A highway construction site state monitoring method and system based on an internet of things

Abstract

This invention relates to the field of status monitoring technology, specifically to an IoT-based method and system for monitoring the status of highway construction sites. The method includes the following steps: collecting and aligning the start, end, and waiting times of multiple construction machines; statistically analyzing the number of start times and frequency of operations to form an operation time series; analyzing the relationship between status changes based on duration and calculating the transition characteristics between operations and waiting; generating a collaborative status by comprehensively considering the dispersion of multiple machine states; analyzing the trend of collaborative status changes according to the monitoring cycle and constructing a status evolution chain; and forming continuously updated status monitoring results for the construction site. In this invention, through the analysis of operation time series and status transition relationships, continuous characterization and correlation identification of the operating status of construction machinery are achieved, forming a dynamic understanding of the collaborative status of multiple machines. By using trend characteristics to reflect the switching patterns between operations and waiting, more detailed status support is provided for the evolution of construction progress, improving the accuracy of judgment and response efficiency in on-site management.

Description

Technical Field

[0001] This invention relates to the field of condition monitoring technology, and in particular to a method and system for condition monitoring of highway construction sites based on the Internet of Things. Background Technology

[0002] The field of status monitoring technology refers to a collection of technologies centered around the continuous acquisition, recording, and judgment of the operational status of objects. Its core aspects include the collection, transmission, storage, and presentation of status information. It typically involves the continuous perception and unified management of on-site environmental parameters, construction activity status, equipment operation status, and personnel work status. This technology field acquires status data by deploying sensing nodes and transmits it to the management end via communication networks for aggregation and display. It is widely used in scenarios such as engineering construction, transportation facility operation, and industrial site management, forming a systematic technology system with the goal of on-site status visualization and process traceability.

[0003] Traditional highway construction site status monitoring methods and systems refer to technical matters such as pavement operation progress, changes in the construction area environment, working status of construction equipment, and personnel distribution during highway construction. Typically, this involves deploying camera devices, environmental monitoring devices, and location acquisition devices at the construction site to obtain specific data such as image information, gas concentration, dust values, noise intensity, and location information. This data is then transmitted to a back-end management platform via wired or wireless communication, allowing managers to monitor and record the construction site status based on the displayed real-time data and historical records.

[0004] Current status monitoring methods primarily rely on traditional methods such as camera devices, environmental monitoring devices, and location acquisition devices for data collection, transmitting the data to a management platform via wired or wireless communication. This approach has several technical drawbacks. First, traditional systems can only provide static environmental information and equipment status, lacking dynamic monitoring and analysis of the construction machinery's operation process, and failing to effectively reflect key operational characteristics such as operating frequency and downtime. Second, the collaborative status between construction machinery is not effectively identified and analyzed, making it difficult for managers to fully grasp the on-site equipment operating efficiency and their mutual influences. Third, existing technologies often cannot predict and assess the evolution trend of construction status in a timely and accurate manner, affecting effective monitoring of construction progress and decision support. These shortcomings make it difficult to identify and handle status changes and collaborative issues during construction in a timely manner, impacting the accuracy and efficiency of construction site management. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies by proposing an Internet of Things-based method and system for monitoring the status of highway construction sites.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a method for monitoring the status of highway construction sites based on the Internet of Things, comprising the following steps:

[0007] S1: Collect the start and end times and waiting durations of multiple construction machines at the highway construction site and align them by timestamps. Extract the number of start times and calculate the frequency of construction machine operation to generate an operation time series.

[0008] S2: Based on the operation time series analysis, analyze the operation duration, extract the relationship between the changes in the state of construction machinery in adjacent time periods, calculate the transition probability between the operation state and the waiting state, and generate a construction operation state sequence.

[0009] S3: Obtain the construction operation state sequence corresponding to multiple construction machines in the construction site, calculate the dispersion of the operation state transition probability and operation duration of multiple construction machines, and generate the construction site collaborative state;

[0010] S4: Sort the collaborative status of the construction site by time according to the continuous monitoring cycle, calculate the slope of status change, analyze the trend characteristics of collaborative status change over time, and generate construction status change data.

[0011] S5: Based on the construction status change data, analyze the construction site status evolution chain with the construction machinery operation status as the node, mark and update the operation status of the construction site in multiple time periods in the status evolution chain, and generate construction site status monitoring results.

[0012] As a further aspect of the present invention, the operation time sequence includes the operation start time, operation end time, waiting duration, number of operation starts, and operation frequency of construction machinery; the construction operation state sequence includes the operation state, waiting state, relationship between the state changes of construction machinery in adjacent time periods, and the transition probability between the operation state and the waiting state; the construction site collaborative state includes the construction machinery operation state sequence and the dispersion of operation duration; the construction state change data includes the state change slope and the trend characteristics of collaborative state changes over time; and the construction site state monitoring results include the state evolution chain, construction site state labeling and updating, and the set of operation states within a time period.

[0013] As a further aspect of the present invention, the step of obtaining the job time sequence specifically includes:

[0014] S101: Collect the start time, end time and waiting time of multiple construction machines at the highway construction site, monitor the changes in the operating status of multiple machines and record the corresponding timestamps, perform sequence correction and position matching of the multi-source timestamps according to a unified time reference, and obtain an aligned time identifier sequence.

[0015] S102: Based on the aligned time identifier sequence, obtain the state switching records of multiple construction machines between adjacent time identifiers, perform counting processing on the start state identifier, calculate the corresponding ratio according to the number of times a single machine is started and the length of the observation period, form a frequency set distinguished by machine, and generate the machine operation frequency.

[0016] S103: Based on the alignment time identifier sequence and the mechanical operation frequency, the time identifier sequence is called to perform position mapping of the start events, embed the frequencies of multiple mechanical start events into the corresponding time periods and rearrange them in time order to generate an operation time sequence.

[0017] As a further aspect of the present invention, the step of obtaining the construction operation state sequence specifically includes:

[0018] S201: Based on the construction site operation time series, extract the start time and end time of adjacent operations of construction machinery and perform difference calculation, analyze the duration of corresponding operations, extract the end time and start time of adjacent operations and calculate the interval duration, use as the state input set to obtain the operation waiting duration;

[0019] S202: Based on the operation waiting duration, obtain the status identifier changes of the construction machinery in adjacent time periods, align the operation status and waiting status with the execution status, calculate the transition probability between the operation status and the waiting status, and generate the state transition probability.

[0020] S203: Invoke the state transition probability, arrange the state nodes according to the time sequence of the construction machinery, and perform sequence mapping between the operation state and the waiting state according to the probability correspondence to establish the construction operation state sequence.

[0021] As a further aspect of the present invention, the step of obtaining the collaborative status of the construction site specifically includes:

[0022] S301: Obtain the construction operation state sequence of multiple construction machines in the construction site, extract the state transition probability and operation duration of each construction machine, and calculate the operation duration deviation based on the state sequence of multiple construction machines;

[0023] S302: Based on the deviation of the operation duration, call the state transition probabilities corresponding to multiple construction machines, compare the state combinations of multiple machines in the same time period, calculate the difference in the state transition probabilities of multiple machines and the difference in the duration deviation, and generate the collaborative probability deviation.

[0024] S303: Based on the aforementioned coordination probability deviation, the state sequences of multiple construction machines at the same time are jointly arranged, the state combinations are integrated according to the coordination probability deviation value, the state mapping results of multi-machine coordination behavior are analyzed, and the coordination state of the construction site is generated.

[0025] As a further aspect of the present invention, the deviation of the operation duration is calculated using the following formula:

[0026] ;

[0027] in, This represents the deviation in the duration of the task. Represents the number of construction machines. Representing the The operating duration of the Taiwanese machinery, This represents the average duration of all mechanical operations. Representing the The state transition probability of the machine. The nonlinear discrete amplification index used when the duration of the operation deviates from the average value. The index parameter represents the weight of the state transition probability on the deviation.

[0028] As a further aspect of the present invention, the step of obtaining the construction status change data specifically includes:

[0029] S401: Based on the collaborative status of the construction site, sort the collaborative status values ​​corresponding to multiple time points according to the continuous monitoring cycle, and perform position verification on the collaborative status values ​​of adjacent cycles to generate a collaborative status time series.

[0030] S402: Based on the time series of the coordinated state, the time series index value and the corresponding coordinated state value are used as input items to calculate the relationship between the time index and the coordinated state value, and to calculate the slope of the coordinated state change.

[0031] S403: Based on the slope of the coordinated state change, and combined with the numerical distribution of multiple time points in the coordinated state time series, the positive and negative directions of the slope and the magnitude of the numerical change are judged, the trend of the coordinated state change over time is sorted out, and construction state change data is generated.

[0032] As a further aspect of the present invention, the steps for obtaining the construction site status monitoring results are specifically as follows:

[0033] S501: Based on the construction status change data, obtain multiple construction machinery operation status records, take the operation status as nodes, connect adjacent states according to the time sequence, mark the time point of the state occurrence and record the duration interval, and generate a state evolution chain sequence.

[0034] S502: Based on the state evolution chain sequence, for multiple time periods covered by the state nodes, collect the running state identifier values ​​within the corresponding time periods, determine the continuation or change of the state nodes between adjacent time periods, replace or retain the state labels, and obtain the state segment labeling results.

[0035] S503: Based on the state segment labeling results and combined with the time span of multiple nodes in the state evolution chain sequence, the updated state labels are summarized in time period order to generate the construction site state monitoring results.

[0036] As a further aspect of the present invention, in the process of connecting adjacent states according to time sequence, marking the time point of state occurrence, and recording the duration interval, when adjacent running state records maintain the same running state identifier value for a continuous interval of not less than a preset fixed duration, they are merged into the same state node, and the time point of the first running state record is taken as the state occurrence time point, and the time difference between the time point of the last running state record and the time point of the first running state record is taken as the duration interval.

[0037] An Internet of Things (IoT)-based highway construction site status monitoring system, wherein the IoT-based highway construction site status monitoring system is used to execute the aforementioned IoT-based highway construction site status monitoring method, the system comprising:

[0038] The data analysis module collects the start and end times and waiting durations of multiple construction machines at the highway construction site, aligns them with timestamps, extracts the number of start times, calculates the operation frequency of construction machines, generates an operation time series, and transmits it to the status analysis module.

[0039] The status analysis module analyzes the duration of the operation based on the operation time series, extracts the relationship between the status changes of construction machinery in adjacent time periods, calculates the transition probability between the operation status and the waiting status, generates a construction operation status sequence, and transmits it to the collaborative computing module.

[0040] The collaborative computing module obtains the construction operation state sequence corresponding to multiple construction machines in the construction site, calculates the dispersion of the operation state transition probability and operation duration of multiple construction machines, generates the collaborative state of the construction site, and transmits it to the trend evaluation module.

[0041] The trend assessment module sorts the collaborative status of the construction site by time according to the continuous monitoring cycle, calculates the slope of status change, analyzes the trend characteristics of collaborative status change over time, generates construction status change data, and transmits it to the status monitoring module.

[0042] The status monitoring module analyzes the construction site status evolution chain with the construction machinery operation status as the node based on the construction status change data, marks and updates the operation status of the construction site in multiple time periods in the status evolution chain, and generates construction site status monitoring results.

[0043] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0044] This invention, through innovative operation time series analysis and construction machinery state transition probability calculation, dynamically captures and accurately analyzes the operating status and changing relationships of machinery during construction. By analyzing the trend characteristics of state changes, it provides in-depth understanding of the collaborative state at the construction site, thereby achieving real-time monitoring of the cooperation between various machines. Through probability analysis of the transition between the operating and waiting states of construction machinery, it accurately reflects the switching between operation and shutdown among different machines, providing more detailed state evolution data support for on-site management, thus more clearly demonstrating the real-time trend of construction progress. The innovation of this method lies in its ability to comprehensively assess the dynamic changes at the construction site through continuous monitoring and data analysis, improving the accuracy, response speed, and decision support capabilities of construction site management. Attached Figure Description

[0045] Figure 1 This is a schematic diagram of the main steps of the present invention;

[0046] Figure 2 This is a flowchart of the operation time sequence of the present invention;

[0047] Figure 3 This is a flowchart of the construction and operation status sequence of the present invention;

[0048] Figure 4 This is a flowchart illustrating the collaborative state of the construction site according to the present invention;

[0049] Figure 5 This is a flowchart of the construction status change data of the present invention;

[0050] Figure 6 This is a flowchart of the construction site status monitoring results of the present invention. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0052] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0053] Please see Figure 1 This invention provides a technical solution: a method for monitoring the status of highway construction sites based on the Internet of Things, comprising the following steps:

[0054] S1: Collect the start and end times and waiting durations of multiple construction machines at the highway construction site and align them by timestamps. Extract the number of start times and calculate the frequency of construction machine operation to generate an operation time series.

[0055] S2: Based on the operation time series analysis, analyze the duration of the operation, extract the relationship between the changes in the state of construction machinery in adjacent time periods, calculate the transition probability between the operation state and the waiting state, and generate a construction operation state sequence.

[0056] S3: Obtain the construction operation state sequence corresponding to multiple construction machines in the construction site, calculate the dispersion of the operation state transition probability and operation duration of multiple construction machines, and generate the construction site collaborative state;

[0057] S4: Sort the collaborative status of the construction site by time according to the continuous monitoring cycle, calculate the slope of status change, analyze the trend characteristics of collaborative status change over time, and generate construction status change data.

[0058] S5: Based on construction status change data, analyze the construction site status evolution chain with the construction machinery operation status as the node, mark and update the operation status of the construction site in multiple time periods in the status evolution chain, and generate construction site status monitoring results.

[0059] The operation time series includes operation start time, operation end time, waiting duration, number of operation starts, and operation frequency of construction machinery. The construction operation status series includes operation status, waiting status, relationship between the status changes of construction machinery in adjacent time periods, and the probability of transition between operation status and waiting status. The construction site coordination status includes the construction machinery operation status sequence and the dispersion of operation duration. The construction status change data includes the status change slope and the trend characteristics of coordination status changes over time. The construction site status monitoring results include the status evolution chain, construction site status labeling and updating, and the set of operation statuses within a time period.

[0060] Please see Figure 2 The specific steps for obtaining the job time series are as follows:

[0061] S101: Collect the start time, end time and waiting time of multiple construction machines at the highway construction site, monitor the changes in the operating status of multiple machines and record the corresponding timestamps, perform sequence correction and position matching of the multi-source timestamps according to a unified time reference, and obtain an aligned time identifier sequence.

[0062] This system collects data on the start and end times of multiple construction machines operating at highway construction sites, such as highway expansion projects. Sensor devices, such as current sensors or GPS modules, are deployed to collect operational data from multiple machines, including excavators, road rollers, and cranes. Specific actions include monitoring the current value of the machine's motors to start; recording the start time when the current exceeds a preset threshold of 400 amps; recording the end time when the current drops below 100 amps; and calculating the operation duration as the difference between the end time and the start time. In the example, excavator A started at 08:00:00 on December 18, 2025, with a start current of 600 amps, and ended at 08:30:00 with a stop current of 80 amps, calculating the operation duration as 30 minutes. Road roller B started at 08:05:00, but its clock was off-clock, so the original timestamp 08:05:00 was recorded. Crane C started at 08:05:00... The system starts recording data from 8:10:00 to 08:10:00, monitoring changes in operating status by reading sensor data in real time. When a status value, such as current, changes from low to high, a timestamp is recorded. Based on a unified time reference, such as synchronizing all devices to a GPS time server, the system performs sequence correction on multi-source timestamps. Specific actions include extracting timestamp values, comparing time values, and adjusting the order. For example, the original timestamp of roller B, 08:05:00, is corrected to GPS time 08:03:00. The sorted timestamp sequence is: excavator A, 08:00:00; roller B, 08:03:00; crane C, 08:10:00. Location matching retrieves the machine's GPS coordinate data, such as excavator A coordinates (116.40°E, 39.90°N), roller B coordinates (116.41°E, 39.91°N), and crane C coordinates (116.42°E, 39.90°N).(92°N), match the timestamp to the corresponding position to form an aligned sequence such as [Time 08:00:00, Excavator A, Status Started], [Time 08:03:00, Road Roller B, Status Started], [Time 08:10:00, Crane C, Status Started]. In the example, the observation period at the construction site is 8 hours, and the collected data includes the start time, end time, and operation duration. In the example, the operation duration is calculated using the time difference, such as Excavator A starting at 08:00:00 and ending at 08:00:00. 08:30:00, 30-minute difference, Roller B starts; 08:03:00, ends; 08:33:00, 30-minute difference, Crane C starts; 08:10:00, ends; 08:40:00, 30-minute difference. The timestamp acquisition process involves direct measurement and recording via sensors. The threshold setting references the typical operating current range of the mechanical motor (300-700 amps). Through experimental testing, a starting threshold of 400 amps is set. Example calculation: The excavator's no-load current is tested at 200 amps, and the load current... A load current of 600 amps is compared to a threshold of 400 amps, indicating a start-up state. The end threshold of 100 amps is determined through similar test settings. The reference value, such as GPS time, is synchronized via satellite signal. The location matching reference is the coordinate system of the construction site map. The judgment interval is when the number of machines is greater than or equal to 2. In this example, 3 machines meet the criteria. Data, such as timestamp values, is obtained through a clock function in seconds. The example acquisition process is as follows: the sensor detects current changes and records the current time, such as 08:00:00. The value range matches the actual construction time. The correction process calls time values, comparing the timestamps of excavator A (168000000 milliseconds), road roller B (original 168030000 milliseconds, corrected 168018000 milliseconds), and crane C (168060000 milliseconds). After sorting, the sequence time values ​​increment. The matching position calls coordinate values, calculating Euclidean distance to confirm machine proximity. For example, a coordinate difference less than 0.01 degrees is considered a match. The alignment sequence is generated by integrating time, machine identifier, and status value to obtain an aligned time identifier sequence.

[0063] S102: Based on the aligned time identifier sequence, obtain the state switching records of multiple construction machines between adjacent time identifiers, perform counting processing on the start state identifier, calculate the corresponding ratio based on the number of times a single machine is started and the length of the observation period, form a frequency set distinguished by machine, and generate the machine operation frequency.

[0064] Based on aligned time stamp sequences, in a highway construction site as shown in the example above, the aligned sequence contains timestamps and machine status data. The goal is to obtain records of state switching between adjacent time stamps for multiple construction machines. Specific actions include extracting consecutive time points from the sequence and comparing changes in state values, such as from start to waiting or end. In the example time stamp sequence [08:00:00 Excavator A starts], [08:30:00 Excavator A ends], the adjacent time stamp from 08:00:00 to 08:30:00 shows the state switching from start to end, and the switching event is recorded. For each machine, a counting process is performed based on its startup status. Records with a "start" status are retrieved, and the number of starts for each machine is counted. In the example, excavator A starts 5 times, road roller B starts 6 times, and crane C starts 4 times during an 8-hour observation period. The ratio of the number of starts for each machine to the length of the observation period is calculated. This involves division, dividing the number of starts by the length of the period in hours. In this example, the observation period is 8 hours. The ratio for excavator A is 5 / 8 = 0.625, for road roller B it is 0.75, and for crane C it is 0.75. 4 is 0.5, forming a frequency set distinguished by machine. This integrates machine identifiers and ratios, such as {Excavator A: 0.625, Road Roller B: 0.75, Crane C: 0.5}. The machine operation frequency is generated as the output frequency set. Data such as the number of starts is obtained by parsing and aligning the sequence. The example acquisition process is as follows: traverse the sequence, filter records with the status "started", count integers, and calculate the observation period length using the difference between the start and end times. For example, if the sequence starts at 08:00:00 and ends at 16:00:00, the difference is 8 hours. The ratio is then calculated and adjusted. Using the number of times and time period values, a division operation is performed. For example, "adjacent" refers to a time interval of less than or equal to 30 minutes. The judgment interval is based on the typical operating cycle of construction machinery. The threshold setting references the state switching judgment. Through experimental observation, a state change threshold is set, such as recording a switch when the current difference is greater than 200 amperes. Example calculation: the test current drops from 600 amperes to 300 amperes. The difference of 300 amperes is greater than the threshold of 200 amperes, and a state switch is judged. Weights or coefficients are not involved. Actual data values ​​are introduced. The number of times the machinery is started and the length of the time period are shown in the table below.

[0065] Table 1 Construction Machinery Operation Data Table

[0066] ;

[0067] Table 1 lists the number of times each machine is started and the length of the observation period, which are used to calculate the operation frequency and generate the machine operation frequency results.

[0068] S103: Based on the alignment time identifier sequence and the mechanical operation frequency, the time identifier order is called to perform position mapping of the start events, embed the frequencies of multiple mechanical start events into the corresponding time periods and rearrange them in time order to generate the operation time sequence;

[0069] Based on the alignment of time stamp sequences and machinery operation frequencies, in a highway construction site example, the alignment sequence contains time points and event data. Machinery operation frequencies are represented by a frequency set {Excavator A: 0.625, Road Roller B: 0.75, Crane C: 0.5}. The time stamp sequence is used to map the location of start events. Specific actions include extracting timestamp values ​​from the sequence, associating start events (e.g., mapping 08:00:00 to the Excavator A start event), embedding the frequencies of multiple machinery start events into corresponding time periods, and assigning frequency values. To define a time interval, for example, embedding a frequency of 0.625 times / hour into the operating time period of excavator A, rearrange the timestamp values ​​according to time order to generate sequences such as [Time 08:00:00, Event Frequency 0.625], [Time 08:03:00, Event Frequency 0.75], [Time 08:10:00, Event Frequency 0.5]. This generates the operating time series as output. In this example, the time interval is divided based on the aligned sequence time points; in the example, the time interval from 08:00:00 to 08:30:00 is allocated... Excavator A has a frequency of 0.625. From 08:03:00 to 08:33:00, it is assigned to road roller B with a frequency of 0.75. From 08:10:00 to 08:40:00, it is assigned to crane C with a frequency of 0.5. After rearranging, the time values ​​of the sequence are in ascending order. Data such as event frequencies are obtained from the frequency set, and timestamps are obtained from the aligned sequence. The example acquisition process is as follows: parse the frequency set to extract the machine identifier and frequency value, call the aligned sequence time point, map the event to the nearest timestamp, embed the frequency value to the time interval, and sort and call the time value comparison. The sequence is generated. "Corresponding" refers to time interval matching. The judgment interval is a time difference of less than or equal to 5 minutes. The threshold setting refers to the location mapping tolerance. Through experiments, the time offset threshold is set to 300 seconds. The calculation example is: the difference between the event time 08:00:00 and the sequence time 08:00:00 is 0 seconds, which is less than the threshold of 300 seconds. Direct mapping is performed. Weights or coefficients are not involved. The actual data values ​​are such as the time value 08:00:00 and the frequency value 0.625. The value range is consistent with the actual construction monitoring. The operation time series results are generated.

[0070] Please see Figure 3 The specific steps for obtaining the construction operation status sequence are as follows:

[0071] S201: Based on the construction site operation time series, extract the start time and end time of adjacent operations of construction machinery and perform difference calculation, analyze the duration of corresponding operations, extract the end time and start time of adjacent operations and calculate the interval duration, use as the state input set to obtain the operation waiting duration;

[0072] Based on the construction site operation time series, the system retrieves sequence data such as operation time series [time 08:00:00, event frequency 0.625], [time 08:03:00, event frequency 0.75], [time 08:10:00, event frequency 0.5], and extracts the start and end times of adjacent operations of construction machinery. Specific actions include retrieving timestamp values ​​such as 08:00:00 and 08:30:00, comparing the time values, and defining adjacent events as those with an interval of less than or equal to 10 minutes. In the example, excavator A starts at 08:00:00 and ends at 08:30:00, while road roller B starts at 08:03:00. At the end time of 08:33:00, perform difference calculation, calculating the difference between the end time and the start time, in minutes. In the example, the difference for excavator A is 08:30:00 - 08:00:00 = 30 minutes, and the difference for roller B is 08:33:00 - 08:03:00 = 30 minutes. Analyze the corresponding operation duration, retrieve the difference result, and record the duration value, such as 30 minutes. Extract the end time and start time of adjacent operations, retrieving the end time, such as excavator A 08:30:00, and the next start time, roller B 08:03:00. However, the time order needs to be adjusted; after sorting the timestamp values ​​in ascending order, the start time corresponding to the end time 08:30:00 is... At 08:33:00 (Road Roller B), the interval duration is calculated by subtracting the start time from the end time. In the example, the difference between the end time of Excavator A (08:30:00) and the start time of Road Roller B (08:33:00) is 3 minutes, which serves as the status input set. All duration data, such as the operation duration of 30 minutes and the interval duration of 3 minutes, are integrated to form a dataset {Operation Duration: 30, Interval Duration: 3}. The "adjacent" judgment interval is based on the typical operation cycle of construction machinery, with a time difference threshold of 600 seconds. Through experimental observation of the machine operation interval, a time difference of less than 500 seconds is considered adjacent. Data such as timestamp values ​​are directly measured by sensors in seconds. The example obtains the flow. Process: The sensor records the time of current change. For example, when the starting current exceeds 400 amps, it records 08:00:00, and when the ending current is below 100 amps, it records 08:30:00. The numerical range conforms to the actual construction. The threshold setting is based on the current change rate. Through experimental testing, the starting threshold is set to 400 amps. The reference value is such as GPS synchronization with the time reference. Weights or coefficients are not involved. Example of actual data values: timestamp 08:00:00, 08:30:00, difference calculated in 30-minute intervals, reasonable value range for operation time is 10-60 minutes, interval is 1-15 minutes. The operation waiting time is obtained as the interval output.

[0073] S202: Based on the duration of the work waiting period, obtain the changes in the status identifier of the construction machinery in adjacent time periods, align the execution status for the work status and the waiting status, calculate the transition probability between the work status and the waiting status, and generate the state transition probability.

[0074] Based on the job waiting duration, the system retrieves the job waiting duration (e.g., an interval of 3 minutes) to obtain the status changes of the construction machinery within adjacent time periods. Specific actions include retrieving status data, such as aligning sequences like [Time 08:00:00, Status: Working] and [Time 08:30:00, Status: Waiting]. It compares status values ​​between adjacent time periods, such as switching from working to waiting. In the example, the status changes from working to waiting during the time period from 08:00:00 to 08:30:00. The system records the change event, performs status alignment for working and waiting states, matches status identifiers like "working" and "waiting" to the same time frame, and calls timestamp alignment. In the example, the working state is aligned to the waiting state at time 08:30:00. The system calculates the transition probability between the working and waiting states, as shown in the formula. In the middle, parameters Represents the time period State transition probability at time step, index Indicates a specific point in time, used to identify the time window for calculation. Representing machinery in time period The status of the task (numerical, 1 represents a task, 0 represents waiting). This represents the waiting time (in minutes) for that period. This represents the machine's operating efficiency (dimensionless ratio) during that time period. This represents the degree of interference in the work environment during that time period (dimensionless index). The summation symbol represents the total number (integer) of consecutive time periods involved in the calculation. Indicates from arrive The parameter values ​​for all time periods are summed, and the absolute value is not specified. To ensure the numerator is positive, the logical approach is to combine the numerator with the product of the work status, waiting time, and work efficiency to reflect the positive factors of state transition. The denominator combines the sum of the waiting time and the degree of disturbance to reflect environmental resistance. The overall formula outputs probability values ​​through division, aiming to quantify the possibility of state changes during construction. Its rationale is to avoid bias from a single parameter; for example, the numerator emphasizes the impact of efficiency, while the denominator incorporates disturbance balance. Parameter values ​​are based on actual monitoring and calculations. The current value (in amperes) is obtained through measurement by a current sensor and compared with a preset threshold of 400 amperes. (This threshold is set with reference to the rated mechanical current; experimental tests show an unloaded current of 200 amperes and a load current of 600 amperes, with the midpoint of 400 amperes used as the threshold. Example calculation: a current of 450 amperes is greater than 400 amperes, so a value is assigned.) ), Get the waiting time (in minutes) from job S201, retrieve the interval duration data, example: interval duration 3 minutes, assign value. , The workload is obtained by dividing the workload by the workload time. The workload is retrieved from GPS displacement sensor measurements (in meters), and the workload time is retrieved from the S201 workload duration (in minutes). Example: Workload 30 meters, workload 30 minutes. Calculation... , For non-numerical data, the quantization process involves calling noise sensor values ​​(in decibels) and comparing them to preset ranges: noise <50dB is assigned a value of 0 (low interference), 50-70dB is assigned a value of 0.5 (medium interference), and >70dB is assigned a value of 1 (high interference). The quantization standard is based on typical noise levels in the construction environment. Example: noise 65dB falls within the 50-70dB range, and a value is assigned... k is obtained by counting the number of consecutive time intervals; example: k=3. The reasonable range of parameter values ​​is based on real-world retrieval. A reasonable interval is 1-15 minutes (typical construction interval). A reasonable range is 0.5-2.0 (efficiency range). The reasonable range is 0-1 (interference index), and k is 1-10 (number of time periods). Set k=3 time periods, and the parameter values ​​are shown in the table below. For 1, 0, 1, (minutes) are 3, 5, 2. The values ​​are 1.0, 0.8, and 1.2. Substitute the values ​​0.5, 0.3, and 0.6 into the formula. :molecular denominator Formula calculation The result indicates that the probability value of 0.4737 is within the preset range of 0.3-0.7 (medium probability), which means that the probability of state transition is moderate. It is associated with the step result as a directly generated state transition probability value for subsequent sequence mapping.

[0075] S203: Invoke the state transition probability, arrange the state nodes according to the time sequence of the construction machinery, map the operation state and waiting state according to the probability correspondence, and establish the construction operation state sequence.

[0076] The process involves calling the state transition probability, such as 0.4737, and arranging state nodes according to the construction machinery's time sequence. Specific actions include calling time-series data such as [Time 08:00:00, Status: Operation], [Time 08:30:00, Status: Waiting], sorting the timestamp values ​​in ascending order, mapping the operation status and waiting status according to their probability correspondence, calling the probability value, and mapping the status to a time point. For example, a probability of 0.4737 represents the likelihood of transitioning from operation to waiting. In this example, the time 08:00:00, status is operation, and based on a probability of 0.4737, it is mapped as a high-probability transition event. A construction operation state sequence is established, and the mapping results are integrated, such as [Time 08:00:00, Status: Operation, Transition Probability 0.4737], [Time 08:30:00, Status: Waiting ... [Probability 0.4737], if the probability value of the "correspondence relationship" judgment interval is 0.3-0.7, it is considered medium. Set the threshold probability to 0.4. Test the transfer frequency through experiments. In the example, the probability 0.4737 is greater than the threshold 0.4, so it is directly mapped. Data such as probability values ​​are obtained from the formula calculation, and timestamps are obtained from the sequence. The example acquisition process is: parse the probability value, call the time point, assign the status, and establish the sequence. The threshold setting reference content is the average probability of historical data. The average probability of the past is calculated to be 0.45 through experiments. Set the threshold to 0.4. Example calculation: the test probability 0.5 is greater than 0.4, and the mapping is determined to be valid. The weight or coefficient is not involved. The actual data values ​​are such as the time value 08:00:00, the probability value 0.4737, and the probability range of the value 0-1. Establish the construction operation status sequence result.

[0077] Please see Figure 4 The specific steps for obtaining the collaborative status of the construction site are as follows:

[0078] S301: Obtain the construction operation status sequence of multiple construction machines in the construction site, extract the state transition probability and operation duration of each construction machine, and calculate the deviation of operation duration based on the state sequence of multiple construction machines;

[0079] This process acquires the operational status sequence of multiple construction machines at the construction site. Sequence data is retrieved, such as the output from S203: [Time 08:00:00, Status: Working, Transition Probability 0.4737], [Time 08:30:00, Status: Waiting, Transition Probability 0.4737]. The state transition probability and operation duration of each construction machine are extracted. Specific actions include parsing the sequence data, retrieving the state transition probability value (e.g., 0.4737), and retrieving the operation duration value (e.g., 30 minutes output from S201). In an example, such as a highway expansion project, the machines include excavator A, road roller B, and crane C. The sequence data is: [Excavator A, Probability 0.4737, Duration 30 minutes], [Road Roller B, Probability 0.5, Duration 35 minutes], [Crane C, Probability 0.6, Duration 40 minutes]. Based on the state sequences of multiple construction machines, the dispersion of the operation duration is calculated. Specific actions include retrieving duration values ​​of 30, 35, and 40 minutes and calculating the average. At each minute, the state transition probability values ​​are 0.4737, 0.5, and 0.6. The deviation is then calculated. The deviation (dimensionless exponent) representing the duration of the operation is used to quantify the degree of fluctuation in the operating time of construction machinery. Represents the number of construction machines (integer), subscript and Indicates the machine serial number (from 1 to n). Representing the The duration of operation of the machine (in minutes). This represents the average duration (in minutes) of all mechanical operations. Representing the The state transition probability of the machine (dimensionless). The nonlinear discrete amplification index (dimensionless) is used to amplify significant deviations when the duration of the operation deviates from the average. The exponential parameter (dimensionless) represents the weight of the state transition probability on the deviation, and the summation sign is given. Indicates from arrive All mechanical parameter values ​​are summed, with the absolute value sign indicating summation. Ensure the difference is positive before performing exponentiation. Weighting of bias, division operation Finding the average value, multiplication operation Combining the numerator and denominator, the square root The standardized output deviation value is calculated logically by using a weighted sum of duration deviations in the numerator (reflecting time fluctuations) and combining the reciprocal weights of state transition probabilities in the denominator (incorporating state influences). The overall formula outputs a dimensionless deviation using a square root. The aim is to comprehensively quantify the fluctuations in construction machinery operation time, avoiding reliance solely on time deviations. This is justified by using the p parameter to amplify significant deviations (e.g., p=2 for variance calculations) and the q parameter to adjust probability weights (e.g., q=1 for linear effects), ensuring the calculation reflects the synergistic effects in actual construction. Parameter values ​​are based on actual monitoring and calculations. The number of machines at the construction site is counted to obtain the sequence data. Example: In a highway expansion project, the output sequence data from S203 is: [Excavator A, probability 0.4737, duration 30 minutes], [Road Roller B, probability 0.5, duration 35 minutes], [Crane C, probability 0.6, duration 40 minutes]. The number of machines counted is n=3. The duration of operation S201 is obtained by measuring the time difference between the start-up and end-of-operation current changes using a current sensor, in minutes. Example: When the sensor detects that the start-up current of excavator A exceeds 500 amps, the time is recorded as 08:00:00; when the end-of-operation current is below 100 amps, the time is recorded as 08:30:00. The difference is calculated and assigned a value of 30 minutes (T1=30). Similarly, for a road roller, BT2=35 minutes, and for a crane, CT3=40 minutes. Obtained by calculating the average value, calling the Ti value, and performing a division operation to divide the sum by n. Example: minute, Obtained from the state transition probability of S202, the probability value is called. Example: S1=0.4737 (excavator A), S2=0.5 (road roller B), S3=0.6 (crane C). p and q are non-preset parameters and need to be set with reference content. p sets the tolerance for the dispersion of the reference construction data. Through experimental testing of historical duration deviation (such as the deviation range of past construction data), p=2 (magnification index) is set. Example calculation: the historical deviation is 5 minutes. When p=2, the amplification is 25, which meets the typical construction fluctuation tolerance (reasonable range p=1-3). q sets the influence weight of the reference state probability. Through experimental calculation of the correlation coefficient between probability and deviation (such as the correlation between the probability difference of past data and the deviation), q=1 (linear weight) is set. Example calculation: the correlation coefficient is 0.6 (moderate correlation). q=1 (reasonable range q=0.5-2). The reasonable range of parameter values ​​is based on reality: n is reasonable 1-10 (typical number range of construction machinery). A reasonable timeframe is 10-60 minutes (standard range for construction time). The reasonable range is 0-1 (probability range), p is reasonable 1-3 (exponential range), and q is reasonable 0.5-2 (weight range). Formula calculation example: Set parameter value n=3. Minutes, p=2, q=1 For 30, 35, 40, The values ​​are 0.4737, 0.5, and 0.6. Substitute these values ​​into the formula. ,calculate ,calculate ,calculate ,calculate ,calculate The results show that the deviation value of 3.254 is within the preset range of 0-5 (low dispersion), indicating that the operation time fluctuates little. It is associated with the step results as a direct generation of the operation duration deviation value (used for the collaborative calculation of S302). The advantage of the formula is that by introducing the state transition probability weight parameter q and the amplification exponent p, the comprehensiveness of the deviation calculation is improved, the influence of mechanical state is avoided, and a more accurate construction collaborative assessment is achieved in the overall system.

[0080] S302: Based on the deviation of the operation duration, call the state transition probabilities corresponding to multiple construction machines, compare the state combinations of multiple machines in the same time period, calculate the difference in state transition probabilities of multiple machines and the difference in duration deviation, and generate the collaborative probability deviation.

[0081] Based on the deviation of the operation duration, the deviation value (e.g., 3.254) is called, along with the state transition probabilities corresponding to multiple construction machines (e.g., probabilities from sequence data: 0.4737, 0.5, 0.6). The state combinations of multiple machines within the same time period are compared. Specific actions include calling a time point (e.g., 08:00:00), extracting all machine state values ​​at that time point (e.g., excavator A is operating, road roller B is waiting, crane C is operating), comparing the differences in state values ​​(in the example, the state combination for the time period 08:00:00 is {operating, waiting, operating}), calculating the difference in state transition probabilities and the difference in duration deviation. Specific actions include calling probability values ​​(0.4737, 0.5, 0.6), calculating the probability difference (e.g., maximum minus minimum 0.6 - 0.4737 = 0.1263), calling the deviation value (3.254), calculating the deviation difference (e.g., referencing a baseline deviation of 5 (preset), the difference is 3.254 - 5 = -1.746), and generating a collaborative probability discrete quantity. The algorithm integrates probability difference and deviation difference, such as {probability difference: 0.1263, deviation difference: -1.746}. The "same time period" is defined as a time difference of less than or equal to 5 minutes, with a threshold of 300 seconds. Synchronization is observed through experiments. In the example, a time difference of 0 seconds (less than the threshold) is directly compared. Data, such as state values, is obtained from sensors, and probability values ​​are obtained from sequences. The example acquisition process involves parsing sequence data, calling the state at a specific time point, comparing strings or values, with a probability difference range of 0-1 and a deviation difference range of -10 to 10 being reasonable. The threshold setting references time alignment tolerance. An experimental time offset threshold of 300 seconds is set. Example calculation: a test time difference of 200 seconds (less than 300 seconds) determines the same time period. Weights or coefficients are not involved. Actual data examples: probability values ​​0.4737, 0.5, 0.6, probability difference 0.1263, deviation value 3.254, deviation difference -1.746, generating a collaborative probability deviation.

[0082] S303: Based on the coordination probability deviation, the state sequences of multiple construction machines at the same time are jointly arranged, the state combinations are integrated according to the coordination probability deviation value, the state mapping results of multi-machine coordination behavior are analyzed, and the coordination state of the construction site is generated.

[0083] Based on the collaborative probability discrete quantity, such as {probability difference: 0.1263, deviation difference: -1.746}, the state sequences of multiple construction machines at the same time are jointly arranged. Specific actions include calling time series data such as [Time 08:00:00, Machine A is operating], [Time 08:00:00, Machine B is waiting], [Time 08:00:00, Machine C is operating], sorting the timestamp values ​​in ascending order, integrating state values, integrating state combinations based on the collaborative probability discrete quantity values, calling the probability difference 0.1263 and deviation difference -1.746, comparing their values; if the probability difference is less than the threshold of 0.2, it is considered low difference. The integrated state combinations, such as mapping {operating, waiting, operating} to "mixed operation state", analyze the state mapping results of multi-machine collaborative behavior, call the integrated state, analyze the behavior pattern, such as low probability difference indicating high collaboration, and generate the construction site collaborative state. For example, in the case of "high-coordination mixed operation," the probability difference between the "coordination" and "coordination" intervals is defined as follows: 0-0.1 indicates high coordination, 0.1-0.3 indicates medium coordination, and above 0.3 indicates low coordination. A threshold of 0.2 is set. Coordination efficiency is tested experimentally. In the example, a probability difference of 0.1263 less than 0.2 indicates high coordination. Data such as the state sequence is obtained from S203, and the coordination quantity is obtained from S302. The example acquisition process is as follows: call the state values ​​at the time point, combine and arrange them to generate an array, call the coordination quantity value, compare the threshold, integrate the state strings, and generate the result. The threshold setting references historical coordination data. The average probability difference is calculated to be 0.15 through experiments, and the threshold is set to 0.2. Example calculation: a probability difference of 0.18 less than 0.2 indicates high coordination. Weights or coefficients are not involved. Actual data values ​​include time value 08:00:00, state value "operation," and coordination quantity value 0.1263. The resulting construction site coordination status is generated.

[0084] Please see Figure 5 The specific steps for obtaining construction status change data are as follows:

[0085] S401: Based on the collaborative status of the construction site, sort the collaborative status values ​​corresponding to multiple time points according to the continuous monitoring cycle, and perform position verification on the collaborative status values ​​of adjacent cycles to generate a collaborative status time series.

[0086] Based on the collaborative status of the construction site, such as the high-coordination mixed operation status value, in the example of the highway expansion project, the numerical quantification standard for collaborative status is: high collaboration is assigned a value of 0.8-1.0, medium collaboration 0.5-0.7, and low collaboration 0-0.4. The average collaborative value of historical data is set to 0.6. The calculation example shows that a test collaboration value of 0.7 falls within the 0.5-0.7 range and is assigned a medium collaboration value. The collaborative status values ​​corresponding to multiple time points are sorted by time according to the continuous monitoring cycle. Specific actions include calling monitoring cycle data, such as recording data from the sensor at time point 08:00:00. At 08:30:00 and 09:00:00, the corresponding collaborative state values, such as 0.8, 0.7, and 0.6, are called. The timestamp values ​​are sorted in ascending order to generate the sequence: [Time 08:00:00, Value 0.8], [Time 08:30:00, Value 0.7], [Time 09:00:00, Value 0.6]. In this example, the time interval is 30 minutes, and the period is defined with a time difference threshold of 600 seconds. The continuity of the period is observed experimentally. A continuous period is determined when the time difference is less than the threshold (500 seconds). The position of the collaborative state values ​​in adjacent periods is also verified. The specific actions include calling adjacent period values, such as 0.8 at 08:00:00 and 0.7 at 08:30:00, comparing the absolute value of the difference (0.1), verifying the continuity of the position, setting a position verification threshold difference of 0.2, and testing value jumps through experiments. The example calculation shows that a difference of 0.15 less than 0.2 indicates continuous position. Data, such as collaborative status values, is obtained from the output of S303, with dimensionless units. The example acquisition process involves parsing the status string for a high-coordination mixed operation, comparing it to a preset interval and assigning a value of 0.8, and measuring the time point using a GPS sensor. Record timestamps, with a reasonable value range of 0-1. The threshold setting is based on the tolerance for numerical fluctuations. Through experiments, the historical average difference is calculated to be 0.18, and the threshold is set to 0.2. Examples of actual data values: time point 08:00:00, value 0.8; time point 08:30:00, value 0.7; difference calculation 0.1; position verification passed; generate a coordinated state time series such as array [time 08:00:00, value 0.8], [time 08:30:00, value 0.7], [time 09:00:00, value 0.6].

[0087] S402: Based on the time series of the coordinated state, the time series index value and the corresponding coordinated state value are used as input items to calculate the relationship between the time index and the coordinated state value, and to calculate the slope of the coordinated state change.

[0088] Based on the time series of co-states, such as {[time index 1, value 0.8], [time index 2, value 0.7], [time index 3, value 0.6]}, the time index values ​​are assigned according to the time sequence, starting from 1 and increasing sequentially. For example, time 08:00:00 index 1, 08:30:00 index 2, 09:00:00 index 3. The time series index values ​​and corresponding co-state values ​​are used as input items. The specific actions include calling the index values ​​1, 2, 3 and the values ​​0.8, 0.7, 0.6 to form input pairs such as (1, 0.8), (2, 0.7), (3, 0.6). The relationship between the time index and the co-state values ​​is calculated through a linear regression model. The specific actions include calculating the slope and calling the index values ​​and values.

[0089] Calculate the index average (1+2+3) / 3=2, the numerical average (0.8+0.7+0.6) / 3=0.7, and the slope formula: numerator sum((index i - index average) x (numerical i - numerical average)), denominator sum((index i - index average)^2). Example: When index i=1, (1-2)(0.8-0.7)=(-1)x0.1=-0.1; when index i=2... (2-2)(0.7-0.7)=0x0=0, when index i=3, (3-2)(0.6-0.7)=1x(-0.1)=-0.1, numerator sum=-0.1+0+-0.1=-0.2, denominator sum((1-2)^2+(2-2)^2+(3-2)^2)=1+0+1=2, slope=-0.2 / 2=-0.1, calculate the slope of the state change. The slope of the cooperative state change is obtained as -0.1. The absolute value of the slope in the change judgment interval is 0-0.05, which is considered stable; 0.05-0.1 is considered slow change; and above 0.1 is considered rapid change. A threshold of 0.05 is set. The slope range is tested through experiments. In the example, the absolute value of the slope -0.1 is greater than 0.05, which is considered a slow change. Data such as index values ​​are obtained sequentially from the time series, and numerical values ​​are obtained from the sequence. The acquisition process of the example is as follows: parse the sequence array, assign incrementing index values, call the numerical values, and calculate the average and difference. The slope unit is dimensionless, and the slope range of -1 to 1 is reasonable. The threshold setting is based on historical slope fluctuations. The average absolute value of the slope is calculated to be 0.08 through experiments, and the threshold is set to 0.05. Example of actual data values: index 1, value 0.8; index 2, value 0.7; index 3, value 0.6; slope calculation -0.1.

[0090] S403: Based on the slope of the collaborative state change, combined with the numerical distribution of multiple time points in the collaborative state time series, the positive and negative directions of the slope and the magnitude of the numerical change are judged, the trend of collaborative state change over time is sorted out, and construction state change data is generated.

[0091] Based on the slope of the cooperative state change, such as -0.1, and combined with the numerical distribution of multiple time points in the cooperative state time series, we call [Time Index 1, Value 0.8], [Time Index 2, Value 0.7], [Time Index 3, Value 0.6], calculate the standard deviation of the numerical distribution, call the values ​​0.8, 0.7, 0.6, the mean is 0.7, and the sum of squared differences is ((0.8-0.7)^2+(0.7-0.7)^2+(0.6-0.7)^2)=0.01+0+0. Given 0.01 = 0.02, and the standard deviation sqrt(0.02 / 3) = sqrt(0.0067) ≈ 0.082, the slope's positive and negative directions and the magnitude of its change are judged. Specific actions include calling the slope -0.1, comparing positive and negative values ​​(negative values ​​indicate a downward trend), calling the absolute value of the change magnitude 0.1, and comparing the magnitude threshold 0.05. Example: an magnitude of 0.1 greater than 0.05 indicates a significant change. The trend of the coordinated state over time is then analyzed, specifically integrating the negative slope direction and magnitude. 0.1, generating a description such as a slow downward trend. In the example, the time series value drops from 0.8 to 0.6, with a consistent trend. For positive and negative direction determination, a slope greater than 0 is positive, and less than 0 is negative. The absolute value of the amplitude is 0-0.05 for small amplitude, 0.05-0.1 for medium amplitude, and above 0.1 for large amplitude. An amplitude threshold of 0.05 is set. Trend stability is tested experimentally. In the example, an amplitude of 0.1 falling within the 0.05-0.1 range is considered a medium-amplitude decline. Data such as the slope is obtained from S402, and the numerical distribution is obtained from the sequence calculation. The example acquisition process is: calling the slope value, calling the numerical array, calculating the distribution index, assigning the trend string, and setting the threshold based on the trend change tolerance. Through experimental observation, the historical average amplitude is 0.07, and the threshold is set to 0.05. Example of actual data values: slope -0.1, values ​​0.8, 0.7, 0.6, amplitude 0.1. This generates construction status change data such as {trend direction: negative, change amplitude: 0.1, trend description: slow decline}.

[0092] Please see Figure 6 The specific steps for obtaining the construction site status monitoring results are as follows:

[0093] S501: Based on construction status change data, acquire multiple construction machinery operation status records, use the operation status as nodes, connect adjacent statuses according to time sequence, mark the time point of the status occurrence and record the duration interval, and generate a state evolution chain sequence.

[0094] Based on construction status change data, a call is made to retrieve the operating status records of multiple construction machines, such as {trend direction: negative, change magnitude: 0.1, trend description: slow decline}. Specific actions include calling data from construction site sensors, such as current sensors recording timestamps and status values ​​(operating or waiting). In the example of a highway expansion project, the machines include excavator A and road roller B. The operating status records are output from S201 as the sequence [Time 08:00:00, Status: Operating, Duration 30 minutes], [Time 08:30:00, Status: Waiting, Duration 5 minutes]. The operating status is used as a node, and the node is defined as a status string value (e.g., assigned to operating status). Value Node 1), connects adjacent states according to time sequence, sorts timestamp values ​​in ascending order, calculates the time difference between the end time of the previous node and the start time of the current node to determine whether continuity is satisfied, sets a time difference threshold of 300 seconds (referring to the construction cycle continuity experiment, testing an interruption interval of 200 seconds less than 300 seconds to determine adjacency; example calculation: the interval between the end of the previous state and the start of the current state is 0 seconds less than 300 seconds, a connected state), marks the state occurrence time point, records the start time such as 08:00:00, records the duration interval, calculates the end time minus the start time (in minutes), data acquisition process: sensor detection current exceeding 500 amperes is... The task begins, ends when the current drops below 100 amperes, and a timestamp is recorded. The duration is calculated based on the difference. Example: Start 08:00:00, End 08:30:00, Difference 30 minutes. Status value quantification standard: Task assigned 1, Waiting assigned 0, value range 0-1 is reasonable. For example, adjacent nodes are considered adjacent if the time difference is 0-300 seconds, and non-adjacent if it is greater than 300 seconds. A threshold of 300 seconds is set. Historical data transmission delays and minor interruptions are observed through experiments. Actual data value example: Node 1 status: Task (08:00:00 to 08:30:00), Node 2 status: Waiting (starting at 08:30:00), calculate the connection time. Difference: The difference between the start time of node 2 (08:30:00) and the end time of node 1 (08:30:00) equals 0 seconds. If the time difference is less than 300 seconds, the nodes are considered adjacent. A connection is successfully established, and a state evolution chain sequence is generated as an array [Node 1: Status Operation, Start Time 08:00:00, End Time 08:30:00, Duration 30 minutes], [Node 2: Status Waiting, Start Time 08:30:00, End Time 09:00:00, Duration 30 minutes]. See Table 2, the construction machinery operation status record table. Table 2 lists the example data, including time points, status values, and durations, for subsequent processing.

[0095] Table 2 Construction Machinery Operation Status Record Sheet

[0096] ;

[0097] Table 2 shows the operational status record data of construction machinery, including time points, status values, and durations, which are used to generate a state evolution chain sequence.

[0098] S502: Based on the state evolution chain sequence, for multiple time periods covered by the state nodes, collect the running state identifier values ​​within the corresponding time periods, determine the continuation or change of the state nodes between adjacent time periods, replace or retain the state labels, and obtain the state segment labeling results.

[0099] Based on the state evolution chain sequence, such as [Node 1: Status: Job, Start 08:00:00, End 08:30:00], [Node 2: Status: Waiting, Start 08:30:00, End 09:00:00], for multiple time periods covered by the state nodes, the time periods are defined as fixed intervals of 30 minutes, such as Time Period 1: 08:00:00-08:30:00, Time Period 2: 08:30:00-09:00:00. The running status identifier values ​​within the corresponding time periods are collected, and the status string is assigned a numerical value (Job assigned 0.8, Waiting assigned 0.3; the quantification standard is based on the status influence weight; example: Job with high influence assigned 0.8, Waiting with low influence assigned 0.3). The continuation or change of the status node between adjacent time periods is determined. The identifier value of Time Period 1 (0.8) and Time Period 2 (0.3) are called, the absolute value of the difference is calculated (0.5), and the status difference threshold is set to 0.1 (referencing historical data). The average data difference is 0.15. Experimental tests show a difference of 0.12, which is greater than 0.1 and indicates a change. Example calculation: a test difference of 0.08 (less than 0.1) indicates continuation; a difference of 0.5 (greater than the threshold of 0.1) indicates a change. Status labels are either replaced or retained. Specific actions include replacing the label with the new status string if a change occurs, otherwise retaining the original label. Example: Time period 1 status "Operating" is retained; Time period 2 status "Waiting" is replaced with "Inefficient Waiting". Data acquisition process: Parse the evolution chain sequence node status strings, assign numerical values ​​(dimensionless, range 0-1 is reasonable). For example, a continuation judgment interval difference of 0-0.1 indicates continuation, and above 0.1 indicates a change. Actual data numerical examples: Time period 1 label value 0.8, Time period 2 label value 0.3, difference 0.5, change judgment, resulting in status segment labeling results such as {Time period 1: Status "Operating", label retained}, {Time period 2: Status "Waiting", label replaced with "Inefficient Waiting"}.

[0100] S503: Based on the state segment labeling results, combined with the time span of multiple nodes in the state evolution chain sequence, the updated state labels are summarized in the order of time periods to generate the construction site state monitoring results.

[0101] Based on the state segment labeling results, such as {Time Segment 1: State Operation, label retained}, {Time Segment 2: State Waiting, label replaced with Inefficient Waiting}, combined with the time span of multiple nodes in the state evolution chain sequence, the S501 output node duration intervals are called, such as excavator A lasting 30 minutes (08:00:00-08:30:00) and road roller B lasting 5 minutes (08:30:00-08:35:00). The minimum time span is calculated to be 5 minutes, the maximum to be 30 minutes, and the difference to be 25 minutes. The updated state labels are then summarized in time segment order and adjusted. Sort the time intervals in ascending order (08:00:00-08:30:00 is segment 1, 08:30:00-09:00:00 is segment 2), integrate the annotation strings, in the example segment 1 is labeled as operation (corresponding to excavator A), segment 2 is labeled as inefficient waiting (corresponding to road roller B), the time span is determined as 0-10 minutes as short span, 10-30 minutes as medium span, and more than 30 minutes as long span, set the span threshold to 10 minutes, and observe the historical average span through experiments to 15 minutes. Actual data example: excavator A span of 30 minutes (medium span, Because 30 minutes is the upper limit of the threshold), the span of roller B is 5 minutes (short span), and the span of crane C is 25 minutes (medium span). Status labels are integrated, and time periods are divided as follows: Segment 1: 08:00:00-08:30:00, Segment 2: 08:30:00-09:00:00. The labels are summarized by time period: Segment 1 label: Operation (corresponding to excavator A), Segment 2 label: Inefficient waiting (corresponding to roller B, but needs to be integrated with crane C data). Span and status are integrated: Segment 1 integration: Operation (medium span), Segment 2 integration: Calculate the percentage of operation: Crane C time. When the length ratio is 25 / (5+25)=83.3% and the dominant state (operation) ratio is >60%, the dominant state is marked, and a fusion label is generated: inefficient waiting and operation mixed (short + medium span), monitoring period 08:00:00-08:30:00, state: operation (medium span), equipment: excavator A continuously operating for 30 minutes, monitoring period 08:30:00-09:00:00, state: inefficient waiting and operation mixed, road roller B inefficient waiting (short span, 5 minutes), crane C operation (medium span, 25 minutes, dominant state).

[0102] An Internet of Things (IoT)-based highway construction site status monitoring system is provided for executing the aforementioned IoT-based highway construction site status monitoring method. The system includes:

[0103] The data analysis module collects the start and end times and waiting durations of multiple construction machines at the highway construction site, aligns them with timestamps, extracts the number of start times, calculates the operation frequency of construction machines, generates an operation time series, and transmits it to the status analysis module.

[0104] The status analysis module analyzes the duration of the operation based on the operation time series, extracts the relationship between the status changes of construction machinery in adjacent time periods, calculates the transition probability between the operation status and the waiting status, generates the construction operation status sequence and transmits it to the collaborative computing module.

[0105] The collaborative computing module obtains the construction operation status sequence corresponding to multiple construction machines in the construction site, calculates the dispersion of the operation status transition probability and operation duration of multiple construction machines, generates the collaborative status of the construction site, and transmits it to the trend evaluation module.

[0106] The trend assessment module sorts the collaborative status of the construction site by time according to the continuous monitoring cycle, calculates the slope of status change, analyzes the trend characteristics of collaborative status change over time, generates construction status change data, and transmits it to the status monitoring module.

[0107] The status monitoring module analyzes the construction site status evolution chain with the operating status of construction machinery as the node based on construction status change data. It marks and updates the operating status of the construction site in multiple time periods in the status evolution chain and generates construction site status monitoring results.

[0108] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for monitoring the status of highway construction sites based on the Internet of Things, characterized in that, Includes the following steps: S1: Collect the start and end times and waiting durations of multiple construction machines at the highway construction site and align them by timestamps. Extract the number of start times and calculate the frequency of construction machine operation to generate an operation time series. S2: Based on the operation time series analysis, analyze the operation duration, extract the relationship between the changes in the state of construction machinery in adjacent time periods, calculate the transition probability between the operation state and the waiting state, and generate a construction operation state sequence. S3: Obtain the construction operation state sequence corresponding to multiple construction machines in the construction site, calculate the dispersion of the operation state transition probability and operation duration of multiple construction machines, and generate the construction site collaborative state; The deviation of the operation duration is calculated using the following formula: ； in, This represents the deviation in the duration of the task. Represents the number of construction machines. Representing the The operating duration of the Taiwanese machinery, This represents the average duration of all mechanical operations. Representing the The state transition probability of the machine. The nonlinear discrete amplification index used when the duration of the operation deviates from the average value. The representative exponent parameter indicates the weight of the state transition probability on the deviation from the exponent. S4: Sort the collaborative status of the construction site by time according to the continuous monitoring cycle, calculate the slope of status change, analyze the trend characteristics of collaborative status change over time, and generate construction status change data. S5: Based on the construction status change data, analyze the construction site status evolution chain with the construction machinery operation status as the node, mark and update the operation status of the construction site in multiple time periods in the status evolution chain, and generate construction site status monitoring results.

2. The method for monitoring the status of highway construction sites based on the Internet of Things according to claim 1, characterized in that, The operation time series includes operation start time, operation end time, waiting duration, number of operation starts, and construction machinery operation frequency. The construction operation state sequence includes operation state, waiting state, relationship between construction machinery state changes in adjacent time periods, and probability of transition between operation state and waiting state. The construction site coordination state includes construction machinery operation state sequence and operation duration dispersion. The construction state change data includes state change slope and trend characteristics of coordination state changes over time. The construction site state monitoring results include state evolution chain, construction site state labeling and updating, and operation state set within a time period.

3. The method for monitoring the status of highway construction sites based on the Internet of Things according to claim 1, characterized in that, The specific steps for obtaining the job time series are as follows: S101: Collect the start time, end time and waiting time of multiple construction machines at the highway construction site, monitor the changes in the operating status of multiple machines and record the corresponding timestamps, perform sequence correction and position matching of the multi-source timestamps according to a unified time reference, and obtain an aligned time identifier sequence. S102: Based on the aligned time identifier sequence, obtain the state switching records of multiple construction machines between adjacent time identifiers, perform counting processing on the start state identifier, calculate the corresponding ratio according to the number of times a single machine is started and the length of the observation period, form a frequency set distinguished by machine, and generate the machine operation frequency. S103: Based on the alignment time identifier sequence and the mechanical operation frequency, the time identifier sequence is called to perform position mapping of the start events, embed the frequencies of multiple mechanical start events into the corresponding time periods and rearrange them in time order to generate an operation time sequence.

4. The method for monitoring the status of highway construction sites based on the Internet of Things according to claim 1, characterized in that, The specific steps for obtaining the construction operation status sequence are as follows: S201: Based on the construction site operation time series, extract the start time and end time of adjacent operations of construction machinery and perform difference calculation, analyze the duration of corresponding operations, extract the end time and start time of adjacent operations and calculate the interval duration, use as the state input set to obtain the operation waiting duration; S202: Based on the operation waiting duration, obtain the status identifier changes of the construction machinery in adjacent time periods, align the operation status and waiting status with the execution status, calculate the transition probability between the operation status and the waiting status, and generate the state transition probability. S203: Invoke the state transition probability, arrange the state nodes according to the time sequence of the construction machinery, and perform sequence mapping between the operation state and the waiting state according to the probability correspondence to establish the construction operation state sequence.

5. The method for monitoring the status of highway construction sites based on the Internet of Things according to claim 1, characterized in that, The specific steps for obtaining the collaborative status of the construction site are as follows: S301: Obtain the construction operation state sequence of multiple construction machines in the construction site, extract the state transition probability and operation duration of each construction machine, and calculate the operation duration deviation based on the state sequence of multiple construction machines; S302: Based on the deviation of the operation duration, call the state transition probabilities corresponding to multiple construction machines, compare the state combinations of multiple machines in the same time period, calculate the difference in the state transition probabilities of multiple machines and the difference in the duration deviation, and generate the collaborative probability deviation. S303: Based on the aforementioned coordination probability deviation, the state sequences of multiple construction machines at the same time are jointly arranged, the state combinations are integrated according to the coordination probability deviation value, the state mapping results of multi-machine coordination behavior are analyzed, and the coordination state of the construction site is generated.

6. The method for monitoring the status of highway construction sites based on the Internet of Things according to claim 1, characterized in that, The specific steps for obtaining the construction status change data are as follows: S401: Based on the collaborative status of the construction site, sort the collaborative status values ​​corresponding to multiple time points according to the continuous monitoring cycle, and perform position verification on the collaborative status values ​​of adjacent cycles to generate a collaborative status time series. S402: Based on the time series of the coordinated state, the time series index value and the corresponding coordinated state value are used as input items to calculate the relationship between the time index and the coordinated state value, and to calculate the slope of the coordinated state change. S403: Based on the slope of the coordinated state change, and combined with the numerical distribution of multiple time points in the coordinated state time series, the positive and negative directions of the slope and the magnitude of the numerical change are judged, the trend of the coordinated state change over time is sorted out, and construction state change data is generated.

7. The method for monitoring the status of highway construction sites based on the Internet of Things according to claim 1, characterized in that, The specific steps for obtaining the construction site status monitoring results are as follows: S501: Based on the construction status change data, obtain multiple construction machinery operation status records, take the operation status as nodes, connect adjacent states according to the time sequence, mark the time point of the state occurrence and record the duration interval, and generate a state evolution chain sequence. S502: Based on the state evolution chain sequence, for multiple time periods covered by the state nodes, collect the running state identifier values ​​within the corresponding time periods, determine the continuation or change of the state nodes between adjacent time periods, replace or retain the state labels, and obtain the state segment labeling results. S503: Based on the state segment labeling results and combined with the time span of multiple nodes in the state evolution chain sequence, the updated state labels are summarized in time period order to generate the construction site state monitoring results.

8. The method for monitoring the status of highway construction sites based on the Internet of Things according to claim 7, characterized in that, In the process of connecting adjacent states according to time sequence, marking the time point of state occurrence, and recording the duration interval, when adjacent running state records maintain the same running state identifier value for a continuous interval of no less than a preset fixed duration, they are merged into the same state node, and the time point of the first running state record is taken as the state occurrence time point, and the time difference between the time point of the last running state record and the time point of the first running state record is taken as the duration interval.

9. A highway construction site status monitoring system based on the Internet of Things, characterized in that, A method for monitoring the status of a highway construction site based on the Internet of Things, according to any one of claims 1-8, wherein the system comprises: The data analysis module collects the start and end times and waiting durations of multiple construction machines at the highway construction site, aligns them with timestamps, extracts the number of start times, calculates the operation frequency of construction machines, generates an operation time series, and transmits it to the status analysis module. The status analysis module analyzes the duration of the operation based on the operation time series, extracts the relationship between the status changes of construction machinery in adjacent time periods, calculates the transition probability between the operation status and the waiting status, generates a construction operation status sequence, and transmits it to the collaborative computing module. The collaborative computing module obtains the construction operation state sequence corresponding to multiple construction machines in the construction site, calculates the dispersion of the operation state transition probability and operation duration of multiple construction machines, generates the collaborative state of the construction site, and transmits it to the trend evaluation module. The trend assessment module sorts the collaborative status of the construction site by time according to the continuous monitoring cycle, calculates the slope of status change, analyzes the trend characteristics of collaborative status change over time, generates construction status change data, and transmits it to the status monitoring module. The status monitoring module analyzes the construction site status evolution chain with the construction machinery operation status as the node based on the construction status change data, marks and updates the operation status of the construction site in multiple time periods in the status evolution chain, and generates construction site status monitoring results.

Images