Initial support force and final resistance identification method and storage medium based on deep neural network
A technology of deep neural network and identification method is applied in the field of automatic identification of the initial support force point and final resistance point of the hydraulic support, which can solve the problems of complex roof pressure and high error rate.
Active Publication Date: 2021-07-02
HENAN POLYTECHNIC UNIV
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Due to the complexity of the roof pressure received by the hydraulic support during the mining process, the current gradient change method has a high error rate
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
specific Embodiment
[0049] (1) Processing of detection data. Detection data is input data, including two types: time information and pressure information. Normalize time information into YYYYMMDDhhmmss format, where YYYY represents year, MM represents month, DD represents day, hh represents hour, mm represents minute, ss represents second, and stores it in float format. The pressure information remains unchanged and is stored in float format.
[0050] (2) Processing of label data. Corresponding to the detection data, an annotation data is introduced. The types of data points are divided into four categories: initial support points, final resistance points, normal points, and abnormal points. The label data can be set to 1000, 0100, 0010, and 0001 in turn. Among them, the abnormal point is the peak point that occurs less than one coal mining cycle, and the normal point is all data points except the initial support point, the end resistance point and the abnormal point.
[0051] (3) Normalizati...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The method for identifying initial support force and final resistance based on a deep neural network provided by the present invention includes the following steps: S1: collecting detection data of hydraulic supports at different times, and each data point of the detected data includes time data and pressure data; S2 : Set label data with four attributes for each data point, the four attributes of the label data are: initial support point, end resistance point, normal point, abnormal point; S3: collected detection data and corresponding Label the data to form training samples of input data, and train the deep neural network based on the training samples; S4: Establish a prediction model based on the deep neural network; S5: Collect the detection data of the hydraulic support on site and send it to the trained deep neural network, and output Classification results for each data point. The invention can automatically extract features from the data of the hydraulic support through a deep neural network, and obtain accurate prediction results through a large number of sample training.
Description
technical field [0001] The invention relates to an automatic identification method of initial support point and final resistance point of a hydraulic support based on a deep neural network. Background technique [0002] At present, the roof control method of the downhole hydraulic support is completed manually, and the operator sets a fixed initial support force according to experience to make the hydraulic column rise. When there are unfavorable situations such as the lowering of the side and top coal, manually operate the cover beam and hydraulic column of the hydraulic support to make corresponding operations. Manual operation cannot comprehensively consider the overall dynamic status of the roof of the working face, and cannot effectively take into account the performance of the working face level, let alone realize the automatic control of the hydraulic support group. The root cause is the lack of a control model at the hydraulic support group system level. [0003] Th...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/24
Inventor 杨艺王宇王科平张旭和李华敏李冰峰李新伟崔立志
Owner HENAN POLYTECHNIC UNIV