the study and application of crane intelligent safety monitoring system is completed through the above work.