CN115758269A - Method and system for determining opening state of safety valve of hydraulic support - Google Patents
Method and system for determining opening state of safety valve of hydraulic support Download PDFInfo
- Publication number
- CN115758269A CN115758269A CN202211482962.6A CN202211482962A CN115758269A CN 115758269 A CN115758269 A CN 115758269A CN 202211482962 A CN202211482962 A CN 202211482962A CN 115758269 A CN115758269 A CN 115758269A
- Authority
- CN
- China
- Prior art keywords
- safety valve
- data
- opening state
- time sequence
- pressure monitoring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000012544 monitoring process Methods 0.000 claims abstract description 81
- 238000000605 extraction Methods 0.000 claims abstract description 35
- 238000010801 machine learning Methods 0.000 claims abstract description 35
- 238000012549 training Methods 0.000 claims abstract description 35
- 238000012360 testing method Methods 0.000 claims description 24
- 230000006870 function Effects 0.000 claims description 20
- 238000003066 decision tree Methods 0.000 claims description 17
- 238000012163 sequencing technique Methods 0.000 claims description 9
- 238000012795 verification Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 230000000007 visual effect Effects 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 abstract description 15
- 238000013106 supervised machine learning method Methods 0.000 abstract description 5
- 239000003245 coal Substances 0.000 description 15
- 230000008569 process Effects 0.000 description 7
- 238000005065 mining Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 238000012216 screening Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Landscapes
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The application provides a method and a system for determining the opening state of a safety valve of a hydraulic support, wherein the method comprises the following steps: acquiring original mine pressure monitoring data of the hydraulic support through a mine pressure monitoring system; manually marking the opening state of a safety valve on original mine pressure monitoring data, and generating a safety valve opening marking time sequence according to a marking result; arranging original mine pressure monitoring data, a safety valve opening state time sequence and target data of a safety valve opening state to be determined into data with equal intervals, and generating a Tsfresh time sequence according to the data with equal intervals; and performing feature extraction and model training on the Tsfresh time sequence through a Tsfresh machine learning framework, and determining the safety valve opening state result of each group of data in the target data through the trained machine learning model. The method combines the manual marking and the supervised machine learning method to realize the analysis of the opening state of the safety valve, and improves the timeliness and the accuracy of the identification of the opening state of the safety valve.
Description
Technical Field
The application relates to the technical field of coal mine equipment, in particular to a method and a system for determining the opening state of a safety valve of a hydraulic support.
Background
At present, the safety of coal mining is more and more concerned. In the stoping process of a coal mine working face, the hydraulic support is used as an important component of fully mechanized mining equipment, is vital to the control of a working face top plate, and is necessary equipment for ensuring mine safety.
The safety valve is an important part in the hydraulic support, whether the safety valve fails or not and whether the set pressure of the safety valve is reasonable or not have significance for reliably and effectively supporting and controlling the working face top plate or not. If the pressure of the safety valve is too low, the supporting strength is reduced, and a roof accident is easy to happen; if the safety valve fails or the set pressure of the safety valve is too high, the cylinder explosion and the injury of people can be caused. Therefore, accurate control of the opening of the safety valve is required.
In the related art, the opening control mode of the safety valve is single, and generally, the opening state of the full valve is manually determined and the opening is controlled according to historical experience. However, in practical applications, only by the single opening state determination method, the opening of the safety valve may not be performed timely in some cases, and the safety valve may not be opened in a more appropriate time period.
Disclosure of Invention
The present application is directed to solving, at least in part, one of the technical problems in the related art.
Therefore, a first object of the present application is to provide a method for determining an opening state of a safety valve of a hydraulic support, which is based on support pressure monitoring time series data acquired by a mine pressure monitoring system, and implements analysis of the opening state of the safety valve by a supervised machine learning method, so as to provide early warning of a support operation state for a coal mine manager.
A second object of the present application is to propose a system for determining the opening condition of a safety valve of a hydraulic support.
A third object of the present application is to propose a non-transitory computer-readable storage medium.
To achieve the above object, a first embodiment of the present application provides a method for determining an open state of a safety valve of a hydraulic support, including the following steps:
acquiring original mine pressure monitoring data of the hydraulic support through a mine pressure monitoring system;
manually marking the opening state of the safety valve on the original mine pressure monitoring data, and generating a safety valve opening marking time sequence according to a marking result;
the original mine pressure monitoring data, the time sequence of the opening state of the safety valve and the target data of the opening state of the safety valve to be determined are arranged into equidistant data, and a Tsfresh time sequence is generated according to the equidistant data;
and performing feature extraction and model training on the Tsfresh time sequence through a Tsfresh machine learning framework, and determining the safety valve opening state result of each group of data in the target data through a trained machine learning model.
Optionally, in an embodiment of the present application, said manually marking the safety valve opening state on the raw mine pressure monitoring data includes: marking the starting point of the opening of each round of safety valve in the graph of the original mine pressure monitoring data; adding a preset time interval to the starting point to determine the starting end point of each wheel of safety valve; and determining the opening pressure value of the safety valve for opening each round of the safety valve according to the ordinate of the curve chart.
Optionally, in an embodiment of the present application, the generating a safety valve open state time series according to a result of the marking includes: generating a plurality of groups according to the starting point, the end point and the safety valve opening pressure value of each safety valve opening; and sequencing the plurality of groups according to the sequence of the occurrence time of the starting point from far to near to generate the time sequence of the opening state of the safety valve.
Optionally, in an embodiment of the application, the sorting the original mine pressure monitoring data, the time sequence of the opening states of the safety valve, and the target data of the opening state of the safety valve to be determined into equidistant data includes: traversing the original mine pressure monitoring data, and segmenting the original mine pressure monitoring data according to the marked safety valve opening time period; intercepting the non-safety valve opening time period between two adjacent safety valve opening time periods according to the time interval, and rejecting data which does not meet the time interval in the intercepted non-safety valve opening time period; and intercepting the target data according to the time interval, and removing the data which do not meet the time interval in the intercepted target data.
Optionally, in an embodiment of the present application, the generating a Tsfresh time series according to the equidistant data includes: determining the safety valve opening state value of each group in a safety valve opening mark time sequence, a non-safety valve opening time sequence and a target data time sequence which are generated after equal interval segmentation; and sequencing each packet containing the state value according to the sequence of the starting time from far to near to generate the Tsfresh time sequence.
Optionally, in an embodiment of the present application, the performing, by using a Tsfresh machine learning framework, feature extraction and model training on the Tsfresh time series includes: setting a type for performing feature extraction through the Tsfresh, and performing feature extraction on the Tsfresh time sequence through a preset feature extraction function to generate a feature time sequence; dividing the characteristic time sequence into a training set, a verification set and a test set through a preset dividing function; and training a preset decision tree classifier through the data in the training set to obtain a trained machine learning model for analyzing the opening state of the safety valve.
Optionally, in an embodiment of the application, the amount of data in the test set is a number of packets in the target data time series, and the determining, by the trained machine learning model, a safety valve opening state result for each set of data in the target data includes: and predicting the data in the test set through the trained decision tree classifier to obtain the safety valve opening state result.
Optionally, in an embodiment of the present application, after the trained machine learning model determines the safety valve opening state result of each set of data in the target data, the method further includes: and drawing the opening state result of the safety valve of each group of data in a support pressure curve graph of the hydraulic support for visual display.
To achieve the above object, a second embodiment of the present application provides a system for determining an open state of a safety valve of a hydraulic support, including the following modules:
the acquisition module is used for acquiring original mine pressure monitoring data of the hydraulic support through the mine pressure monitoring system;
the marking module is used for manually marking the opening state of the safety valve on the original mine pressure monitoring data and generating a safety valve opening marking time sequence according to a marking result;
the generating module is used for collating the original mine pressure monitoring data, the time sequence of the opening state of the safety valve and target data of the opening state of the safety valve to be determined into data with equal intervals, and generating a Tsfresh time sequence according to the data with equal intervals;
and the determining module is used for performing feature extraction and model training on the Tsfresh time sequence through a Tsfresh machine learning framework and determining the safety valve opening state result of each group of data in the target data through the trained machine learning model.
In order to achieve the above embodiments, the third aspect of the present application further proposes a non-transitory computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the method for determining the open state of the safety valve of the hydraulic support in the above embodiments.
The technical scheme provided by the embodiment of the application at least has the following beneficial effects: according to the method, firstly, the support pressure monitoring time sequence data are obtained based on a mine pressure monitoring system, then, the safety valve opening state is marked on the data in a period of time manually, the original data and the marked data are processed into a time sequence required by a machine learning frame tsfresh, then, the safety valve opening state of the hydraulic support is automatically analyzed through a machine learning method, the method comprises the steps of characteristic extraction, characteristic screening, training, prediction and the like, and the safety valve opening state analysis is realized through a supervised machine learning method. From this, this application has richened the mode of relief valve opening state analysis, can be timely and the time period that the accurate definite relief valve opened to can in time provide early warning information for the colliery staff, guarantee that the relief valve opens on time at suitable time period, improve the timeliness and the accuracy that the relief valve opened state discernment, be favorable to improving colliery roof management level, improve the security of coal mining.
Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.
Drawings
The above and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which
Fig. 1 is a flowchart of a method for determining an open state of a safety valve of a hydraulic bracket according to an embodiment of the present disclosure;
FIG. 2 is a schematic diagram of a manual marking result of an opened state of a safety valve according to an embodiment of the present disclosure;
fig. 3 is a flowchart of a method for generating equidistant data according to an embodiment of the present disclosure;
fig. 4 is a schematic diagram of a data interception method according to an embodiment of the present application;
fig. 5 is a flowchart of a method for predicting an opening state result of a safety valve according to an embodiment of the present disclosure;
fig. 6 is a schematic structural diagram of a system for determining an opening state of a safety valve of a hydraulic mount according to an embodiment of the present application.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the accompanying drawings are illustrative and intended to explain the present invention and should not be construed as limiting the present invention.
The following describes a method and a system for determining an open state of a safety valve of a hydraulic bracket according to an embodiment of the present invention with reference to the accompanying drawings.
Fig. 1 is a flowchart of a method for determining an opening state of a safety valve of a hydraulic mount according to an embodiment of the present application, as shown in fig. 1, the method includes the following steps:
and S101, acquiring original mine pressure monitoring data of the hydraulic support through a mine pressure monitoring system.
The hydraulic support is a structure for controlling the mine pressure of the coal face, the mine pressure of the coal face acts on the hydraulic support in an external load mode, and the hydraulic support is composed of various components such as a hydraulic cylinder, a bearing structural member and a pushing device. The hydraulic support safety valve can keep constant working resistance of the upright column, and when the pressure of the top plate exceeds the working resistance of the support, liquid in the support working cavity is released to perform multiple functions such as overload protection and the like. Therefore, the opening state of the safety valve needs to be accurately analyzed, and the safety valve is opened at a proper time interval.
The mine pressure monitoring system can monitor the pressure applied to equipment such as a hydraulic support in real time and upload the monitored pressure data to a ground monitoring host computer in real time. Therefore, the original mine pressure monitoring data of the hydraulic support are obtained by means of the existing mine pressure monitoring system. The original mine pressure monitoring data are unprocessed, the pressure data of the support directly acquired by the mine pressure monitoring system at different moments are received, and the original mine pressure monitoring data are time sequence data.
In specific implementation, the original mine pressure monitoring data can be obtained in different modes. As a first example, the mine pressure monitoring data can be acquired by performing data interaction with an existing mine pressure monitoring system of a coal mine through various communication protocols, wherein the various communication protocols comprise file transfer protocol ftp, webService and the like. In terms of WebService, webService is a cross-programming-language and cross-operating-system-platform remote calling technology, and can remotely call data of the mine pressure monitoring system on terminal equipment executing the safety valve opening state determining method, without considering the difference of programming languages of different equipment programs of a coal mine, so that the data acquisition is convenient.
As a second example, a database for storing mine pressure monitoring data of the mine pressure monitoring system may also be preset, and the data in the database of the mine pressure monitoring system may be directly read when the data needs to be acquired.
In an embodiment of the application, for a stope face of a non-electrohydraulic control system, a support pressure monitoring sensor is generally installed on a mine pressure monitoring system at intervals of n supports, and a set of support lists on which the sensor is installed is L = < n,2n, …, m × n >, where m is a grouping number of the supports. In this embodiment, the acquired monitoring raw data of each stent are sorted from small to large according to the time value, and the acquired set is as follows:
S m×n ＝<(t 1 ,x 1 ),(t 2 ,x 2 )…,(t i ,x i )>，
wherein the set is the original mine pressure monitoring data of the obtained m multiplied by n groups of brackets, (t) i ,x i ) Is that the support is at monitoring time t i Corresponding support pressure monitoring raw data x i And i is a data record serial number in the analysis time period of the support.
And S102, manually marking the opening state of the safety valve for the original mine pressure monitoring data, and generating a safety valve opening marking time sequence according to a marking result.
The safety valve opening state comprises state data such as opening time and ending time of the safety valve, and opening pressure of the safety valve at the opening time.
Specifically, any one hydraulic support is selected from all supports, and the opening state of the safety valve is marked manually according to original mine pressure monitoring data of the support. The method can be used for sequentially determining the opening state of the safety valve for each support under the mine according to the modes recorded in the step and the subsequent steps. Specifically, when manual marking is performed, the opening state of the safety valve can be marked by combining various factors such as historical operating data (including actual opening conditions before the safety valve), expert knowledge, personal experience and the like of the bracket.
In one embodiment of the present application, the manual marking may be performed by: firstly, marking the starting point of the opening of each round of safety valve in a curve graph of original mine pressure monitoring data; adding a preset time interval to the starting point to determine the starting end point of each wheel of the safety valve; and determining the opening pressure value of the safety valve for opening each round of the safety valve according to the ordinate of the curve chart.
Specifically, in the present embodiment, a graph of the change of the pillar pressure of the bracket with time as shown in fig. 2 is generated according to the obtained raw mine pressure monitoring data, and then a safety valve opening starting point t is selected from the raw data history curve at a predetermined time interval, for example, at 30-minute intervals as an example s Marking the end point t of the opening of the safety valve e ＝t s +30, whereby all safety valve opening occurrences start time, end time and safety valve opening pressure within a period of time of the stent resulting from manual marking of the stent.
As shown in fig. 2, the horizontal transverse line (T) at the position of the curve in the figure is marked 1 To T 8 ) And for the safety valve opening result obtained by manual marking, the starting position and the ending position of the transverse line respectively represent the starting time and the ending time of the opening of the safety valve, and the scale value of the longitudinal axis corresponding to the transverse line is the opening pressure value of the safety valve.
And further, generating a safety valve opening marking time sequence according to the result of manual marking, namely the marked opening state of the safety valve.
In one embodiment of the present application, generating a safety valve open state time series from the flagged results includes: generating a plurality of groups according to the starting point and the end point of the opening of each round of safety valves and the opening pressure value of the safety valves; and sequencing the groups according to the sequence of the occurrence time of the starting point from far to near to generate a time sequence of the opening state of the safety valve.
Specifically, the safety valve opening marking results are sorted, the marking results of each safety valve opening, that is, the data corresponding to each horizontal line in fig. 2, are recombined into packet data in the form of (start time, end time, and safety valve opening pressure), the marking results of each safety valve opening correspond to one packet data, and the packets are arranged according to the sequence of the occurrence time of the start point from far to near, that is, the numerical values of the start time are from small to large, so as to obtain a new time sequence as shown below:
T＝<(t s1 ,t e1 ,P 1 ),(t s2 ,t e2 ,P 2 )…,(t si ,t ei ,P i )>，
wherein the sequence represents each time the safety valve opens at a time t si To the end time t of the safety valve ei Corresponding opening pressure P of safety valve i . The data represented by the sequence may also be arranged in a table format.
Step S103, arranging the original mine pressure monitoring data, the safety valve opening state time sequence and the target data of the safety valve opening state to be determined into equidistant data, and generating a Tsfresh time sequence according to the equidistant data.
Specifically, the tsfresh machine learning framework is adopted to perform feature extraction, training and prediction on the generated time sequence. Since the tsfresh time series processing framework requires that the time series to be processed must be equidistant data, the acquired raw mine pressure monitoring data N of the stent, the generated safety valve opening mark time series T and the data P to be analyzed for the opening state of the safety valve need to be processed into equidistant data required by the tsfresh time series processing algorithm.
It should be noted that the target data of the opening state of the safety valve to be determined may be data which is not manually marked in the obtained mine pressure monitoring data, or may also be pressure monitoring data of the bracket which is currently obtained in real time, and may be determined according to an actual safety valve opening state analysis requirement.
In an embodiment of the present application, in order to more clearly describe a specific implementation process of the present application for processing each data into required equidistant data, an exemplary description is given below by using a data processing method proposed in the embodiment of the present application, and fig. 3 is a flowchart of a method for generating equidistant data proposed in the embodiment of the present application, as shown in fig. 3, the method includes the following steps:
step S201, original mine pressure monitoring data are traversed, and the original mine pressure monitoring data are segmented according to the marked safety valve opening time period.
Specifically, the data in the original mine pressure monitoring data N is cyclically traversed, and the original mine pressure monitoring data is segmented by the safety valve opening time period marked in step S102 (i.e., the time period between the starting point and the ending point of the opening of each safety valve), so that the segmented original mine pressure monitoring data is composed of a plurality of segments of data, including the data of each segment of safety valve opening time period and the data of the non-open safety valve opening time period between two adjacent segments of safety valve opening time periods.
And S202, intercepting the non-safety valve opening time interval between two adjacent safety valve opening time intervals according to the time interval, and rejecting data which do not meet the time interval in the intercepted non-safety valve opening time interval.
Specifically, according to the preset time interval adopted in the manual marking, for example, 30 minutes, the data of the opening time period of each non-opening safety valve is intercepted at an interval of 30 minutes, and part of the data of less than 30 minutes left after the previous interception is discarded, that is, the data of the divided non-opening time period of the safety valve is intercepted in the manner shown in fig. 4. In the figure N 1 And N 2 The packet data is intercepted by the opening time period of the non-opening safety valve.
And step S203, intercepting the target data according to time intervals, and removing data which do not meet the time intervals in the intercepted target data.
Specifically, in the above-described cutting method in step S202, the data P to be subjected to the safety valve opening analysis is similarly cut at intervals of 30 minutes, and a part of the data less than 30 minutes is discarded, so that 1 or more time-series groups are obtained.
Therefore, the original mine pressure monitoring data N, the generated safety valve opening mark time sequence T and the target data P to be analyzed for the opening state of the safety valve are processed into the equal-interval data which are divided at the preset intervals during manual marking.
Further, a Tsfresh time series is generated from the data of the equal intervals obtained by the division.
In one embodiment of the present application, generating the Tsfresh time series comprises the steps of: and determining the safety valve opening state value of each group in the safety valve opening mark time sequence, the non-safety valve opening time sequence and the target data time sequence which are generated after equal interval division, and then sequencing each group containing the state value according to the sequence of the starting time from far to near to generate the Trefresh time sequence.
Specifically, for the safety valve opening flag time series T after the equal pitch division, the corresponding safety valve opening state is set to True, for the non-safety opening flag time series N, the corresponding safety valve opening state is set to False, and for the target tertiary time series P, the safety valve opening state is temporarily assumed to False, so that the safety valve opening state value for each group in each time series can be determined. And sequencing the groups according to the sequence of the starting time from far to near, and combining the groups into a new time sequence Ts (namely the Tsfresh time sequence). Referring to the data splitting case shown in fig. 4 in the above embodiment, one possible Ts sequence is as follows:
Ts＝<(T 1 ,True),(N 1 ,False),(N 2 ,False)…(T 2 ,True),…,(P 1 ,False)>
and S104, performing feature extraction and model training on the Tsfresh time sequence through a Tsfresh machine learning framework, and determining the safety valve opening state result of each group of data in the target data through the trained machine learning model.
The Tsfresh (TimeSeries Fresh) is a Python third-party tool kit, can automatically calculate a large number of time series data features, and also comprises a feature importance evaluation and feature selection method.
Specifically, a supervised machine learning framework is constructed based on Tsfresh, and feature extraction, training and prediction are carried out on a new time sequence Ts through Tsfresh. And analyzing to obtain the safety valve opening state result of each group of data of the sequence P to be analyzed.
In an embodiment of the present application, in order to more clearly illustrate a specific implementation process of the present application for feature extraction and model training on a Tsfresh time series, an exemplary method for predicting an opening state result of a safety valve, which is provided in the embodiment of the present application, is described below, and fig. 5 is a flowchart of a method for predicting an opening state result of a safety valve, which is provided in the embodiment of the present application, and as shown in fig. 5, the method includes the following steps:
step S301, setting a type of feature extraction performed by Tsfresh, and performing feature extraction on the Tsfresh time sequence by using a preset feature extraction function to generate a feature time sequence.
Specifically, in practical applications, the amount of data for feature extraction by Tsfresh is large, and the calculation amount for extracting time-series features is large, which may take a long time. For this purpose, the present application sets parameters of feature extraction and determines the type of feature extraction. In this embodiment, the feature extraction setting type is defined as comprehensive fcparameters, that is, an exhaustive method is adopted to perform an exhaustive list of all possible parameters, and feature reduction may also be performed when extracting features.
Further, the Tsfresh time series is subjected to feature extraction through a preset feature extraction function, the feature extraction function may be selected as needed, and as an example, the Ts time series may be subjected to feature extraction through an extract _ reduce _ features function, so as to obtain a new feature time series X _ filtered, that is, a feature time series.
In specific implementation, as a possible implementation manner, the following codes are used for implementation:
“extract_settings＝ComprehensiveFCParameters()
X_filtered＝extract_relevant_features(TsX,TsY,column_id＝'id',
column_sort＝'time',default_fc_parameters＝extract_settings)”。
step S302, the characteristic time sequence is divided into a training set, a verification set and a test set through a preset division function.
In this embodiment of the application, the splitting function may be a train _ test _ split function, and the feature time sequence X _ filtered after the Ts time sequence conversion is split by the train _ test _ split function, so as to automatically split the training set, the verification set, and the test set. The data in the training set is used for training a machine learning model, and the data in the testing set is target data to be predicted, namely, the data amount in the testing set is the packet number in the target data time sequence.
In specific implementation, as a possible implementation manner, the following codes are used for implementation:
“X_train,X_test,X_filtered_train,X_filtered_test,y_train,
y_test＝train_test_split(TsX,X_filtered,TsY,test_size＝4)”。
where the number of test _ size parameters is the number of packets of the sequence P to be analyzed, which in this example is 4.
Step S303, training a preset decision tree classifier through the data in the training set to obtain a trained machine learning model for analyzing the opening state of the safety valve.
Among them, the decision tree classifier (decision tree classifier) is a classifier form that provides a set of attributes and classifies data by making a series of decisions based on the set of attributes. Decision trees are tree-type structures in which each branch represents a test output, and decision tree learning is trained using a top-down recursive approach.
Specifically, the types of the safety valve opening state analysis results output by the decision tree classifier are determined to include (True and False), and the data in the training set is used as training data to train the decision tree classifier. The training process of the specific decision tree classifier may refer to a training mode in the related art, and details are not repeated here.
And S304, predicting the data in the test set through the trained decision tree classifier to obtain a safety valve opening state result.
Specifically, after the training of the decision tree classifier is completed, the test set is predicted to obtain the safety valve opening state analysis result. Namely, the trained decision tree classifier model outputs the result type of the opening state analysis of each grouped data safety valve in the target data of the test set, wherein the result type comprises True or False, and meanwhile, if True, the predicted safety valve opening pressure value and the like can be output.
In specific implementation, as a possible implementation manner, the following codes are used for implementation:
“cl＝DecisionTreeClassifier()
cl.fit(X_train,y_train)
cl.predict(X_test)”。
therefore, the safety valve opening state of each grouped data in the target data to be analyzed can be obtained, so that whether the safety valve needs to be opened or not can be determined.
Further, in order to show the identification result of the open state of the safety valve more clearly, in an embodiment of the present application, after determining the open state result of the safety valve for each set of data in the target data through the trained machine learning model, the method further includes: and drawing the opening state result of the safety valve of each group of data in a support pressure curve graph of the hydraulic support for visual display.
Specifically, the graph of stent pressure may be a graph showing stent pressure at different times, with time as the abscissa and stent pressure as the ordinate, as shown in fig. 2. In the embodiment, a curve graph is generated according to the stent pressure data at different moments, and the predicted opening state result of the safety valve is displayed in the curve graph in the form of a horizontal line shown in fig. 2, so that the visual display mode is performed through chart drawing, and the relevant workers can conveniently and visually check the recognition result.
Furthermore, in an embodiment of the application, early warning information can be provided for relevant workers in a coal mine according to the determined opening state result of the safety valve. For example, the terminal device executing the method for determining the open state of the safety valve according to the present application may store account information of a mobile terminal (e.g., a mobile phone) of a relevant legal worker in advance, verify the validity of the mobile terminal after establishing a wireless connection with the mobile terminal of the relevant worker in practical applications, and send warning information to the relevant mobile terminal through a wireless network after the verification is passed. For example, when the opening state of the safety valve of the hydraulic support at a certain moment is determined to be True and the safety valve needs to be opened, early warning information can be sent to the mobile terminal in advance to remind a worker to open the safety valve in time. Therefore, the more accurate safety valve opening state determined through machine learning assists workers in controlling the safety valve, a new means is added for safety valve opening analysis, and the coal mine roof management level is improved.
In summary, according to the method for determining the opening state of the safety valve of the hydraulic support, the time sequence data of the support pressure monitoring is obtained based on the mine pressure monitoring system, then the safety valve opening state marking is performed on the data within a period of time manually, the original data and the marked data result are processed into the time sequence required by the machine learning frame tsfresh, the opening state of the safety valve of the hydraulic support is automatically analyzed through a machine learning method, the method comprises the steps of feature extraction, feature screening, training, prediction and the like, and the opening state analysis of the safety valve is achieved through a supervised machine learning method. Therefore, the method enriches the analysis mode of the opening state of the safety valve, and can timely and accurately determine the opening time period of the safety valve, so that early warning information can be timely provided for coal mine workers, the safety valve can be ensured to be opened on time in a proper time period, the timeliness and the accuracy of the identification of the opening state of the safety valve are improved, the management level of a coal mine roof is favorably improved, and the safety of coal mining is improved.
In order to realize the embodiment, the application further provides a system for determining the opening state of the safety valve of the hydraulic support. Fig. 6 is a schematic structural diagram of a system for determining an opening state of a safety valve of a hydraulic mount according to an embodiment of the present application.
As shown in fig. 6, the system includes an acquisition module 100, a marking module 200, a generation module 300, and a determination module 400.
The acquisition module 100 is configured to acquire original mine pressure monitoring data of the hydraulic support through a mine pressure monitoring system.
And the marking module 200 is used for manually marking the opening state of the safety valve on the original mine pressure monitoring data and generating a safety valve opening marking time sequence according to a marking result.
The generating module 300 is configured to sort the original mine pressure monitoring data, the time sequence of the opening state of the safety valve, and the target data of the opening state of the safety valve to be determined into data with equal intervals, and generate a Tsfresh time sequence according to the data with equal intervals.
And the determining module 400 is configured to perform feature extraction and model training on the Tsfresh time sequence through a Tsfresh machine learning framework, and determine the safety valve opening state result of each group of data in the target data through the trained machine learning model.
Optionally, in an embodiment of the present application, the marking module 200 is specifically configured to: marking the starting point of the opening of each round of safety valve in a curve graph of original mine pressure monitoring data; adding a preset time interval to the starting point to determine the starting end point of each round of safety valve; and determining the opening pressure value of the safety valve for opening each round of the safety valve according to the ordinate of the graph.
Optionally, in an embodiment of the present application, the marking module 200 is further configured to: generating a plurality of groups according to the starting point and the end point of the opening of each round of safety valves and the opening pressure value of the safety valves; and sequencing the plurality of groups according to the sequence of the occurrence time of the starting point from far to near to generate a time sequence of the opening state of the safety valve.
Optionally, in an embodiment of the present application, the generating module 300 is specifically configured to: traversing original mine pressure monitoring data, and segmenting the original mine pressure monitoring data according to the marked safety valve opening time period; intercepting the non-safety valve opening time interval between two adjacent safety valve opening time intervals according to a time interval, and rejecting data which do not meet the time interval in the intercepted non-safety valve opening time interval; and intercepting the target data according to a time interval, and removing data which do not meet the time interval in the intercepted target data.
Optionally, in an embodiment of the present application, the generating module 300 is further configured to: determining the safety valve opening state value of each group in a safety valve opening mark time sequence, a non-safety valve opening time sequence and a target data time sequence which are generated after equal interval segmentation; and sequencing each packet containing the state value according to the sequence of the starting time from far to near to generate a Tsfresh time sequence.
Optionally, in an embodiment of the present application, the determining module 400 is specifically configured to: setting a type for performing feature extraction through the Tsfresh, and performing feature extraction on the Tsfresh time sequence through a preset feature extraction function to generate a feature time sequence; dividing the characteristic time sequence into a training set, a verification set and a test set through a preset division function; and training a preset decision tree classifier through the data in the training set to obtain a trained machine learning model for analyzing the opening state of the safety valve.
Optionally, in an embodiment of the present application, the data amount in the test set is a number of packets in the target data time series, and the determining module 400 is further configured to: and predicting the data in the test set through the trained decision tree classifier to obtain the safety valve opening state result.
Optionally, in an embodiment of the present application, the system further includes a display module, configured to draw the safety valve opening state result of each set of data in a rack pressure graph of the hydraulic rack for visual display.
It should be noted that the foregoing description of the embodiment of the method for determining the open state of the safety valve of the hydraulic bracket is also applicable to the system of this embodiment, and the implementation principle is the same, and is not repeated here.
In summary, the system for determining the opening state of the safety valve of the hydraulic support in the embodiment of the application firstly obtains the time sequence data of the monitoring of the support pressure based on the mine pressure monitoring system, then manually marks the opening state of the safety valve on the data within a period of time, processes the original data and the marked data result into a time sequence required by a machine learning framework tsfresh, automatically analyzes the opening state of the safety valve of the hydraulic support by a machine learning method, and comprises the steps of characteristic extraction, characteristic screening, training, prediction and the like, and realizes the analysis of the opening state of the safety valve by a supervised machine learning method. From this, this system has richened the mode that the safety valve opened the state analysis, can be timely and the time interval that the accurate definite safety valve opened to can in time provide early warning information for the colliery staff, guarantee that the safety valve opens on time at suitable time interval, improve the promptness and the accuracy that the state discernment was opened to the safety valve, be favorable to improving colliery roof management level, improve the security of coal mining.
In order to achieve the above embodiments, the present invention further proposes a non-transitory computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the method for determining the open state of the hydraulic support safety valve according to the embodiment of the first aspect of the present application.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In the present specification, if a schematic expression of the above-described terms is employed in a plurality of embodiments or examples, it does not mean that the embodiments or examples are the same. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Moreover, various embodiments or examples and features of various embodiments or examples described in this specification can be combined and combined by one skilled in the art without being mutually inconsistent.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the embodiments of the present application.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Further, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.
Claims (10)
1. A method for determining the opening state of a safety valve of a hydraulic support is characterized by comprising the following steps:
acquiring original mine pressure monitoring data of the hydraulic support through a mine pressure monitoring system;
manually marking the opening state of the safety valve on the original mine pressure monitoring data, and generating a safety valve opening marking time sequence according to a marking result;
arranging the original mine pressure monitoring data, the time sequence of the opening state of the safety valve and target data of the opening state of the safety valve to be determined into equidistant data, and generating a Tsfresh time sequence according to the equidistant data;
and performing feature extraction and model training on the Tsfresh time sequence through a Tsfresh machine learning framework, and determining the safety valve opening state result of each group of data in the target data through a trained machine learning model.
2. The method of claim 1, wherein said manually marking the raw mine pressure monitoring data for the open state of the safety valve comprises:
marking the starting point of the opening of each round of safety valve in the graph of the original mine pressure monitoring data;
adding a preset time interval to the starting point to determine the starting end point of each wheel of safety valve;
and determining the safety valve opening pressure value for opening each round of the safety valve according to the ordinate of the graph.
3. The method of determining the open state of a hydraulic mount safety valve according to claim 2, wherein the generating a safety valve open state time series from the result of the marking comprises:
generating a plurality of groups according to the starting point, the end point and the safety valve opening pressure value of each round of safety valve opening;
and sequencing the plurality of groups according to the sequence of the occurrence time of the starting point from far to near to generate the time sequence of the opening state of the safety valve.
4. The method for determining the open state of a safety valve of a hydraulic support according to claim 2, wherein the step of arranging the original mine pressure monitoring data, the time sequence of the open state of the safety valve and the target data of the open state of the safety valve to be determined into the data with equal intervals comprises the following steps:
traversing the original mine pressure monitoring data, and segmenting the original mine pressure monitoring data according to the marked safety valve opening time period;
intercepting the non-safety valve opening time interval between two adjacent safety valve opening time intervals according to the time interval, and rejecting data which do not meet the time interval in the intercepted non-safety valve opening time interval;
and intercepting the target data according to the time interval, and removing the data which do not meet the time interval in the intercepted target data.
5. The method of claim 4, wherein the generating a Tsfresh time series from the equidistant data comprises:
determining the safety valve opening state value of each group in a safety valve opening mark time sequence, a non-safety valve opening time sequence and a target data time sequence which are generated after equal interval segmentation;
and sequencing each packet containing the state value according to the sequence of the starting time from far to near to generate the Tsfresh time sequence.
6. The method of claim 5, wherein the performing feature extraction and model training on the Tsfresh time series through the Tsfresh machine learning framework comprises:
setting a type for performing feature extraction through the Tsfresh, and performing feature extraction on the Tsfresh time sequence through a preset feature extraction function to generate a feature time sequence;
dividing the characteristic time sequence into a training set, a verification set and a test set through a preset division function;
and training a preset decision tree classifier through the data in the training set to obtain a trained machine learning model for analyzing the opening state of the safety valve.
7. The method for determining the open state of a safety valve of a hydraulic support according to claim 6, wherein the data volume in the test set is the number of groups in the target data time series, and the determining the open state result of the safety valve of each group of data in the target data through the trained machine learning model comprises:
and predicting the data in the test set through the trained decision tree classifier to obtain the safety valve opening state result.
8. The method for determining the open state of a safety valve of a hydraulic support according to claim 1, after the determining the open state result of the safety valve of each set of the target data through the trained machine learning model, further comprising:
and drawing the opening state result of the safety valve of each group of data in a support pressure curve graph of the hydraulic support for visual display.
9. A system for determining the opening state of a safety valve of a hydraulic support is characterized by comprising the following modules:
the acquisition module is used for acquiring original mine pressure monitoring data of the hydraulic support through the mine pressure monitoring system;
the marking module is used for manually marking the opening state of the safety valve on the original mine pressure monitoring data and generating a safety valve opening marking time sequence according to a marking result;
the generating module is used for collating the original mine pressure monitoring data, the time sequence of the opening state of the safety valve and target data of the opening state of the safety valve to be determined into data with equal intervals, and generating a Tsfresh time sequence according to the data with equal intervals;
and the determining module is used for performing feature extraction and model training on the Tsfresh time sequence through a Tsfresh machine learning framework and determining the safety valve opening state result of each group of data in the target data through the trained machine learning model.
10. A non-transitory computer-readable storage medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements the method of determining the open state of a hydraulic mount safety valve according to any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211482962.6A CN115758269A (en) | 2022-11-24 | 2022-11-24 | Method and system for determining opening state of safety valve of hydraulic support |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211482962.6A CN115758269A (en) | 2022-11-24 | 2022-11-24 | Method and system for determining opening state of safety valve of hydraulic support |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115758269A true CN115758269A (en) | 2023-03-07 |
Family
ID=85337089
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211482962.6A Pending CN115758269A (en) | 2022-11-24 | 2022-11-24 | Method and system for determining opening state of safety valve of hydraulic support |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115758269A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117743988A (en) * | 2024-02-20 | 2024-03-22 | 太原理工大学 | Instant prediction method for pressure-bearing state of hydraulic support after initial support |
-
2022
- 2022-11-24 CN CN202211482962.6A patent/CN115758269A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117743988A (en) * | 2024-02-20 | 2024-03-22 | 太原理工大学 | Instant prediction method for pressure-bearing state of hydraulic support after initial support |
| CN117743988B (en) * | 2024-02-20 | 2024-04-19 | 太原理工大学 | Instant prediction method for pressure-bearing state of hydraulic support after initial support |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112015838B (en) | Road test data processing method and system and server | |
| CN108965340B (en) | Industrial control system intrusion detection method and system | |
| CN105487970B (en) | A kind of method for showing interface and device | |
| CN115758269A (en) | Method and system for determining opening state of safety valve of hydraulic support | |
| CN114511026A (en) | Fault diagnosis method and device, terminal equipment and storage medium | |
| CN115545058A (en) | Water meter data analysis method and system and readable storage medium | |
| CN112861706A (en) | Road state monitoring method, device, equipment and storage medium | |
| CN113502870A (en) | Excavator working condition judging method and device | |
| CN115609112A (en) | Welding parameter determination method, device and medium thereof | |
| CN110134040B (en) | Method and system for processing operation data of industrial equipment | |
| CN115712834A (en) | Alarm false alarm detection method, device, equipment and storage medium | |
| CN116074215A (en) | Network quality detection method, device, equipment and storage medium | |
| CN116739317B (en) | Mining winch automatic management and dispatching platform, method, equipment and medium | |
| CN111062827B (en) | Engineering supervision method based on artificial intelligence mode | |
| CN111815941A (en) | Frequent congestion bottleneck identification method and device based on historical road conditions | |
| CN107506832B (en) | Hidden danger mining method for assisting monitoring tour | |
| CN117852893A (en) | Engineering security risk detection method and system based on artificial intelligence technology | |
| CN113313304A (en) | Power grid accident abnormity analysis method and system based on big data decision tree | |
| CN112506754A (en) | System performance monitoring method and platform | |
| CN115604747A (en) | Index abnormity closed-loop management method and device based on 5GC | |
| CN114357185A (en) | Manufacturing equipment fault monitoring model training method based on knowledge graph | |
| CN113256256A (en) | Work order early warning method, device, equipment and storage medium | |
| CN110969721A (en) | Data processing method, device and equipment and touch screen | |
| CN117473347B (en) | Ore dressing full-flow data processing method and system based on rule engine | |
| CN118740892A (en) | Industrial gas analysis method based on Internet of things |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |