A fuzzy and uncertain reliability analysis method, device and storage medium

Abstract

The present invention discloses a fuzzy and uncertain reliability analysis method, device and storage medium, and the fuzzy and uncertain reliability analysis method includes the following steps: S1, collecting basic spatial data, business time series data and vehicle trajectory spatial data of the target area; S2, performing uncertainty analysis on the collected data to obtain deterministic data and uncertain data; S3, performing reliability analysis on the uncertain data to obtain application decision models of different business combinations; S4, constructing an error propagation cumulative estimation model to correct the errors generated during the data analysis and processing; S5, setting the trigger threshold of the application decision model, and outputting the warning / disposal / judgment application results. The present invention constructs a traceable data model, associates the construction site vehicle monitoring data itself and the related business parties involved, and uses the influence of multiple targets on the main monitoring targets to judge the reliability of its system data.

Description

Technical Field

[0001] The present invention relates to the technical field of discrete data reliability analysis, and in particular to a fuzzy and uncertain reliability analysis method. Background Art

[0002] Construction site vehicles play an increasingly important role in urban operations, and the supervision of construction site vehicles directly affects the efficiency of projects and operations.

[0003] The disadvantage of the existing technology is that the vehicle monitoring system only monitors a single target, and the collected data lacks representativeness. The reliability analysis of the system using simulation method is short in cycle and large in error. It is difficult to judge the uncertainty of single data, and the uncertainty of error analysis causes many unstable factors in the reliability analysis process. Summary of the invention

[0004] The present invention proposes a fuzzy uncertain reliability analysis method, device and storage medium. By expanding a single monitoring target to multi-target monitoring and combining spatial data uncertainty theory, fuzzy modeling theory and error propagation accumulation theory, a traceable data model is constructed. On the basis of the numerical method, by constructing a fuzzy model, the construction site vehicle monitoring data itself and the relevant business parties involved are associated, and the influence of multiple targets on the main monitoring targets is used to judge the reliability of its system data.

[0005] A fuzzy and uncertain reliability analysis method comprises the following steps: S1, collecting basic spatial data, business time series data and vehicle trajectory spatial data of a target area; S2, performing uncertainty analysis on the basic spatial data, the business time series data and the vehicle trajectory spatial data to obtain deterministic data and uncertain data, wherein the uncertain data includes random data and fuzzy data; S3, performing reliability analysis on the uncertain data to obtain application decision models for different business combinations; S4, constructing an error propagation cumulative estimation model to correct errors generated during data analysis and processing; S5, setting a trigger threshold of the application decision model and outputting application results.

[0006] Furthermore, the basic spatial data includes remote sensing images, traffic vector maps and regional weather data; the business time series data includes enterprise, construction site, regulatory agency and traffic rules data; and the vehicle trajectory spatial data includes vehicle GPS or GIS location data.

[0007] Furthermore, step S2 includes: S21, event definition uncertainty analysis; S22, event attribute definition uncertainty analysis; S23, data collection uncertainty analysis; S24, data processing-analysis-storage uncertainty analysis.

[0008] Furthermore, step S3 includes: S31, dimensionlessly processing the uncertain data; S32, constructing a continuous random model to process the random data, and the output data has fuzziness; S33, establishing a corresponding fuzzy matrix, performing cluster analysis on the fuzzy data and the random data processed by step S32, and analyzing the correlation between the data; S34, establishing a target fuzzy comprehensive evaluation index model based on the data correlation obtained in step S33, setting the weight coefficient of the corresponding business, and obtaining an application decision model for different business combinations.

[0009] Furthermore, the dimensionless processing method described in step S31 includes min-max normalization method, average normalization method, nonlinear normalization method, norm normalization method, and standard deviation / zero mean normalization method.

[0010] Further, step S34 is repeated to obtain a multi-level target fuzzy comprehensive evaluation index model.

[0011] Further, subsequent uncertain analysis nodes are added after step S34.

[0012] Furthermore, the application decision model includes early warning, disposal, and analysis.

[0013] A fuzzy and uncertain reliability analysis device, comprising:

[0014] A data acquisition module is used to collect basic spatial data, business time series data and vehicle trajectory spatial data of the target area; a spatial uncertainty filter is used to filter the basic spatial data, the business time series data and the vehicle trajectory spatial data to obtain deterministic data and uncertain data respectively; a reliability analysis filter is used to process the uncertain data to obtain application decision models for different business combinations; an error propagation accumulation estimator is used to correct data errors in the process processing; and a result output module is used to output application results.

[0015] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the fuzzy and uncertain reliability analysis steps described in any of the above schemes.

[0016] The beneficial effects of the present invention are:

[0017] The present invention constructs a data traceability model for multiple related targets such as vehicles, enterprises, transportation, construction sites, etc. in the target area, establishes a decision evaluation model for vehicle-related business using fuzzy comprehensive decision-making method, associates the construction site vehicle monitoring data itself and related business parties involved, and uses the impact of multiple targets on the main monitoring targets to judge the reliability of its system data. The reliability analysis method based on the present invention has good stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flow chart of a fuzzy and uncertain reliability analysis method according to one embodiment of the present invention;

[0019] Figure 2 It is a block diagram of a fuzzy and uncertain reliability analysis device according to one embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. The preferred embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to understand the disclosure of the present invention more thoroughly and comprehensively.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used herein in the specification of the present invention are only for describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0022] See also Figure 1 , Figure 1 The figure is a flow chart of a fuzzy and uncertain reliability analysis method according to one embodiment of the present invention, and the method comprises the following steps:

[0023] S1. Collect basic spatial data, business time series data and vehicle trajectory spatial data of the target area;

[0024] Among them, basic spatial data include remote sensing images, traffic vector maps and regional weather data; business time series data include enterprise, construction site, regulatory agency and traffic rules data; vehicle trajectory spatial data include vehicle GPS or GIS location data.

[0025] S2. Input the collected basic spatial data, business time series data and vehicle trajectory spatial data into the uncertainty analysis filter for processing. The uncertainty analysis process is as follows:

[0026] S21, event definition uncertainty analysis, used to define the certain and uncertain attributes of business events;

[0027] S22, event attribute definition uncertainty analysis, used to analyze the definite and uncertain types of event attribute definitions;

[0028] S23, data collection uncertainty analysis, used to analyze the uncertainties in the data collection process;

[0029] S24, data processing-analysis-storage uncertainty analysis, is used to analyze the uncertainty factors in the data processing process.

[0030] After the uncertainty analysis of the sequential process of steps S21-S24, three types of data are obtained:

[0031]

[0032] S3. Perform reliability analysis on fuzzy data and random data. The steps are as follows:

[0033] S31. Dimensionless processing is performed on fuzzy data and random data to make different indicators comparable. Linear function normalization method (Min-Max Scaling) is used to linearly transform the original data so that the result is mapped to the range of [0,1] to achieve geometric scaling of the original data. The normalization formula is as follows:

[0034]

[0035] S32, constructing a continuous random model to process random data;

[0036] S321. Select a vehicle monitoring data x that has been classified and dimensionless processed by an uncertainty analysis filter as a sample space Ω, the function ξ(ω) is a random variable of the sample space Ω, and the probability distribution function of the sample space Ω is as follows:

[0037] F(x)=P(ξ(ω)<x),x∈(-∞,∞)

[0038] The probability density function p(x) is:

[0039]

[0040] S322, let ξ1,ξ2,…,ξ n is the sample space of all collected data, and its n-dimensional distribution function is:

[0041] F(x1,x2,…,x n )=P(ξ1 <x1,ξ2<x2,…,ξ n <x n )

[0042] The corresponding density is:

[0043]

[0044] S323. The two-dimensional normal distribution of the two-dimensional random variable (ξ,η) transformed from the sample space is:

[0045]

[0046] (ξ,η)~N(μ1,μ2,σ1 2 ,σ2 2 ,ρ)

[0047] Among them -∞<μ1,μ2<+∞,σ1>0,σ2>0,|ρ|<1

[0048] S324. After calculating the normal distribution under multiple random conditions, the mathematical expectation of ξ under the condition of (η=y) is calculated:

[0049]

[0050] S33, the expected result of the random type obtained according to step S32 is fuzzy. According to the fuzzy mathematics theory, a corresponding fuzzy matrix is ​​established, cluster analysis is performed on the data, and the correlation between the data is statistically calculated. The method is as follows:

[0051] S331, take λ1=1 (maximum value), for each x i Similar category: [x i ] R ={x j |r ij =1}, satisfying r ij =1 i and x j Treated as one category, forming a similar category;

[0052] S332, take λ2 (λ2<λ1) as the next largest value, and directly find the element pair (x) whose similarity is λ2 from R i ,x j ), that is, r ij =λ2, and correspondingly, the equivalent classification corresponding to λ1=1 i , x j The classes are merged into one class, so that we can get the equivalent class at the λ2 level;

[0053] S333, take λ1>λ2>λ3>… in sequence, and follow the method of S332 in sequence until X is merged into the last category, and a dynamic clustering graph can be obtained.

[0054] S34: According to the data correlation obtained in step S33, a target fuzzy comprehensive evaluation index model is established, and the weight coefficient of the corresponding business is set to obtain an application decision model for different business combinations. The construction method is as follows:

[0055] S341, establish a set X{x1,x2,…,x k}

[0056] S342, Setting the indicator systemi = {u (i) 1,u (i) 2,…,u (i) ni},i=1,2,…,s

[0057] in:

[0058] S343, calculate the comment set V = {v1, v2, ..., v m}

[0059] S344, by U i and V, with evaluation matrix

[0060] S345, for each U i Setting factor weights in When R i is a single factor matrix, with a single factor evaluation vector set

[0061] S346, each U i Considered as a factor, let U = {U1,U2,…,U s}, the single factor judgment matrix of U is:

[0062]

[0063] S4. Construct an error propagation cumulative estimation model to correct data errors in the process of process processing. The method is as follows:

[0064] S41, establish independent observation objects x1, x2, ..., x n Function Z:

[0065] Z = f(x1, x2, ..., x n )

[0066] S42, construct the mean error m1, m2, ..., m of the corresponding observation object value n , the mean error of function Z is:

[0067]

[0068] in is the partial derivative of the function Z with respect to variable x

[0069] S5. After processing steps S1-S4, a more accurate fuzzy decision model is obtained, and its comprehensive index value is reflected between [0,1]. The output values ​​of early warning, disposal, and judgment are set to be x1, x2, and x3, and the corresponding y1, y2, y3, and y4 are the set trigger thresholds. The corresponding application results can be output according to the following calculation rules:

[0070]

[0071] In other embodiments, the data dimensionless processing method in step S31 may also be an average normalization method, a nonlinear normalization method, a norm normalization method, or a standard deviation / zero mean normalization method.

[0072] In other embodiments, steps S345-S346 are repeated to obtain a multi-level target fuzzy comprehensive evaluation index model.

[0073] In an exemplary embodiment, a fuzzy uncertain reliability analysis device (S00) is also provided, including a data acquisition module (S100), a spatial uncertainty filter (S200), a reliability analysis filter (S300), an error propagation accumulation estimator (S400) and a result output module (S500).

[0074] In an exemplary embodiment, a storage medium is also provided, and the storage medium may be a computer-readable storage medium.

[0075] The above is a specific description of the preferred implementation of the present invention, but the invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A fuzzy and uncertain reliability analysis method, characterized in that: The method comprises the following steps: S1. Collect basic spatial data, business time series data and vehicle trajectory spatial data of the target area; S2, performing uncertainty analysis on the basic spatial data, the business time series data and the vehicle trajectory spatial data to obtain deterministic data and uncertain data, wherein the uncertain data includes random data and fuzzy data; S3, performing reliability analysis on the uncertain data, comprising the following steps: S31, performing dimensionless processing on the uncertain data to make different indicators comparable; S32, constructing a continuous random model to process random data to obtain the mathematical expectation of the random data; S33. According to fuzzy mathematics theory, a corresponding fuzzy matrix is ​​established to perform cluster analysis on random data and calculate the correlation between the random data; S34: According to the data correlation obtained in step S33, a target fuzzy comprehensive evaluation index model is established, and the weight coefficient of the corresponding business is set. Obtain application decision models for different business combinations; S4. Construct an error propagation cumulative estimation model to correct the errors generated during data analysis and processing; S5. Setting a trigger threshold of the application decision model and outputting an application result.

2. A fuzzy and uncertain reliability analysis method according to claim 1, characterized in that: The basic spatial data include remote sensing images, traffic vector maps and regional weather data; the business time series data include enterprise, construction site, regulatory agency and traffic rules data; the vehicle trajectory spatial data include vehicle GPS or GIS location data.

3. A fuzzy and uncertain reliability analysis method according to claim 1, characterized in that: Step S2 includes: S21, event definition uncertainty analysis; S22, event attribute definition uncertainty analysis; S23, data collection uncertainty analysis; S24, data processing-analysis-storage uncertainty analysis.

4. A fuzzy and uncertain reliability analysis method according to claim 1, characterized in that: The dimensionless processing method described in step S31 includes linear normalization method, average normalization method, nonlinear normalization method, norm normalization method, and standard deviation / zero mean normalization method.

5. A fuzzy and uncertain reliability analysis method according to claim 1, characterized in that: Repeat step S34 to obtain a multi-level target fuzzy comprehensive evaluation index model.

6. A fuzzy and uncertain reliability analysis method according to claim 1, characterized in that: Subsequent uncertain analysis nodes are added after step S34.

7. A fuzzy and uncertain reliability analysis method according to claim 1, characterized in that: The application decision model includes early warning, disposal, and analysis.

8. A fuzzy and uncertain reliability analysis device, characterized in that: include: Data collection module, used to collect basic spatial data, business time series data and vehicle trajectory spatial data of the target area; A spatial uncertainty filter, used to filter the basic spatial data, the business time series data and the vehicle trajectory spatial data to obtain deterministic data and uncertain data respectively; The reliability analysis filter is used to process the uncertain data, and comprises the following steps: S31, performing dimensionless processing on the uncertain data to make different indicators comparable; S32, constructing a continuous random model to process random data to obtain the mathematical expectation of the random data; S33. According to fuzzy mathematics theory, a corresponding fuzzy matrix is ​​established to perform cluster analysis on random data and calculate the correlation between the random data; S34, according to the data correlation obtained in step S33, establish a target fuzzy comprehensive evaluation index model, set the weight coefficient of the corresponding business, and obtain an application decision model for different business combinations; Error propagation accumulation estimator, used to correct data errors during process processing; The result output module is used to output application results.

9. A storage medium, characterized in that: The storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the method according to any one of claims 1 to 7.

Images