Sensor-based water-stable paver operation data acquisition and segregation early warning system

Abstract

This invention relates to the field of early warning technology, specifically to a sensor-based data acquisition and segregation early warning system for water-stabilized paver operation. The system includes a processor and a memory. The processor executes a computer program stored in the memory to perform the following steps: acquiring the temperature vibration coupling intensity, segregation confidence level, target area, and corresponding persistent area sequence of the target area at the current monitoring time; obtaining the trend persistence characterization value of the target area based on the persistent area sequence; and obtaining the segregation risk value of the target area based on the trend persistence characterization value, segregation confidence level, area, temperature vibration coupling intensity, and total data quantity in the target area. Based on the segregation risk value, a segregation early warning is provided for the water-stabilized paver's paving process. Furthermore, this invention can improve the accuracy and reliability of segregation early warning during water-stabilized paver paving.

Description

Technical Field

[0001] This invention relates to the field of early warning technology, specifically to a sensor-based system for collecting and analyzing operational data of a water-stabilized paver and providing early warning of segregation. Background Technology

[0002] Segregation during the paving process of cement-stabilized aggregate is closely related to the quality of the paved road. Segregation significantly reduces the density, strength, and durability of the base layer, thus affecting the overall performance of the pavement. Therefore, to ensure road quality, it is crucial to collect operational data from the cement-stabilized aggregate paver and implement segregation early warning during the paving process. Segregation refers to the uneven particle size distribution and separation of coarse and fine aggregates in cement-stabilized crushed stone mixtures due to external forces or improper construction techniques. Specifically, it manifests as increased porosity in areas with concentrated coarse aggregates and insufficient filling of fine aggregates, resulting in localized loose or banded distributions.

[0003] Current technologies typically rely on comparing paving temperature data collected by sensors with a preset temperature range to achieve segregation warnings. That is, a segregation warning is issued when the collected paving temperature data exceeds the preset range. However, segregation during water-stabilized paving is a complex physical process influenced by multiple factors, such as paver speed, vibration, and paving temperature changes. Existing methods that rely on only a single dimension of data cannot comprehensively and accurately reflect the true segregation situation, leading to low accuracy and reliability in segregation warnings. For example, a paving temperature outside the preset range might be a normal phenomenon caused by varying material cooling rates due to changes in paver speed, rather than true segregation. Therefore, relying solely on paving temperature for segregation warnings can easily result in false alarms. Improving the accuracy and reliability of segregation warnings is therefore a pressing issue that needs to be addressed. Summary of the Invention

[0004] To address the aforementioned problems, this invention provides a sensor-based system for acquiring and monitoring segregation in the operation data of a water-stabilized paver. The specific technical solution adopted is as follows: One embodiment of the present invention provides a sensor-based system for collecting and monitoring the operational data of a water-stabilized paver and for early warning of segregation, comprising a processor and a memory. The processor executes a computer program stored in the memory to perform the following steps: During the paving process, the current paving temperature data, current vibration data, current paving speed data, and current temperature baseline threshold, current vibration baseline threshold, and speed baseline threshold range are acquired. Based on the current temperature baseline threshold and current vibration baseline threshold, the abnormal amplitude of the current paving temperature data and current vibration data is obtained. Based on the abnormal amplitude, the current abnormal data is obtained, which includes abnormal paving temperature data and abnormal vibration data. Based on the number of current abnormal data within a preset neighborhood of the sensor's location coordinates that collects current abnormal data, the area to be analyzed at the current monitoring time is obtained. Based on the abnormal paving temperature data and abnormal vibration data in the area to be analyzed at the current monitoring time, the temperature-vibration coupling intensity of the area to be analyzed at the current monitoring time is obtained. Based on the proportion of different data, temperature-vibration coupling intensity, current paving speed data, speed baseline threshold range, and abnormal amplitude in the area to be analyzed at the current monitoring time, the segregation confidence level of the area to be analyzed at the current monitoring time is obtained. Based on the intersection of each region to be analyzed at the current monitoring time and each region to be analyzed at the historical monitoring time, the target region at the current monitoring time and the corresponding continuous region sequence are obtained. The trend continuity characterization value of the target region is obtained based on the continuous region sequence. Based on the trend continuity characterization value of the target region, the segregation confidence, the area and temperature vibration coupling strength, and the total number of data in the target region, the segregation risk value of the target region is obtained. Based on the segregation risk value, a segregation warning is issued for the paving process of the water-stabilized paver.

[0005] Beneficial effects: This invention first obtains the area to be analyzed at the current monitoring time based on the number of current abnormal data within a preset neighborhood range of the sensor's location coordinates that collects current abnormal data. Then, based on the abnormal paving temperature and vibration data within the area to be analyzed at the current monitoring time, the temperature-vibration coupling intensity of the area to be analyzed at the current monitoring time is obtained. Finally, based on the proportion of different data, temperature-vibration coupling intensity, current paving speed data, speed baseline threshold range, and abnormal amplitude within the area to be analyzed at the current monitoring time, the segregation confidence level of the area to be analyzed at the current monitoring time is obtained. Next, based on the intersection between each area to be analyzed at the current monitoring time and each area to be analyzed at historical monitoring times, the target area and its corresponding continuous area sequence are obtained. The trend continuity characterization value of the target area is obtained from the continuous area sequence. Based on the trend continuity characterization value, segregation confidence level, area, temperature-vibration coupling intensity, and the total number of data in the target area, the segregation risk value of the target area is obtained. Finally, based on the segregation risk value, a segregation warning is issued for the paving process of the water-stabilized paver. Furthermore, the segregation risk value obtained by this invention based on parameters such as trend persistence characterization value, segregation confidence level, area and temperature vibration coupling strength can improve the accuracy and reliability of segregation warning when performing segregation warning during the paving process of water-stabilized paver. Attached Figure Description

[0006] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0007] Figure 1 This is a flowchart of a sensor-based method for collecting and predicting segregation in the operation data of a water-stabilized paver according to the present invention. Detailed Implementation

[0008] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the protection scope of the embodiments of the present invention.

[0009] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art.

[0010] This embodiment provides a sensor-based system for collecting and monitoring the operational data of a water-stabilized paver, including a processor and a memory. The processor executes a computer program stored in the memory to implement a sensor-based method for collecting and monitoring the operational data of a water-stabilized paver, such as... Figure 1 As shown, the sensor-based method for collecting operational data and providing early warning of segregation in water-stabilized pavers includes the following steps: Step S001: During the paving process, acquire the current paving temperature data, current vibration data, current paving speed data, and the current temperature baseline threshold, current vibration baseline threshold, and speed baseline threshold range. Based on the current temperature baseline threshold and the current vibration baseline threshold, obtain the abnormal amplitude of the current paving temperature data and the current vibration data. Based on the abnormal amplitude, obtain the current abnormal data, which includes abnormal paving temperature data and abnormal vibration data.

[0011] This embodiment mainly obtains the segregation risk value by fusing and analyzing data from three dimensions: paving temperature, vibration, and paving speed. Based on the segregation risk value, it provides segregation warnings during the paving process of the water-stabilized paver, thereby improving the accuracy and reliability of segregation warnings. Furthermore, for ease of understanding, this embodiment only provides segregation warnings during the paving process of any single water-stabilized paver, meaning that all subsequent data collected are from sensors on that water-stabilized paver.

[0012] Based on the above description, this embodiment needs to use the segregation risk value to provide segregation warning during the paving process of the water-stabilized paver. Paving temperature, vibration, and paving speed are the data support for obtaining the segregation risk value. Therefore, this embodiment needs to first obtain the paving temperature data, vibration data, and paving speed data of the water-stabilized paver at each monitoring moment during the paving process. The specific process for obtaining the paving temperature data, vibration data, and paving speed data at each monitoring moment is as follows: First, the sensors on the water-stabilized paver are configured. Temperature measurement points are evenly spaced along the paving width, 0.5 meters in front of the paver's screed. An infrared non-contact temperature sensor is placed at each measurement point to collect the paving temperature of the material. If the material being paved by the water-stabilized paver in this embodiment is cement-stabilized crushed stone, then the paving temperature of the material is the same as that of the cement-stabilized crushed stone. Vibration sensors are placed at key vibration points, such as the screed's vibration shaft, the tamping hammer, and the conveyor belt support, to collect vibration data during paving. These vibration sensors are triaxial accelerometers. A... A velocity sensor is used to collect the speed of the water-stabilized paver during paving. Then, data is collected from different sensors on the water-stabilized paver at different monitoring times. The data collected by the vibration sensor at each monitoring time is recorded as the vibration data to be processed at that time. The data collected by the sensor for material paving temperature at each monitoring time is recorded as the paving temperature data to be processed at that time. The data collected by the velocity sensor at each monitoring time is recorded as the paving speed data to be processed at that time. Then, the paving temperature data to be processed and the data to be processed are... The vibration data and the paving speed data to be processed are preprocessed. The preprocessed data are recorded as paving temperature data, vibration data, and paving speed data at the corresponding monitoring time, respectively. The paving temperature data, vibration data, and paving speed data at the current monitoring time are recorded as current paving temperature data, current vibration data, and current paving speed data, respectively. The number of current paving temperature data and current vibration data is greater than 1, that is, the current paving temperature data, current vibration data, and current paving speed data are the results of preprocessing the data collected by the sensors on the water-stabilized paver at the current monitoring time. Moreover, the various sensors on the water-stabilized paver... At the current monitoring time, each sensor collecting the material paving temperature can obtain a current paving temperature data; the speed sensor on the water-stabilized paver can obtain a current paving speed data; and each vibration sensor on the water-stabilized paver can obtain a current vibration data. If a certain current paving temperature data is the result of preprocessing the data collected by a certain temperature sensor on the water-stabilized paver used to collect the material paving temperature at the current monitoring time, then the sensor that collects the current paving temperature data is that temperature sensor. The same applies to the sensors that collect other data.

[0013] Preprocessing includes, but is not limited to, data filtering, data cleaning, and dimensionless removal. Dimensionless removal does not change the magnitude of the data values. In this embodiment, the sensors are set to acquire data synchronously. Additionally, the position coordinates of each sensor on the water-stabilized paver are obtained in real time through the GPS positioning system on the paver. In this embodiment, the coordinates of the sensor position coordinates are in the geodetic coordinate system. Alternatively, in other real-time methods, the coordinates of the sensor position coordinates can be considered in the world coordinate system. Furthermore, the coordinate values ​​of the sensor position coordinates in this embodiment change as the water-stabilized paver operates; in other words, the sensor position coordinates in this embodiment are not relative to the water-stabilized paver.

[0014] Segregation is a complex physical process influenced by a combination of factors, such as paver speed, vibration, and paving temperature changes. Segregation can occur when these factors deviate from normal values. Therefore, after obtaining the current paving temperature, vibration, and paving speed data, it's necessary to analyze any abnormal data within these data. The confidence level for segregation is then determined based on this abnormal data. However, due to varying construction conditions across different regions and seasons, relying on fixed thresholds for anomaly identification would result in low reliability. Therefore, to improve the reliability of anomaly identification, this embodiment will acquire thresholds for anomaly identification in real-time based on actual working conditions. These thresholds are the current temperature baseline threshold, current vibration baseline threshold, and current speed baseline threshold range. The speed baseline threshold range remains constant regardless of changes in working conditions. The specific process for acquiring these thresholds is as follows: The process of obtaining the current temperature baseline threshold: Due to heat exchange in the environment, the temperature changes over time, so the temperature baseline threshold should also change in real time. Therefore, the current temperature baseline threshold is determined by considering three factors: ambient temperature, heat exchange, and so on. First, the temperature of the material paved by the water-stabilized paver in an ideal environment is obtained and recorded as the first influencing value at the current monitoring time. The temperature of the material paved by the water-stabilized paver in an ideal environment is generally within a range. To ensure good workability and compaction of the material, the midpoint of the range is usually taken as the temperature of the material paved by the water-stabilized paver in an ideal environment. For example, if the paving temperature range of cement-stabilized crushed stone in an ideal environment is required to be 140 to 150 degrees Celsius, to ensure good workability and compaction of the material, the midpoint of 140 to 150 degrees Celsius, 145 degrees Celsius, is usually taken to represent the temperature of cement-stabilized crushed stone in an ideal environment. This value may vary slightly depending on the material ratio and the region. Then, the ambient temperature data at the current monitoring time is collected using a temperature sensor that collects ambient temperature data. The ambient temperature data at the current monitoring time is the ambient temperature at the location of the water-stabilized paver at the current monitoring time. The absolute value of the difference between the ambient temperature data at the current monitoring time and the preset reference ambient temperature is calculated and multiplied by the preset heat exchange coefficient. This result is recorded as the second influence value at the current monitoring time. That is, since there is direct heat exchange between the environment and the material, it will cause temperature changes. Since radiative heat exchange is approximately linear in the normal temperature range, the above is simplified using a linear model to obtain the second influence value for ease of calculation. In this embodiment, the preset reference ambient temperature and the preset heat exchange coefficient are empirical values. For example, the preset reference ambient temperature is taken as 25 degrees Celsius and the preset heat exchange coefficient is taken as 0.8. Next, the time interval from the completion of production of the paving material used by the water-stabilized paver to the current monitoring time is obtained. The unit of time interval here is minutes. The completion of paving material production usually refers to the cement-stabilized crushed stone (water-stabilized) pavement base material reaching a state that can be transported to the paving site after the mixing plant has completed the proportioning and mixing. The product of the time interval from the completion of production of the paving material used at the current monitoring time to the current monitoring time and the preset temperature change rate is calculated and used as the third influence value at the current monitoring time. That is, as the construction time goes by, the material temperature will gradually decrease, so the third influence value is obtained based on the time interval from the completion of production of the paving material used at the current monitoring time and the preset temperature change rate. The preset temperature change rate is taken as an empirical value, such as 0.1 degrees Celsius per minute. Finally, the sum of the first influence value, the second influence value and the third influence value at the current monitoring time is calculated and used as the current temperature baseline threshold.

[0015] The process of obtaining the current vibration baseline threshold is as follows: The time period formed by the current monitoring time and the preset historical monitoring time period before the current monitoring time is obtained and recorded as the time period to be analyzed. The mean and standard deviation of all vibration data at all monitoring times within the time period to be analyzed are recorded as the current vibration mean and current vibration standard deviation, respectively. The sum of the current vibration mean and a preset multiple of the current vibration standard deviation is used as the current vibration baseline threshold. The current vibration baseline threshold is determined based on statistical methods. In this embodiment, approximately 85% of the normal vibration data is selected. Therefore, the result of adding 1.5 times the current vibration standard deviation to the current vibration mean is the current vibration baseline threshold. In specific applications, the implementer needs to set the preset historical monitoring time period before the current monitoring time according to the actual situation. For example, in this embodiment, 60 seconds before the current monitoring time is used as the preset historical monitoring time period before the current monitoring time.

[0016] The process of obtaining the speed baseline threshold range: Obtain standard paving speed data, and denote the range formed by the preset values ​​of plus or minus a certain percentage of the standard paving speed data as the speed baseline threshold range. The preset value needs to be set by the implementer according to the actual situation. For example, in this embodiment, the preset value can be set to 15, then the speed baseline threshold range is... s represents the standard paving speed data. If the standard paving speed of a water-stabilized paver during operation is 1 to 3 meters per minute, then the median value of 1.5 meters is taken as the standard paving speed.

[0017] Therefore, this embodiment can obtain the current temperature baseline threshold, the current vibration baseline threshold, and the velocity baseline threshold range through the above calculation. After obtaining the current temperature baseline threshold, the current vibration baseline threshold, and the velocity baseline threshold range, the abnormal amplitude of the current paving temperature data and the abnormal amplitude of the current vibration data are obtained based on the current temperature baseline threshold and the current vibration baseline threshold. The absolute value of the difference between any current paving temperature data and the current temperature baseline threshold is the abnormal amplitude of the current paving temperature data, and the difference between any current vibration data and the current vibration baseline threshold is the abnormal amplitude of the current vibration data. Based on the abnormal amplitude of the current paving temperature data and the abnormal amplitude of the current vibration data, the abnormal paving temperature data and abnormal vibration data at the current monitoring time are obtained. The abnormal paving temperature data and abnormal vibration data at the current monitoring time are both current abnormal data.

[0018] The specific process for acquiring abnormal paving temperature data and abnormal vibration data at the current monitoring time is as follows: For any current paving temperature data, determine whether the abnormal amplitude of the current paving temperature data is greater than a preset first abnormal amplitude threshold. If it is greater, then the current paving temperature data is recorded as an abnormal paving temperature data at the current monitoring time. For any current vibration data, determine whether the abnormal amplitude of the current vibration data is greater than a preset second abnormal amplitude threshold. If it is greater, then the current vibration data is recorded as an abnormal vibration data at the current monitoring time. The implementer can set the preset first abnormal amplitude threshold according to engineering practice experience. For example, based on engineering practice experience, it is known that a temperature deviation exceeding 5 degrees Celsius will significantly affect the compaction quality. Therefore, in this embodiment, the preset first abnormal amplitude threshold can be set to 5 degrees Celsius. Since when the current vibration data exceeds the current vibration baseline threshold, it indicates that the current vibration data deviates from the normal vibration range. Therefore, in this embodiment, the preset second abnormal amplitude threshold is set to 0. That is to say, when the current vibration data is greater than the current vibration baseline threshold, the corresponding current vibration data is abnormal vibration data. In addition, since speed is a global quantity, it needs to be processed separately and is not involved in the subsequent multi-sensor joint analysis. That is, the current paving speed data and speed baseline threshold range are mainly used for the subsequent calculation of mechanical segregation matching degree.

[0019] Therefore, this embodiment can obtain all current abnormal data at the current monitoring time through the above process, or in other words, it can obtain all abnormal paving temperature data and all abnormal vibration data at the current monitoring time.

[0020] Step S002: Based on the number of current abnormal data within a preset neighborhood of the sensor's location coordinates that collected the current abnormal data, the area to be analyzed at the current monitoring time is obtained. Based on the abnormal paving temperature data and abnormal vibration data in the area to be analyzed at the current monitoring time, the temperature-vibration coupling intensity of the area to be analyzed at the current monitoring time is obtained. Based on the proportion of different data, temperature-vibration coupling intensity, current paving speed data, speed baseline threshold range, and abnormal amplitude in the area to be analyzed at the current monitoring time, the segregation confidence level of the area to be analyzed at the current monitoring time is obtained.

[0021] Since the occurrence of segregation is regional, meaning that the necessity for analysis is high when the number of abnormal data in a certain region is relatively dense, while analyzing a single abnormal data point is computationally difficult and not particularly necessary, this embodiment requires a multi-sensor joint analysis. This involves performing a joint analysis of multiple current abnormal data points to obtain the region to be analyzed at the current monitoring time and the temperature-vibration coupling strength of that region. The temperature-vibration coupling strength reflects the compatibility with different types of segregation and is a crucial parameter for obtaining the segregation confidence level. Based on the above description, this embodiment requires a joint analysis of multiple current abnormal data points to obtain the region to be analyzed at the current monitoring time. Specifically, this embodiment requires determining the region to be analyzed at the current monitoring time based on the number of current abnormal data points within a preset neighborhood of the sensor's location coordinates. The specific process for obtaining the region to be analyzed at the current monitoring time is as follows: First, the position coordinates of the sensors collecting each current abnormal data at the current monitoring time are recorded as the sensor position points corresponding to the current abnormal data. That is, the coordinate values ​​of the sensor position points for the current abnormal data are composed of the position coordinates of the sensors collecting the corresponding current abnormal data at the current monitoring time. Then, each current abnormal data is traversed to obtain all unprocessed regions corresponding to the current monitoring time. Specifically, for any current abnormal data *r*, the number of current abnormal data within a preset first neighborhood of the sensor position point of *r* is obtained, and it is determined whether the number of current abnormal data within the preset first neighborhood of the sensor position point of *r* is greater than or equal to a preset data quantity threshold. If it is greater, the preset first neighborhood of the sensor position point of *r* is taken as an unprocessed region corresponding to the current monitoring time. Furthermore, in specific applications, the implementer can adjust the parameters according to the transmission... The sensor sets a preset first neighborhood range and a preset data quantity threshold based on actual conditions such as distance during sensor time. For example, in this embodiment, the preset first neighborhood range can be set to a range with a radius of 1 meter, and the preset data quantity threshold can be set to 2. Then, all regions that have intersections in the regions to be processed corresponding to the current monitoring time are merged, and all regions obtained after merging are recorded as regions to be analyzed at the current monitoring time. There is no intersection between regions to be analyzed at the current monitoring time. For example, if the regions to be processed corresponding to the current monitoring time are region 1, region 2, region 3, and region 4, where there is an intersection between region 1 and region 2, an intersection between region 3 and region 4, and no intersection between region 1 and region 2 and between region 3 and region 4, then the region after merging region 1 and region 2 can be recorded as the region to be analyzed, and the region after merging region 3 and region 4 can also be recorded as the region to be analyzed.

[0022] After obtaining the area to be analyzed at the current monitoring time, the current abnormal data in the area to be analyzed at the current monitoring time is obtained, that is, the abnormal paving temperature data and abnormal vibration data in the area to be analyzed at the current monitoring time are obtained. And there is only the current abnormal data in the area to be analyzed at the current monitoring time, and there is no other data. For example, for the abnormal paving temperature data v1, abnormal paving temperature data v2, and abnormal vibration data v3 at the current monitoring time, the position coordinates of the sensors that collected the abnormal paving temperature data v1, abnormal paving temperature data v2, and abnormal vibration data v3 at the current monitoring time are all located in the area to be analyzed at the current monitoring time. Then, the abnormal paving temperature data v1 and abnormal paving temperature data v2 are the abnormal paving temperature data in the area to be analyzed A, and the abnormal vibration data v3 is the abnormal vibration data in the area to be analyzed A. Then, based on the abnormal paving temperature data and abnormal vibration data of the area to be analyzed at the current monitoring time, the temperature-vibration coupling strength of the area to be analyzed at the current monitoring time is obtained. The temperature-vibration coupling strength is a key parameter for obtaining the segregation confidence level. Therefore, the specific process for obtaining the temperature-vibration coupling strength of the area to be analyzed at the current monitoring time is as follows: The comprehensive coupling strength of each abnormal paving temperature data in any analysis area A at the current monitoring time is obtained, and the average of the comprehensive coupling strengths of all abnormal paving temperature data in analysis area A is taken as the temperature vibration coupling strength of analysis area A; and the specific process for obtaining the comprehensive coupling strength of any abnormal paving temperature data a in analysis area A is as follows: First, within the analysis area A, all abnormal vibration data located within a preset second neighborhood of the sensor location point of the abnormal paving temperature data a and belonging to the analysis area A are acquired and recorded as the associated abnormal vibration data corresponding to the abnormal paving temperature data a. That is, the associated abnormal vibration data corresponding to the abnormal paving temperature data a belongs to the analysis area A. In specific applications, the implementer needs to set the preset second neighborhood range based on engineering experience and other practical conditions. For example, in this embodiment, a range with a radius of 2 meters can be used as the preset second neighborhood range. Then, based on the Gaussian kernel function between the sensor location point of the abnormal paving temperature data a and the sensor location points of each associated abnormal vibration data corresponding to the abnormal paving temperature data a, and the absolute value of the abnormal amplitude of the abnormal paving temperature data a... The coupling strength between abnormal paving temperature data a and its associated abnormal vibration data is obtained by taking the absolute values ​​of the abnormal amplitudes of the abnormal paving temperature data a and the associated abnormal vibration data corresponding to the abnormal paving temperature data a. In this embodiment, the coupling strength between abnormal paving temperature data a and its associated abnormal vibration data b is recorded as the product of the Gaussian kernel function between the sensor location point of abnormal paving temperature data a and the sensor location point of the b-th associated abnormal vibration data corresponding to abnormal paving temperature data a, the absolute value of the abnormal amplitude of abnormal paving temperature data a, and the absolute value of the abnormal amplitude of the b-th associated abnormal vibration data. Therefore, the specific expression for calculating the coupling strength between abnormal paving temperature data a and its associated abnormal vibration data b is as follows: in, The coupling strength between abnormal paving temperature data a and the bth associated abnormal vibration data; The Gaussian kernel function is used to represent the relationship between the sensor location point of abnormal paving temperature data 'a' and the sensor location point of the corresponding b-th associated abnormal vibration data 'a', representing the influence of spatial distance on coupling strength. exp() is an exponential function with a base of e. This is the distance between the sensor location point for abnormal paving temperature data 'a' and the sensor location point for the corresponding b-th associated abnormal vibration data. This distance can be calculated using Euclidean distance. For spatially related length, Generally, calibration is achieved through experiments, such as using a 0.5-meter guideline. The abnormal amplitude of abnormal paving temperature data 'a'. Let be the anomalous amplitude of the b-th associated anomalous vibration data. The closer the sensor location of the anomalous paving temperature data 'a' is to the sensor location of the b-th associated anomalous vibration data, the stronger the coupling effect. The larger it should be; A larger value indicates more severe segregation and a greater contribution to the coupling strength. The larger The larger; A larger value indicates a more pronounced vibration anomaly and a greater contribution to the coupling strength. The larger Also bigger.

[0023] Since segregation is mainly divided into temperature segregation, particle size segregation, and mechanical segregation, temperature segregation is mainly manifested as abnormal paving temperature, and secondarily as normal or slightly abnormal vibration, with temperature-vibration coupling usually being low; particle size segregation is mainly manifested as abnormal vibration, and secondarily as potentially abnormal paving temperature, with temperature-vibration coupling strength usually being high; mechanical segregation is mainly manifested as abnormal paving speed, and secondarily as potentially abnormal vibration and paving temperature, with its spatial distribution related to the machine's direction of travel. Based on the above analysis, this embodiment, based on the temperature-vibration coupling strength of each area to be analyzed at the current monitoring time, further obtains the matching degree between the area to be analyzed at the current monitoring time and different types of segregation according to parameters such as the proportion of different data in the area to be analyzed, the current paving speed data, the speed baseline threshold range, and the abnormal amplitude. The segregation confidence of the area to be analyzed at the current monitoring time is finally obtained by matching the degree between different types of segregation. Based on the above description, this embodiment will first obtain the temperature segregation matching degree, particle size segregation matching degree, and mechanical segregation matching degree of the area to be analyzed at the current monitoring time based on the temperature vibration coupling strength of the area to be analyzed at the current monitoring time, the proportion of abnormal paving temperature data and abnormal vibration data in the area to be analyzed at the current monitoring time, the abnormal amplitude of abnormal paving temperature data and abnormal vibration data in the area to be analyzed at the current monitoring time, the current paving speed data, and the speed baseline threshold range. Then, the maximum matching degree among the temperature segregation matching degree, particle size segregation matching degree, and mechanical segregation matching degree of any area A to be analyzed at the current monitoring time will be selected as the segregation confidence of area A to be analyzed. The larger the segregation confidence degree, the greater the probability or risk of segregation.

[0024] Furthermore, in this embodiment, the specific process for obtaining the temperature segregation matching degree, particle size segregation matching degree, and mechanical segregation matching degree of the region to be analyzed at the current monitoring time is as follows: For the region to be analyzed A at the current monitoring time: First, obtain , , The weighted sum is used as the temperature segregation matching degree corresponding to region A to be analyzed, and the results are obtained. , , The weighted sum is used as the granularity separation matching degree corresponding to the region A to be analyzed.

[0025] The specific formula for calculating the temperature segregation matching degree corresponding to the region A to be analyzed is as follows: in, Temperature segregation matching degree corresponding to region A to be analyzed. , , As the first percentage value, The first proportion is the ratio of the number of abnormal paving temperature data points in region A to the total number of data points in region A. The second proportion is the ratio of the number of abnormal vibration data points in region A to the total number of data points in region A. `min()` is a minimum value function, designed to ensure that the value range is between 0 and 1. This represents the mean of the absolute values ​​of the abnormal amplitudes of all abnormal paving temperature data in region A to be analyzed. Let z be the mean of the absolute values ​​of the abnormal amplitudes of all abnormal vibration data in region A to be analyzed, and z be the preset second temperature anomaly threshold. To preset the second vibration anomaly threshold, To determine the temperature vibration coupling strength of the region A to be analyzed, c0 is a preset coupling strength reference threshold, and W1, W2, and W3 are the first, second, and third weights, respectively; the first term... It is mainly determined by the proportion of different types, reflecting the dominant type of abnormality, the second item. It is mainly determined by the magnitude of the anomaly, which reflects the severity of the anomaly. (The third item...) Determined by the temperature-vibration coupling strength, it reflects the correlation between anomalies. Since the proportion of different types and the magnitude of anomalies provide a wealth of information, the first item... Second item The weight values ​​should be relatively large. For example, in this embodiment, W1, W2, and W3 can be set to 0.35, 0.35, and 0.3, respectively. The preset second temperature anomaly threshold and the preset second vibration anomaly threshold are empirical values ​​determined based on engineering experience. That is, since a temperature deviation exceeding 8 degrees Celsius usually indicates severe segregation, the preset second temperature anomaly threshold can be set to 8 in this embodiment. A vibration deviation exceeding 0.1g usually indicates a significant anomaly, so the preset second vibration anomaly threshold is set to 0.1g. In this embodiment, the vibration data is collected by a triaxial accelerometer. The unit of vibration data collected by the triaxial accelerometer is usually meters per second², and the commonly used unit of gravitational acceleration is g (approximately 9.8 m / s²). The preset coupling strength reference threshold c0 can also be set based on engineering experience, such as... When the value is greater than 0.3, vibration is likely the main cause of temperature anomalies, so in this embodiment, c0 is set to 0.3.

[0026] because The larger and The smaller, The larger the value, the more it conforms to the characteristics of abnormal paving temperature and normal or slightly abnormal vibration in temperature segregation. The larger the value, the greater the match between the analyzed region A and the temperature separation, or in other words... The larger; In The larger and The smaller, The larger the value, the more it conforms to the characteristics of abnormal paving temperature and normal or slightly abnormal vibration in temperature segregation. The larger the value, the greater the match between the analyzed region A and the temperature separation, or in other words... The larger; In The smaller, The larger the value, the more it conforms to the characteristic of lower temperature vibration coupling strength in temperature segregation. The larger the value, the greater the match between the analyzed region A and the temperature separation, or in other words... The larger it is, the more we can know The bigger, The larger and When it is larger, The larger, and A larger value indicates a higher degree of matching between the analyzed region A and temperature segregation, and vice versa. The smaller the value, the lower the match between the analyzed region A and the temperature segregation.

[0027] The specific formula for calculating the particle size segregation matching degree corresponding to the region A to be analyzed is as follows: in, The particle size separation matching degree corresponds to region A to be analyzed. , ,and The formula is primarily based on particle size segregation characteristics; The larger, The larger and When it is larger, The larger, and A larger value indicates a higher degree of matching between the analyzed region A and the particle size distribution, and vice versa. The smaller the size, the lower the match between the analyzed region A and the particle size separation. Formula Logic AND The logic is similar, so it will not be described in detail.

[0028] Next, it is determined whether the current paving speed data is within the speed baseline threshold range. If not, a constant of 1 is used as the first index value; otherwise, 0 is used as the first index value. Then, the absolute value of the difference between the temperature vibration coupling strength of the area to be analyzed (A) and the preset coupling strength reference threshold is obtained and negatively correlated, and this result is used as the second index value. Finally, the weighted sum of the first and second index values ​​is used as the mechanical segregation matching degree corresponding to the area to be analyzed (A). The specific calculation expression for the mechanical segregation matching degree corresponding to the area to be analyzed (A) is as follows: in, To determine the mechanical segregation matching degree corresponding to region A to be analyzed, F1 is the first index value, W4 is the fourth weight value, and W5 is the fifth weight value. Since velocity provides a significant amount of information, W4 is required to be greater than W5. For example, W4 can be set to 0.6 and W5 to 0.4. In actual engineering... The rounding range is usually between 0 and 1. The purpose of subtracting 1 from the result of comparing it with 3 is to make the second index value... It falls approximately within the range of 0.7 to 1, thus ensuring that the matching accuracy does not drop to zero due to extreme deviations, preserving the baseline value, and conforming to the deviations in engineering. Smaller values ​​indicate a higher degree of matching. Furthermore, in mechanical segregation, it is usually not caused by a single factor, but rather by the combined effects of velocity, vibration, and temperature anomalies. If velocity anomalies are accompanied by vibration-temperature coupling anomalies, the confidence level of mechanical segregation is higher. Additionally, velocity instability alters the material's residence time under the paver, affecting heat dissipation and temperature distribution. As a correlation indicator of vibration and temperature anomalies, it can capture this combined effect, improving the reliability of diagnosis. Furthermore, by combining... This can filter out some accidental situations, such as when constructing on curves or slopes, the speed may be adjusted normally, but if there is no coupling abnormality, the risk of mechanical segregation is low.

[0029] Therefore, this embodiment can obtain the separation confidence level of each region to be analyzed at the current monitoring time through the above process.

[0030] Step S003: Based on the intersection between each region to be analyzed at the current monitoring time and each region to be analyzed at the historical monitoring time, obtain the target region at the current monitoring time and the continuous region sequence corresponding to the target region. Obtain the trend continuity characterization value of the target region based on the continuous region sequence. Obtain the separation risk value of the target region based on the trend continuity characterization value of the target region, the separation confidence level, the area and temperature vibration coupling strength, and the total number of data in the target region.

[0031] Since segregation is a process, relying solely on parameter values ​​at a single moment for early warning will affect its accuracy. To further ensure the accuracy of subsequent early warnings, this embodiment will perform trend analysis, specifically analyzing whether each region to be analyzed at the current monitoring moment meets the persistence condition. If so, the trend persistence characteristic value will be obtained. Specifically, this embodiment will first obtain the target region and its corresponding persistence region sequence based on the intersection between each region to be analyzed at the current monitoring moment and each region to be analyzed at historical monitoring moments. Then, the trend persistence characteristic value of the target region at the current monitoring moment will be obtained based on the persistence region sequence. The trend persistence characteristic value is a crucial parameter for determining the segregation risk value. The specific process of obtaining the target region and its corresponding persistence region sequence at the current monitoring moment based on the intersection between each region to be analyzed at the current monitoring moment and each region to be analyzed at historical monitoring moments is as follows: For any region A to be analyzed at the current monitoring time: First, obtain the regions to be analyzed at each monitoring time before the current monitoring time, and the method for obtaining the regions to be analyzed at any monitoring time is the same as the method for obtaining the regions to be analyzed at the current monitoring time. Record the current monitoring time as the t-th monitoring time. Then, determine whether the region to be analyzed at the (t-1)-th monitoring time intersects with region A. If so, record the region to be analyzed at the (t-1)-th monitoring time that intersects with region A as the first historical region corresponding to region A. If the region to be analyzed, n, at a given monitoring time point intersects with the region to be analyzed, then region n is the first historical region corresponding to region A. The process continues by checking if the region to be analyzed at the (t-2)th monitoring time point intersects with the first historical region. If so, the region to be analyzed at the (t-2)th monitoring time point that intersects with the first historical region is recorded as the second historical region corresponding to region A. This process continues until all regions to be analyzed at a given monitoring time point no longer intersect with any historical region. The process then stops, and the obtained region to be analyzed corresponding to region A is recorded. The sequence consisting of all historical regions and the region to be analyzed, A, is denoted as the region sequence corresponding to region A. The first region in the region sequence corresponding to region A is region A to be analyzed, and the (d+1)th region in the region sequence corresponding to region A is the d-th historical region corresponding to region A. That is, the second region in the region sequence corresponding to region A is the first historical region corresponding to region A. Then, the time interval between the monitoring times corresponding to the first and last regions in the region sequence corresponding to region A is obtained and recorded as the duration of the region sequence corresponding to region A. After that, the duration of the region sequence is determined. If the duration of the region sequence corresponding to region A is greater than a preset duration threshold, then region A is considered to have a risk of segregation, making it necessary to calculate the segregation risk value. Therefore, if the duration of the region sequence corresponding to region A is greater than the preset duration threshold, region A is designated as the target region, and the region sequence of region A is taken as the corresponding persistent region sequence. All regions in the persistent region sequence are recorded as persistent regions. That is, if a target region is region A, then the persistent region sequence corresponding to that target region is the region sequence corresponding to region A. In specific applications, the implementer needs to set the preset duration threshold according to the actual situation. For example, in this embodiment, the preset duration threshold can be set to 10 seconds.

[0032] Here's an example of obtaining the region sequence corresponding to region A to be analyzed and the duration of the region sequence corresponding to region A to be analyzed: If region u1 to be analyzed intersects with region A to be analyzed at monitoring time t-1, region u2 to be analyzed intersects with region u1 to be analyzed at monitoring time t-2, region u3 to be analyzed intersects with region u2 to be analyzed at monitoring time t-3, and all regions to be analyzed at monitoring time t-4 do not intersect with region u3 to be analyzed, then the region sequence corresponding to region A to be analyzed is {region A to be analyzed, region u1 to be analyzed, region u2 to be analyzed, region u3 to be analyzed}. The duration of the region sequence corresponding to region A to be analyzed is the time interval between the time corresponding to monitoring time t and the time corresponding to monitoring time t-3. Region A to be analyzed is the region to be analyzed at the current monitoring time, and the current monitoring time is recorded as the tth monitoring time.

[0033] In this embodiment, the specific process of obtaining the trend persistence characterization value of the target region at the current monitoring time based on the persistent region sequence is as follows: For any target region B: First, the normalized duration of the persistent region sequence corresponding to target region B is denoted as the time duration representation value Z1 of target region B; and the time duration representation value Z1 of target region B is... , T0 is the duration of the continuous region sequence corresponding to the target region B, in seconds. T0 is the preset duration. According to engineering experience, if the duration exceeds 60 seconds, emergency handling is required. Therefore, in this embodiment, T0 is set to 60 seconds.

[0034] Then, in the continuous region sequence corresponding to target region B, regions are marked at preset intervals, and each marked region is recorded as a marked region of target region B. The new sequence formed by the marked regions of target region B is recorded as the continuous region subsequence of target region B. The positional relationship of regions in the continuous region subsequence corresponds to the positional relationship of regions in the continuous region sequence corresponding to target region B. For example, the a-th marked region in the continuous region subsequence is the a-th marked region in the continuous region sequence corresponding to target region B. In practical applications, the implementer needs to set the preset number according to the actual situation, such as setting the preset number to 5. The process of obtaining the persistent region subsequence of target region B is as follows: If the preset number is 5 and the number of regions in the persistent region sequence corresponding to target region B is 11, then the first region in the persistent region sequence corresponding to target region B is marked first, then the first (1+5)th region in the persistent region sequence corresponding to target region B is marked, and then the first (1+5+5)th region in the persistent region sequence corresponding to target region B is marked. Then the persistent region subsequence is composed of the first, sixth, and eleventh regions in the persistent region sequence corresponding to target region B. The minimum value function in this embodiment is used to achieve normalization and saturation effect.

[0035] Next, based on the area of ​​each region in the continuous subsequence of target region B and the abnormal amplitude of abnormal paving temperature data in each region, the characteristic values ​​of area change and abnormal amplitude change corresponding to target region B are obtained. The specific process for obtaining these characteristic values ​​is as follows: Obtain the area change rate sequence corresponding to the continuous subsequence; record the mean of the area change rate sequence as the target area change rate corresponding to target region B; record the normalized result of the target area change rate as the characteristic value of area change corresponding to target region B; and record the k-th area change rate in the area change rate sequence. , This represents the absolute value of the difference between the area of ​​the k-th region and the area of ​​the (k+1)-th region in a continuous subsequence. The time interval between the monitoring time corresponding to the k-th region and the monitoring time corresponding to the (k+1)-th region is defined as follows: if a region is a region at a certain monitoring time, then the monitoring time corresponding to that region is that monitoring time; obtain the abnormal amplitude change rate sequence corresponding to the continuous region subsequence, and denote the mean of the abnormal amplitude change rate sequence as the target abnormal amplitude change rate corresponding to the target region B; denote the result of normalizing the target abnormal amplitude change rate as the abnormal amplitude change characterization value corresponding to the target region B; the h-th abnormal amplitude change rate in the abnormal amplitude change rate sequence is defined as follows: , This is the absolute value of the difference between the mean of the anomalous magnitudes of all anomalous paving temperature data in the h-th region of the continuous subsequence and the mean of the anomalous magnitudes of all anomalous paving temperature data in the (h+1)-th region. Z2 represents the time interval between the monitoring time corresponding to the h-th region and the monitoring time corresponding to the (h+1)-th region. If any region is a region at a certain time, then the time at which all data is collected in that region is that time. Finally, the weighted sum of the time persistence characterization value, the area change characterization value, and the abnormal amplitude change characterization value corresponding to target region B is used as the trend persistence characterization value of target region B. The area change characterization value Z2 corresponding to target region B is... , The target area change rate corresponding to target region B. This represents the maximum value of the target area change rate, or the maximum value among the target area change rates for all target areas at all monitoring times. Its function is to... After normalization, the abnormal amplitude change characterization value Z3 corresponding to the target region B is: , R0 represents the target anomaly amplitude change rate corresponding to target region B, where R0 is the maximum value of the target anomaly amplitude change rate or the maximum value among the target anomaly amplitude change rates corresponding to all target regions at all monitoring times. Its function is to... Normalization is performed.

[0036] The expression for calculating the trend persistence characteristic value of target region B is: Wherein, QB is the trend persistence representation value of target region B, v1, v2, and v3 are the first weighting factor, the second weighting factor, and the third weighting factor, respectively, Z1 is the time persistence representation value of target region B, Z2 is the area change representation value of target region B, and Z3 is the abnormal amplitude change representation value of target region B. Since the duration better reflects the segregation situation, this embodiment requires a larger value for v1. For example, v1, v2, and v3 can be set to 0.4, 0.3, and 0.3, respectively. Moreover, the larger Z1, Z2, and Z3 are, the more rapidly the anomaly is deteriorating, and the greater the corresponding segregation risk. In other words, the larger QB is, the greater the segregation risk, and vice versa.

[0037] After obtaining the trend persistence characterization value of the target area at the current monitoring time, the segregation risk value of the target area at the current monitoring time is obtained based on the trend persistence characterization value, segregation confidence, area and temperature vibration coupling strength of the target area at the current monitoring time, and the total number of data in the target area at the previous monitoring time. The specific process of obtaining the segregation risk value of the target area at the current monitoring time is as follows: For target area B, the value obtained by multiplying the trend persistence characterization value, segregation confidence, and total number of data in target area B, the area of ​​target area B, and the temperature vibration coupling strength of target area B is taken as the segregation risk value of target area B.

[0038] The specific calculation expression for obtaining the segregation risk value of target region B is as follows: Where RB is the segregation risk value of target region B, CB is the segregation confidence level of target region B, QB is the trend persistence characteristic value of target region B, NB is the total number of data in target region B, ln() is the logarithmic function with base e, and AB is the area of ​​target region B. The temperature vibration coupling strength is denoted as CB; CB indicates that if the system determines that the current anomaly belongs to a certain type of segregation, then the necessity of taking targeted measures is greater. Therefore, the segregation risk value should be proportional to CB. QB represents the development trend of the anomaly; the larger the trend persistence value, the more rapidly the anomaly is deteriorating. The impact of the outlier size is represented by logarithmic growth because as the number of outliers increases, the risk does not increase linearly but gradually saturates. Mapping NB with a logarithmic function avoids the problem of this term becoming too large when there are too many outliers. Adding a constant of 1 prevents this term from becoming 0. AB represents the area of ​​influence of the outlier; the larger the area, the larger the area of ​​influence and the higher the risk. Multiplying AB by 0.1 and adding 1 ensures that the area term increases linearly with the increase of AB, and that the contribution of the area term to the risk value is relatively moderate. 0.1 is a scaling factor to avoid the area having an excessive impact. A larger value indicates that segregation may be caused by vibration, which can be more complex and therefore poses a higher risk. To increase the influence of temperature vibration coupling strength on segregation risk value, 0.5 is the weighting factor; and CB, QB, NB, AB and The larger the value, the greater the RB value, and a larger RB value indicates a greater risk of segregation.

[0039] Therefore, this embodiment can obtain the segregation risk value of the target area at the current monitoring time.

[0040] Step S004: Based on the segregation risk value, a segregation warning is issued for the paving process of the water-stabilized paver.

[0041] After obtaining the segregation risk value, this embodiment uses the obtained segregation risk value of the target area at the current monitoring time to provide a segregation warning for the paving process of the water-stabilized paver, specifically: For any target area at the current monitoring time, the risk level of the target area is determined based on its segregation risk value. If the risk level of the target area is medium or high, a segregation warning is immediately issued, and corresponding measures are taken. These measures may include adjusting the screed heating temperature, checking the uniformity of material supply, or focusing on monitoring temperature changes in abnormal areas; adjusting vibration parameters, checking material gradation, monitoring vibration anomalies and material separation; optimizing speed, assisting in checking mechanical condition, and monitoring speed stability and paving uniformity. Furthermore, a segregation risk value less than a preset first threshold is classified as low risk; a segregation risk value greater than or equal to the preset first threshold but less than a preset second threshold is classified as medium risk; and a segregation risk value greater than or equal to the preset second threshold is classified as high risk. In other words, if the segregation risk value of a target area is greater than the preset second threshold, then the risk level of that target area is high risk. In specific applications, implementers need to set the preset first and second thresholds according to the actual situation; for example, the preset first and second thresholds can be set to 5 and 15 respectively.

[0042] Thus, this embodiment completes the data collection and segregation early warning of the water-stabilized paver operation. Moreover, the segregation risk value obtained by this embodiment based on multi-dimensional parameters can provide segregation early warning relatively accurately and reliably. It should be noted that all parameters involved in the summation and multiplication of the formulas in this embodiment have been dimensionless.

[0043] In summary, this embodiment first obtains the area to be analyzed at the current monitoring time based on the number of current abnormal data within a preset neighborhood of the sensor's location coordinates that collects the current abnormal data. Then, based on the abnormal paving temperature and vibration data within the area to be analyzed at the current monitoring time, the temperature-vibration coupling intensity of the area to be analyzed at the current monitoring time is obtained. Finally, based on the proportion of different data, temperature-vibration coupling intensity, current paving speed data, speed baseline threshold range, and abnormal amplitude within the area to be analyzed at the current monitoring time, the segregation confidence level of the area to be analyzed at the current monitoring time is obtained. Next, based on the intersection between each area to be analyzed at the current monitoring time and each area to be analyzed at historical monitoring times, the target area and its corresponding continuous area sequence are obtained. The trend continuity characterization value of the target area is obtained from the continuous area sequence. Based on the trend continuity characterization value, segregation confidence level, area, temperature-vibration coupling intensity, and the total number of data points in the target area, the segregation risk value of the target area is obtained. Finally, based on the segregation risk value, a segregation warning is issued for the paving process of the water-stabilized paver. Furthermore, the segregation risk value obtained by this embodiment based on parameters such as trend persistence characterization value, segregation confidence level, area and temperature vibration coupling strength can improve the accuracy and reliability of segregation warning when performing segregation warning during the paving process of water-stabilized paver.

[0044] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A sensor-based data acquisition and segregation early warning system for water-stabilized paver operation, comprising a processor and a memory, characterized in that, The processor executes the computer program stored in the memory to perform the following steps: During the paving process, the current paving temperature data, current vibration data, current paving speed data, and current temperature baseline threshold, current vibration baseline threshold, and speed baseline threshold range are acquired. Based on the current temperature baseline threshold and current vibration baseline threshold, the abnormal amplitude of the current paving temperature data and current vibration data is obtained. Based on the abnormal amplitude, the current abnormal data is obtained, which includes abnormal paving temperature data and abnormal vibration data. Based on the number of current abnormal data within a preset neighborhood of the sensor's location coordinates that collects current abnormal data, the area to be analyzed at the current monitoring time is obtained. Based on the abnormal paving temperature data and abnormal vibration data in the area to be analyzed at the current monitoring time, the temperature-vibration coupling intensity of the area to be analyzed at the current monitoring time is obtained. Based on the proportion of different data, temperature-vibration coupling intensity, current paving speed data, speed baseline threshold range, and abnormal amplitude in the area to be analyzed at the current monitoring time, the segregation confidence level of the area to be analyzed at the current monitoring time is obtained. Based on the intersection of each region to be analyzed at the current monitoring time and each region to be analyzed at the historical monitoring time, the target region at the current monitoring time and the corresponding continuous region sequence are obtained. The trend continuity characterization value of the target region is obtained based on the continuous region sequence. Based on the trend continuity characterization value of the target region, the segregation confidence, the area and temperature vibration coupling strength, and the total number of data in the target region, the segregation risk value of the target region is obtained. Based on the segregation risk value, a segregation warning is issued for the paving process of the water-stabilized paver.

2. The sensor-based water-stabilized paver operation data acquisition and segregation early warning system as described in claim 1, characterized in that, The methods for obtaining the current temperature baseline threshold, the current vibration baseline threshold, and the velocity baseline threshold range include: The temperature of the paving material under ideal conditions is recorded as the first influence value at the current monitoring time. The absolute value of the difference between the ambient temperature data at the current monitoring time and the preset reference ambient temperature is multiplied by the preset heat exchange coefficient and recorded as the second influence value at the current monitoring time. The product of the time interval from the completion of production of the paving material used at the current monitoring time to the current monitoring time and the preset temperature change rate is used as the third influence value at the current monitoring time. The sum of the first influence value, the second influence value and the third influence value is used as the current temperature baseline threshold. The mean and standard deviation of all vibration data collected during the time period formed by the current monitoring time and the preset historical monitoring time period before the current monitoring time are recorded as the current vibration mean and the current vibration standard deviation. The sum of the current vibration mean and the current vibration standard deviation by a preset multiple is used as the current vibration baseline threshold. The interval formed by the preset values ​​of plus or minus a certain percentage of the standard paving speed data is denoted as the speed baseline threshold interval.

3. The sensor-based water-stabilized paver operation data acquisition and segregation early warning system as described in claim 1, characterized in that, The absolute value of the difference between the current paving temperature data and the current temperature baseline threshold is the abnormal amplitude of the current paving temperature data. The result of subtracting the current vibration baseline threshold from the current vibration data is the abnormal amplitude of the current vibration data. The abnormal amplitude of the current abnormal data is greater than the corresponding threshold.

4. The method for acquiring the area to be analyzed at the current monitoring time in the sensor-based water-stabilized paver operation data acquisition and segregation early warning system as described in claim 1 includes: The location coordinates of the sensor that collects the current abnormal data at the current monitoring time are recorded as the sensor location point corresponding to the current abnormal data. For any current abnormal data, determine whether the number of current abnormal data within a preset first neighborhood of the sensor location point of the current abnormal data is greater than or equal to a preset data number threshold. If so, the preset first neighborhood of the sensor location point of the current abnormal data is taken as a processing area corresponding to the current monitoring time. Merge any overlapping regions among all regions to be processed at the current monitoring time, and record all regions obtained after merging as regions to be analyzed at the current monitoring time.

5. The sensor-based water-stabilized paver operation data acquisition and segregation early warning system as described in claim 4, characterized in that, The method for obtaining the temperature-vibration coupling intensity of the region to be analyzed at the current monitoring time includes: For any abnormal paving temperature data a in any area A to be analyzed at the current monitoring time: In the region A to be analyzed, abnormal vibration data within a preset second neighborhood of the sensor location point of the abnormal paving temperature data a are acquired and recorded as the associated abnormal vibration data corresponding to the abnormal paving temperature data a. The Gaussian kernel function between the sensor location point of the abnormal paving temperature data a and the sensor location point of the b-th associated abnormal vibration data corresponding to the abnormal paving temperature data a, the absolute value of the abnormal amplitude of the abnormal paving temperature data a, and the absolute value of the abnormal amplitude of the b-th associated abnormal vibration data are multiplied and recorded as the coupling strength between the abnormal paving temperature data a and the b-th associated abnormal vibration data. The average value of the coupling strength between the abnormal paving temperature data a and all associated abnormal vibration data corresponding to the abnormal paving temperature data a is recorded as the comprehensive coupling strength of the abnormal paving temperature data a. The average of the combined coupling strength of all abnormal paving temperature data in the region A to be analyzed is taken as the temperature vibration coupling strength of the region A to be analyzed.

6. The sensor-based water-stabilized paver operation data acquisition and segregation early warning system as described in claim 1, characterized in that, Methods for obtaining the segregation confidence level of the region to be analyzed at the current monitoring time include: For any region A to be analyzed at the current monitoring time: Will , , The weighted sum is used as the temperature segregation matching degree corresponding to the region A to be analyzed. , , The weighted sum is used as the particle size separation matching degree corresponding to the region A to be analyzed; As the first percentage value, The second proportion value is the ratio of the number of abnormal paving temperature data in the area to be analyzed to the total number of data in the area to be analyzed, and the second proportion value is the ratio of the number of abnormal vibration data in the area to be analyzed to the total number of data in the area to be analyzed. , min() is the function that finds the minimum value. The mean of the absolute values ​​of the abnormal amplitudes of all abnormal paving temperature data in the area to be analyzed, A. Let z be the mean of the absolute values ​​of the abnormal amplitudes of all abnormal vibration data in the region A to be analyzed, and z be a preset second temperature anomaly threshold. To preset the second vibration anomaly threshold; , , The temperature vibration coupling strength of the region A to be analyzed is given by c0, where c0 is a preset coupling strength reference threshold. Determine whether the current paving speed data is within the speed baseline threshold range. If it is not, use a constant 1 as the first index value; otherwise, use 0 as the first index value. Use the result of negative correlation mapping between the absolute value of the difference between the temperature vibration coupling intensity of the area to be analyzed A and the preset coupling intensity reference threshold as the second index value. Use the weighted sum of the first index value and the second index value as the mechanical segregation matching degree corresponding to the area to be analyzed A. The maximum value among the temperature segregation matching degree, particle size segregation matching degree, and mechanical segregation matching degree is selected as the segregation confidence degree of the region A to be analyzed.

7. The sensor-based water-stabilized paver operation data acquisition and segregation early warning system as described in claim 1, characterized in that, The method for obtaining the target area at the current monitoring time and the corresponding continuous region sequence includes: For any region A to be analyzed at the current monitoring time: Let the current monitoring time be denoted as the t-th monitoring time. Determine whether the region to be analyzed at the (t-1)-th monitoring time intersects with region A. If so, record the region to be analyzed that intersects with region A as the first historical region corresponding to region A. Continue determining whether the region to be analyzed at the (t-2)-th monitoring time intersects with the first historical region. If so, record the region to be analyzed that intersects with the first historical region as the second historical region corresponding to region A, and so on, until all regions to be analyzed at any given monitoring time no longer intersect with any historical region. Then, record all historical regions corresponding to region A. The sequence consisting of the historical region and the region to be analyzed A is denoted as the region sequence corresponding to the region to be analyzed A. The first region in the region sequence corresponding to the region to be analyzed A is the region to be analyzed A, and the (d+1)th region in the region sequence corresponding to the region to be analyzed A is the dth historical region corresponding to the region to be analyzed A. The time interval between the monitoring times corresponding to the first and last regions in the region sequence is denoted as the duration of the region sequence. It is determined whether the duration of the region sequence is greater than a preset duration threshold. If it is greater, the region to be analyzed A is denoted as the target region, and the region sequence of the region to be analyzed A is taken as the continuous region sequence corresponding to the target region.

8. The sensor-based water-stabilized paver operation data acquisition and segregation early warning system as described in claim 1, characterized in that, The method for obtaining the trend persistence characterization value of the target region includes: For any target region B: the normalized result of the time interval between the monitoring times corresponding to the first and last regions in the continuous region sequence corresponding to the target region B is recorded as the time duration characterization value; in the continuous region sequence corresponding to the target region B, a preset number of regions are marked once, and the marked regions are all recorded as the marked regions of the target region B, and the new sequence formed by the marked regions of the target region B is recorded as the continuous region subsequence; based on the area of ​​each region in the continuous region subsequence and the abnormal amplitude of the abnormal paving temperature data in each region, the region area change characterization value and the abnormal amplitude change characterization value are obtained, and the weighted sum of the time duration characterization value, the region area change characterization value, and the abnormal amplitude change characterization value is used as the trend duration characterization value of the target region B.

9. The sensor-based water-stabilized paver operation data acquisition and segregation early warning system as described in claim 8, characterized in that, The methods for obtaining the regional area change characterization value and the abnormal amplitude change characterization value include: Obtain the area change rate sequence corresponding to the persistent region subsequence, and record the normalized result of the mean of the area change rate sequence as the region area change characterization value. The k-th area change rate in the area change rate sequence is... , It is the absolute value of the area difference between the k-th region and the (k+1)-th region in the continuous region subsequence. The time interval between the monitoring time corresponding to the k-th region and the monitoring time corresponding to the (k+1)-th region is defined. The abnormal amplitude change rate sequence corresponding to the continuous region subsequence is obtained, and the normalized result of the mean of the abnormal amplitude change rate sequence is recorded as the abnormal amplitude change characterization value. The h-th abnormal amplitude change rate in the abnormal amplitude change rate sequence is defined as... , It is the absolute value of the difference between the mean of the abnormal amplitudes of all abnormal paving temperature data in the h-th region of the continuous region subsequence and the mean of the abnormal amplitudes of all abnormal paving temperature data in the (h+1)-th region. The time interval between the monitoring time corresponding to the h-th region and the monitoring time corresponding to the (h+1)-th region is given.

10. The sensor-based water-stabilized paver operation data acquisition and segregation early warning system as described in claim 1, characterized in that, The result of multiplying the trend persistence characterization value of the target region, the segregation confidence level, the total number of data in the target region, the area of ​​the target region, and the temperature vibration coupling strength of the target region are the segregation risk values ​​of the corresponding target region.

Images