Road pollution evaluation method and system based on AHP and grey clustering

Abstract

The invention discloses a road pollution evaluation method and system based on AHP and grey clustering, which are applied to the technical field of intelligent pollution evaluation, and the method comprises the steps: constructing an expert database model and an expert directory; acquiring feature data of a target road; inputting the feature data and the change of the expert into an expert database model to obtain an evaluation weight; selecting a plurality of experts with the maximum evaluation weight as practically invited experts; and analyzing the evaluation result according to the evaluation weight to generate pollution evaluation. According to the method, appropriate experts can be effectively screened out for invitation evaluation in road pollution evaluation, evaluation errors caused by subjective evaluation of the experts are effectively reduced, influence factors about road construction pavement construction pollution are fully considered, a set of scientific and reasonable road construction pavement pollution index system is constructed, and the evaluation efficiency is improved. The method can provide a certain theoretical basis for evaluation of the road surface pollution level in highway engineering construction, and provides reference for formulation of pollution prevention and control measures in highway construction.

Description

Technical Field

[0001] The invention relates to intelligent pollution evaluation technology, and in particular to a road pollution evaluation method and system based on AHP and grey clustering. Background Art

[0002] Green highway is one of the sustainable practices introduced in highway development, including transportation and environmental sustainability. On the one hand, the impact on the environment is judged by analyzing the characteristics of pollutants, such as noise, runoff pollution, habitat disturbance, land use, air quality, climate change, vibration, and pollution to plants and wildlife. On the other hand, in the field of transportation systems, more and more advanced construction technologies and management methods help to establish a safe, efficient, economical, and environmentally friendly sustainable "zero pollution" pavement. To achieve the "zero pollution" construction process, the application of new technologies and pollution prevention and control and evaluation during construction are crucial, and the importance of pollution problems during pavement construction is obvious. This not only involves environmental pollution problems during operations, but also affects the quality of the pavement, which has an important impact on the safety of passing vehicles, risk prevention and control, and ensuring a high level of user comfort and safety.

[0003] In the existing technology, the evaluation of road pollution is mainly carried out by expert evaluation; however, different experts' pollution evaluations on different road conditions are based on subjective factors, and the evaluation levels are uneven. How to select suitable experts to accurately evaluate road pollution has become a pain point in the industry. Summary of the invention

[0004] In order to at least overcome the above-mentioned deficiencies in the prior art, the purpose of the present application is to provide a road pollution evaluation method and system based on AHP and grey clustering.

[0005] In a first aspect, an embodiment of the present application provides a road pollution evaluation method based on AHP and grey clustering, comprising: Acquire historical expert evaluation data of road pollution, and construct an expert database model and an expert list based on the historical expert evaluation data; the historical expert evaluation data is data generated by experts who have conducted road pollution evaluations; When performing pollution evaluation on a target road, characteristic data of the target road is obtained; the characteristic data includes material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics; Obtain a list of experts to be invited, and obtain a number corresponding to each expert in the expert list according to the expert list; Inputting the feature data and the serial number into the expert database model, and obtaining the evaluation weight corresponding to each expert output by the expert database model; Selecting multiple experts with the largest evaluation weights as the experts actually invited; The pollution of the target road is evaluated by the experts actually invited to generate an evaluation result, and the evaluation result is subjected to grey clustering analysis according to the evaluation weight to generate a pollution grade as the final pollution evaluation of the target road.

[0006] In a possible implementation, building an expert database model and an expert directory includes: Obtaining road characteristic data, expert names and evaluation weights at each expert evaluation from the historical expert evaluation data; the evaluation weights are obtained by AHP calculation; the road characteristic data include material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics; Numbering each expert to form a correspondence between the expert name and the number as the expert list; The road feature data and the numbers corresponding to the expert names are used as input data, and the evaluation weights are used as output data to train a neural network model to form the expert database model.

[0007] In a possible implementation, calculating the evaluation weight by AHP includes: Acquire the AHP scale scoring data of each expert in each expert evaluation from the historical expert evaluation data; the AHP scale is a 1-9 scale or an index scale; Calculate the subjective weight value of each expert in this evaluation based on the AHP scale scoring data of each expert in the same expert evaluation; The subjective weight value is converted into an objective weight value by the entropy weight method as the evaluation weight of the expert in this expert evaluation.

[0008] In a possible implementation, the calculation of the subjective weight value includes: Acquire evaluation data of each element at each expert evaluation from the historical expert evaluation data; After comparing each element pairwise, the judgment matrix is ​​constructed by assigning values ​​through the AHP scale; the unit in the judgment matrix is ​​the importance ratio scale between two elements; Checking the consistency of the judgment matrix, and after the consistency check passes, normalizing the row vectors of the judgment matrix and calculating the arithmetic mean to form a weight vector; The weight vector is constructed as a matrix to form the subjective weight value.

[0009] In a possible implementation, converting the subjective weight value into an objective weight value as the evaluation weight of the expert in this expert evaluation by using an entropy weight method includes: Assume that there is an ideal optimal expert in the same expert evaluation; Calculating the difference between the subjective weight value and the optimal weight value as difference data; the optimal weight value is the subjective weight value of the optimal expert; constructing an entropy model of expert evaluation results based on the difference data; Calculating the entropy weight of each expert according to the entropy model; The subjective weight value is weighted according to the entropy weight to generate the objective weight value of each expert.

[0010] In a possible implementation, performing grey cluster analysis on the evaluation results according to the evaluation weights to generate a pollution level as the final pollution evaluation of the target road includes: The corresponding indicators in the evaluation results are divided into multiple gray classes according to the evaluation requirements, and the center point of each gray class, the turning point of the front gray class and the turning point of the tail gray class are calculated as grayscale parameters; Constructing a whitening weight function according to the grayscale parameter, and calculating the degree of membership of each gray class according to the whitening weight function and the center point; The clustering coefficient is calculated according to the membership degree and the evaluation weight, and the evaluation result is divided into corresponding gray classes according to the clustering coefficient to form a pollution evaluation result.

[0011] In a second aspect, the embodiment of the present application further provides a road pollution evaluation system based on AHP and grey clustering, comprising: A modeling unit is configured to obtain historical expert evaluation data of road pollution, and to construct an expert database model and an expert list according to the historical expert evaluation data; the historical expert evaluation data is data generated by experts who have evaluated road pollution; An acquisition unit is configured to acquire characteristic data of the target road when performing pollution evaluation on the target road; the characteristic data includes material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics; A computing unit configured to obtain a list of experts to be invited, and obtain a number corresponding to each expert in the expert list according to the expert list; Inputting the feature data and the serial number into the expert database model, and obtaining the evaluation weight corresponding to each expert output by the expert database model; Selecting multiple experts with the largest evaluation weights as the experts actually invited; The evaluation unit is configured to evaluate the pollution of the target road by actually inviting experts to generate an evaluation result, and to perform grey cluster analysis on the evaluation result according to the evaluation weight to generate a pollution level as a final pollution evaluation of the target road.

[0012] In a possible implementation, the modeling unit is further configured to: Obtaining road characteristic data, expert names and evaluation weights at each expert evaluation from the historical expert evaluation data; the evaluation weights are obtained by AHP calculation; the road characteristic data include material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics; Numbering each expert to form a correspondence between the expert name and the number as the expert list; The road feature data and the numbers corresponding to the expert names are used as input data, and the evaluation weights are used as output data to train a neural network model to form the expert database model.

[0013] In a possible implementation, the modeling unit is further configured to: Acquire the AHP scale scoring data of each expert in each expert evaluation from the historical expert evaluation data; the AHP scale is a 1-9 scale or an index scale; Calculate the subjective weight value of each expert in this evaluation based on the AHP scale scoring data of each expert in the same expert evaluation; The subjective weight value is converted into an objective weight value by the entropy weight method as the evaluation weight of the expert in this expert evaluation.

[0014] In a possible implementation, the modeling unit is further configured to: Acquire evaluation data of each element at each expert evaluation from the historical expert evaluation data; After comparing each element pairwise, the judgment matrix is ​​constructed by assigning values ​​through the AHP scale; the unit in the judgment matrix is ​​the importance ratio scale between two elements; Checking the consistency of the judgment matrix, and after the consistency check passes, normalizing the row vectors of the judgment matrix and calculating the arithmetic mean to form a weight vector; The weight vector is constructed as a matrix to form the subjective weight value.

[0015] Compared with the prior art, the present invention has the following advantages and beneficial effects: The present invention can effectively screen out suitable experts for invited evaluation in road pollution evaluation, effectively reduce the evaluation error caused by subjective evaluation of experts, and fully consider the influencing factors of road pavement construction pollution in highway construction, and construct a set of scientific and reasonable road pavement pollution index system for highway construction, which can provide a certain theoretical basis for the assessment of road pavement pollution level in highway engineering construction, and provide a reference for the formulation of pollution prevention and control measures in highway construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings: Figure 1 This is a schematic diagram of the method steps of an embodiment of the present application. DETAILED DESCRIPTION

[0017] To make the purpose, technical scheme and advantages of the embodiment of the present application clearer, the technical scheme in the embodiment of the present application will be clearly and completely described below in conjunction with the drawings in the embodiment of the present application. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the embodiment of the present application. It should be understood that the operations of the flowchart can be implemented out of order, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart under the guidance of the content of the present application, or remove one or more operations from the flowchart.

[0018] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0019] Please refer to Figure 1 , which is a flow chart of a road pollution evaluation method based on AHP and gray clustering provided in an embodiment of the present invention. Furthermore, the road pollution evaluation method based on AHP and gray clustering may specifically include the contents described in the following steps S1 to S6.

[0020] S1: Obtain historical expert evaluation data on road pollution, and construct an expert database model and an expert list based on the historical expert evaluation data; the historical expert evaluation data is data generated by experts conducting road pollution evaluation; S2: when performing pollution evaluation on a target road, obtaining characteristic data of the target road; the characteristic data includes material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics; S3: Obtain a list of experts to be invited, and obtain a number corresponding to each expert in the expert list according to the expert list; S4: inputting the feature data and the serial number into the expert database model, and obtaining the evaluation weight corresponding to each expert output by the expert database model; S5: Selecting multiple experts with the largest evaluation weights as the experts actually invited; S6: The experts actually invited evaluate the pollution of the target road to generate an evaluation result, and perform grey cluster analysis on the evaluation result according to the evaluation weight to generate a pollution grade as the final pollution evaluation of the target road.

[0021] When implementing the embodiment of the present application, it is necessary to first obtain the historical expert evaluation data of the road pollution evaluation that has been conducted before for the construction of the expert database model and the expert list; wherein the expert list is the correspondence between the expert name and the number; the expert database model is used to identify the evaluation ability of experts under different road conditions. When conducting an actual road assessment, the weight data corresponding to each expert can be identified in the expert database model through the characteristics of the target road, such as material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics, so as to accurately evaluate the most suitable experts for the target road for invitation. It should be understood that in order to accurately express the characteristics of the target road, it is necessary to digitize the material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics. In the embodiment of the present application, the material pollution feature B1 includes four features: mortar concrete pollution C11, soil and stone pollution C12, asphalt impurity pollution C13 and road heavy metal pollution C14; the operation pollution feature B2 includes six features: cross-construction pollution C21, construction waste pollution C22, auxiliary construction pollution C23, greening project pollution C24, operation dust pollution C25 and road field test pollution C26; the mechanical pollution feature B3 includes four features: mechanical vehicle rolling pollution C31, vehicle exhaust pollution C32, oil pollution C33 and water pollution C34; the environmental pollution feature B4 includes three features: construction temperature pollution C41, wind pollution C42 and rain and snow pollution C43. In the process of dataization, it is necessary to evaluate whether the corresponding road has the above-mentioned pollution. If it exists, the corresponding pollution is assigned a value of 1, and the non-existent pollution is assigned a value of 0, and then the values ​​of the material pollution feature, the operation pollution feature, the mechanical pollution feature and the environmental pollution feature are calculated. At this time, the corresponding features of the road can be represented as an array {B1, B2, B3, B4}, that is: {C11+C12+C13+C14, C21+C22+C23+C24+C25+C26, C31+C32+C33+C34, C41+C42+C43}. For example, a road has mortar concrete pollution, soil and stone pollution, and asphalt impurity pollution, but no heavy metal pollution on the road surface. At this time, the corresponding B1 value is 3; this method can effectively characterize the characteristics of different roads, which is convenient for expert screening.

[0022] When evaluating the pollution of the target road, that is, the road that needs to be evaluated, it is necessary to first obtain the corresponding feature data and the list of experts to be invited. The experts in the expert list need to be converted into corresponding numbers through the expert list, and together with the feature data, form an array to input the expert library model, and then obtain the evaluation weight corresponding to the number of each expert. The higher the evaluation weight, the more accurate the pollution assessment of the target road by this expert; at this time, selecting multiple experts with the largest evaluation weight as the actual invited experts can effectively improve the accuracy of the final evaluation result; the final pollution evaluation result can be obtained by gray clustering analysis of these expert evaluation results. This application can effectively screen out suitable experts for invitation evaluation in road pollution evaluation through the above technical scheme, effectively reduce the evaluation error caused by the subjective evaluation of experts, and fully consider the influencing factors of road construction pollution, and construct a set of scientific and reasonable road construction road pollution index system, which can provide a certain theoretical basis for the assessment of road pollution level in highway engineering construction, and provide a reference for the formulation of pollution prevention and control measures in highway construction.

[0023] In a possible implementation, building an expert database model and an expert directory includes: Obtaining road characteristic data, expert names and evaluation weights at each expert evaluation from the historical expert evaluation data; the evaluation weights are obtained by AHP calculation; the road characteristic data include material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics; Numbering each expert to form a correspondence between the expert name and the number as the expert list; The road feature data and the numbers corresponding to the expert names are used as input data, and the evaluation weights are used as output data to train a neural network model to form the expert database model.

[0024] When implementing the embodiment of the present application, after obtaining a large amount of road feature data, expert names and evaluation weights, a neural network model is used to construct an expert library model, wherein the input data is the road feature data and the expert number, and the output data is the corresponding evaluation weight. For example, in order to realize the construction of the expert library model, the input layer in the trained neural network model needs to include at least 5 input neurons, corresponding to 4 road feature data and 1 expert number, and the output layer needs to include 1 output neuron corresponding to the evaluation weight.

[0025] In a possible implementation, calculating the evaluation weight by AHP includes: Acquire the AHP scale scoring data of each expert in each expert evaluation from the historical expert evaluation data; the AHP scale is a 1-9 scale or an index scale; Calculate the subjective weight value of each expert in this evaluation based on the AHP scale scoring data of each expert in the same expert evaluation; The subjective weight value is converted into an objective weight value by the entropy weight method as the evaluation weight of the expert in this expert evaluation.

[0026] When implementing the embodiment of the present application, in order to digitize the evaluation weights in the historical expert evaluation data, the AHP scale scoring is adopted, and the subjective weights are converted into objective weights through the entropy weight method, which can effectively characterize the evaluation performance of different experts in different scenarios.

[0027] In a possible implementation, the calculation of the subjective weight value includes: Acquire evaluation data of each element at each expert evaluation from the historical expert evaluation data; After comparing each element pairwise, the judgment matrix is ​​constructed by assigning values ​​through the AHP scale; the unit in the judgment matrix is ​​the importance ratio scale between two elements; Checking the consistency of the judgment matrix, and after the consistency check passes, normalizing the row vectors of the judgment matrix and calculating the arithmetic mean to form a weight vector; The weight vector is constructed as a matrix to form the subjective weight value.

[0028] When implementing the embodiment of the present application, the AHP scale may adopt a 1-9 scale or an index scale as shown in the following table: Table 1 AHP ​​index scale For example, in the construction of the judgment matrix, if there are m experts participating in the scoring, according to the opinions of the experts, the elements of the same layer in the evaluation object are compared two by two, and the values ​​are assigned using the exponential scale to construct the judgment matrix A=(aij ) nxn In the formula, a ij is the importance ratio scale of the ith element and the jth element. The specific comparison method is shown in Table 1.

[0029] In the consistency check, it is necessary to first calculate the judgment matrix consistency index CL according to the following formula: in is the maximum eigenvalue of the judgment matrix, n is the matrix order; Then calculate the random consistency ratio CR according to the following formula: In the formula, is the average consistency index value. When CR < 0.1, the judgment matrix is ​​accepted, otherwise it is not accepted.

[0030] When the weight vector is constructed as a matrix to form the subjective weight value, the arithmetic mean of the normalized n row vectors of the judgment matrix is ​​approximated as the weight vector according to the following formula: And finally form the subjective weight vector of m experts: .

[0031] In a possible implementation, converting the subjective weight value into an objective weight value as the evaluation weight of the expert in this expert evaluation by using an entropy weight method includes: Assume that there is an ideal optimal expert in the same expert evaluation; Calculating the difference between the subjective weight value and the optimal weight value as difference data; the optimal weight value is the subjective weight value of the optimal expert; constructing an entropy model of expert evaluation results based on the difference data; Calculating the entropy weight of each expert according to the entropy model; The subjective weight value is weighted according to the entropy weight to generate the objective weight value of each expert.

[0032] When the embodiment of the present application is implemented, the subjective weight value can be converted into an objective weight value by the entropy weight method. In the conversion process, it is necessary to first assume that there is an ideal optimal expert S*, whose subjective weight vector is , by comparing the subjective weight results of each expert with the subjective weight results of the best expert S*, the quality of the subjective weight is measured by the size of the difference. The subjective weight level vector E of the kth expert is k for: The entropy model constructed is: in: Among them, H K The value of represents the uncertainty of the expert evaluation results. K The larger the value of is, the lower the expert's evaluation level is. K The smaller the value, the opposite is true.

[0033] The entropy weight of each expert is calculated according to the following formula: Finally, the subjective weight value is weighted according to the entropy weight to generate the objective weight value of each expert according to the following formula: .

[0034] In a possible implementation, performing grey cluster analysis on the evaluation results according to the evaluation weights to generate a pollution level as the final pollution evaluation of the target road includes: The corresponding indicators in the evaluation results are divided into multiple gray classes according to the evaluation requirements, and the center point of each gray class, the turning point of the front gray class and the turning point of the tail gray class are calculated as grayscale parameters; Constructing a whitening weight function according to the grayscale parameter, and calculating the degree of membership of each gray class according to the whitening weight function and the center point; The clustering coefficient is calculated according to the membership degree and the evaluation weight, and the evaluation result is divided into corresponding gray classes according to the clustering coefficient to form a pollution evaluation result.

[0035] When implementing the embodiment of the present application, the grayscale clustering method is used for the final pollution evaluation; for example, a gray clustering evaluation model is constructed based on the improved center point triangle whitening weight function. Its definition and steps are as follows: Definition 1: Suppose there are n decision objects, m clustering indicators, and s different gray classes. According to the observation value x of object i (i=1,2....n) on indicator j (j=1,2...,m) ij (i=1,2…n,j1.,2..,m) classifies object i into gray class k (k∈.,...,s), which is called gray clustering.

[0036] Definition 2: For a gray class k (k∈{2,3... ,s-1}), the point that most likely belongs to gray class k is called the center point of the gray class.

[0037] Based on the above definition, the specific steps are as follows: Set the value range of index j to [a j ,b j ], divided into s gray classes according to the review requirements, and the turning points of gray class 1 are determined respectively , the turning point of gray class s , and the center point of gray class k .

[0038] For gray class 1 and gray class s, construct the corresponding lower limit measure whitening weight function and the upper limit measure function , at this time, let the evaluation value of evaluation index j be x, when x∈ or x∈ The membership degree of gray class 1 can be calculated according to the following formula: and the membership degree of gray class s : For the gray class k (k∈{2,3... ,s-1}), the connection points and the center point of the gray class k-1 as well as and the center point of the gray class K+1 , we get the triangular whitening weight function of index j about gray class k: For an observation value x of index j, when k=2, 3, ..., s-1, the membership degree of the gray class k (k∈{2,3..., s-1}) can be calculated by the following formula: : The clustering coefficient can be calculated according to the membership degree and the evaluation weight by the following formula: In the formula, is the whitening weight function of index k, W j is the weight of index j in comprehensive clustering.

[0039] Depend on Determine whether the object 𝑖 belongs to the grey class k*.

[0040] Based on the same inventive concept, the embodiment of the present application also provides a road pollution evaluation system based on AHP and grey clustering, including: A modeling unit is configured to obtain historical expert evaluation data of road pollution, and to construct an expert database model and an expert list according to the historical expert evaluation data; the historical expert evaluation data is data generated by experts who have evaluated road pollution; An acquisition unit is configured to acquire characteristic data of the target road when performing pollution evaluation on the target road; the characteristic data includes material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics; A computing unit configured to obtain a list of experts to be invited, and obtain a number corresponding to each expert in the expert list according to the expert list; Inputting the feature data and the serial number into the expert database model, and obtaining the evaluation weight corresponding to each expert output by the expert database model; Selecting multiple experts with the largest evaluation weights as the experts actually invited; The evaluation unit is configured to evaluate the pollution of the target road by actually inviting experts to generate an evaluation result, and to perform grey cluster analysis on the evaluation result according to the evaluation weight to generate a pollution level as a final pollution evaluation of the target road.

[0041] In a possible implementation, the modeling unit is further configured to: Obtaining road characteristic data, expert names and evaluation weights at each expert evaluation from the historical expert evaluation data; the evaluation weights are obtained by AHP calculation; the road characteristic data include material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics; Numbering each expert to form a correspondence between the expert name and the number as the expert list; The road feature data and the numbers corresponding to the expert names are used as input data, and the evaluation weights are used as output data to train a neural network model to form the expert database model.

[0042] In a possible implementation, the modeling unit is further configured to: Acquire the AHP scale scoring data of each expert in each expert evaluation from the historical expert evaluation data; the AHP scale is a 1-9 scale or an index scale; Calculate the subjective weight value of each expert in this evaluation based on the AHP scale scoring data of each expert in the same expert evaluation; The subjective weight value is converted into an objective weight value by the entropy weight method as the evaluation weight of the expert in this expert evaluation.

[0043] In a possible implementation, the modeling unit is further configured to: Acquire evaluation data of each element at each expert evaluation from the historical expert evaluation data; After comparing each element pairwise, the judgment matrix is ​​constructed by assigning values ​​through the AHP scale; the unit in the judgment matrix is ​​the importance ratio scale between two elements; Checking the consistency of the judgment matrix, and after the consistency check passes, normalizing the row vectors of the judgment matrix and calculating the arithmetic mean to form a weight vector; The weight vector is constructed as a matrix to form the subjective weight value.

[0044] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0045] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, or it can be an electrical, mechanical or other form of connection.

[0046] The units described as separate components may or may not be physically separated. As units, it is obvious that a person of ordinary skill in the art can realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0047] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0048] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a grid device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0049] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A road pollution evaluation method based on AHP and grey clustering, characterized in that: include: Obtaining historical expert evaluation data on road pollution, and constructing an expert database model and an expert list based on the historical expert evaluation data; The historical expert evaluation data is data generated by experts who have conducted road pollution evaluations; When performing pollution evaluation on a target road, characteristic data of the target road is obtained; the characteristic data includes material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics; Obtain a list of experts to be invited, and obtain a number corresponding to each expert in the expert list according to the expert list; Inputting the feature data and the serial number into the expert database model, and obtaining the evaluation weight corresponding to each expert output by the expert database model; Selecting multiple experts with the largest evaluation weights as the experts actually invited; The pollution of the target road is evaluated by the experts actually invited to generate an evaluation result, and the evaluation result is subjected to grey clustering analysis according to the evaluation weight to generate a pollution grade as the final pollution evaluation of the target road.

2. A road pollution evaluation method based on AHP and grey clustering according to claim 1, characterized in that: Building the expert database model and expert directory includes: Obtaining road characteristic data, expert names and evaluation weights at each expert evaluation from the historical expert evaluation data; the evaluation weights are obtained by AHP calculation; the road characteristic data include material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics; Numbering each expert to form a correspondence between the expert name and the number as the expert list; The road feature data and the numbers corresponding to the expert names are used as input data, and the evaluation weights are used as output data to train a neural network model to form the expert database model.

3. A road pollution evaluation method based on AHP and grey clustering according to claim 2, characterized in that: Calculating the evaluation weights by AHP includes: Acquire the AHP scale scoring data of each expert in each expert evaluation from the historical expert evaluation data; the AHP scale is a 1-9 scale or an index scale; Calculate the subjective weight value of each expert in this evaluation based on the AHP scale scoring data of each expert in the same expert evaluation; The subjective weight value is converted into an objective weight value by the entropy weight method as the evaluation weight of the expert in this expert evaluation.

4. A road pollution evaluation method based on AHP and grey clustering according to claim 3, characterized in that: The calculation of the subjective weight value includes: Acquire evaluation data of each element at each expert evaluation from the historical expert evaluation data; After comparing each element pairwise, the judgment matrix is ​​constructed by assigning values ​​through the AHP scale; the unit in the judgment matrix is ​​the importance ratio scale between two elements; Checking the consistency of the judgment matrix, and after the consistency check passes, normalizing the row vectors of the judgment matrix and calculating the arithmetic mean to form a weight vector; The weight vector is constructed as a matrix to form the subjective weight value.

5. The road pollution evaluation method based on AHP and grey clustering according to claim 3 is characterized in that: The subjective weight value is converted into an objective weight value by the entropy weight method as the evaluation weight of the expert in this expert evaluation, including: Assume that there is an ideal optimal expert in the same expert evaluation; Calculating the difference between the subjective weight value and the optimal weight value as difference data; the optimal weight value is the subjective weight value of the optimal expert; constructing an entropy model of expert evaluation results based on the difference data; Calculating the entropy weight of each expert according to the entropy model; The subjective weight value is weighted according to the entropy weight to generate the objective weight value of each expert.

6. The road pollution evaluation method based on AHP and grey clustering according to claim 1 is characterized in that: Performing grey cluster analysis on the evaluation results according to the evaluation weights to generate a pollution level as the final pollution evaluation of the target road includes: The corresponding indicators in the evaluation results are divided into multiple gray classes according to the evaluation requirements, and the center point of each gray class, the turning point of the front gray class and the turning point of the tail gray class are calculated as grayscale parameters; Constructing a whitening weight function according to the grayscale parameter, and calculating the degree of membership of each gray class according to the whitening weight function and the center point; The clustering coefficient is calculated according to the membership degree and the evaluation weight, and the evaluation result is divided into corresponding gray classes according to the clustering coefficient to form a pollution evaluation result.

7. A road pollution evaluation system based on AHP and grey clustering, characterized in that: include: a modeling unit configured to obtain historical expert evaluation data of road pollution and construct an expert database model and an expert list based on the historical expert evaluation data; The historical expert evaluation data is data generated by experts who have conducted road pollution evaluations; An acquisition unit is configured to acquire characteristic data of the target road when performing pollution evaluation on the target road; the characteristic data includes material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics; A computing unit configured to obtain a list of experts to be invited, and obtain a number corresponding to each expert in the expert list according to the expert list; Inputting the feature data and the serial number into the expert database model, and obtaining the evaluation weight corresponding to each expert output by the expert database model; Selecting multiple experts with the largest evaluation weights as the experts actually invited; The evaluation unit is configured to evaluate the pollution of the target road by actually inviting experts to generate an evaluation result, and to perform grey cluster analysis on the evaluation result according to the evaluation weight to generate a pollution level as a final pollution evaluation of the target road.

8. The road pollution evaluation system based on AHP and grey clustering according to claim 7 is characterized in that: The modeling unit is further configured to: Obtaining road characteristic data, expert names and evaluation weights at each expert evaluation from the historical expert evaluation data; the evaluation weights are obtained by AHP calculation; the road characteristic data include material pollution characteristics, operation pollution characteristics, mechanical pollution characteristics and environmental pollution characteristics; Numbering each expert to form a correspondence between the expert name and the number as the expert list; The road feature data and the numbers corresponding to the expert names are used as input data, and the evaluation weights are used as output data to train a neural network model to form the expert database model.

9. The road pollution evaluation system based on AHP and grey clustering according to claim 8 is characterized in that: The modeling unit is further configured to: Acquire the AHP scale scoring data of each expert in each expert evaluation from the historical expert evaluation data; the AHP scale is a 1-9 scale or an index scale; Calculate the subjective weight value of each expert in this evaluation based on the AHP scale scoring data of each expert in the same expert evaluation; The subjective weight value is converted into an objective weight value by the entropy weight method as the evaluation weight of the expert in this expert evaluation.

10. The road pollution evaluation system based on AHP and grey clustering according to claim 9, characterized in that: The modeling unit is further configured to: Acquire evaluation data of each element at each expert evaluation from the historical expert evaluation data; After comparing each element pairwise, the judgment matrix is ​​constructed by assigning values ​​through the AHP scale; the unit in the judgment matrix is ​​the importance ratio scale between two elements; Checking the consistency of the judgment matrix, and after the consistency check passes, normalizing the row vectors of the judgment matrix and calculating the arithmetic mean to form a weight vector; The weight vector is constructed as a matrix to form the subjective weight value.

Images