An equipment safety monitoring system for highway construction based on the Internet of Things

Abstract

The present invention discloses a safety monitoring system for highway construction equipment based on the Internet of Things, comprising: a device status data acquisition module for obtaining device location data, device operation status data, and device interaction data in real time through a distributed Internet of Things sensor network; a central data processing unit for receiving and storing the highway construction equipment operation status data set in real time; an ant colony heuristic path search module for generating a global risk assessment path for abnormal device operation status; a genetic tabu search module for dynamically generating local risk optimization results; a risk assessment information fusion module for fusing the global risk assessment path and the local risk optimization results; and a safety warning module for generating device abnormal warning signals in real time according to the comprehensive risk assessment information. The present invention greatly reduces the risk of safety accidents caused by the failure to detect device abnormalities in a timely manner.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment monitoring, and particularly to a safety monitoring system for highway construction equipment based on the Internet of Things. Background Art

[0002] With the rapid development of Internet of Things technology, intelligent monitoring systems have been gradually introduced in various industries to improve the operation safety and management efficiency of equipment. In the field of highway construction, a large number of large-scale mechanical equipment are involved in the construction process, and the operation status of these equipment directly affects the project progress and construction safety. However, due to the complex construction site environment and various types of equipment, the existing safety monitoring methods are difficult to meet the needs of modern highway construction. The existing monitoring means mainly rely on manual inspections and traditional sensor monitoring, which have certain technical defects and limitations.

[0003] At present, the safety management of highway construction equipment mainly relies on the following methods: on the one hand, construction workers regularly inspect the equipment, and judge the operation status of the equipment through manual observation, recording and experience. However, this method relies on subjective judgment of personnel, has low efficiency, and there are problems of omission and delay; on the other hand, some construction sites have installed monitoring systems based on single sensors, such as GPS positioning devices, vibration sensors or temperature sensors, to obtain partial operation status data of the equipment. However, single monitoring means are difficult to achieve all-round and real-time monitoring of the equipment. Especially in the case of multi-equipment collaborative operation, drastic changes in equipment status or sudden abnormal events, the reliability and accuracy of the existing monitoring systems are relatively low.

[0004] In recent years, with the improvement of data processing capabilities, some highway construction enterprises have begun to introduce remote monitoring systems based on wireless communication, upload equipment operation data to the management platform, and combine simple data analysis algorithms to judge abnormalities. However, such systems still have the following defects in practical applications: First, the data collection means are relatively single and difficult to comprehensively reflect the real-time status of the equipment, resulting in a lag in risk warning; Second, the risk assessment mechanism of the existing system is relatively rough, usually based on threshold judgment, and an alarm is triggered when the equipment parameters exceed the preset range, but the mutual influence between equipment and complex abnormal patterns are not fully considered, resulting in a high false alarm rate. In addition, the existing monitoring systems mainly use fixed rules for abnormal identification, lack the ability of adaptive optimization, and are difficult to dynamically adjust the monitoring strategy, resulting in difficulty in effectively coping with changes in the construction environment.

[0005] In summary, the existing safety monitoring technologies for highway construction equipment have obvious deficiencies in aspects such as the comprehensiveness of data collection, the accuracy of risk assessment, and the real-time nature of abnormal early warning, and cannot meet the requirements of modern highway construction for efficient and intelligent safety monitoring. Therefore, there is an urgent need for an intelligent monitoring method based on the Internet of Things to improve the real-time perception ability of the equipment operation status, and to enhance construction safety and management efficiency by optimizing risk assessment and early warning strategies. Summary of the Invention

[0006] An object of the present invention is to propose a safety monitoring system for highway construction equipment based on the Internet of Things, and the present invention greatly reduces the risk of safety accidents caused by the failure to timely detect equipment abnormalities.

[0007] A safety monitoring system for highway construction equipment based on the Internet of Things according to an embodiment of the present invention includes the following modules:

[0008] The equipment status data collection module is used to obtain equipment location data, equipment operation status data, and equipment interaction data in real time through a distributed Internet of Things sensor network, construct a highway construction equipment operation status data set, and transmit it to the central data processing unit;

[0009] The central data processing unit is used to receive and store the highway construction equipment operation status data set in real time;

[0010] The ant colony heuristic path search module performs ant colony heuristic path search based on a dynamic node abnormal risk level function, a dynamic path cost function, and a path heuristic function, and dynamically constructs a global risk assessment path for equipment operation status abnormalities;

[0011] The genetic tabu search module is used to further perform local refined search on the preliminary potential abnormal equipment area based on the global risk assessment path, and dynamically generate a local risk optimization result;

[0012] The risk assessment information fusion module is used to fuse the global risk assessment path and the local risk optimization result, and form comprehensive risk assessment information including risk level, risk distribution range, and abnormal status confirmation information;

[0013] The safety early warning module is used to generate equipment abnormal early warning signals in real time according to the comprehensive risk assessment information, and transmit the equipment abnormal early warning signals to the highway construction site safety management terminal to guide on-site personnel to respond to and handle the abnormal status in a timely manner.

[0014] A safety monitoring method for highway construction equipment based on the Internet of Things, which is applied to a safety monitoring system for highway construction equipment based on the Internet of Things, includes:

[0015] S1. Deploy a distributed Internet of Things sensor network at the highway construction site to collect the operating status data set of highway construction equipment in real time;

[0016] S2. Transmit the operating status data set of highway construction equipment to the central data processing unit in real time through a wireless communication network, and establish a real-time data monitoring database in the central data processing unit;

[0017] S3. Conduct preliminary processing on the operating status data set of highway construction equipment to form an initial state set of the operating data of highway construction equipment;

[0018] S4. Use the ant colony heuristic path search algorithm to globally search the initial state set of the operating data of highway construction equipment, predict the risk areas of highway construction equipment, and generate a global risk assessment path;

[0019] S5. Take the global risk assessment path as an input parameter, apply the genetic tabu search algorithm to conduct local detailed search and analysis on the potential abnormal areas determined in step S4, and generate a local risk optimization result;

[0020] S6. Integrate the global risk assessment path and the local risk optimization result to form comprehensive risk assessment information, and the comprehensive risk assessment information clarifies the risk level, risk distribution range, and abnormal state confirmation information of the operating status of highway construction equipment;

[0021] S7. According to the comprehensive risk assessment information, generate and release equipment abnormal warning signals in real time through a preset safety warning mechanism, and transmit the equipment abnormal warning signals to the safety management terminal at the highway construction site, thereby realizing the safety monitoring between highway construction equipment.

[0022] Optionally, S1 includes the following steps:

[0023] S11. Divide the monitoring area at the highway construction site, defined as the monitoring area set , and each monitoring area in the monitoring area set corresponds to the on-site construction scope of the i-th highway construction;

[0024] S12. Deploy multiple Internet of Things sensor nodes in each monitoring area according to the types of highway construction equipment, construction characteristics, and safety monitoring requirements to form an Internet of Things sensor network set :

[0025] ;

[0026] Among them, represents the th monitoring area, and the A sensor node, where n represents the total number of monitoring areas divided in the highway construction site, represents the th monitoring area;

[0027] S13. Through the sensor node Collect real-time equipment location data, equipment operation status data and equipment interaction data to form a highway construction equipment operation status data set :

[0028] ;

[0029] Among them, represents at time , the th monitoring area, the th sensor node collects highway construction equipment operation status data, is the time series of real-time data collected by the monitoring system;

[0030] S14. Transmit the real-time collected highway construction equipment operation status data set to the central data processing unit through the wireless communication network.

[0031] Optionally, the S3 includes the following steps:

[0032] S31. Clean the real-time collected highway construction equipment operation status data set , remove the abnormal data and invalid data generated due to transmission failures, sensor anomalies or environmental interference in the data set, and form a cleaned highway construction equipment operation status data set;

[0033] S32. Perform data normalization processing on the cleaned highway construction equipment operation status data set to eliminate the analysis errors caused by dimensional differences between data collected by different sensor nodes, and generate a normalized highway construction equipment operation status data set ;

[0034] S33. Extract features from the normalized highway construction equipment operation status data set , select the state feature vector according to the highway construction equipment type and construction safety requirements, and construct an initial state set of highway construction equipment operation data .

[0035] Optionally, the S4 includes the following steps:

[0036] S41. Each initial state feature vector in the initial state set of highway construction equipment operation data Defined as the path search node in the ant colony heuristic path search algorithm, forming a set of search nodes :

[0037] ;

[0038] Among them, represents the path search node corresponding to the th sensor node in the th monitoring area, represents the initial state feature vector in the set of initial states of the operating data of the highway construction equipment corresponding to the th moment, the th sensor node in the th monitoring area;

[0039] S42. For each path search node , combined with the difference degree of equipment operating state parameters, the change trend of equipment state and the equipment anomaly history record, define the node anomaly risk level function:

[0040] ;

[0041] Among them, represents the risk level in the global search stage of the path search node , , and are weight factors, represents the difference degree of the initial state feature vector relative to the reference state feature vector , represents the change trend of the initial state feature vector over time, is the cumulative function of the anomaly history record of the path search node;

[0042] S43. Based on the risk level in the global search stage, construct a dynamic adaptive path cost function between path search nodes, which is used to measure the risk transfer cost between any two path search nodes and :

[0043] ;

[0044] Among them, is the risk transfer cost between the path search node and the path search node , which is used to describe the transfer degree of the equipment state anomaly risk between any two path search nodes, and respectively represent the path search nodes With the path search node The global search phase risk level at time t 、 respectively represent adjacent path search nodes is the Euclidean distance of the feature difference between path search nodes is the sensitivity adjustment index, and the dynamic adaptive path cost function reflects the transmission probability of equipment operation abnormal risk among different equipment nodes. The greater the sum of risk levels, the higher the possibility of abnormal propagation

[0045] S44. Construct a path search heuristic function based on the node abnormal risk level Perform dynamic prediction of the risk area. The path search heuristic function is used to guide the path search

[0046] ;

[0047] Among them is the sensitivity index of the feature difference, which is used to dynamically adjust the influence of the feature difference between nodes. When the node abnormal risk level is higher than the preset value and the feature difference is higher than the preset value, the value of the path search heuristic function becomes larger, guiding the ant colony heuristic path search algorithm to preferentially explore the risk area of the potential abnormal equipment operation state

[0048] S45. Synthesize the pheromone concentration and the path search heuristic function , and adopt an improved path selection probability formula to dynamically determine the path selection probability in the global path search , which is used to initially locate the risk area of the potential abnormal equipment operation state

[0049] ;

[0050] Among them and are time-varying weight factors that change with the dynamic data of the highway construction site is a dynamic adaptation factor based on the fluctuation of the recent highway construction equipment operation data represents the path search node 's set of adjacent nodes, and r represents all possible path targets

[0051] S46. Dynamically adapt to the construction environment change according to the path selection probability, and optimize and form a global risk assessment path in real time .

[0052] Optionally, the S5 includes the following steps

[0053] S51. The global risk assessment path The determined potential abnormal area is used as the initial input of the genetic tabu search algorithm to form a local search population set , each individual in the population set corresponds to a set of device nodes to be optimized within the potential abnormal area:

[0054] ;

[0055] Among them, represents the th individual in the local search population set , represents the scale of the local search population;

[0056] S52. For the individuals in the local search population set , define the local abnormal risk fitness function:

[0057] ;

[0058] Among them, , and are the weight coefficients of the fitness function, represents the cumulative value of the historical risk assessment of the population individual , is the local abnormal risk fitness level;

[0059] S53. According to the local abnormal risk fitness level , perform genetic tabu search algorithm operations on the current local search population set to generate the next generation of population individuals ;

[0060] S54. Construct a dynamic tabu list of the genetic tabu search algorithm according to the local abnormal risk fitness level , and the dynamic tabu list is used for tabu constraints in high fitness areas:

[0061] ;

[0062] Among them, is the dynamic risk threshold:

[0063] ;

[0064] Among them, is the reference risk coefficient, is the risk amplification factor, is the preset risk reference value;

[0065] After meeting the preset termination condition, select the individual with the highest local anomaly risk fitness from the local search population set outside the dynamic taboo list as the local risk optimization result according to the following formula :

[0066] ;

[0067] where represents the diversity adjustment factor of the individuals in the local search population set in the to evaluate the distribution difference degree of the state feature vectors of the device nodes inside the individual:

[0068] ;

[0069] where represents the number of device nodes included in the individual , is the central vector of all path search nodes inside the individual , represents a single device operating state node inside the local search population individual is the selected diversity adjustment coefficient, used to balance the trade-off between individual risk and internal node diversity represents the exponential function

[0070] Optionally, the S6 includes the following steps:

[0071] S61. Perform feature fusion on the global risk assessment path and the local risk optimization result to construct comprehensive risk assessment information, and define the fusion function as:

[0072] ;

[0073] where , and are fusion weight coefficients represents the overlapping confidence level that the path search node is identified as a high-risk node in both global search and local search is the comprehensive risk assessment information

[0074] S62. Define the comprehensive risk level evaluation index based on the comprehensive risk assessment information , and divide the device nodes into low-risk, medium-risk, and high-risk levels according to the preset risk threshold:

[0075] ;

[0076] Among them, and are the lower threshold and the high-risk critical threshold for distinguishing the abnormal risk level of device nodes respectively;

[0077] S63. According to the comprehensive risk level evaluation index rank the risk levels of device nodes to determine the high-risk node set :

[0078] ;

[0079] Among them, is the dynamic threshold parameter of the comprehensive risk level, which is jointly determined by the statistical distribution characteristics of the historical device abnormal risk levels at the highway construction site and the real-time risk level fluctuation situation;

[0080] S63. Based on the high-risk node set construct the comprehensive risk area set of highway construction equipment , and locate and dynamically track the abnormal device operation state area within the construction site:

[0081] ;

[0082] Among them, represents the comprehensive risk area set of highway construction equipment determined by the comprehensive risk assessment, represents the i-th monitoring area, that is, the i-th specific construction scope divided within the highway construction site. The monitoring area corresponds to the actual geographical monitoring area where the abnormal state node of the highway construction equipment is located, and is used to clarify the specific spatial position of the abnormal device when the device is abnormal, so as to realize the positioning and dynamic tracking of the abnormal device operation state.

[0083] The beneficial effects of the present invention are:

[0084] The present invention adopts the ant colony heuristic path search algorithm. In the monitoring of the operation state of highway construction equipment, by constructing a node abnormal risk level function, a dynamic path cost function and a path heuristic function to establish a global risk assessment path, it can dynamically adjust the path search direction according to the change of the device operation state, making the identification of abnormal devices more accurate. In addition, the path search process combines multi-dimensional factors such as historical abnormal records and state change trends, avoiding the false alarms and missed alarms caused by the traditional method due to fixed rule judgment, and greatly reducing the risk of safety accidents caused by the failure to detect device abnormalities in time.

[0085] The present invention further performs local optimization search on high-risk areas on the basis of the global risk assessment path by introducing a genetic tabu search algorithm, deeply analyzes the devices in the potential abnormal areas by using a local abnormal risk fitness function, and combines a dynamic tabu list to optimize the search process to avoid local optimal traps, improve the accurate positioning ability of abnormal areas, reduce redundant alarms, and improve the accuracy of construction site safety management.

[0086] The present invention adopts a multi-level information fusion mechanism to fuse the global risk assessment path and the local risk optimization results, constructs comprehensive risk assessment information, and dynamically generates safety warning signals based on risk levels, risk distribution ranges, and abnormal status confirmation information. By introducing an adaptive risk weight adjustment mechanism, it ensures that the warning system can dynamically adjust the risk assessment criteria according to the changes in real-time data, reduce the occurrence of false alarms and missed alarms, and effectively improve the ability to prevent construction site safety incidents in advance. Brief Description of the Drawings

[0087] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention, but do not constitute a limitation to the present invention. In the drawings:

[0088] Figure 1 It is a flowchart of a device safety monitoring system for highway construction based on the Internet of Things proposed by the present invention. Detailed Embodiments

[0089] Now, the present invention will be further described in detail with reference to the drawings. These drawings are all simplified schematic diagrams, only illustrating the basic structure of the present invention in a schematic manner, so they only show the components related to the present invention.

[0090] Refer to Figure 1 , a device safety monitoring system for highway construction based on the Internet of Things, includes the following modules:

[0091] A device status data acquisition module, which is used to obtain device location data, device operation status data, and device interaction data in real time through a distributed Internet of Things sensor network, construct a highway construction equipment operation status data set, and transmit it to the central data processing unit;

[0092] The central data processing unit is used to receive and store the highway construction equipment operation status data set in real time;

[0093] An ant colony heuristic path search module, which performs ant colony heuristic path search based on a dynamic node abnormal risk level function, a dynamic path cost function, and a path heuristic function, and dynamically constructs a global risk assessment path for abnormal device operation status;

[0094] A genetic tabu search module, which is used to further perform local refined search on the preliminary potential abnormal device area based on the global risk assessment path and dynamically generate local risk optimization results;

[0095] A risk assessment information fusion module, which is used to fuse the global risk assessment path and the local risk optimization results and form comprehensive risk assessment information including risk level, risk distribution range and abnormal state confirmation information;

[0096] A safety warning module, which is used to generate device abnormal warning signals in real time according to the comprehensive risk assessment information and transmit the device abnormal warning signals to the safety management terminal of the highway construction site to guide on-site personnel to respond to and handle the abnormal state in a timely manner.

[0097] A method for monitoring the safety of highway construction equipment based on the Internet of Things, which is applied to a safety monitoring system for highway construction equipment based on the Internet of Things, including:

[0098] S1. Deploy a distributed Internet of Things sensor network at the highway construction site to collect the operation status data set of highway construction equipment in real time;

[0099] S2. Transmit the operation status data set of highway construction equipment to the central data processing unit in real time through a wireless communication network, and establish a real-time data monitoring database in the central data processing unit. The data storage structure includes time stamp, device identifier, device status parameters and historical abnormal records, and perform data integrity verification and redundancy removal on the received data as the data basis for global risk assessment and local optimization analysis;

[0100] S3. Perform preliminary processing on the operation status data set of highway construction equipment to form an initial state set of the operation data of highway construction equipment;

[0101] S4. Use the ant colony heuristic path search algorithm to perform global search on the initial state set of the operation data of highway construction equipment, and predict the risk area of highway construction equipment to generate a global risk assessment path;

[0102] S5. Use the genetic tabu search algorithm to perform local detailed search and analysis on the potential abnormal area determined in step S4 with the global risk assessment path as the input parameter, and generate local risk optimization results;

[0103] S6. Fuse the global risk assessment path and the local risk optimization results to form comprehensive risk assessment information, and the comprehensive risk assessment information clarifies the risk level, risk distribution range and abnormal state confirmation information of the operation status of highway construction equipment;

[0104] S7. Based on the comprehensive risk assessment information, construct an abnormal warning signal for the equipment. When the comprehensive risk level exceeds the set threshold, trigger the equipment abnormal warning mechanism, transmit the equipment abnormal warning signal to the safety management terminal at the highway construction site. The safety management terminal provides the equipment operation abnormal report and response suggestions to the management personnel according to the equipment abnormal warning signal, combined with the equipment operation status at the construction site, realizes the dynamic safety monitoring of highway construction equipment, and at the same time stores the warning feedback information into the risk assessment database for subsequent optimization of the equipment monitoring strategy, so as to realize the safety monitoring between highway construction equipment.

[0105] In this embodiment, S1 includes the following steps:

[0106] S11. Divide the monitoring areas at the highway construction site and define them as the monitoring area set , and each monitoring area in the monitoring area set corresponds to the on-site construction scope of the i-th highway construction;

[0107] S12. Deploy multiple Internet of Things sensor nodes in each monitoring area according to the highway construction equipment type, construction characteristics and safety monitoring requirements to form an Internet of Things sensor network set :

[0108] ;

[0109] Among them, represents the th sensor node in the th monitoring area, n represents the total number of monitoring areas divided at the highway construction site, represents the number of sensor nodes in the th monitoring area;

[0110] S13. Real-time collect the equipment position data, equipment operation status data and equipment interaction data through the sensor nodes to form a highway construction equipment operation status data set :

[0111] ;

[0112] Among them, represents the highway construction equipment operation status data collected by the th sensor node in the th monitoring area at time , is the time series of the data collected by the monitoring system in real time;

[0113] S14. The highway construction equipment operation status data set collected in real time Transmitted to the central data processing unit through a wireless communication network.

[0114] In this embodiment, by deploying a distributed Internet of Things sensor network at the highway construction site, accurate collection and real-time transmission of equipment status data are achieved. By reasonably arranging the monitoring area and the sensor network, all-round monitoring of the operating status of all key equipment at the construction site is ensured. Using wireless communication technology, the equipment operating status data can be transmitted to the central data processing unit with low latency, providing real-time support for subsequent analysis. By removing invalid and abnormal data and performing standardization and normalization processing, the reliability of data analysis is improved.

[0115] In this embodiment, S3 includes the following steps:

[0116] S31. For the real-time collected dataset of the operating status of highway construction equipment Perform data cleaning to remove abnormal data and invalid data in the dataset caused by transmission failures, sensor anomalies, or environmental interference, and form a cleaned dataset of the operating status of highway construction equipment;

[0117] S32. Perform data normalization processing on the cleaned dataset of the operating status of highway construction equipment to eliminate the analysis errors caused by the dimensionality differences between the data collected by different sensor nodes, and generate a normalized dataset of the operating status of highway construction equipment ;

[0118] S33. Perform feature extraction on the normalized dataset of the operating status of highway construction equipment Select the state feature vectors according to the highway construction equipment type and construction safety requirements to construct an initial state set of the operating data of highway construction equipment .

[0119] In this embodiment, by initially processing the equipment operating status dataset to form an initial state set of the operating data of highway construction equipment, the accuracy of data analysis is improved. Through data cleaning, standardization, and feature extraction, the usability of equipment data is enhanced, avoiding the influence of noise interference on the analysis results. Constructing feature vectors enables the equipment status information to be stored and processed in a structured manner, performing dimensionality reduction processing on the data, reducing redundant information, and ensuring the real-time response ability of the equipment monitoring system.

[0120] In this embodiment, S4 includes the following steps:

[0121] S41. Define each initial state feature vector in the initial state set of the operating data of highway construction equipment as a path search node in the ant colony heuristic path search algorithm to form a search node set :

[0122] ;

[0123] wherein, denotes the path search node corresponding to the th sensor node in the th monitoring area, denotes the initial state feature vector in the set of initial operating data of the highway construction equipment corresponding to the th sensor node in the th monitoring area at time ;

[0124] S42. For each path search node , define a node abnormal risk level function by combining the difference degree of equipment operating state parameters, the change trend of equipment state, and the equipment abnormal history record:

[0125] ;

[0126] wherein, denotes the risk level of the global search stage of the path search node , , and are weight factors, denotes the difference degree of the initial state feature vector relative to the reference state feature vector , denotes the change trend of the initial state feature vector over time, is the cumulative function of the abnormal history record of the path search node;

[0127] S43. Construct a dynamic adaptive path cost function between path search nodes based on the risk level of the global search stage, which is used to measure the risk transfer cost between any two path search nodes and :

[0128] ;

[0129] wherein, is the risk transfer cost between the path search node and the path search node , which is used to describe the transfer degree of the equipment state abnormal risk between any two path search nodes, and respectively denote the risk levels of the global search stage of the path search node and the path search node at time t, , respectively represent adjacent path search nodes, is the Euclidean distance of the feature difference between path search nodes, is the sensitivity adjustment index. The dynamic adaptive path cost function reflects the transmission probability of equipment operation abnormal risks between different equipment nodes. The greater the sum of risk levels, the higher the possibility of abnormal propagation;

[0130] S44. Construct a path search heuristic function based on the node abnormal risk level Perform dynamic prediction of the risk area. The path search heuristic function is used to guide path search:

[0131] ;

[0132] Among them, is the sensitivity index of feature difference, which is used to dynamically adjust the influence of feature difference between nodes. When the node abnormal risk level is higher than the preset value and the feature difference is higher than the preset value, the value of the path search heuristic function becomes larger, guiding the ant colony heuristic path search algorithm to preferentially explore the risk area of the potential abnormal equipment operation state;

[0133] S45. Synthesize the pheromone concentration and the path search heuristic function , and use the improved path selection probability formula to dynamically determine the path selection probability in the global path search , which is used to initially locate the risk area of the potential abnormal equipment operation state:

[0134] ;

[0135] Among them, and are time-varying weight factors that change with the dynamic data of the highway construction site, is a dynamic adaptation factor based on the fluctuation of recent highway construction equipment operation data, represents the adjacent node set of the path search node , and r represents all possible path targets;

[0136] S46. Dynamically adapt to the change of the construction environment according to the path selection probability, and optimize and form a global risk assessment path in real time .

[0137] This embodiment uses an improved ant colony heuristic path search algorithm to globally search the device operation status data, achieving accurate prediction of potential abnormal risk areas. By constructing a dynamic path cost and pheromone update mechanism, the global search ability of the ant colony search algorithm in complex environments is improved, avoiding falling into local optima. Combining the node abnormal risk level and dynamic path selection probability, the algorithm can be adaptively adjusted according to the real-time changes of the device status, improving the accuracy of abnormal detection. The pheromone propagation and reinforcement learning mechanism are used to dynamically optimize the device operation status, improving the search efficiency of the global risk assessment path.

[0138] In this embodiment, S5 includes the following steps:

[0139] S51. Use the potential abnormal area determined by the global risk assessment path as the initial input of the genetic tabu search algorithm to form a local search population set , and each individual in the population set corresponds to a set of device nodes to be optimized within the potential abnormal area:

[0140] ;

[0141] Among them, represents the th individual in the local search population set , and represents the scale of the local search population;

[0142] S52. For the individuals in the local search population set , define a local abnormal risk fitness function:

[0143] ;

[0144] Among them, , and are the weight coefficients of the fitness function, represents the cumulative value of the historical risk assessment of the population individual , and is the local abnormal risk fitness level;

[0145] S53. According to the local abnormal risk fitness level perform genetic tabu search algorithm operations on the current local search population set to generate the next generation of population individuals ;

[0146] S54. According to the local abnormal risk fitness level Constructing a dynamic taboo table for the genetic taboo search algorithm , the dynamic taboo table is used for taboo constraints in high fitness regions:

[0147] ;

[0148] Among them, is the dynamic risk threshold:

[0149] ;

[0150] Among them, is the benchmark risk coefficient, is the risk amplification factor, is the preset risk reference value;

[0151] S55. After meeting the preset termination conditions, select the individual with the highest local abnormal risk fitness from the local search population set outside the dynamic taboo table as the local risk optimization result according to the following formula :

[0152] ;

[0153] Among them, represents the diversity adjustment factor of the individuals in the local search population set in to evaluate the distribution difference degree of the state feature vectors of the device nodes inside the individual:

[0154] ;

[0155] Among them, represents the number of device nodes included in the individual in is the center vector of all path search nodes inside the individual represents a single device operation state node inside the local search population individual is the selected diversity adjustment coefficient to balance the trade-off between individual risk and internal node diversity, represents the exponential function.

[0156] In this embodiment, an improved genetic tabu search algorithm is used to conduct a local detailed search on the abnormal area determined by the global risk assessment path, making the determination of the abnormal state more accurate. Based on the global risk assessment, a fitness function is used for local risk assessment, enabling the algorithm to more accurately determine the scope and degree of the abnormal state of the device. By introducing a dynamic tabu list, the search process can effectively avoid the trap of the historical optimal solution, improving the diversity of local search. Cross-over and mutation operations are used to enable the local search population to converge to the optimal abnormal area during continuous optimization, improving the accuracy and stability of fault detection.

[0157] In this embodiment, S6 includes the following steps:

[0158] S61. Feature fusion is performed on the global risk assessment path and the local risk optimization result to construct comprehensive risk assessment information. The fusion function is defined as:

[0159] ;

[0160] Among them, , and are fusion weight coefficients, represents the overlapping confidence level that the path search node is simultaneously identified as a high-risk node in global search and local search, is the comprehensive risk assessment information;

[0161] S62. Based on the comprehensive risk assessment information a comprehensive risk level evaluation index is defined, and the device nodes are divided into low-risk, medium-risk, and high-risk levels according to a preset risk threshold:

[0162] ;

[0163] Among them, , are respectively the lower threshold and the high-risk critical threshold for distinguishing the abnormal risk levels of device nodes;

[0164] S63. According to the comprehensive risk level evaluation index the risk levels of the device nodes are sorted to determine the high-risk node set :

[0165] ;

[0166] Among them, is the dynamic threshold parameter of the comprehensive risk level, which is jointly determined by the statistical distribution characteristics of the historical equipment abnormal risk level at the highway construction site and the real-time risk level fluctuation situation;

[0167] S63. Based on the high-risk node set Construct the comprehensive risk area set of highway construction equipment , and locate and dynamically track the abnormal equipment operation state area within the construction site:

[0168] ;

[0169] Among them, represents the comprehensive risk area set of highway construction equipment determined by the comprehensive risk assessment, represents the i-th monitoring area, that is, the i-th specific construction scope divided within the highway construction site. The monitoring area corresponds to the actual geographical monitoring area where the abnormal state node of the highway construction equipment is located, and is used to clarify the specific spatial position of the abnormal equipment when the equipment is abnormal, so as to realize the positioning and dynamic tracking of the abnormal equipment operation state. In this embodiment, by integrating the global risk assessment path and the local risk optimization result, the comprehensive risk assessment information is constructed, making the equipment status monitoring more accurate and systematic. Through multi-dimensional information fusion, the results of global search and local search are comprehensively analyzed to improve the overall accuracy of risk assessment. The equipment risk status is divided into low-risk, medium-risk and high-risk levels, making the safety management more refined, which helps to formulate targeted safety measures. By continuously tracking the changes in equipment status, the comprehensive risk assessment information can be updated at any time to ensure that the risk assessment results always conform to the actual situation of the current construction environment.

[0170] Example Example

[0171] At 9:30 am on May 15, 2024, at the construction site of a provincial highway expansion project, the system monitored an excavator numbered PL-EXC2023, and the operation state data of which showed abnormal fluctuations. The sensor data showed that the engine temperature of the equipment remained at 98°C for 15 consecutive minutes, exceeding the safety threshold of the normal operating temperature of 85°C - 95°C. At the same time, the pressure of the hydraulic system fluctuated abnormally, and the pressure value suddenly dropped from 320 bar to 210 bar, and then quickly rose to 340 bar in a short time, forming an unstable pressure pulse curve. Based on the ant colony heuristic path search algorithm, the system marked the equipment as a high-risk equipment and generated a warning report at 9:32 to notify the construction site management personnel.

[0172] At 9:35, the safety monitoring platform of the construction management center received an abnormal warning signal from the device. The system automatically generated a risk assessment report on the device status, recording the start time of the anomaly, the duration of the anomaly, the anomaly characteristic curve, and the comparison results of the historical device operation status data. The analysis report showed that the hydraulic system of the device had briefly exhibited similar abnormal fluctuations in the past three days, but it did not reach the level of triggering an alarm at that time. Combining historical data analysis, the system determined that there was a risk of wear or leakage in the hydraulic system of the device. If not dealt with in a timely manner, it could lead to the failure of the device's hydraulic pump, causing the device to suddenly stop working and even affecting the coordinated operation of surrounding construction equipment.

[0173] At 9:40, according to the system warning instructions, the construction management personnel assigned maintenance personnel to go to the site to check the device.

[0174] At 9:45, after the maintenance personnel arrived at the device location, they checked the abnormal report generated by the handheld terminal system and found that there was a slight leakage at a joint part of the hydraulic pipeline, and the hydraulic oil temperature was higher than the normal level. It was initially judged to be a micro-leakage caused by seal aging.

[0175] At 9:50, the maintenance personnel used a special detection instrument to conduct a pressure test on the hydraulic system and found that the pressure of the device's hydraulic system remained unstable and was accompanied by slight bubbles, further verifying the leakage problem.

[0176] At 10:10, the maintenance personnel replaced the damaged sealing ring and replenished the hydraulic oil. Subsequently, a system pressure stability test was conducted, and it was confirmed that the pressure had returned to the normal range of 320 bar to 330 bar, and the hydraulic temperature had dropped back to 92°C.

[0177] The system conducted an automatic re-inspection at 10:20, and the device returned to the normal operation state. The risk level dropped from high risk to normal, and the construction continued.

[0178] The second case: Abnormal detection of data interaction between devices

[0179] At 14:15 on the afternoon of June 2, 2024, in the construction area of a highway bridge section, the system detected that during the operation of a roller PL-COM2025, there was an abnormality in the data interaction with another paver PL-PAV3008. The system monitoring data showed that:

[0180] The position coordinates of the PL-COM2025 device drifted abnormally 3 times within 5 minutes, and the position deviation exceeded 1.2 meters, far exceeding the allowable deviation (±0.3 meters) within the normal construction range of the device.

[0181] The operation rate data feedback by the PL-PAV3008 suddenly decreased by 35%, but no mechanical failure was detected in the device itself.

[0182] At 14:18, the genetic taboo search algorithm intervened in the analysis. After combining the global risk assessment path calculation, the system determined that there was a risk of device signal interference or sensor failure, and then triggered the device interaction abnormality alarm. The system generated an abnormality analysis report:

[0183] Abnormal start time: 2024 June 2, 14:15;

[0184] Affected equipment: PL-COM2025 (road roller), PL-PAV3008 (paving machine);

[0185] Abnormal data:

[0186] Abnormal position coordinates: deviation 1.2 meters; equipment data interaction failure rate: 45%; paver operation rate decrease: 35%;

[0187] Possible causes:

[0188] GPS module data is lost; wireless communication between devices is disturbed; sensor hardware failure;

[0189] At 14:20, after receiving the alarm, the construction safety manager assigned technicians to go to the site with portable debugging equipment to troubleshoot the problem.

[0190] At 14:35, technicians found that the construction site was close to high-voltage transmission lines, causing electromagnetic interference to the equipment's wireless communications and affecting the stability of the GPS signal.

[0191] At 14:45, the technicians adjusted the equipment's communication frequency band and recalibrated the GPS receiver. Testing found that the signal had returned to normal, the equipment data interaction had returned to normal levels, the system alarm was automatically lifted, and construction continued.

[0192] In order to further verify the advantages of the method of the present invention, we compared and analyzed the application data of this system with the traditional method. The specific data are as follows:

[0193] Index Traditional method (manual inspection) Method of the present invention (intelligent monitoring + dynamic optimization) Improvement amplitude Equipment anomaly discovery time Average 5 hours Average 20 minutes -92% False alarm rate 18.2% 4.7% -13.5% Equipment fault recovery time Average 6 hours Average 2 hours -66.7% Equipment communication anomaly recovery time Approximately 2 hours 30 minutes -75% Construction progress delay time 64 hours (90 days) 12 hours (90 days) -52 hours Equipment shutdown times (within 90 days) 17 times 4 times -76.5%

[0194] This embodiment verifies its effectiveness in the safety monitoring of highway construction equipment through actual construction cases. It collects data based on IoT sensors and combines ant colony heuristic path search for global risk assessment. It uses genetic taboo search for local optimization to more accurately identify equipment anomalies and provide timely warnings. By comparing data from real cases, it can be seen that the equipment anomaly detection time, false alarm rate, and fault recovery time are superior to existing methods, which effectively improves the safety management level of the construction site, reduces equipment downtime, and improves construction efficiency.

[0195] The present invention adopts an ant colony heuristic path search algorithm. In the monitoring of the operating status of highway construction equipment, a global risk assessment path is established by constructing a node abnormal risk level function, a dynamic path cost function, and a path heuristic function, which can dynamically adjust the path search direction according to the changes in the equipment operating status, making the identification of abnormal equipment more accurate. In addition, historical abnormal records and multi-dimensional factors such as the trend of status changes are combined in the path search process, avoiding the false alarms and missed alarms caused by the fixed rule judgment of traditional methods, and greatly reducing the risk of safety accidents caused by the failure to detect equipment abnormalities in a timely manner.

[0196] The present invention further performs a local optimization search for high-risk areas on the basis of the global risk assessment path by introducing a genetic tabu search algorithm. A local abnormal risk fitness function is used to deeply analyze the equipment in potential abnormal areas, and the search process is optimized by combining a dynamic tabu list to avoid local optimal traps, improve the accurate positioning ability of abnormal areas, and reduce redundant alarms, thereby improving the accuracy of safety management at the construction site.

[0197] The present invention adopts a multi-level information fusion mechanism to fuse the global risk assessment path and the local risk optimization results, construct comprehensive risk assessment information, and dynamically generate safety warning signals based on the risk level, risk distribution range, and abnormal status confirmation information. By introducing an adaptive risk weight adjustment mechanism, it is ensured that the warning system can dynamically adjust the risk assessment criteria according to the changes in real-time data, reduce the occurrence of false alarms and missed alarms, and effectively improve the ability to prevent safety incidents in advance at the construction site.

[0198] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and all should be covered by the protection scope of the present invention.

Claims

1. An equipment safety monitoring system for highway construction based on the Internet of Things, characterized in that, It includes the following modules: The device status data acquisition module is used to obtain device location data, device operation status data, and device interaction data in real time through a distributed Internet of Things sensor network, construct a dataset of the operation status of highway construction equipment, and transmit it to the central data processing unit; The central data processing unit is used to receive and store the dataset of the operation status of highway construction equipment in real time; The ant colony heuristic path search module performs ant colony heuristic path search based on the dynamic node anomaly risk level function, the dynamic adaptive path cost function, and the path search heuristic function, and dynamically constructs a global risk assessment path for abnormal operation status of the equipment. Among them, the construction methods of the dynamic node anomaly risk level function, the dynamic adaptive path cost function, and the path search heuristic function include: Defining each initial state feature vector in the initial state set of highway construction equipment operation data as a path search node in the ant colony heuristic path search algorithm to form a search node set; for each path search node, defining a dynamic node anomaly risk level function in combination with the degree of difference in equipment operation status parameters, the trend of equipment status change, and the equipment anomaly history record; constructing a dynamic adaptive path cost function between path search nodes based on the risk level in the global search stage to measure the risk transfer cost between any two path search nodes; constructing a path search heuristic function based on the node anomaly risk level for dynamic prediction of the risk area, and the path search heuristic function is used to guide the path search; The genetic tabu search module is used to further perform local refined search on the preliminary potential abnormal equipment area based on the global risk assessment path, and dynamically generate a local risk optimization result; The risk assessment information fusion module is used to fuse the global risk assessment path and the local risk optimization result, and form comprehensive risk assessment information including risk level, risk distribution range, and abnormal status confirmation information; The safety warning module is used to generate an equipment anomaly warning signal in real time according to the comprehensive risk assessment information, and transmit the equipment anomaly warning signal to the on-site safety management terminal of the highway construction site to guide on-site personnel to respond to and handle the abnormal status in a timely manner.

2. A method for monitoring the safety of highway construction equipment based on the Internet of Things, which is applied to the safety monitoring system for highway construction equipment based on the Internet of Things described in claim 1, and is characterized in that, It includes: S1. Deploy a distributed Internet of Things sensor network at the highway construction site to collect the dataset of the operation status of highway construction equipment in real time; S2. Transmit the dataset of the operation status of highway construction equipment to the central data processing unit in real time through a wireless communication network, and establish a real-time data monitoring database in the central data processing unit; S3. Perform preliminary processing on the dataset of the operation status of highway construction equipment to form an initial state set of highway construction equipment operation data; S4. Use the ant colony heuristic path search algorithm to perform a global search on the initial state set of highway construction equipment operation data, and predict the risk area of highway construction equipment to generate a global risk assessment path; S5. Use the global risk assessment path as an input parameter, and apply the genetic tabu search algorithm to perform a local detailed search and analysis on the potential abnormal area determined in step S4 to generate a local risk optimization result; S6. Integrate the global risk assessment path with the local risk optimization results to form comprehensive risk assessment information, which clarifies the risk level, risk distribution range, and abnormal status confirmation information of the operation status of highway construction equipment; S7. According to the comprehensive risk assessment information, generate and real-time publish equipment abnormal warning signals through a preset safety warning mechanism, and transmit the equipment abnormal warning signals to the on-site safety management terminal of highway construction, so as to realize the safety monitoring among highway construction equipment.

3. The method for safely monitoring equipment used in highway construction based on the Internet of Things according to claim 2, wherein The said S1 includes the following steps: S11. Divide the monitoring area at the highway construction site and define it as the set of monitoring areas , and each monitoring area in the set of monitoring areas corresponds to the on-site construction scope of the i-th highway construction; S12. In each monitoring area deploy multiple Internet of Things sensor nodes according to the types of highway construction equipment, construction characteristics, and safety monitoring requirements to form a collection of Internet of Things sensor networks : ； Among them, represents the th sensor node in the th monitoring area, n represents the total number of monitoring areas divided at the highway construction site, represents the number of sensor nodes in the S13. Through the sensor nodes Real-time collect the device location data, device operation status data, and device interaction data among devices to form a dataset of the operation status of highway construction equipment : ； Among them, represents the operation status data of highway construction equipment collected by the th sensor node in the th monitoring area at the moment, is the time series of data collected by the monitoring system in real time; S14. Transmit the real-time collected dataset of the operating status of highway construction equipment to the central data processing unit through a wireless communication network.

4. The safety monitoring method for highway construction equipment based on the Internet of Things according to claim 3, characterized in that, The said S3 includes the following steps: S31. Clean the dataset of the operating status of highway construction equipment collected in real time to remove the abnormal data and invalid data in the dataset caused by transmission failures, sensor anomalies, or environmental interference, and form a cleaned dataset of the operating status of highway construction equipment; S32. Normalize the operation status data set of the road construction equipment after cleaning to eliminate the analysis errors caused by the dimensional differences in the data collected by different sensor nodes, and generate a normalized operation status data set of the road construction equipment ; S33. Extract features from the standardized dataset of the operating status of highway construction equipment and select state feature vectors according to the types of highway construction equipment and construction safety requirements to construct an initial set of operating data for highway construction equipment .

5. The security monitoring method for highway construction equipment based on the Internet of Things according to claim 4, wherein, The said S4 includes the following steps: S41. Initialize the set of initial states of highway construction equipment operation data Each initial state feature vector in is defined as a path search node in the ant colony heuristic path search algorithm, forming a set of search nodes : ； Among them, represents the path search node corresponding to the th sensor node in the th monitoring area, represents the initial state feature vector in the set of initial states of the operation data of the highway construction equipment corresponding to the th th sensor node in the th monitoring area; S42. For each path search node , combining the degree of difference in device operation state parameters, the trend of device state change, and the device exception history record, define a dynamic node exception risk level function: ； Among them, represents the risk level in the global search phase of the path search node , , and are weight factors, represents the degree of difference between the initial state feature vector and the reference state feature vector , represents the change trend of the initial state feature vector over time, is the cumulative function of the abnormal history record of the path search node; S43. Based on the risk level in the global search phase Construct a dynamic adaptive path cost function between path search nodes, which is used to measure the risk transfer cost between any two path search nodes and therebetween: ； Among them, is the path search node and the risk transfer cost between path search nodes is used to describe the degree of risk transfer of abnormal device states between any two path search nodes. and respectively represent the risk levels of the path search node and the path search node in the global search stage at time t. 、 respectively represent adjacent path search nodes. is the Euclidean distance of feature differences between path search nodes. is the sensitivity adjustment index. The dynamic adaptive path cost function reflects the transfer probability of abnormal device operation risks between different device nodes. The greater the sum of risk levels, the higher the probability of abnormal propagation. S44. Construct a path search heuristic function based on the node abnormal risk level Perform dynamic prediction of the risk area, and the path search heuristic function is used to guide the path search: ； Among them, is the sensitivity index of feature differences, which is used to dynamically adjust the influence of feature differences between nodes. When the abnormal risk level of a node is higher than the preset value and the feature difference is higher than the preset value, the value of the path search heuristic function increases, guiding the ant colony heuristic path search algorithm to preferentially explore the risk area of the operating state of potential abnormal devices; S45. Comprehensive pheromone concentration and the path search heuristic function , and the improved path selection probability formula is used to dynamically determine the path selection probability in the global path search , which is used to preliminarily locate the risk area of the potential abnormal device operation state: ； Among them, and are time-varying weight factors that change with the dynamic data of the highway construction site, is a dynamic adaptation factor based on the fluctuation of the operation data of highway construction equipment in the near future, represents the set of adjacent nodes of the path search node , and r represents all possible path targets; S46. Dynamically adapt to changes in the construction environment according to the path selection probability, and optimize and form a global risk assessment path in real time .

6. The method for safely monitoring equipment for highway construction based on the Internet of Things according to claim 5, characterized in that, The said S5 includes the following steps: S51. Take the potential abnormal area determined by the global risk assessment path as the initial input of the genetic taboo search algorithm to form a local search population set , and each individual in the population set corresponds to the set of equipment nodes to be optimized in the potential abnormal area: ； Among them, represents the $i$-th individual in the local search population set where $N$ represents the size of the local search population; S52. For individuals in the local search population set , define the local anomaly risk fitness function: ； Among them, , and are the weight coefficients of the fitness function, represents the cumulative value of the historical risk assessment of the population individual , is the local abnormal risk fitness level; S53. Adapt according to the local anomaly risk fitness level Perform genetic tabu search algorithm operations on the current local search population set to generate the next generation of population individuals ; S54. Adapt to the local abnormal risk fitness level Construct a dynamic taboo table for the genetic taboo search algorithm , and the dynamic taboo table is used for taboo constraints in high fitness regions: ； Among them, is the dynamic risk threshold: ； Among them, is the reference risk coefficient, is the risk amplification factor, is the preset risk reference value; S55. After satisfying the preset termination condition, select the individual with the highest local abnormal risk fitness from the local search population set outside the dynamic taboo table as the local risk optimization result according to the following formula : ； Among them, represents the local search population set in which the individual is a diversity regulation factor for evaluating the distribution difference degree of the state feature vectors of the internal device nodes of the individual: ； in, Represents an individual The number of device nodes contained in For individuals The center vector of all path search nodes within, Represents a single device operation status node within a local search population individual, To select the diversity adjustment coefficient, it is used to balance the trade-off between individual risk and internal node diversity. Represents an exponential function.

7. A method for safely monitoring equipment used in highway construction based on the Internet of Things according to claim 6, characterized in that, The said S6 includes the following steps: S61. Incorporate the global risk assessment path with the local risk optimization result to perform feature fusion, construct comprehensive risk assessment information, and define the fusion function as: ； Among them, , and are fusion weight coefficients, represents the overlap confidence that the path search node is simultaneously recognized as a high-risk node in global search and local search, is the comprehensive risk assessment information; S62. Based on the comprehensive risk assessment information Define the evaluation indexes of the comprehensive risk level , and divide the device nodes into low-risk, medium-risk and high-risk levels according to the preset risk thresholds: ； Among them, and are the lower threshold and the high-risk critical threshold for distinguishing the abnormal risk level of device nodes, respectively; S63. According to the comprehensive risk level evaluation index Rank the risk levels of equipment nodes and determine the set of high-risk nodes : ； Among them, is the dynamic threshold parameter of the comprehensive risk level, which is jointly determined by the statistical distribution characteristics of the historical equipment abnormal risk level at the highway construction site and the real-time risk level fluctuation situation; S63. According to the high-risk node set Construct a comprehensive risk area set of highway construction equipment , locate and dynamically track the areas with abnormal equipment operation status within the construction site: ； Among them, represents the set of comprehensive risk areas of highway construction equipment determined by comprehensive risk assessment, represents the i-th monitoring area, that is, the i-th specific construction scope divided within the highway construction site. The monitoring area corresponds to the actual geographical monitoring area where the abnormal state node of the highway construction equipment is located, and is used to clarify the specific spatial position of the abnormal equipment when the equipment is abnormal, so as to realize the positioning and dynamic tracking of the operating state of the abnormal equipment.

Images