CN115018343A - System and method for recognizing and processing abnormity of mass mine gas monitoring data - Google Patents
System and method for recognizing and processing abnormity of mass mine gas monitoring data Download PDFInfo
- Publication number
- CN115018343A CN115018343A CN202210691021.7A CN202210691021A CN115018343A CN 115018343 A CN115018343 A CN 115018343A CN 202210691021 A CN202210691021 A CN 202210691021A CN 115018343 A CN115018343 A CN 115018343A
- Authority
- CN
- China
- Prior art keywords
- data
- monitoring
- value
- risk
- characteristic
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 205
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012545 processing Methods 0.000 title claims abstract description 23
- 230000005856 abnormality Effects 0.000 claims abstract description 22
- 238000010276 construction Methods 0.000 claims abstract description 13
- 238000010219 correlation analysis Methods 0.000 claims abstract description 10
- 230000004044 response Effects 0.000 claims abstract description 7
- 238000013139 quantization Methods 0.000 claims abstract description 4
- 230000008569 process Effects 0.000 claims abstract description 3
- 230000002159 abnormal effect Effects 0.000 claims description 39
- 230000007774 longterm Effects 0.000 claims description 32
- 230000035772 mutation Effects 0.000 claims description 18
- 230000008859 change Effects 0.000 claims description 17
- 230000007246 mechanism Effects 0.000 claims description 17
- 238000005422 blasting Methods 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 13
- 239000003245 coal Substances 0.000 claims description 13
- 238000007405 data analysis Methods 0.000 claims description 12
- 238000005553 drilling Methods 0.000 claims description 10
- 238000013500 data storage Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 8
- 238000005065 mining Methods 0.000 claims description 8
- 230000000630 rising effect Effects 0.000 claims description 8
- 238000012512 characterization method Methods 0.000 claims description 6
- 238000009826 distribution Methods 0.000 claims description 6
- 238000000518 rheometry Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 5
- 238000005259 measurement Methods 0.000 claims description 4
- 230000001174 ascending effect Effects 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 2
- BTCSSZJGUNDROE-UHFFFAOYSA-N gamma-aminobutyric acid Chemical compound NCCCC(O)=O BTCSSZJGUNDROE-UHFFFAOYSA-N 0.000 claims 1
- 229940084388 gammar Drugs 0.000 claims 1
- 230000002045 lasting effect Effects 0.000 claims 1
- 238000010223 real-time analysis Methods 0.000 abstract description 3
- 230000002354 daily effect Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000009412 basement excavation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000005111 flow chemistry technique Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000000869 mutational effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/23—Updating
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2455—Query execution
- G06F16/24568—Data stream processing; Continuous queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Databases & Information Systems (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Data Mining & Analysis (AREA)
- Marketing (AREA)
- General Engineering & Computer Science (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Agronomy & Crop Science (AREA)
- Animal Husbandry (AREA)
- Marine Sciences & Fisheries (AREA)
- Mining & Mineral Resources (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention relates to a system and a method for identifying and disposing abnormity of mass mine gas monitoring data, which belong to the field of gas monitoring and comprise the following steps: the characteristic map construction model is used for constructing a characteristic map of the data fluctuation characteristic, the data mean deviation characteristic, the data constant and invariable characteristic and the data trend variation characteristic of the gas monitoring data; the anomaly identification and treatment model is used for fusing different dimension characteristic quantization values in long, medium and short periods according to the characteristic map, and monitoring, identifying and treating the data; the flow type correlation analysis processing module is used for collecting and storing data, loading a characteristic map construction model and an abnormality identification and disposal model, carrying out real-time analysis on gas monitoring data, formulating risk cooperative response disposal strategies of different levels and different levels, solidifying a business process, constructing an abnormality disposal mode based on an electronic work order, and carrying out abnormality identification and rapid disposal on mass gas monitoring data.
Description
Technical Field
The invention belongs to the field of gas monitoring, and relates to a system and a method for identifying and processing abnormity of mass mine gas monitoring data.
Background
Aiming at monitoring data such as gas, dust, roof and the like which are generated by coal mines in a jurisdiction every day and are less than ten thousand and more than million, a medium-short term (generally within two years) data storage system is constructed on the basis of a relational database, real-time alarm is carried out through a preset threshold value, a 3-5-day minute statistic curve is manually inquired, and trend deviation judgment and analysis are carried out. In the mode, the requirements of convenient query, multi-dimensional mining and real-time processing are difficult to meet from the technical aspect, and the intelligent level of the data change characteristics, trend and rule judgment is extremely low; the method has the advantages of huge workload, obvious time lag and poor effectiveness from the management aspect, and is difficult to migrate the accident prevention nodes to realize effective supervision.
The existing gas anomaly identification is mainly based on a cause mechanism, is complex in model, needs to acquire geological conditions of mine gas and dynamically determine coal rock parameters, is suitable for gas identification and supervision of a single coal mine, is poor in universality and cannot meet supervision of large-scale gas data. Therefore, it is urgently needed to construct a data characteristic map based on collectable monitoring data by adopting a big data technology and a big data method, effectively reveal the time-space evolution characteristics and rules of the monitoring data of the mine excavation key area, and realize automatic locking, identification and early warning of abnormal changes of monitoring values in a large range.
Disclosure of Invention
In view of the above, the invention aims to provide a data-driven coal mine gas monitoring data mathematical characteristic map construction method, which is used for dynamically identifying large-scale gas monitoring data abnormity, solving the problems of automatic locking of mass gas monitoring data abnormity, automatic release of risk information, dynamic tracking and rapid disposal, and realizing machine-assisted on-duty supervision.
In order to achieve the purpose, the invention provides the following technical scheme:
on one hand, the invention provides a system for identifying and disposing the abnormity of mass mine gas monitoring data, which comprises a characteristic map construction model, an abnormity identification and disposal model and a flow-type correlation analysis processing module;
the characteristic map construction model is used for constructing a characteristic map of the data fluctuation characteristic, the data mean deviation characteristic, the data constant and invariable characteristic and the data trend variation characteristic of the gas monitoring data;
the anomaly identification and handling model is used for fusing long, medium and short-term different dimension characteristic quantization values according to a characteristic map, and monitoring, identifying and handling the data;
the flow type correlation analysis processing module is used for collecting and storing data, loading a characteristic map construction model and an abnormality identification and disposal model, carrying out real-time analysis on gas monitoring data, formulating risk cooperative response disposal strategies of different levels and different levels, solidifying a business process, constructing an abnormality disposal mode based on an electronic work order, and carrying out abnormality identification and rapid disposal on mass gas monitoring data.
Further, the streaming correlation analysis processing module comprises a data acquisition unit, a data storage and analysis module and a risk handling module;
the data acquisition unit is used for acquiring gas monitoring data, comprises a plurality of data acquisition modes, acquires interface data in the forms of text files, databases and WebAPI (web application program interfaces) by acquiring and configuring UI (user interface), converts the acquired data into a standardized format and submits the standardized format to a kafka data bus; the collected data comprises monitoring equipment definition data, real-time data, historical minute data and mine basic data; the equipment definition data comprises equipment address types and working faces to which the equipment address types belong; the real-time data and the historical minute data comprise value states, whether the data are in adjustment or not and alarm reasons; the mine basic data comprises coal mining methods, cycle operation shift and gas grade of each working face;
the data storage and analysis module is used for storing the acquired original data into HBase through data analysis service and mapping and storing the acquired original data into Hive; according to collected historical data, storing hot data into an Sqlserver relational database according to data cold and hot characteristics, wherein the data cold and hot characteristics are divided according to the service cycle and frequency of the data; the data storage and analysis module is also used for acquiring a data stream of a time window from Kafka in real time through a spark streaming processing framework by loading historical data in Hive, calling a feature map model on one hand to realize construction and updating of a feature map, storing a result in an Sqlserver database, calling an exception identification and handling model on the other hand to perform data exception identification, and pushing an exception identification result to the risk handling module;
the risk handling module comprises a risk handling policy base and a risk pushing handling module; the risk disposal strategy base establishes various exception disposal mechanisms according to the exception identification result, wherein the exception disposal mechanisms comprise monitoring failure, slow rise of medium and long term trends, data mutation exception, alarming and power failure; determining a risk level and a response mechanism according to the risk value R;
the risk pushing and handling module calls a risk handling strategy according to the abnormal identification result and the risk grade, automatically generates an electronic work order, pushes risk information to a supervisor, and reminds the supervisor through a short message and a mobile phone terminal.
On the other hand, the invention provides a method for identifying and disposing the abnormity of mass mine gas monitoring data, which comprises the following steps:
s1: collecting gas monitoring data and preprocessing the data;
s2: storing the acquired data, and constructing and updating a feature map through a feature map model;
s3: performing anomaly identification on the data through an anomaly identification and treatment model;
s4: and (4) carrying out risk grade division on the abnormal recognition result, giving a corresponding abnormal handling mechanism, and pushing the abnormal handling mechanism to related personnel.
Further, in step S1, collecting gas monitoring data with different form interfaces, converting the collected data into a standardized format, and submitting the standardized format to a kafka data bus; the interface form comprises a text file, a database and a WebAPI; the gas monitoring data comprises monitoring equipment definition data, real-time data, historical minute data and mine basic data; the equipment definition data comprises equipment address types and working faces to which the equipment address types belong; the real-time data and the historical minute data comprise value states, whether the data are in adjustment or not and alarm reasons; the mine basic data comprise coal mining methods of all working surfaces, cycle operation shift and gas grades.
Further, the storing the collected data in step S2 specifically includes:
the acquired original data is stored in HBase through data analysis service and is mapped and stored in Hive; aiming at the collected historical data, storing the hot data into an Sqlserver relational database according to the cold and hot characteristics of the data simultaneously, so that the query efficiency of the data is improved; the data cold and hot characteristics are divided according to the use period and frequency of the data.
Further, in step S2, the feature map constructed by the feature map model includes a data fluctuation feature, a data mean deviation feature, a data constancy feature and a data trend variation feature;
the method for calculating the data fluctuation characteristics comprises the following steps:
the actual fluctuation amplitude mean value is adopted to represent the fluctuation degree of the gas concentration, if a plurality of monitoring data exist in the monitoring period t, the difference between the maximum value and the minimum value is the actual fluctuation amplitude TR t Namely:
TR t ＝x t,high -x t,low
in the formula: x is the number of t,high For monitoring the maximum value in the period t, x t,low Is the minimum value in the monitoring period t;
the real fluctuation amplitude mean value is the real fluctuation amplitude mean value of a plurality of monitoring periods, and the real fluctuation amplitude mean value is calculated by adopting a deformation form of exponential moving average:
wherein N is the number of monitoring periods;
for the first true fluctuation amplitude mean value, the calculation is started in the following way:
adopting the stream processing frame to respectively calculate the fluctuation characteristics of alarming, blasting or drilling and normal date to form a data fluctuation characteristic map, which specifically comprises the following steps:
counting the data fluctuation characteristic map in the alarm period, and selecting the fluctuation characteristics of three stages of 30min before alarm, 30min after alarm duration and 30min after alarm;
analyzing a data fluctuation characteristic map during blasting or drilling according to a period of 30min, and calculating fluctuation characteristics of a monitoring point which lasts for the whole time period after blasting or drilling is started;
analyzing the data fluctuation characteristic map of the normal date according to a period of 30min, and setting the average value of fluctuation amplitude as a cycle operation time;
the method for calculating the data mean deviation characteristics comprises the following steps:
if there are n monitoring data in a certain monitoring point statistical period t, then the mean value SMA in the period t Comprises the following steps:
if the mean value of the previous time is known, the moving average value of the gas concentration at the current time is:
respectively calculating the mean values of the latest 1 day, the latest 3 days and the latest 7 days of each monitoring point to form a mean value characteristic spectrum;
the method for calculating the data constant feature comprises the following steps:
(4) firstly, setting a threshold value zeta according to the minimum detection range of each sensor;
(5) and calculating the maximum time interval in which the difference between the maximum value and the minimum value of the continuous sampling values of each monitoring point is less than the threshold value in the last 1 month, and recording as the constant duration of the data.
(6) And clustering according to the mine gas grade and the address type of each monitoring point to obtain the data constant duration distribution probability characteristic of each address type monitoring point.
The data trend change characteristics comprise a long-term trend and a short-term trend, the long-term trend reflects the overall change condition of gas emission, and the short-term trend is used for revealing whether the monitoring point exceeds the limit or not;
the long-term trend calculation method comprises the following steps:
a1: calculating the daily average value of the monitoring points in the statistical period t aiming at each monitoring point;
a2: obtaining a gas concentration time sequence within a statistical period t according to the daily average value;
a3: fitting the time sequence by adopting a least square method to obtain a fitting curve with the shape of y ═ ax + b, wherein a is the trend characterization of the monitoring point;
calculating the trend characterization a value of each monitoring point according to the steps A1-A3 to form a medium and long term trend change characteristic map;
the short-term trend calculation method comprises the following steps:
b1: and (3) constructing a rank sequence aiming at n data at t moments before each monitoring point:
wherein
k＝1,2,…,n，x i The ith monitoring value of the gas concentration sequence at the latest t time;
b2: calculating the variance of S:
if each monitoring value is unique within t time, the variance is as follows:
if the data of each monitoring value is not unique within the time t, the variance is as follows:
wherein p is the number of repeating numbers, g is the number of unique numbers, t p The number of repetitions of the p-th repetition number;
b3: calculating a trend threshold Z MK ：
B4: and (3) trend judgment:
setting the fault tolerance rate as beta, wherein beta is more than 0 and less than 0.5, and the confidence coefficient of the trend judgment result is 1-beta/2;
when the temperature is higher than the set temperature
When the time is long, the trend is not good;
when in use
And Z is mk When the ratio is more than 0, the trend is increased;
when in use
And Z is mk When the average molecular weight is less than 0, the trend is reduced;
wherein ppf (1-beta/2) is the value of an x axis when the confidence level in normal distribution is 1-beta/2;
using steps B1-B4, Z30 minutes prior to the historical alert data is calculated mk And forming a trend rising measurement characteristic map 30 minutes before the alarm.
Further, the abnormality identification and treatment model in step S3 includes:
based on the constant and invariable characteristic map of the data, identifying and monitoring failure abnormity specifically comprises the following steps:
c1: setting a constant duration threshold gamma according to constant characteristic maps of various monitoring address types of mines with different gas grades, and determining that suspected monitoring is invalid when the continuous micro-variation time of monitoring points is greater than gamma; wherein γ is greater than 80% of the monitoring point minimum change time;
c2: if the micro-change detection is passed, carrying out data fluctuation detection for identifying monitoring failure; setting a minimum fluctuation threshold lambda according to a data fluctuation feature map of the monitoring point, wherein lambda is less than the fluctuation amplitude mean value of 80%; when the data fluctuation values of n consecutive days are all smaller than lambda, the suspected monitoring failure abnormality is judged;
the method for identifying the slow rise of the middle and long term trend specifically comprises the following steps:
and (3) judging the trend of a single monitoring point: firstly, setting a trend deviation threshold, if historical gas outburst accident data exists, calculating a long-term trend representation value of the outburst site by taking the outburst accident time as a reference, and setting the threshold according to the calculation result and the safety coefficientValue theta 1 Otherwise, setting a threshold value theta according to the medium-long term trend characteristic map 1 A step of,; when the current middle and long term trend characteristic value gamma of the monitoring point>θ 1 If yes, judging the abnormal condition, and determining an abnormal risk value R according to the value of gamma; according to the risk rheology mutation theory, an exponential function is adopted to calculate the abnormal risk value, which is shown as the following formula:
in the formula R o Is gamma-theta 1 The risk value corresponding to a lowest risk level;
meanwhile, based on the acquired data, classification and identification are carried out according to the physical positions of the monitoring points, the sensors on the same working face of the mine are divided into an area, and theta is set 2 As a threshold value, the medium-long term trend characteristic value gamma of each monitoring point in the area>θ 2 When it is determined that the region is abnormal, where θ 2 <θ 1 Determining a regional anomaly risk value R according to the value of gamma Region(s) (ii) a According to a risk rheology mutation theory, calculating a regional abnormal risk value by adopting an exponential function:
wherein R is Region(s) As regional anomaly risk value, R o Is gamma-theta 2 The value of the risk of (c) is,
the weighted average of the medium and long-term characteristic values of each monitoring point in the area is obtained; the weight of each monitoring point is calculated according to a power failure limit given by a coal mine safety regulation, and the lower the power failure limit is, the higher the weight is, as shown in the following formula:
in the above formula: w is a k Is the weight of the kth monitoring point in the area, l k The power failure limit of the kth monitoring point is defined, and n is the number of monitoring points in the area;
identifying data mutation abnormalities specifically comprises:
calculating the data fluctuation characteristics of the monitoring points in the last 30 minutes, selecting a monitoring period t for 1min, and counting the period for 30min to obtain a data fluctuation amplitude mean value RT;
identifying whether the data has an ascending trend or not by adopting an MK trend identification method based on the data of the monitoring point in the last 30 minutes;
calculating the deviation degree of the current monitoring value of the monitoring point and the n-day average value, wherein the number of days with the maximum deviation degree is selected as a reference according to the deviation significance degree of the average value characteristics of the alarm data in the first 30 minutes and the 1-day average value, the 3-day average value and the 7-day average value;
integrating the three indexes to judge whether data mutation abnormality exists or not, and setting a threshold theta according to the fluctuation characteristic, the trend rising characteristic and the mean deviation characteristic of gas monitoring data of an alarm, blasting or drilling event 3 、θ 4 、θ 5 (ii) a When any condition is met, determining that the condition is abnormal; determining a risk value R according to the size of the exceeding threshold value, and accumulating the risk values according to the indexes;
the risk values are accumulated with reference to a risk matrix method, as follows:
R'＝R 1 *R 2 *R 3
in the above formula: r' is the accumulated risk value, R 1 For data fluctuation characteristic risk values, R 2 To raise the characteristic risk value according to trend, R 3 To mean the deviation from the characteristic risk value, the risk value R is set to 1 when the indicator does not exceed the threshold.
Further, in step S4, determining a risk level according to the risk value R, wherein the risk value that just reaches the threshold corresponds to the lowest risk level, establishing an exception handling mechanism, which includes monitoring failure, slow rising of medium and long term trends, data mutation exception, alarming, power off, and establishing a risk handling policy library; and after the abnormal recognition result given in the step S3 is obtained, finding a corresponding abnormal handling mechanism from the risk handling strategy library, generating an electronic work order, and pushing the risk information to related personnel.
The invention has the beneficial effects that: according to the method, based on a data driving method, different scene characteristic maps such as alarming, blasting, overrun and the like are established in different dimensions such as a middle and long-term trend slow rise, a recent fluctuation amplitude mean value, data constancy and invariance and the like, a risk handling strategy is formulated, a flow processing framework is used for analyzing and processing gathered massive monitoring data in real time, abnormity is dynamically identified, the time-space evolution characteristics and the law of the monitoring data of the coal mine in a key area are effectively revealed, the abnormal change of the monitoring value in a large range is automatically locked, identified and early warned, the monitoring energy efficiency is greatly improved, and the labor cost is reduced.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
For the purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:
fig. 1 is a structural diagram of an abnormal identification and disposal system for mass mine gas monitoring data according to the invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.
Wherein the showings are for the purpose of illustrating the invention only and not for the purpose of limiting the same, and in which there is shown by way of illustration only and not in the drawings in which there is no intention to limit the invention thereto; to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of an actual product; it will be understood by those skilled in the art that certain well-known structures in the drawings and descriptions thereof may be omitted.
The same or similar reference numerals in the drawings of the embodiments of the present invention correspond to the same or similar components; in the description of the present invention, it should be understood that if there is an orientation or positional relationship indicated by terms such as "upper", "lower", "left", "right", "front", "rear", etc., based on the orientation or positional relationship shown in the drawings, it is only for convenience of description and simplification of description, but it is not an indication or suggestion that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and therefore, the terms describing the positional relationship in the drawings are only used for illustrative purposes, and are not to be construed as limiting the present invention, and the specific meaning of the terms may be understood by those skilled in the art according to specific situations.
Referring to fig. 1, the system for recognizing and disposing the abnormality of the mass mine gas monitoring data of the present invention includes a feature map building model, an abnormality recognizing and disposing module, and a streaming correlation analysis processing module. The characteristic map building model builds a data change mathematical characteristic library under different scenes from dimensions such as gas monitoring data mutation, trend deviation, multi-time window micro fluctuation and the like; the anomaly identification model is used for fusing different dimension characteristic quantization values in long, medium and short periods according to the characteristic map model and establishing a monitoring data anomaly identification model; the flow type correlation analysis processing module is responsible for collecting and storing data, loading a characteristic map construction model and an abnormality identification model, carrying out real-time analysis on mass gas monitoring data, formulating risk cooperative response disposal strategies of different levels and different levels, solidifying a service flow, constructing an abnormality disposal mode based on an electronic work order, and carrying out abnormality identification and rapid disposal on the mass monitoring data.
The gas monitoring data characteristic map comprises a data fluctuation characteristic, a data mean deviation characteristic, a data constant characteristic and a data trend variation characteristic. The feature calculation method is as follows:
1) data fluctuation characteristics
And representing the fluctuation degree of the gas concentration by adopting a real fluctuation amplitude mean value. If there are multiple monitoring data in the monitoring period t, the difference between the maximum value and the minimum value is the true fluctuation range TR t . Namely:
TR t ＝x t,high -x t,low
in the formula: x is the number of t,high For monitoring the maximum value within the period t
x t,low Is the minimum value within the monitoring period t.
The real fluctuation amplitude mean value is the real fluctuation amplitude mean value of a plurality of monitoring periods, and the real fluctuation amplitude mean value is calculated by adopting a deformation form of exponential moving average
Wherein N is the number of monitoring periods.
For the first true fluctuation amplitude mean value, the calculation is started in the following way:
and respectively calculating the fluctuation characteristics of the alarm, blasting and normal date by adopting a streaming processing frame to form a data fluctuation characteristic map.
a: and (3) during alarming, the data fluctuation characteristic map is used for counting the fluctuation characteristics of three stages of 30min before alarming, 30min after alarming and 30min after alarming in a period selection. The monitoring period is selected to be 1 min. I.e. N is 30.
And b, calculating the fluctuation characteristics of the monitoring points after blasting for the whole time period when blasting or drilling and other excavation activities which easily cause gas accumulation. And blasting or drilling is identified through the uploaded operation shift. And for the selection of the monitoring period t, the analysis is recommended to be carried out according to a period of 30min, and the duration of the whole operation shift is selected in a statistical period.
And c, data fluctuation characteristics in normal period. For the selection of the monitoring period t, since the latest fluctuation situation is very important, the analysis is recommended to be performed according to the period of 30 min. Since the concentration of the underground gas is mainly influenced by the mining activity, the statistical time of the mean values of the two fluctuation amplitudes should contain approximately the same mining activity. I.e. the average value of the amplitude of the fluctuation should preferably be set to one cycle operation time. For convenience, the statistical period of the characteristic index of all the monitored parameters can be uniformly set to 1 d. When the monitoring period t is 30min, N is 24.
2) Data mean deviation characterization
The moving average may smooth out short term fluctuations, reflecting long term trends or periods, which may be used to demonstrate the near term level of the medium and short term monitored data. If n monitoring data exist in a certain monitoring point statistical period t, the mean value SMA in the period t Comprises the following steps:
if the mean value of the previous time is known, the moving average value of the gas concentration at the current time can be calculated according to the following formula:
according to the formula, the average value of the nearest 1d, 3d and 7d of each monitoring point can be respectively calculated to form an average value characteristic map.
3) Data invariant feature
The constant data refers to the maximum time interval during which the continuous sampling value is unchanged in the statistical period t. In order to reduce data fluctuations due to sensor measurement errors, sampling is done with small variations instead of being constant. I.e. the maximum time interval within the statistical period t in which the difference between the maximum value and the minimum value of the continuous sampling values is smaller than the threshold value.
According to the method, the constant duration time of the data of the latest 1 month of each monitoring point is calculated respectively. And clustering is carried out according to the mine gas grades and the address types of all the monitoring points to obtain the data constant duration distribution probability of the monitoring points of the address types with different gas grades.
4) Data trend change characteristics
The data trend change characteristics are divided into long-term trends and short-term trends. The long-term trend reflects the overall change condition of gas emission, and the short-term trend is used for revealing whether the monitoring point exceeds the limit or not. And the long-term trend is identified by fitting by using a least square method, and the short-term trend is identified by using an MK test method.
The long-term trend calculation method is as follows:
firstly, calculating the daily average value of the monitoring points in the statistical period t aiming at each monitoring point. Here, the statistical period t should be selected appropriately for a longer period, such as 3 months or 6 months.
Secondly, obtaining a gas concentration time sequence within a statistical period t according to the daily average value;
and thirdly, fitting the time sequence by adopting a least square method to obtain a fitting curve with the shape of y as ax + b. and a is the trend representation of the monitoring point.
And calculating the trend characterization a value of each monitoring point according to the method to form a medium and long term trend change characteristic map.
The short-term trend calculation method is as follows:
the method comprises the following steps of firstly, constructing a rank sequence aiming at n (n >10) data at t time before each monitoring point:
wherein
k＝1,2,…,n，x i The ith monitoring value of the gas concentration sequence at the latest time t.
② calculating the variance of S
If each monitoring value is unique within t time, the variance is as follows:
if the data of each monitoring value in the time t is not unique, the following steps are carried out:
wherein p is the number of repeating numbers, g is the number of unique numbers, t p The number of repetitions of the p-th repetition number.
Calculating z value
Trend judgment
And setting the fault tolerance rate to be alpha, wherein 0< alpha <0.5, and the confidence coefficient of the trend judgment result is 1-alpha/2.
When in use
There was no tendency.
When in use
And Z is mk Trend rises > 0.
When in use
And Z is mk The tendency is decreased <0.
Wherein ppf (1-. alpha./2) is the value of the x-axis with confidence level of 1-. alpha./2 in the normal distribution.
By adopting the method, Z30 minutes before the historical alarm data is calculated mk And forming a trend rising measurement characteristic map 30 minutes before the alarm.
(2) Anomaly identification and handling model
1) Monitoring failure identification
And identifying and monitoring failure abnormity based on the multi-time window micro-variation characteristic map. The details are as follows:
setting a threshold value alpha according to a tiny variation characteristic map of each monitoring address type of different gas-grade mines, and judging that suspected monitoring is invalid when the duration of tiny variation of monitoring points is longer than alpha. Where alpha should be greater than eighty percent of the monitoring point minimum change time. The longer the duration, the greater the risk. This approach primarily identifies equipment failure or monitoring failures caused by equipment being blocked.
And if the micro change detection is passed, carrying out data fluctuation detection to identify monitoring failure caused by dragging positions. And setting a minimum fluctuation threshold value beta (wherein beta is less than the average value of 80% of fluctuation amplitude) according to the data fluctuation feature map of the monitoring points. And when the data fluctuation value of n consecutive days (wherein n >3) is less than beta, determining that the monitoring is abnormal. The risk is greater with longer duration.
2) Medium and long term slow-rising trend identification
And (3) judging the trend of a single monitoring point: firstly, setting a trend deviation threshold, if historical gas outburst accident data exists, calculating a long-term trend characteristic value of the outburst site by taking the outburst accident time as a reference, and setting a threshold theta according to the calculation result and a safety coefficient 1 Otherwise, setting a threshold value theta according to the medium-long term trend characteristic map 1 B, carrying out the following steps of; when the current middle and long term trend characteristic value gamma of the monitoring point>θ 1 If yes, judging the system to be abnormal, and determining an abnormal risk value R according to the value of gamma; according to the risk rheology mutation theory, an exponential function is adopted to calculate the abnormal risk value, which is shown as the following formula:
in the formula R o Is gamma-theta 1 The risk value of (a), the risk value corresponding to a minimum risk level;
meanwhile, based on the collected data, according to the position of the monitoring pointClassifying and identifying physical positions, dividing sensors on the same working face of a mine into a region, and setting theta 2 As a threshold value, the characteristic value gamma of the medium-long term trend of each monitoring point in the area>θ 2 When it is determined that the region is abnormal, where θ 2 <θ 1 Determining a regional anomaly risk value R according to the value of gamma Region(s) (ii) a According to a risk rheology mutation theory, calculating a region abnormal risk value by adopting an exponential function:
wherein R is Region(s) Is a regional abnormal risk value, R o Is gamma-theta 2 The value of the risk of (c) is,
the weighted average of the medium and long-term characteristic values of each monitoring point in the area is obtained; the weight of each monitoring point is calculated according to a power failure limit given by a coal mine safety regulation, and the lower the power failure limit is, the higher the weight is, as shown in the following formula:
in the above formula: w is a k Is the weight of the kth monitoring point in the area, l k Is the power-off limit of the kth monitoring point, and n is the number of the monitoring points in the area.
3) Data mutational abnormalities
The specific method for identifying the data mutation abnormality is as follows:
firstly, calculating the data fluctuation characteristics of the monitoring points in the last 30 minutes (the monitoring period t is selected to be 1min, and the statistical period is 30min), and obtaining a data fluctuation amplitude mean value RT.
And secondly, identifying whether the data has an ascending trend or not by adopting an MK trend identification method based on the data of the monitoring point in the last 30 minutes.
And thirdly, calculating the deviation degree of the current monitoring value of the monitoring point and the n day average value, wherein the day with the maximum deviation degree is selected as a reference according to the deviation significance degree of the average value characteristics of the alarm data in the first 30 minutes and the 1d average value, the 3d average value and the 7d average value. Here the 3d mean is taken temporarily.
Integrating the three indexes to judge whether data mutation abnormality exists, setting a threshold value theta according to the fluctuation characteristic, the trend rising characteristic and the mean deviation characteristic of gas monitoring data of events such as alarming, blasting and the like 1 、θ 2 、θ 3 . When either condition is satisfied, it is determined to be abnormal. The more conditions that are satisfied, the greater the risk. Determining a risk value R according to the size of the exceeding threshold value, and accumulating the risk values by adopting a risk matrix method according to the indexes; as follows:
R'＝R 1 *R 2 *R 3
in the above formula: r' is the accumulated risk value, R 1 For data fluctuation characteristic risk values, R 2 To raise the characteristic risk value according to trend, R 3 To mean the deviation from the characteristic risk value, the risk value R is set to 1 when the indicator does not exceed the threshold.
(3) Stream type correlation analysis processing module
1) Data acquisition
The data acquisition is completed by a data acquisition unit. The data acquisition unit is internally provided with a plurality of data acquisition modes, realizes the acquisition of different forms of interface data such as text files, databases, WebAPI and the like through acquisition configuration UI, converts the acquired data into a standardized format and submits the standardized format to a kafka data bus.
The collected data includes monitoring equipment definition data, real-time data, historical minute data, mine basic data, and the like. The device definition data comprises a device address type, a working face and the like; the real-time data and the historical minute data comprise value states, whether the data are in adjustment or not, alarm reasons and the like; the mine basic data comprises coal mining methods of all working surfaces, cycle operation shift, gas grade and the like.
2) Data storage and analysis
And storing the acquired original data into HBase through data analysis service, and mapping and storing the acquired original data into Hive. And aiming at the collected historical data, storing the hot data into the Sqlserver relational database according to the cold and hot characteristics of the data, so that the query efficiency of the data is improved. The data cold and hot characteristics are divided according to the use period and frequency of the data. Here, the historical data of the last 1 month is classified as thermal data.
The characteristic map is constructed by compiling HiveSQL, the processing framework loads historical data in the Hive, the characteristic map model is called, the characteristic map is constructed and updated, and the result is stored in the Sqlserver database.
The system acquires the data stream of a time window from Kafka in real time through a spark streaming processing framework, on one hand, calls a feature map model to realize the construction and the updating of the feature map, stores the feature map into an Sqlserver database, on the other hand, calls an exception identification and handling model to perform exception identification, and automatically pushes the exception identification result to a risk handling module.
3) Risk handling module
The risk handling module comprises a risk handling strategy library and a risk pushing handling module.
And the risk disposal strategy library establishes various abnormal disposal mechanisms such as monitoring failure, slow rise of medium and long term trends, abnormal data mutation, alarm, power failure and the like according to the abnormal identification result, namely the risk level. And determining a risk level and a response mechanism according to the risk value R.
And the risk pushing and handling module calls a risk handling strategy according to the abnormal identification result and the risk grade, automatically generates an electronic work order, pushes risk information to a supervisor, and reminds the supervisor through a short message, a mobile phone terminal and the like.
Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.
Claims (8)
1. The utility model provides a mine magnanimity gas monitoring data anomaly identification and processing system which characterized in that: the method comprises a feature map construction model, an abnormality identification and treatment model and a streaming correlation analysis processing module;
the characteristic map construction model is used for constructing a characteristic map of the data fluctuation characteristic, the data mean deviation characteristic, the data constant and invariable characteristic and the data trend variation characteristic of the gas monitoring data;
the anomaly identification and handling model is used for fusing long, medium and short-term different dimension characteristic quantization values according to a characteristic map, and monitoring, identifying and handling the data;
the flow type correlation analysis processing module is used for collecting and storing data, loading a characteristic map construction model and an exception identification and handling model, analyzing gas monitoring data in real time, formulating risk cooperative response handling strategies of different levels and different grades, solidifying a business process, constructing an exception handling mode based on an electronic work order, and performing exception identification and rapid handling on massive gas monitoring data.
2. The mine mass gas monitoring data anomaly identification and disposal system of claim 1, wherein: the stream-type correlation analysis processing module comprises a data acquisition unit, a data storage and analysis module and a risk handling module;
the data acquisition unit is used for acquiring gas monitoring data, comprises a plurality of data acquisition modes, acquires and configures UI (user interface) to realize the acquisition of interface data in the forms of text files, databases and WebAPI (web application program interfaces), converts the acquired data into a standardized format and submits the standardized format to a kafka data bus; the collected data comprises monitoring equipment definition data, real-time data, historical minute data and mine basic data; the equipment definition data comprises equipment address types and working faces to which the equipment address types belong; the real-time data and the historical minute data comprise value states, whether the data are in adjustment or not and alarm reasons; the mine basic data comprises coal mining methods, cycle operation shift and gas grade of each working face;
the data storage and analysis module is used for storing the acquired original data into the HBase through data analysis service and mapping and storing the acquired original data into the Hive; according to collected historical data, storing hot data into an Sqlserver relational database according to data cold and hot characteristics, wherein the data cold and hot characteristics are divided according to the service cycle and frequency of the data; the data storage and analysis module is also used for acquiring a data stream of a time window from Kafka in real time through a spark streaming processing framework by loading historical data in Hive, calling a feature map model on one hand to realize construction and updating of a feature map, storing a result in an Sqlserver database, calling an exception identification and handling model on the other hand to perform data exception identification, and pushing an exception identification result to the risk handling module;
the risk handling module comprises a risk handling policy base and a risk pushing handling module; the risk disposal strategy base establishes various exception disposal mechanisms according to the exception identification result, wherein the exception disposal mechanisms comprise monitoring failure, slow rise of medium and long term trends, data mutation exception, alarming and power failure; determining a risk level and a response mechanism according to the risk value R;
and the risk pushing and handling module calls a risk handling strategy according to the abnormal identification result and the risk grade, automatically generates an electronic work order, pushes risk information to a supervisor, and reminds the supervisor through a short message and a mobile phone terminal.
3. A mine mass gas monitoring data abnormity identification and disposal method is characterized by comprising the following steps: the method comprises the following steps:
s1: collecting gas monitoring data and preprocessing the data;
s2: storing the acquired data, and constructing and updating a characteristic map through a characteristic map model;
s3: performing anomaly identification on the data through an anomaly identification and treatment model;
s4: and (4) carrying out risk grade division on the abnormal recognition result, giving a corresponding abnormal handling mechanism, and pushing the abnormal handling mechanism to related personnel.
4. The mine mass gas monitoring data anomaly identification and disposal method according to claim 3, wherein: in step S1, collecting gas monitoring data of different form interfaces, converting the collected data into a standardized format and submitting the standardized format to a kafka data bus; the interface form comprises a text file, a database and a WebAPI; the gas monitoring data comprises monitoring equipment definition data, real-time data, historical minute data and mine basic data; the equipment definition data comprises equipment address types and working faces to which the equipment address types belong; the real-time data and the historical minute data comprise value states, whether the data are in adjustment or not and alarm reasons; the mine basic data comprise coal mining methods of all working surfaces, cycle operation shift and gas grades.
5. The mine mass gas monitoring data anomaly identification and disposal method according to claim 3, wherein: storing the acquired data in step S2, specifically including:
the acquired original data is stored in HBase through data analysis service and is mapped and stored in Hive; aiming at the collected historical data, storing the hot data into an Sqlserver relational database according to the cold and hot characteristics of the data simultaneously, so that the query efficiency of the data is improved; the data cold and hot characteristics are divided according to the use period and frequency of the data.
6. The mine mass gas monitoring data anomaly identification and disposal method according to claim 3, wherein: in step S2, the feature map constructed by the feature map model includes a data fluctuation feature, a data mean deviation feature, a data constant feature and a data trend variation feature;
the method for calculating the data fluctuation characteristics comprises the following steps:
the true fluctuation amplitude mean value is adopted to represent the fluctuation degree of the gas concentration, if a plurality of monitoring data exist in the monitoring period t, the difference between the maximum value and the minimum value is the true fluctuation amplitude TR t Namely:
TR t ＝x t,high -x t,low
in the formula: x is the number of t,high For monitoring the maximum value within the period t, x t,low Is the minimum value in the monitoring period t;
the real fluctuation amplitude mean value is the real fluctuation amplitude mean value of a plurality of monitoring periods, and the real fluctuation amplitude mean value is calculated by adopting a deformation form of exponential moving average:
wherein N is the number of monitoring periods;
for the first true fluctuation amplitude mean value, the calculation is started in the following way:
adopting the stream processing frame to respectively calculate the fluctuation characteristics of alarming, blasting or drilling and normal date to form a data fluctuation characteristic map, which specifically comprises the following steps:
counting the data fluctuation characteristic map in the alarm period, and selecting the fluctuation characteristics of three stages of 30min before alarm, 30min after alarm duration and 30min after alarm;
analyzing the data fluctuation characteristic map during blasting or drilling according to a 30min period, and calculating the fluctuation characteristic of the monitoring point lasting for the whole time period after blasting or drilling is started;
analyzing the data fluctuation characteristic map of the normal date according to a period of 30min, and setting the average value of fluctuation amplitude as a cycle operation time;
the method for calculating the data mean deviation features comprises the following steps:
if there are n monitoring data in a certain monitoring point statistical period t, then the mean value SMA in the period t Comprises the following steps:
if the mean value of the previous time is known, the moving average value of the gas concentration at the current time is:
respectively calculating the average values of the latest 1 day, 3 days and 7 days of each monitoring point to form an average characteristic map;
the method for calculating the data constant feature comprises the following steps:
(1) firstly, setting a threshold value zeta according to the minimum detection range of each sensor;
(2) calculating the maximum time interval that the difference between the maximum value and the minimum value of the continuous sampling values of each monitoring point is less than the threshold value in the last 1 month, and recording as the constant and invariable duration of the data;
(3) clustering according to the mine gas grade and the address type of each monitoring point to obtain the data constant duration distribution probability characteristic of each address type monitoring point;
the data trend change characteristics comprise a long-term trend and a short-term trend, the long-term trend reflects the overall change condition of gas emission, and the short-term trend is used for revealing whether the monitoring point exceeds the limit or not;
the long-term trend calculation method comprises the following steps:
a1: calculating the daily average value of the monitoring points in the statistical period t aiming at each monitoring point;
a2: obtaining a gas concentration time sequence within a statistical period t according to the daily average value;
a3: fitting the time sequence by adopting a least square method to obtain a fitting curve with the shape of y ═ ax + b, wherein a is the trend characterization of the monitoring point;
calculating the trend characterization a value of each monitoring point according to the steps A1-A3 to form a medium and long term trend change characteristic map;
the short-term trend calculation method comprises the following steps:
b1: and (3) constructing a rank sequence aiming at n data at t moments before each monitoring point:
wherein
(j＝1,2,...,i；i＝1,2,...,n)，k＝1,2,…,n，x i The ith monitoring value of the gas concentration sequence at the latest t time;
b2: calculating the variance of S:
if each monitoring value is unique within t time, the variance is as follows:
if data of each monitoring value in the time t are not unique, the variance is as follows:
wherein p is the number of repeating numbers, g is the number of unique numbers, t p The number of repetitions of the p-th repetition number;
b3: calculating a trend threshold Z MK ：
B4: and (3) trend judgment:
setting the fault tolerance rate as beta, wherein beta is more than 0 and less than 0.5, and the confidence coefficient of the trend judgment result is 1-beta/2;
when in use
When the method is used, no trend exists;
when in use
And Z is mk When the ratio is more than 0, the trend is increased;
when in use
And Z is mk When the average molecular weight is less than 0, the trend is reduced;
wherein ppf (1-beta/2) is the value of an x axis when the confidence level in normal distribution is 1-beta/2;
using steps B1-B4, Z30 minutes prior to the historical alert data is calculated mk And forming a trend rising measurement characteristic map 30 minutes before the alarm.
7. The mine mass gas monitoring data anomaly identification and disposal method according to claim 6, wherein: the anomaly identification and treatment model in step S3 includes:
based on the constant and invariable characteristic map of the data, identifying and monitoring failure abnormity specifically comprises the following steps:
c1: setting a constant duration threshold gamma according to constant characteristic maps of various monitoring address types of mines with different gas grades, and determining that suspected monitoring is invalid when the continuous micro-variation time of monitoring points is greater than gamma; wherein γ is greater than 80% of the monitoring point minimum change time;
c2: if the micro-change detection is passed, carrying out data fluctuation detection for identifying monitoring failure; setting a minimum fluctuation threshold lambda according to a data fluctuation feature map of the monitoring point, wherein lambda is less than the fluctuation amplitude mean value of 80%; when the data fluctuation values of n consecutive days are all smaller than lambda, the suspected monitoring failure abnormality is judged;
the method for identifying the slow rise of the middle and long term trend specifically comprises the following steps:
and (3) judging the trend of a single monitoring point: firstly, setting a trend deviation threshold, if historical gas outburst accident data exists, calculating a long-term trend characteristic value of the outburst site by taking the outburst accident time as a reference, and setting a threshold theta according to the calculation result and a safety coefficient 1 Otherwise, setting a threshold value theta according to the medium-long term trend characteristic map 1 B, carrying out the following steps of; when the current middle and long term trend characteristic value gamma of the monitoring point>θ 1 If so, determining the abnormal condition, and determining an abnormal risk value according to the value of gammaR; according to the risk rheology mutation theory, an exponential function is adopted to calculate the abnormal risk value, which is shown as the following formula:
in the formula R o Is gamma-theta 1 The risk value corresponding to a lowest risk level;
meanwhile, based on the acquired data, classification and identification are carried out according to the physical positions of the monitoring points, the sensors on the same working face of the mine are divided into an area, and theta is set 2 As a threshold value, the medium-long term trend characteristic value gamma of each monitoring point in the area>θ 2 When it is determined that the region is abnormal, where θ 2 <θ 1 Determining a regional anomaly risk value R according to the value of gamma Region(s) (ii) a According to a risk rheology mutation theory, calculating a regional abnormal risk value by adopting an exponential function:
wherein R is Region(s) As regional anomaly risk value, R o Is gamma-theta 2 The value of the risk of (c) is,
the weighted average of the medium and long-term characteristic values of each monitoring point in the area is obtained; the weight of each monitoring point is calculated according to a power failure limit given by a coal mine safety regulation, and the lower the power failure limit is, the higher the weight is, as shown in the following formula:
in the above formula: w is a k Is the weight of the kth monitoring point in the area, l k The power failure limit of the kth monitoring point is defined, and n is the number of monitoring points in the area;
identifying data mutation abnormalities specifically comprises:
calculating the data fluctuation characteristics of the monitoring points in the last 30 minutes, selecting a monitoring period t for 1min, and counting the period for 30min to obtain a data fluctuation amplitude mean value RT;
identifying whether the data has an ascending trend or not by adopting an MK trend identification method based on the data of the monitoring point in the last 30 minutes;
calculating the deviation degree of the current monitoring value of the monitoring point and the n-day average value, wherein the number of days with the maximum deviation degree is selected as a reference according to the deviation significance degree of the average value characteristics of the alarm data in the first 30 minutes and the 1-day average value, the 3-day average value and the 7-day average value;
integrating the three indexes to judge whether data mutation abnormality exists or not, and setting a threshold theta according to the fluctuation characteristic, the trend rising characteristic and the mean deviation characteristic of gas monitoring data of an alarm, blasting or drilling event 3 、θ 4 、θ 5 (ii) a When any condition is met, judging the condition to be abnormal; determining a risk value R according to the size of the exceeding threshold value, and accumulating the risk values according to the indexes;
the risk values are accumulated with reference to a risk matrix method, as follows:
R'＝R 1 *R 2 *R 3
in the above formula: r' is the accumulated risk value, R 1 For data fluctuation characteristic risk values, R 2 To raise the characteristic risk value according to trend, R 3 To mean the deviation from the characteristic risk value, the risk value R is set to 1 when the indicator does not exceed the threshold.
8. The mine mass gas monitoring data anomaly identification and disposal method according to claim 7, characterized by comprising the following steps: in step S4, determining risk levels according to the risk values R, wherein the risk value which just reaches a threshold corresponds to the lowest risk level, establishing an exception handling mechanism which comprises monitoring failure, slowly rising of medium and long term trends, data mutation exception, alarming, power failure and establishing a risk handling strategy library; and after the abnormal recognition result given in the step S3 is obtained, finding a corresponding abnormal handling mechanism from the risk handling strategy library, generating an electronic work order, and pushing the risk information to related personnel.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210691021.7A CN115018343A (en) | 2022-06-09 | 2022-06-09 | System and method for recognizing and processing abnormity of mass mine gas monitoring data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210691021.7A CN115018343A (en) | 2022-06-09 | 2022-06-09 | System and method for recognizing and processing abnormity of mass mine gas monitoring data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115018343A true CN115018343A (en) | 2022-09-06 |
Family
ID=83075577
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210691021.7A Pending CN115018343A (en) | 2022-06-09 | 2022-06-09 | System and method for recognizing and processing abnormity of mass mine gas monitoring data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115018343A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116046063A (en) * | 2023-01-05 | 2023-05-02 | 安徽建筑大学 | Method for monitoring prestress anchor bolt support of deep soft rock roadway |
| CN116112292A (en) * | 2023-04-12 | 2023-05-12 | 湖南丛茂科技有限公司 | Abnormal behavior detection method, system and medium based on network flow big data |
| CN116510223A (en) * | 2023-06-29 | 2023-08-01 | 欣灵电气股份有限公司 | Self-identification fire pump electrical parameter inspection monitoring system based on Internet of things |
| CN116860563A (en) * | 2023-09-05 | 2023-10-10 | 山东捷瑞数字科技股份有限公司 | Cloud platform-based database server monitoring method and system |
| CN117554882A (en) * | 2023-11-14 | 2024-02-13 | 浙江金卡电力科技有限公司 | Meter box monitoring method and device based on Internet of things communication device |
-
2022
- 2022-06-09 CN CN202210691021.7A patent/CN115018343A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116046063A (en) * | 2023-01-05 | 2023-05-02 | 安徽建筑大学 | Method for monitoring prestress anchor bolt support of deep soft rock roadway |
| CN116046063B (en) * | 2023-01-05 | 2023-07-07 | 安徽建筑大学 | Method for monitoring prestress anchor bolt support of deep soft rock roadway |
| CN116112292A (en) * | 2023-04-12 | 2023-05-12 | 湖南丛茂科技有限公司 | Abnormal behavior detection method, system and medium based on network flow big data |
| CN116112292B (en) * | 2023-04-12 | 2023-06-09 | 湖南丛茂科技有限公司 | Abnormal behavior detection method, system and medium based on network flow big data |
| CN116510223A (en) * | 2023-06-29 | 2023-08-01 | 欣灵电气股份有限公司 | Self-identification fire pump electrical parameter inspection monitoring system based on Internet of things |
| CN116510223B (en) * | 2023-06-29 | 2023-09-01 | 欣灵电气股份有限公司 | Self-identification fire pump electrical parameter inspection monitoring system based on Internet of things |
| CN116860563A (en) * | 2023-09-05 | 2023-10-10 | 山东捷瑞数字科技股份有限公司 | Cloud platform-based database server monitoring method and system |
| CN116860563B (en) * | 2023-09-05 | 2023-12-15 | 山东捷瑞数字科技股份有限公司 | Cloud platform-based database server monitoring method and system |
| CN117554882A (en) * | 2023-11-14 | 2024-02-13 | 浙江金卡电力科技有限公司 | Meter box monitoring method and device based on Internet of things communication device |
| CN117554882B (en) * | 2023-11-14 | 2024-06-07 | 浙江金卡电力科技有限公司 | Meter box monitoring method and device based on Internet of things communication device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115018343A (en) | System and method for recognizing and processing abnormity of mass mine gas monitoring data | |
| CN106022592B (en) | Electricity consumption behavior abnormity detection and public security risk early warning method and device | |
| CN113325153A (en) | Water quality multi-parameter monitoring comprehensive information management system | |
| CN112817280A (en) | Implementation method for intelligent monitoring alarm system of thermal power plant | |
| CN112084684B (en) | Bridge health visual monitoring system based on Internet of things | |
| CN113036913B (en) | Method and device for monitoring state of comprehensive energy equipment | |
| CN113173104B (en) | New energy vehicle power battery early warning method and system | |
| CN117610322B (en) | Digital twinning-based intelligent water affair dynamic monitoring system and monitoring method | |
| CN106874671A (en) | A kind of algorithm of the data-driven for spontaneous fire in coal mine big data platform | |
| CN116258241A (en) | Space-time correlation-based supervision environment risk prediction method and system | |
| CN114519923A (en) | Intelligent diagnosis and early warning method and system for power plant | |
| CN112668873A (en) | Mine safety situation analysis and prediction early warning method | |
| CN116579615A (en) | Vegetation coverage monitoring system based on unmanned aerial vehicle remote sensing | |
| CN116740883A (en) | Security monitoring alarm system based on cloud computing | |
| CN117370919A (en) | Remote monitoring system for sewage treatment equipment | |
| CN117192042A (en) | Intelligent mine gas inspection system and method | |
| CN117691722B (en) | Lithium battery charging safety monitoring and early warning method and system | |
| CN111460027A (en) | Intelligent dynamic monitoring method and system suitable for energy Internet | |
| CN114969068B (en) | Urban pressure pipe network real-time flow monitoring data analysis method and system | |
| CN112765553B (en) | Engineering project management system based on big data | |
| CN115578841A (en) | Shield abnormal data detection method and system based on Z-score model | |
| CN110515365B (en) | Industrial control system abnormal behavior analysis method based on process mining | |
| CN112884000A (en) | Power utilization inspection intelligent diagnosis method and diagnosis system based on data mining | |
| CN101923605A (en) | Wind pre-warning method for railway disaster prevention | |
| CN116992390B (en) | Configuration and display method of abnormal data |
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 |