Multi-granularity coal mine gas risk prediction method based on cloud model
A technology for coal mine gas and risk prediction, which is applied in measurement devices, complex mathematical operations, instruments, etc., can solve problems such as the lack of research on the mutual conversion relationship between quantitative analysis and qualitative evaluation, and the difficulty in effectively guiding quantitative production practice in coal mine risk assessment.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0058] This embodiment mainly illustrates the improved algorithm-adaptive hybrid cloud transformation algorithm (A_MCT algorithm) proposed by the present invention on the adaptive Gaussian cloud transformation algorithm (A-GCT algorithm) that is suitable for coal mine safety big data, and mainly involves steps S1~ S3, through the dataset sample set {x i |i=1,2,...,N}, the upper limit of concept ambiguity β, the threshold of data distribution skewness γ, and finally output qualitative concepts represented by various cloud models, specifically including the following steps:
[0059] Step1: Calculate the skewness of the data distribution and perform logarithmic transformation
[0060] Set a data distribution skewness threshold γ to consider the overall data distribution and count the frequency distribution of the original data p(x i ), calculate the skewness Δp of the original data distribution, if Δp>γ, then transform the frequency distribution of the original data, expressed a...
Embodiment 2
[0071] This embodiment mainly illustrates the construction of time granules and space granules in the present invention.
[0072] Step1: time grain
[0073] Formulate granular standards, that is, the minute granularity is the real-time change of the sensor monitoring value, the hour granularity is the situation fluctuation and trend within the current hour period, the day granularity is the situation distribution in each period of the current day, and the monthly granularity is the current month. The situation fluctuates, and the annual granularity is the overall concentration situation of the year;
[0074] Based on the granulation standard, the original data is divided into p time slices {T 1 , T 2 ,...,T i ,...,T p}; where ω is the window size of the time slice, and its different time slice sizes correspond to different time granularities; the real-time data of the coal mine gas concentration obtained is one node per minute, and the width of the time slice is controlled...
Embodiment 3
[0084] This embodiment proposes a calculation method of the degree of membership.
[0085] Before calculating the degree of membership, it is necessary to judge the representation cloud of each time granular layer Whether the expected value Ex of the middle time grain belongs to obtain the original data macro-concept interval according to embodiment 1, that is, in the interval formed by the expected maximum value and minimum value of multiple hybrid clouds of the original data;
[0086] If it belongs, that is, the time grain belongs to the domain of discourse interval of the corresponding concept, calculate its expected degree of membership in the concept of membership μ=exp(-(Ex'-Ex)) 2 / 2*(En') 2 , get multi-granularity representation;
[0087] Among them, Ex is the expectation of time grain, Ex' is the expectation of the macro concept to which it belongs, En' is the entropy of the macro concept to which it belongs, Ex k is the expectation of a Gaussian cloud, En k is the...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


