Self-adaptive road surface pollution monitoring method and system

Abstract

The invention discloses a self-adaptive road surface pollution monitoring method and system, and the method comprises the steps: collecting the elongation at break and tensile strength data of a polyurethane adhesive, and employing a self-adaptive neural network to carry out the self-adaptive algorithm to evaluate the performance characteristics of the polyurethane adhesive; whether the road surface pollution degree belongs to a healthy range or not is judged according to the performance characteristics of the polyurethane adhesive, an unhealthy early warning level is set for the road surface pollution degree belonging to an unhealthy range, the judged road surface pollution degree is rechecked, the accuracy of a judgment result is checked, and then the rechecked data is output and displayed. Meanwhile, an alarm is given according to the non-health early warning level triggering the non-health early warning mechanism, serious pollution on the road surface is treated in time, the road surface pollution problem can be effectively monitored, and the purpose of monitoring the road surface pollution degree is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of road monitoring, and in particular to an adaptive road surface pollution monitoring method and system. Background Art

[0002] The Chinese patent publication number: CN114858670A discloses a method and related products for monitoring road dust pollution. The invention obtains the measured value of the road surface dust concentration and the measured value of the air dust concentration at a certain speed, calculates the difference between the two, and compares them with the second difference corresponding to the same speed in the road dust standard to analyze the dust pollution situation of the road section.

[0003] In the existing technology, the degree of road surface pollution is assessed by monitoring road surface factors such as dust, water and heavy metals on the road surface. However, rutting, cracks and potholes on asphalt roads can also affect road surface pollution.

[0004] The present invention proposes a solution to the above-mentioned shortcomings: according to the technical indicators of the elongation at break and the tensile strength of the polyurethane adhesive, the performance characteristics of the polyurethane adhesive are analyzed in combination with an adaptive neural network algorithm, the degree of pollution of the road surface is judged according to the performance characteristics of the polyurethane adhesive, and the severity of the pollution of the unhealthy road surface is judged by setting an early warning level. Summary of the invention

[0005] In order to solve the above-mentioned technical problems, the present invention provides an adaptive road surface pollution monitoring method and system.

[0006] The technical solution of the present invention is implemented as follows: an adaptive road surface pollution monitoring method, comprising:

[0007] S1, marking the elongation at break G and tensile strength P of the polyurethane adhesive, presetting the initial range thresholds of the marked technical indicators as T_G and T_P respectively, collecting the polyurethane adhesive, obtaining the technical indicators of the elongation at break G1 and tensile strength P1 of the polyurethane adhesive and transmitting them to the neural network module;

[0008] S2, the neural network module uses the obtained technical indicators of elongation at break G1 and tensile strength P1 to evaluate the performance characteristics of the polyurethane adhesive through an adaptive neural network and transmits them to the data analysis and judgment module;

[0009] S3, the data analysis and judgment module judges the degree of road surface pollution according to the performance characteristics of the polyurethane adhesive output by the neural network module, and sets the unhealthy warning level if the road surface pollution degree belongs to the unhealthy range: the first level severe pollution warning Y1, the second level severe pollution warning Y2;

[0010] S4, the review and inspection module reviews the healthy range, the first-level severe pollution warning Y1 and the second-level severe pollution warning Y2 output by the data analysis and judgment module, and transmits the completed values ​​back to the data analysis and judgment module;

[0011] S5. The data analysis and judgment module transmits the road surface pollution degree result to the output and alarm module, and the output and alarm module outputs and displays the road surface pollution degree, and alarms after triggering the unhealthy early warning mechanism.

[0012] Furthermore, in the step S1, the initial range thresholds of the elongation at break and the tensile strength of the polyurethane adhesive are set, and the technical indicators of the elongation at break (25°C) and the tensile strength (25°C) of the national standard GB / T 528 are set as the initial thresholds: the elongation at break is greater than 50% and satisfies the elongation at break is equal to 50%, and the tensile strength is greater than 20 MPa and satisfies the tensile strength is equal to 20 MPa.

[0013] Furthermore, the steps for collecting the polyurethane adhesive are as follows: using a sampler to collect a sample of the polyurethane adhesive mixture at the surface of the asphalt road, grinding and sieving the sample, mixing the sieved particles with a solvent, N-methylpyrrolidone, and stirring to promote the dissolution of the polyurethane adhesive, filtering to separate the solvent in which the polyurethane adhesive is dissolved and the undissolved solid particles, and purifying the solvent containing the polyurethane adhesive by a chemical separation technology of distillation. After purification, a treated polyurethane adhesive is obtained, and the technical indicators of the elongation at break G1 and the tensile strength P1 of the polyurethane adhesive are obtained.

[0014] Furthermore, in step S2, the adaptive neural network performs the adaptive algorithm steps as follows:

[0015] S201, training neural network model: collecting a large amount of historical data of technical indicators (G, P) of elongation at break G and tensile strength P of polyurethane adhesive, as well as corresponding excellent, medium and poor performance characteristic labels of polyurethane adhesive, and removing outliers from the historical data;

[0016] Input: Dataset D, which is a large amount of historical data of technical indicators (G, P) of elongation at break G and tensile strength P of polyurethane adhesives collected through preprocessing, with preset initial range thresholds T_G and T_P, and output: excellent, medium and poor performance characteristics of polyurethane adhesives;

[0017] The specific steps of neural network model training are as follows:

[0018] The input layer presets 2 nodes:

[0019] Node 11: Receive the preset initial range threshold T_G of the elongation at break and the preset initial range threshold T_P of the tensile strength of the polyurethane adhesive as technical indicators;

[0020] Node 12: Receive the elongation at break G and the tensile strength P of the polyurethane adhesive as technical indicators;

[0021] Hidden layer: Preset 4 cross nodes. f(G, P) = 1 indicates that the technical indicators (G, P) of the elongation at break G and the tensile strength P of the input polyurethane adhesive meet the range conditions, and perform the calculation of this node; f(G, P) = 0 indicates that the technical indicators (G, P) of the elongation at break G and the tensile strength P of the input polyurethane adhesive do not meet the range conditions, and do not perform the calculation of this node. The preset calculation formula of the hidden layer is as follows:

[0022] Node 21: If G ≥ T_G and P ≥ T_P, then f1(G, P) = 1; otherwise f1(G, P) = 0;

[0023] Node 22: If G ≥ T_G and P < T_P, then f2(G, P) = 1; otherwise f2(G, P) = 0;

[0024] Node 23: If G < T_G and P ≥ T_P, then f3(G, P) = 1; otherwise f3(G, P) = 0;

[0025] Node 24: If G < T_G and P < T_P, then f4(G, P) = 1; otherwise f4(G, P) = 0;

[0026] The output layer outputs 3 nodes:

[0027] Node 31: The technical indicators (G, P) of the elongation at break G and the tensile strength P meet the conditions of Node 21, and output the excellent performance characteristics of the polyurethane adhesive;

[0028] Node 32: The technical indicators (G, P) of the elongation at break G and the tensile strength P meet the conditions of Node 22 or Node 23, and output the medium performance characteristics of the polyurethane adhesive;

[0029] Node 33: The technical indicators (G, P) of the elongation at break G and the tensile strength P meet the conditions of Node 24, and output the poor performance characteristics of the polyurethane adhesive;

[0030] S202. Performance prediction of the neural network model: Input the new data of the elongation at break G and the tensile strength P into the trained neural network. If it is verified that the performance characteristics of the polyurethane adhesive output by the neural network are inconsistent with the predicted performance characteristics of the polyurethane adhesive, return to the previous step S201. If it is verified that the performance characteristics of the polyurethane adhesive output by the neural network are consistent with the predicted performance characteristics of the polyurethane adhesive, perform the next step S203;

[0031] S203, application of neural network model: deploy the neural network model that has been trained to achieve the expected results to actual applications, input the obtained technical indicators of elongation at break G1 and tensile strength P1 of the polyurethane adhesive, and use the adaptive neural network model that has been trained to achieve the expected results from steps S201 to S202 to perform an adaptive algorithm to evaluate the performance characteristics of the polyurethane adhesive, and then output whether the performance characteristics of the polyurethane adhesive are excellent, medium or low.

[0032] Furthermore, in the step S3, based on the excellent output of the polyurethane adhesive performance characteristic, the road surface pollution degree is judged to be within the healthy range; based on the medium output of the polyurethane adhesive performance characteristic, the road surface pollution degree is judged to be within the unhealthy range as the first level severe pollution warning Y1; based on the poor output of the polyurethane adhesive performance characteristic, the road surface pollution degree is judged to be within the unhealthy range as the second level severe pollution warning Y2.

[0033] Furthermore, in the S4 step, the review and inspection module checks whether the output of the road surface pollution correctly triggers the excellent, medium, and poor thresholds of the polyurethane adhesive performance characteristics based on the healthy range and unhealthy warning level of the road surface pollution output by the data analysis and judgment module, and then confirms the healthy range and unhealthy warning level of the road surface pollution, and further manually reviews the warning level automatically output.

[0034] Furthermore, in the step S5, an adaptive neural network is used to perform an adaptive algorithm to output the performance characteristics of the polyurethane adhesive, determine the healthy degree of road surface pollution and the unhealthy warning level, and then output the road surface pollution level on a computer display screen through a review test, and at the same time trigger an unhealthy warning mechanism and then sound an alarm.

[0035] The system solution of the present invention is implemented as follows: an adaptive road surface pollution monitoring system, comprising:

[0036] An input control module, connected to the neural network module, is used to preset the initial range threshold of the basic technical indicators of the polyurethane adhesive of the road material, and obtain the technical indicators and mark them;

[0037] The data analysis and judgment module is connected with the input control module, the neural network module, the review and inspection module, and the output and alarm module, and is used to receive the preset initial range threshold value, the acquired technical indicators and information tags of the input control module, use the numerical range calculated by the neural network module to judge the healthy range and unhealthy warning level of road surface pollution, cooperate with the review and inspection module to inspect the range of road surface pollution, and transmit the data analysis completed by the review and inspection to the data analysis and judgment module for calculation;

[0038] A neural network module, connected to the input control module and the data analysis and judgment module, is used to calculate and evaluate the performance characteristics of the polyurethane adhesive;

[0039] A review and inspection module, connected to the data analysis and judgment module, is used to review the data analysis and judgment module's judgment on the healthy range and unhealthy warning level of road surface pollution;

[0040] The output and alarm module is connected to the data analysis and judgment module, and is used to output and display the degree of road surface pollution and to alarm after triggering the unhealthy early warning mechanism.

[0041] Beneficial Effects

[0042] In view of the fact that the existing technology of evaluating the degree of road surface pollution by monitoring dust, water accumulation and heavy metals on the road surface can no longer meet the actual demand for monitoring the degree of road surface pollution, the present invention uses the performance characteristics of the polyurethane adhesive to detect and then obtain the degree of road surface pollution, thereby solving the problem in the existing technology that rutting, cracking and potholes on the asphalt road surface affect the degree of road surface pollution. At the same time, an adaptive neural network is used to perform an adaptive algorithm to evaluate the performance characteristics of the polyurethane adhesive, which can accurately capture the relationship between the input polyurethane adhesive performance characteristics and the elongation at break G1 and the tensile strength P1, and the characteristics of parallel processing and generalization capabilities of the adaptive neural network are used to accelerate the calculation speed, making data processing more efficient and accurate. In addition, by setting an early warning level for the unhealthy range of the monitored road surface pollution level, the response efficiency of the system is further improved and the waste of system resources is reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A structural block diagram of an adaptive road surface pollution monitoring method in an embodiment of the present invention;

[0044] Figure 2 A flowchart of the steps of an adaptive road surface pollution monitoring method according to an embodiment of the present invention;

[0045] Figure 3 A flowchart of the steps of performing an adaptive algorithm for an adaptive neural network in an embodiment of the present invention;

[0046] Figure 4 The structure block diagram of an adaptive road surface pollution monitoring system in an embodiment of the present invention. DETAILED DESCRIPTION

[0047] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0048] The preferred implementation methods of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these implementation methods are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0049] It can be understood that the terms "first", "second", etc. used in the present application can be used in this article to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from another element.

[0050] It should be noted that when an element is considered to be "connected" to another element, it can be directly connected to the other element, or connected to the other element through an intermediate element. In addition, the "connection" in the following embodiments should be understood as "electrical connection", "communication connection", etc. if there is transmission of electrical signals or data between the connected objects.

[0051] When used herein, the singular forms "a", "an", and "said / the" may also include plural forms, unless the context clearly indicates otherwise. It should also be understood that the terms "include / comprise" or "have" and the like specify the presence of stated features, wholes, steps, operations, components, parts, or combinations thereof, but do not exclude the possibility of the presence or addition of one or more other features, wholes, steps, operations, components, parts, or combinations thereof. At the same time, the term "and / or" used in this specification includes any and all combinations of the relevant listed items.

[0052] See also Figure 1-Figure 4 As shown, an adaptive road surface pollution monitoring method comprises:

[0053] S1, marking the elongation at break G and tensile strength P of the polyurethane adhesive, presetting the initial range thresholds of the marked technical indicators as T_G and T_P respectively, collecting the polyurethane adhesive, obtaining the technical indicators of the elongation at break G1 and tensile strength P1 of the polyurethane adhesive and transmitting them to the neural network module;

[0054] S2, the neural network module uses the obtained technical indicators of elongation at break G1 and tensile strength P1 to evaluate the performance characteristics of the polyurethane adhesive through an adaptive neural network and transmits them to the data analysis and judgment module;

[0055] S3, the data analysis and judgment module judges the degree of road surface pollution according to the performance characteristics of the polyurethane adhesive output by the neural network module, and sets the unhealthy warning level for the road surface pollution degree that belongs to the unhealthy range: the first level severe pollution warning Y1, the second level severe pollution warning Y2;

[0056] S4, the review and inspection module reviews the healthy range, the first-level severe pollution warning Y1 and the second-level severe pollution warning Y2 output by the data analysis and judgment module, and transmits the completed values ​​back to the data analysis and judgment module;

[0057] S5. The data analysis and judgment module transmits the road surface pollution degree result to the output and alarm module, and the output and alarm module outputs and displays the road surface pollution degree, and alarms after triggering the unhealthy early warning mechanism.

[0058] like Figure 1-Figure 2 As shown, in the step S1, the input control module marks two technical indicators of the elongation at break G and the tensile strength P of the polyurethane adhesive, and presets the initial range thresholds of the marked technical indicators as T_G and T_P, respectively. By collecting the polyurethane adhesive mixture of road materials, the processed polyurethane adhesive is obtained to obtain the technical indicators of the elongation at break G1 and the tensile strength P1 of the polyurethane adhesive, wherein the collected road materials include plain areas, plateau areas, mountainous areas, hilly areas and basin areas.

[0059] Specifically, the elongation at break G1 is obtained by a tensile test. During the test, the polyurethane adhesive solid sample is fixed on the test machine, and the length at this time is recorded as the original length of the test solid sample. The tensile force is gradually increased. When the internal or surface damage area of ​​the solid sample plate is more than 50%, the damage state occurs in the molded plate. At this time, it is recorded as the polyurethane adhesive fracture. The length at this time is recorded as the total length of the test solid sample when it fractures. The elongation at break is calculated by the ratio of the elongation at the fracture of the solid sample to the original length, referring to the following formula:

[0060]

[0061] ε: elongation at break, expressed in %;

[0062] L1: total length of the solid sample when it breaks, in mm;

[0063] L0: original length of the solid sample tested, in mm;

[0064] The tensile strength P1 is obtained by tensile test. During the test, the polyurethane adhesive solid sample is fixed on the test machine, and the length of the polyurethane adhesive solid sample at this time is recorded as the initial bonding surface length, and the thickness at this time is recorded. The tensile force is gradually increased. When the internal or surface damage area of ​​the solid sample plate is more than 50%, the damage state occurs in the molded plate. At this time, it is recorded as the polyurethane adhesive fracture. The maximum load at the fracture is recorded. The tensile strength is the load per unit area of ​​the solid sample. The tensile strength is calculated by the following formula:

[0065]

[0066] σ: tensile strength, in MPa;

[0067] F: Maximum load when the solid sample is broken, in N;

[0068] B: initial bonding surface length of the tested solid sample, in mm;

[0069] D: Thickness of the solid sample under test, in mm.

[0070] like Figure 1-Figure 3 As shown, the steps of using the adaptive neural network algorithm in step S2 are:

[0071] S201, training neural network model: collecting a large amount of historical data of technical indicators (G, P) of elongation at break G and tensile strength P of polyurethane adhesive, as well as corresponding excellent, medium and poor performance characteristic labels of polyurethane adhesive, and removing outliers from the historical data;

[0072] Input: Dataset D, which is a large amount of historical data of technical indicators (G, P) of elongation at break G and tensile strength P of polyurethane adhesives collected through preprocessing, with preset initial range thresholds T_G and T_P, and output: excellent, medium and poor performance characteristics of polyurethane adhesives;

[0073] The specific steps of neural network model training are as follows:

[0074] The input layer presets 2 nodes:

[0075] Node 11: receiving a preset initial range threshold value T_G of the elongation at break and a preset initial range threshold value T_P of the tensile strength of the polyurethane adhesive;

[0076] Node 12: receiving the technical indicators of elongation at break G and tensile strength P of polyurethane adhesive;

[0077] Hidden layer: 4 cross nodes are preset, f(G, P) = 1 means that the input polyurethane adhesive elongation at break G and tensile strength P technical indicators (G, P) meet the range conditions, and the calculation of this node is performed; f(G, P) = 0 means that the input polyurethane adhesive elongation at break G and tensile strength P technical indicators (G, P) do not meet the range conditions, and the calculation of this node is not performed. The calculation formula preset in the hidden layer is as follows:

[0078] Node 21: If G ≥ T_G and P ≥ T_P, then f1(G, P) = 1; otherwise f1(G, P) = 0;

[0079] Node 22: If G ≥ T_G and P < T_P, then f2(G, P) = 1; otherwise f2(G, P) = 0;

[0080] Node 23: If G < T_G and P ≥ T_P, then f3(G, P) = 1; otherwise f3(G, P) = 0;

[0081] Node 24: If G < T_G and P < T_P, then f4(G, P) = 1; otherwise f4(G, P) = 0;

[0082] The output layer outputs 3 nodes:

[0083] Node 31: When the technical indicators (G, P) of the elongation at break G and tensile strength P meet the conditions of Node 21, the excellent performance characteristics of the polyurethane adhesive are output;

[0084] Node 32: When the technical indicators (G, P) of the elongation at break G and tensile strength P meet the conditions of Node 22 or Node 23, the medium performance characteristics of the polyurethane adhesive are output;

[0085] Node 33: When the technical indicators (G, P) of the elongation at break G and tensile strength P meet the conditions of Node 24, the poor performance characteristics of the polyurethane adhesive are output;

[0086] S202. Performance prediction of the neural network model: Input the new data of the elongation at break G and tensile strength P into the trained neural network. If it is verified that the performance characteristics of the polyurethane adhesive output by the neural network are inconsistent with the predicted performance characteristics of the polyurethane adhesive, return to the previous step S201. If it is verified that the performance characteristics of the polyurethane adhesive output by the neural network are consistent with the predicted performance characteristics of the polyurethane adhesive, perform the next step S203;

[0087] S203. Application of the neural network model: Deploy the neural network model that has reached the expectation in training to the actual application. Input the technical indicators of the elongation at break G1 and tensile strength P1 of the obtained polyurethane adhesive, and use the adaptive neural network model that has reached the expectation in training from step S201 to step S202 to evaluate the performance characteristics of the polyurethane adhesive through the adaptive algorithm, and then output that the performance characteristics of the polyurethane adhesive belong to excellent, medium or low grade.

[0088] As Figure 1-Figure 3 shown, in the S3 step, the data analysis and judgment module determines whether the degree of road surface pollution belongs to the healthy range according to the performance characteristics of the polyurethane adhesive output by the neural network module. For the degree of road surface pollution belonging to the non - healthy range, set non - healthy warning levels: the first - level severe pollution warning Y1, the second - level severe pollution warning Y2;

[0089] Specifically, the specific process of setting the unhealthy warning level is: based on the excellent performance characteristics of the polyurethane adhesive, the road surface pollution level is judged to be within the healthy range; based on the medium output of the polyurethane adhesive performance characteristics, the road surface pollution level is judged to be within the unhealthy range, which is the first level severe pollution warning Y1; based on the poor output of the polyurethane adhesive performance characteristics, the road surface pollution level is judged to be within the unhealthy range, which is the second level severe pollution warning Y2.

[0090] like Figure 1-Figure 3 As shown, in the step S4, the determined healthy range, the first level severe pollution warning Y1 of the non-health warning level and the second level severe pollution warning Y2 are reviewed and tested;

[0091] Specifically, in step S4, it is checked whether the output of the healthy range of road surface pollution and the unhealthy warning level have correctly triggered the excellent, medium and poor thresholds of the corresponding polyurethane adhesive performance characteristics, thereby confirming the unhealthy warning level of road surface pollution, and further manually reviewing the automatically output warning level, including whether a higher level of road surface pollution develops from a lower level of road surface pollution.

[0092] like Figure 1-Figure 3 As shown, in the step S5, the road surface pollution degree is output, and the output and alarm module outputs the display result and an alarm is issued after the early warning mechanism is triggered;

[0093] Specifically, in step S5, the polyurethane adhesive of the road material component is collected, the elongation at break G1 and the tensile strength P1 technical indicators of the polyurethane adhesive are detected, and the performance characteristics of the polyurethane adhesive are evaluated by using an adaptive neural network adaptive algorithm, so as to judge the healthy range and unhealthy range of road surface pollution, set an early warning level for the unhealthy range of road surface pollution, further verify and calculate the output of the healthy degree and unhealthy early warning level of road surface pollution, display the road surface pollution degree on the computer display screen, and at the same time, trigger an early warning mechanism for the unhealthy range and then give an alarm;

[0094] The setting of the early warning mechanism: when the output shows that the pollution degree of the road surface belongs to the unhealthy range, the early warning mechanism is triggered, and the relevant departments are reminded to take corresponding measures to clean the road surface pollution according to the output of the first serious pollution warning Y1 and the second serious pollution warning Y2.

[0095] like Figure 1-Figure 3 As shown, the present invention solves the shortcomings of the prior art that the assessment of the degree of road surface pollution by monitoring dust, water and heavy metals on the road surface can no longer meet the demand for monitoring the degree of road surface pollution. The present invention works as follows:

[0096] Data Tags:

[0097] Mark the elongation at break G and tensile strength P of the polyurethane adhesive, preset the initial range thresholds of the marked technical indicators as T_G and T_P respectively, and obtain the technical indicators of the elongation at break G1 and tensile strength P1 of the polyurethane adhesive by collecting the polyurethane adhesive;

[0098] Initial range thresholds T_G, T_P:

[0099] T_G: elongation at break (25°C) (%), greater than 50% and meeting the elongation at break equal to 50%;

[0100] The elongation at break G is expressed as a percentage: %;

[0101] T_P: tensile strength (25℃) (MPa), greater than 20MPa and meeting the tensile strength equal to 20MPa;

[0102] The unit of tensile strength P is: MPa (megapascal);

[0103] Data collection:

[0104] A sampler is used to collect a polyurethane adhesive mixture sample on the asphalt road surface, the polyurethane adhesive is extracted, and the technical indicators of the elongation at break G1 and tensile strength P1 of the polyurethane adhesive are obtained through a tensile test;

[0105] Neural network calculations:

[0106] The adaptive neural network adaptive algorithm is used to analyze the data of the technical indicators of elongation at break G1 and tensile strength P1 of polyurethane adhesive;

[0107] The input layer receives the technical indicators of elongation at break G1 and tensile strength P1 of the polyurethane adhesive;

[0108] The hidden layer uses convolutional layers, recurrent layers, and fully connected layers to extract and transform features of the input data;

[0109] The output layer outputs the performance characteristics of the polyurethane adhesive, including whether the performance characteristics of the polyurethane adhesive are excellent, medium and poor;

[0110] Pollution warning level judgment:

[0111] According to the performance characteristics of polyurethane adhesive, the degree of road surface pollution is judged as healthy range and unhealthy range. Then, according to the excellent output of polyurethane adhesive performance characteristics, the degree of road surface pollution is judged to be in the healthy range. If the degree of road surface pollution is in the unhealthy range, an unhealthy range warning level is set: the medium output of polyurethane adhesive performance characteristics is judged as the first severe pollution warning Y1, and the poor output of polyurethane adhesive performance characteristics is judged as the second severe pollution warning Y2;

[0112] Verification inspection:

[0113] Review and verify whether the road pollution level is within the healthy range or the unhealthy warning level, and further verify the output warning level through manual review;

[0114] The output shows:

[0115] Outputting the healthy range or unhealthy warning level of road surface pollution that has passed the manual review and inspection on a computer display screen;

[0116] Early warning mechanism setting:

[0117] When the output shows that the pollution level of the road surface is within the unhealthy range, the early warning mechanism is triggered, and the relevant departments are reminded to take corresponding measures to clean up the road surface pollution according to the output of the first severe pollution early warning Y1 and the second severe pollution early warning Y2.

[0118] See also Figure 4 As shown, an adaptive road surface pollution monitoring system comprises:

[0119] An input control module, connected to the neural network module, is used to preset the initial range threshold of the basic technical indicators of the polyurethane adhesive of the road material, and obtain the technical indicators and mark them;

[0120] The data analysis and judgment module is connected with the input control module, the neural network module, the review and inspection module, and the output and alarm module, and is used to receive the preset initial range threshold value, the acquired technical indicators and information tags of the input control module, use the numerical range calculated by the neural network module to judge the healthy range and unhealthy warning level of road surface pollution, cooperate with the review and inspection module to inspect the range of road surface pollution, and transmit the data analysis completed by the review and inspection to the data analysis and judgment module for calculation;

[0121] A neural network module, connected to the input control module and the data analysis and judgment module, is used to calculate and evaluate the performance characteristics of the polyurethane adhesive;

[0122] A review and inspection module, connected to the data analysis module, is used to review the judgment of the data analysis and judgment module on the degree of road surface pollution, and to review the judged unhealthy warning level;

[0123] The output and alarm module is connected to the data analysis and judgment module, and is used to output and display the degree of road surface pollution and to alarm after triggering the unhealthy early warning mechanism.

[0124] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0125] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and rules of the present invention shall be included in the protection scope of the present invention.

Claims

1. An adaptive road surface pollution monitoring method, characterized in that: S1, marking the elongation at break G and tensile strength P of the polyurethane adhesive, presetting the initial range thresholds of the marked technical indicators as T_G and T_P respectively, collecting the polyurethane adhesive, obtaining the technical indicators of the elongation at break G1 and tensile strength P1 of the polyurethane adhesive and transmitting them to the neural network module; S2, the neural network module uses the obtained technical indicators of elongation at break G1 and tensile strength P1 to evaluate the performance characteristics of the polyurethane adhesive through an adaptive neural network and transmits them to the data analysis and judgment module; S3, the data analysis and judgment module judges the degree of road surface pollution according to the performance characteristics of the polyurethane adhesive output by the neural network module, and sets the unhealthy warning level if the road surface pollution degree belongs to the unhealthy range: the first level severe pollution warning Y1, the second level severe pollution warning Y2; S4, the review and inspection module reviews the healthy range, the first-level severe pollution warning Y1 and the second-level severe pollution warning Y2 output by the data analysis and judgment module, and transmits the completed values ​​back to the data analysis and judgment module; S5. The data analysis and judgment module transmits the road surface pollution degree result to the output and alarm module, and the output and alarm module outputs and displays the road surface pollution degree, and alarms after triggering the unhealthy early warning mechanism.

2. The adaptive road surface pollution monitoring method according to claim 1, characterized in that: The polyurethane adhesive collection steps are as follows: using a sampler to collect a polyurethane adhesive mixture sample at the surface of an asphalt road, grinding and sieving the sample, mixing the sieved particles with a solvent, N-methylpyrrolidone, and stirring to promote the dissolution of the polyurethane adhesive, filtering to separate the solvent in which the polyurethane adhesive is dissolved and the undissolved solid particles, and purifying the solvent containing the polyurethane adhesive by a chemical separation technology of distillation. After the purification process, a processed polyurethane adhesive is obtained, and technical indicators of the elongation at break G1 and the tensile strength P1 of the polyurethane adhesive are obtained.

3. The adaptive road surface pollution monitoring method according to claim 1, characterized in that: The adaptive neural network performs the adaptive algorithm steps as follows: S201, training neural network model: collecting historical data of technical indicators (G, P) of elongation at break G and tensile strength P of polyurethane adhesive, as well as excellent, medium and poor performance characteristic labels of polyurethane adhesive, and removing outliers from the historical data; Input: Dataset D, which is the historical data of the technical indicators (G, P) of the elongation at break G and tensile strength P of polyurethane adhesive collected through preprocessing, with preset initial range thresholds T_G and T_P, and output: excellent, medium and poor performance characteristics of polyurethane adhesive; The specific steps of neural network model training are as follows: The input layer presets 2 nodes: Node 11: receiving a preset initial range threshold value T_G of the elongation at break and a preset initial range threshold value T_P of the tensile strength of the polyurethane adhesive; Node 12: receiving the technical indicators of elongation at break G and tensile strength P of polyurethane adhesive; Hidden layer: Preset 4 cross nodes. f(G, P) = 1 indicates that the technical indicators (G, P) of the elongation at break G and tensile strength P of the input polyurethane adhesive meet the range conditions, and the calculation of this node is executed; f(G, P) = 0 indicates that the technical indicators (G, P) of the elongation at break G and tensile strength P of the input polyurethane adhesive do not meet the range conditions, and the calculation of this node is not executed. The preset calculation formula of the hidden layer is as follows: Node 21: If G ≥ T_G and P ≥ T_P, then f1(G, P) = 1; otherwise f1(G, P) = 0; Node 22: If G ≥ T_G and P < T_P, then f2(G, P) = 1; otherwise f2(G, P) = 0; Node 23: If G < T_G and P ≥ T_P, then f3(G, P) = 1; otherwise f3(G, P) = 0; Node 24: If G < T_G and P < T_P, then f4(G, P) = 1; otherwise f4(G, P) = 0; The output layer outputs 3 nodes: Node 31: When the technical indicators (G, P) of the elongation at break G and tensile strength P meet the conditions of Node 21, the excellent performance characteristics of the polyurethane adhesive are output; Node 32: When the technical indicators (G, P) of the elongation at break G and tensile strength P meet the conditions of Node 22 or Node 23, the medium performance characteristics of the polyurethane adhesive are output; Node 33: When the technical indicators (G, P) of the elongation at break G and tensile strength P meet the conditions of Node 24, the poor performance characteristics of the polyurethane adhesive are output; S202. Performance prediction of the neural network model: Input the new data of the elongation at break G and tensile strength P into the trained neural network. If it is verified that the performance characteristics of the polyurethane adhesive output by the neural network are inconsistent with the predicted performance characteristics of the polyurethane adhesive, return to the previous step S201. If it is verified that the performance characteristics of the polyurethane adhesive output by the neural network are consistent with the predicted performance characteristics of the polyurethane adhesive, execute the next step S203; S203. Application of the neural network model: Deploy the neural network model that has reached the expectation in training to the actual application, input the technical indicators G1 of the elongation at break and P1 of the tensile strength of the obtained polyurethane adhesive, and use the adaptive neural network model that has reached the expectation in training from step S201 to step S202 to evaluate the performance characteristics of the polyurethane adhesive through the adaptive algorithm, and then output that the performance characteristics of the polyurethane adhesive belong to excellent, medium or low grade.

4. The adaptive road surface pollution monitoring method according to claim 1, characterized in that: When the performance characteristics of the polyurethane adhesive are output as excellent, it is judged that the degree of road surface pollution belongs to the healthy range. When the performance characteristics of the polyurethane adhesive are output as medium, it is judged that the degree of road surface pollution belongs to the first-level severe pollution warning Y1 of the unhealthy range. When the performance characteristics of the polyurethane adhesive are output as poor, it is judged that the degree of road surface pollution belongs to the second-level severe pollution warning Y2 of the unhealthy range.

5. The adaptive road surface pollution monitoring method according to claim 1, characterized in that: The review and inspection module checks the thresholds of excellent, medium and poor performance characteristics of the polyurethane adhesive triggered by the output of the road surface pollution according to the healthy range and unhealthy warning level of the road surface pollution output by the data analysis and judgment module, and then confirms the healthy range and unhealthy warning level of the road surface pollution, and further manually reviews the warning level automatically output.

6. The adaptive road surface pollution monitoring method according to claim 1, characterized in that: The adaptive neural network is used to perform an adaptive algorithm to output the performance characteristics of the polyurethane adhesive, determine the healthy degree of road surface pollution and the unhealthy warning level, and then output the road surface pollution level on a computer display screen through a review test, and at the same time trigger an alarm after the unhealthy warning mechanism is triggered.

7. An adaptive road surface pollution monitoring system, characterized in that: An input control module is used to preset the initial range threshold of the basic technical indicators of the polyurethane adhesive of the road material, and obtain the technical indicators and mark them; A data analysis and judgment module is used to receive the preset initial range threshold, the acquired technical indicators and information tags, use the calculated polyurethane adhesive performance characteristic value range to judge the healthy range of road surface pollution and the unhealthy warning level, cooperate with the inspection and judgment of the road surface pollution degree range, and calculate the data analysis completed by the review inspection; A neural network module for calculating and evaluating the performance characteristics of the polyurethane adhesive; A review and inspection module is used to review and inspect the judgment of the healthy range and unhealthy warning level of road surface pollution; The output and alarm module is used to output and display the pollution degree of the road surface and to alarm after triggering the unhealthy early warning mechanism.

Images