Research On Dynamic Assessment Of Collision Risk Of Workers And Mobile Machinery On Construction Site

The construction industry is one of the most dangerous industries in the world with frequent accidents.Among all the accidents,the collision accident is considered to be one of the four most fatal construction injuries.In order to reduce the collision risk and improve the efficiency of safety management,this paper constructs a dynamic evaluation framework of collision risk based on computer vision and fuzzy comprehensive evaluation model,which aims to dynamically monitor the unsafe state of construction site,evaluate the safety risk level of workers and construction machinery in real time,explore potential risks,and achieve timely warning,risk reduction and event reduction purpose.In this paper,the research background and achievements are combed through the literature research method,and the current research deficiencies are summarized.The research problem of this paper is how to make up for the static attributes of traditional risk assessment methods and the limitations of existing dynamic risk assessment methods,so as to improve the monitoring efficiency of collision risk and achieve the purpose of automatic risk assessment process.In this paper,computer vision technology and security risk assessment model are combined.Firstly,a dynamic risk assessment framework based on computer vision is built.Computer vision technology is used as the flow channel of security information,and the integrated automatic processing process from data collection to risk assessment and early warning is realized.Secondly,by identifying the collision risk assessment indicators of workers and mobile machinery,the risk assessment mechanism and early warning mechanism are designed,and a fuzzy comprehensive assessment model based on literature research and expert questionnaire method is constructed.Finally,the feasibility of the technology and the validity of the model are verified by experimental research.The main research results of this paper are divided into three aspects.First,this paper introduces congestion as an important evaluation index of collision risk,integrates two indexes of approach degree and crowding degree,and improves the index system of collision risk.Secondly,a risk assessment mechanism is designed to comprehensively consider the interaction of all entities around the monitoring object,distinguish the working attributes of workers and construction machinery,and improve the reliability of risk assessment.Thirdly,the combination of computer vision technology and security risk assessment model realizes the automation of risk assessment,which is helpful to improve the efficiency of security management and reduce the occurrence of collision accidents.The main conclusions of this paper are as follows: first,noisy and crowded environment will affect the safety of workers,so the impact risk of workers should not be ignored noise interference and the interaction of surrounding environment.Secondly,the comprehensive risk assessment framework based on computer vision constructed in this paper can automatically and effectively identify the collision risks of workers and construction machinery,increase the controllability of safety risks,improve management efficiency,and have great potential in assisting safety decision-making and reducing the possibility of collision accidents.Thirdly,combined with the experimental results,the construction unit shall reasonably arrange the entry sequence of people,materials and machines,standardize the stacking of materials,fine site layout,and standardize the orderly construction environment,which will contribute to the safety management of the construction site and the successful application of information technology.Fourthly,computer vision technology can realize the dynamic monitoring of safety risk,and it is one of the effective tools to assist the safety management of construction site.At present,its limitation mainly lies in the problem of occlusion and light.In the future research,it can be combined with other technical means to ensure the reliability of risk assessment.