CN115567562A - Mine multi-sensor data acquisition, cleaning and fault discrimination system - Google Patents
Mine multi-sensor data acquisition, cleaning and fault discrimination system Download PDFInfo
- Publication number
- CN115567562A CN115567562A CN202211338736.0A CN202211338736A CN115567562A CN 115567562 A CN115567562 A CN 115567562A CN 202211338736 A CN202211338736 A CN 202211338736A CN 115567562 A CN115567562 A CN 115567562A
- Authority
- CN
- China
- Prior art keywords
- data
- sensor
- queue
- abnormal
- time
- 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
- 238000004140 cleaning Methods 0.000 title claims abstract description 40
- 230000002159 abnormal effect Effects 0.000 claims abstract description 83
- 238000012544 monitoring process Methods 0.000 claims abstract description 83
- 238000000034 method Methods 0.000 claims abstract description 26
- 230000004044 response Effects 0.000 claims abstract description 18
- 239000011435 rock Substances 0.000 claims abstract description 15
- 238000001514 detection method Methods 0.000 claims abstract description 14
- 238000004519 manufacturing process Methods 0.000 claims abstract description 13
- 238000003860 storage Methods 0.000 claims abstract description 7
- 238000005070 sampling Methods 0.000 claims description 65
- 239000011159 matrix material Substances 0.000 claims description 33
- 230000005540 biological transmission Effects 0.000 claims description 27
- 238000013507 mapping Methods 0.000 claims description 20
- 238000013500 data storage Methods 0.000 claims description 13
- 230000008569 process Effects 0.000 claims description 10
- 238000011156 evaluation Methods 0.000 claims description 8
- 238000007726 management method Methods 0.000 claims description 6
- 230000035945 sensitivity Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000012217 deletion Methods 0.000 claims description 4
- 230000037430 deletion Effects 0.000 claims description 4
- 230000002085 persistent effect Effects 0.000 claims description 3
- 238000005406 washing Methods 0.000 claims description 2
- 238000013480 data collection Methods 0.000 claims 4
- 238000010586 diagram Methods 0.000 description 11
- 230000008447 perception Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000009466 transformation Effects 0.000 description 4
- 238000006073 displacement reaction Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
-
- 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
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/06—Management of faults, events, alarms or notifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L43/00—Arrangements for monitoring or testing data switching networks
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- Medical Informatics (AREA)
- Environmental & Geological Engineering (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The invention provides a mine multi-sensor data acquisition, cleaning and fault discrimination system, belonging to the technical field of real-time intelligent sensing of mine engineering environment; the method comprises the following steps: the data acquisition subsystem is used for carrying out real-time acquisition and storage on information cloud configuration and mechanical response information of the rock mechanical response monitoring equipment in the mine production process; the method comprises the steps that a dynamic queue is constructed on the basis of a data acquisition subsystem after monitoring equipment information configuration and rock mass mechanics response information data in the mine production process acquired in real time, queue data classification is realized according to a binary classification principle, and a multi-sensor abnormal data cleaning subsystem in an abnormal state of latest acquired monitoring data is judged according to a classification result of queue tail elements; and the sensor fault detection subsystem is used for judging the fault state of the sensor in operation by adopting a dynamic time window mode and a data threshold value or data acquisition time interval mode.
Description
Technical Field
The invention relates to the technical field of real-time intelligent sensing of mine engineering environments, in particular to a system for data acquisition, cleaning and fault judgment of a mine multi-sensor.
Background
In recent years, the rapid development of subversive technologies such as 5G, internet +, cloud computing, artificial intelligence, big data and the like draws a sequence of the fourth industrial revolution, accelerates the revolution of the traditional industry of mining engineering, enables the transformation and upgrade of the mining industry by using an intelligent technology, and ensures that the digitization and the intelligent transformation of mines are the inevitable trend of the development of the mines.
A diversified sensor is installed in a mine, so that the real-time perception of the environment is realized, and the method is a basis for developing the digital and intelligent transformation of the mine. At present, diversified monitoring equipment is installed in a large number of mines in China, real-time perception of mine environment is achieved preliminarily, but some problems still exist. Firstly, due to the difference of communication protocols of all devices, the unified management of multi-element heterogeneous data cannot be realized; secondly, the harsh environment of the mine can affect the data collected by the sensor, so that the sensor cannot accurately acquire real environment data, and the problem of sensing data interruption is more likely to occur. The existence of the problems can cause remarkable influence on the continuity and accuracy of mine environment perception, and influence the digital and intelligent transformation process of the mine. Therefore, it is very necessary to develop a method and a system for acquiring, cleaning and judging faults of mine multi-sensor data, realize one-key access and unified management of the multi-sensor data, and intelligently clean monitoring data and automatically detect sensor abnormality.
Disclosure of Invention
In order to solve the problems, the invention provides a mine multi-sensor data acquisition, cleaning and fault discrimination system, which has the specific technical scheme that:
a mine multi-sensor data acquisition, cleaning and discrimination system comprises:
the data acquisition subsystem is used for carrying out real-time acquisition and storage on information cloud configuration and mechanical response information of the rock mechanical response monitoring equipment in the mine production process;
the abnormal data cleaning subsystem of the multi-element sensor is used for constructing a dynamic queue based on the data acquisition subsystem after information configuration of the monitoring equipment and rock mass mechanics response information data acquired in real time in the mine production process, realizing queue data classification according to a binary classification principle, judging the abnormal state of newly acquired monitoring data according to the classification result of elements at the tail of the queue and completing data cleaning of the multi-element sensor;
the sensor fault detection subsystem is used for dynamically updating the running state of the multi-element sensor based on the cleaning subsystem, and judging the fault state of the sensor in running by adopting a dynamic time window mode and a data threshold value or data acquisition time interval mode.
Further, the data acquisition subsystem includes:
the multivariate environment sensing equipment is used for sensing information such as rock mechanical response and the like in real time in the production process of a mine;
the system comprises a cloud database, a monitoring data storage position, a basic information table and a monitoring data storage position, wherein the cloud database is used for storing basic information and diversified monitoring data of a sensor in a persistent manner, and the basic information table of the sensor comprises a transmission protocol, a measuring point mapping code, a sensor name, a monitoring area, a responsible person, a sampling period, latest sampling time, abnormal sampling quantity and a monitoring data storage position;
the diversified monitoring data table mainly stores sensor numbers, data acquisition time, acquired data and data states;
the cloud service management system is used for visually configuring and persistently storing parameters such as a data transmission protocol, a measuring point mapping code, a sensor sampling period, a monitoring area and a person in charge of the sensor to a sensor basic information table, and storing sensing data of the multi-sensor to the multi-monitoring data table in real time on the basis, so that the cloud service of data transmission parameter configuration of the monitoring data to a cloud database in real time is realized.
Further: the multi-element sensor is configured through a data transmission protocol cloud end, and real-time transmission of multi-element data is achieved.
Further: the data acquisition subsystem after information configuration based on the monitoring equipment, rock mass mechanics response information data in the mine production process acquired in real time, dynamic queues are constructed, queue data classification is realized according to a binary classification principle, the abnormal state of the latest acquired monitoring data is judged according to a classification structure of queue tail elements, and the process of completing data cleaning of the multi-element sensor is as follows:
step 1.1, reading a measuring point mapping code and a real-time data storage position field in a sensor definition table, and constructing a sensor definition matrix;
step 1.2, traversing a sensor definition matrix, reading m pieces of normal data in a specified data table as initial data to form a dynamic queue matrix, wherein m is the length of a dynamic queue and can be specified independently;
step 1.3, starting data real-time acquisition, extracting a measuring point mapping code and sensing data in the data aiming at the latest acquired transmission data, finding an index aiming at the sensor in a sensor definition matrix by taking the mapping code as a retrieval condition, and pushing the sensing data into a dynamic queue matrix index queue;
step 1.4, find the maximum value and the minimum value in the dynamic queue matrix index queue, and generate two unequal randoms between the maximum value and the minimum valueCenter point k 1 ,k 2 As the initial center point, calculating the distance k between each element in the queue and the center point 1 ,k 2 The Manhattan distance of the queue element is divided into two types under the condition of the minimum Manhattan distance to obtain a classification result;
step 1.5, combining the classification results and generating a new central point k according to a midpoint calculation formula 1 ',k 2 ', when k 1 ',k 2 ' and k 1 ,k 2 When they are completely equal, step 1.6 is entered, otherwise k is assigned 1 ',k 2 ' values assigned to k 1 ,k 2 And here back to step 1.4;
step 1.6, calculating the number of elements of the category where the latest monitoring data are located according to the classification result, so as to judge the state of the latest acquired monitoring data, if the number of elements of the category where the tail elements of the queue are located is less than k, determining that the state is an abnormal state, removing the abnormal state from the queue matrix index queue, and otherwise, determining that the state is a normal state; dequeuing the queue head element of the queue matrix index queue, and enqueuing the monitoring data to realize dynamic update of a time window; and k is an abnormal state evaluation threshold of the monitored data, when k =1, the data is in a completely abnormal state, and the evaluation criterion of the abnormal data is lower as k is increased.
Further: the value of the abnormal data is between [1,4], and the optimal value of the abnormal data is 2.
Further: acquiring the current latest data acquisition time so as to update the latest sampling time data in the sensor definition table;
when the monitoring data are judged to be in an abnormal state, updating the abnormal sampling number in the sensor definition table to be the original number plus one, storing the monitoring data into a data table recorded in a queue matrix index queue, and setting a deleted field to be 1 to indicate that the monitoring data are in a logic deletion state;
and when the monitoring data is in a normal state, storing the data into a data table recorded in the queue matrix index queue, and setting the deleted field to be 0.
Further: the process of dynamically updating the running state of the multi-element sensor based on the cleaning subsystem adopts a dynamic time window form and a data threshold value or data acquisition time interval mode to realize the judgment of the fault state of the sensor in the running process as follows:
step 2.1, setting a dynamic time window, and dynamically reading data of a responsible person, a sampling period, latest sampling time and abnormal sampling quantity in a sensor definition table in a timing task mode;
and 2.2, calculating the interval between the last sampling time and the current time based on the latest sampling time and the sampling period data which are obtained dynamically and by combining the current time, and indicating that the sensor is in a broken line state when the interval exceeds n times of the sampling period. Wherein n is a user-specified parameter used to characterize the sensitivity of the sensor disconnection detection of the system;
and 2.3, judging whether the abnormal sampling quantity data is larger than an abnormal sampling threshold value k or not based on the abnormal sampling quantity data in the step 2.1, and if so, indicating that the data acquisition of the sensor is abnormal, wherein k is a user-defined parameter and is used for representing the sensitivity of abnormal data detection.
And 2.4, based on the judgment results of the step 2.2 and the step 2.3, when the sensor is detected to be in a broken line state or a data acquisition abnormal state, sending short messages and mail alarm information by combining the information of the responsible person obtained in the step 2.1, informing the responsible person of checking and maintaining the sensor, and updating the abnormal sampling number in the definition of the sensor to be 0.
The mine multi-sensor data acquisition, cleaning and fault discrimination system has the beneficial effects that:
1) According to the method, the relevant information of data transmission of the multi-element sensor is configured in a cloud visualization mode, one-click access and unified management of multi-element environment perception data can be realized, and a continuous data source is provided for cleaning of subsequent data;
2) The invention adopts a dynamic queue and a second classification mode, realizes the judgment of the abnormal state of the real-time sensing data, can realize the intelligent elimination of the abnormal data and the automatic update of the sensor state, ensures the reliability of the sensing data and provides a basis for the fault detection of the sensor;
3) The invention completes the construction of the dynamic time window by adopting the timing task, completes the detection of the fault state of the sensor, realizes the function of sending alarm information of the abnormal working state of the sensor to a responsible person by adopting a mail and short message mode, and furthest ensures the continuity of data acquisition of the sensor.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a diagram of a mine multivariate sensor data acquisition, cleaning and fault discrimination system architecture according to the present invention;
FIG. 2 is a flow chart of data cleansing according to an embodiment of the present invention;
FIG. 3 is an interface diagram of a data acquisition subsystem according to an embodiment of the present invention;
FIG. 4 is a background log graph of data real-time acquisition according to an embodiment of the present invention;
FIG. 5 is a graph comparing before and after data cleansing according to an embodiment of the present invention;
FIG. 6 is a diagram of a sensor fault alarm in accordance with an embodiment of the present invention;
FIG. 7 is a diagram illustrating raw data in a classification method used in the present invention;
FIG. 8 is a diagram illustrating the processing results of the binary classification method used in the present invention.
Detailed Description
In order to more clearly illustrate the technical scheme and the implementation example of the present invention, the algorithm in the embodiment or the prior art is introduced.
FIG. 1 is a diagram of a mine multivariate sensor data acquisition, cleaning and fault discrimination system architecture according to the present invention;
the utility model provides a mine multi-element sensor data acquisition, washing and discrimination system which characterized in that: the method comprises the following steps:
the data acquisition subsystem is used for carrying out real-time acquisition and storage on information cloud configuration and mechanical response information of the rock mechanical response monitoring equipment in the mine production process;
based on the data acquisition subsystem after monitoring equipment information configuration, real-time acquired rock mass mechanics response information data in the mine production process are used for constructing a dynamic queue, queue data classification is realized according to a binary classification principle, and an abnormal state of newly acquired monitoring data is judged according to a classification result of queue tail elements;
the data acquisition subsystem only acquires one piece of data at a time, and the judged data is the abnormal state of the piece of data, namely simply, the latest data is judged to be not in fault through the previous historical data, so that the data cleaning of the multi-element sensor is completed, and the abnormal data cleaning subsystem of the multi-element sensor is obtained;
the information of the monitoring equipment is configured on the above, the rock mechanics information can be sensed regularly, then a dynamic queue is constructed based on the real-time rock mechanics information, for example, in 12;
the sensor fault detection subsystem is used for dynamically updating the running state of the multi-element sensor based on the cleaning subsystem, and judging the fault state of the sensor in running by adopting a dynamic time window mode and a data threshold value or data acquisition time interval mode.
Based on the dynamic state of the cleaning subsystem, for example, at 12. Whether the operation state information of the monitoring equipment is abnormal or not, the operation state information of the monitoring equipment is updated, the last sampling time field is mainly updated, and the time of the latest acquired data is used for updating
Then the cleaning system continuously operates to continuously update the state of the multi-element sensor, and then the fault of the sensor is judged through the last sampling time and the acquired abnormal data volume of the multi-element sensor
The data acquisition subsystem includes: the multivariate environment sensing equipment is used for sensing information such as rock mass mechanical response and the like in real time in the mine production process; the multi-element environment sensing equipment comprises a multi-point displacement meter, an anchor rod dynamometer, a deep hole inclinometer and the like;
the system comprises a cloud database, a monitoring data storage unit, a data storage unit and a data processing unit, wherein the cloud database is used for persistently storing basic information of a sensor and diversified monitoring data, and a basic information table of the sensor comprises a transmission protocol, a measuring point mapping code, a sensor name, a monitoring area, a responsible person, a sampling period, latest sampling time, abnormal sampling quantity and a monitoring data storage position;
the diversified monitoring data table is mainly used for storing sensor numbers, data acquisition time, acquired data and data states;
the cloud service management system is used for visually configuring and persistently storing parameters of a sensor such as a data transmission protocol, a measuring point mapping code, a sensor sampling period, a monitoring area and a responsible person to a sensor basic information table, and storing sensing data of a multi-element sensor to the multi-element monitoring data table in real time on the basis so as to realize real-time cloud service configuration of the monitoring data to data transmission parameters of a cloud database.
The multi-element sensor realizes real-time transmission of multi-element data through data transmission protocol cloud configuration.
Further: the method comprises the following steps of constructing a dynamic queue based on acquired real-time multivariate data, realizing queue data classification according to a classification principle, judging data of abnormal states in latest acquired monitoring data according to a classification structure of queue tail elements, and completing the data cleaning process of a multivariate sensor as follows:
step 1.1, reading a measuring point mapping code and a real-time data storage position field in a sensor definition table when service is started for the first time, and constructing a sensor definition matrix SensorList;
step 1.2, traversing a sensor definition matrix SensorList, reading m pieces of normal data in a specified data table as initial data to form a dynamic queue matrix SensorQueue, wherein m is the length of a dynamic queue and can be specified independently;
step 1.3, starting data real-time acquisition, extracting a measuring point mapping code and sensing data in the data aiming at the latest acquired transmission data, using the measuring point mapping code and the sensing data as retrieval conditions, finding an index aiming at a sensor in a sensor definition matrix SensorList, and pushing the sensing data into a dynamic queue matrix index queue SensorQueue [ index ];
step 1.4, find dynamic queue matrix index queue SensorQueue [ index ]]Between the maximum and minimum values, two unequal random center points k are generated 1 ,k 2 As the initial center point, calculating the distance k between each element in the queue and the center point 1 ,k 2 The queue elements are divided into two types under the condition of the minimum Manhattan distance to obtain a classification result;
step 1.5, combining the classification result and generating a new central point k according to a central point calculation formula 1 ',k 2 ', when k 1 ',k 2 ' and k 1 ,k 2 When they are completely equal, go to step 1.6, otherwise k is 1 ',k 2 ' values assigned to k 1 ,k 2 And here back to step 1.4;
step 1.6, calculating the number of elements of the category where the latest monitoring data are located according to the classification result, so as to judge the state of the latest acquired monitoring data, if the number of elements of the category where the tail elements of the queue are located is less than k, the latest acquired monitoring data are in an abnormal state, and the latest acquired monitoring data are removed from a queue matrix index queue SensorList [ index ], otherwise, the latest acquired monitoring data are in a normal state; dequeuing the head element of a queue matrix sensorList [ index ], enqueuing the monitoring data, and realizing dynamic updating of a time window; and k is an abnormal state evaluation threshold of the monitored data, when k =1, the data is in a completely abnormal state, and the evaluation criterion of the abnormal data is lower as k is increased.
The value of the abnormal data is between [1,4], and the optimal value of the abnormal data is 2.
The method also comprises the steps of obtaining the current latest data acquisition time so as to update the latest sampling time data in the sensor definition table;
when the monitoring data are judged to be in an abnormal state, updating the abnormal sampling number in the sensor definition table to be the original number plus one, storing the monitoring data in a data table recorded in a queue matrix index queue SensorList [ index ], and setting a deleted field to be 1 to indicate that the monitoring data are in a logic deletion state;
when the monitoring data is judged to be in a normal state, the data is stored in a data table recorded in a queue matrix index queue SensorList [ index ], and a deleted field of the data table is set to be 0.
FIG. 2 is a flow chart of data cleansing according to an embodiment of the present invention;
further: the process of dynamically updating the running state of the multi-element sensor based on the cleaning subsystem adopts a dynamic time window form and a data threshold value or data acquisition time interval mode to realize the judgment of the fault state of the sensor in the running process as follows:
step 2.1, setting a dynamic time window, and dynamically reading data of a responsible person, a sampling period, latest sampling time and abnormal sampling quantity in a sensor definition table in a timing task mode;
and 2.2, calculating the interval between the last sampling time and the current time based on the dynamically acquired latest sampling time and sampling period data and combining the current time, and indicating that the sensor is in a broken line state when the interval exceeds n times of the sampling period. Wherein n is a user-specified parameter used to characterize the sensitivity of the sensor disconnection detection of the system;
and 2.3, judging whether the abnormal sampling quantity data is larger than an abnormal sampling threshold value k or not based on the abnormal sampling quantity data in the step 2.1, and if so, indicating that the data acquisition of the sensor is abnormal, wherein k is a user-defined parameter and is used for representing the sensitivity of abnormal data detection.
And 2.4, based on the judgment results of the step 2.2 and the step 2.3, when the sensor is detected to be in a broken line state or a data acquisition abnormal state, sending short messages and mail alarm information by combining the information of the responsible person obtained in the step 4.1, informing the responsible person of checking and maintaining the sensor, and updating the abnormal sampling number in the definition of the sensor to be 0.
Examples
In the embodiment, a gold civil engineering JTM-V7000 series multipoint displacement meter and a JTM-V1000B series vibrating wire type anchor rod dynamometer are used as environment sensing equipment, and sensor parameter cloud configuration is adopted, so that intelligent cleaning of sensing data in real time and sensing data and automatic detection of sensor faults are realized.
A mine multi-sensor data cleaning method and a fault distinguishing method realize one-key access of sensor monitoring data, intelligent cleaning of the monitoring data and automatic detection of sensor faults, and the specific implementation scheme is as follows:
step 1: and (3) building a data acquisition subsystem, wherein the data acquisition subsystem comprises a Jin Tumu JTM-V7000 series multipoint displacement meter and a JTM-V1000B series vibrating wire anchor dynamometer as environment sensing equipment, a cloud database and data transmission parameter configuration cloud service.
FIG. 3 is an interface diagram of a data acquisition subsystem according to an embodiment of the present invention;
specifically, the data transmission parameter configuration cloud service shown in fig. 3 is opened, a new or modified button is clicked, parameters such as a sensor measurement point name, a measurement point mapping code, a database table name, a resolving formula and a transmission protocol are input, and then a storage button is clicked, so that information can be stored in a sensor definition table, configuration of relevant indexes of sensor data transmission is realized, and cloud-up of monitoring data is completed;
FIG. 4 is a diagram of an example data real-time collection background log of the present invention;
fig. 4 shows a background response log of the cloud on the monitoring data after the cloud configuration is completed. The name of a measuring point of the sensor is commonly used for representing the monitoring position of the sensor, measuring point mapping codes, a resolving formula and a transmission protocol are determined after being communicated with a sensing equipment supplier, and the protocol is an MQTT protocol for data transmission in the embodiment; the database name is used for determining the storage position of the real-time acquired data, so that the data can be stored persistently after being clouded;
step 2: based on the dynamic queue and the classification principle, the intelligent judgment of the abnormal state of the real-time monitoring data is realized, as shown in fig. 3. The step 2 specifically comprises the following steps:
step 2.1, reading a measuring point mapping code and a real-time data storage position field in a sensor definition table when the service is started for the first time, and constructing a sensor definition matrix SensorList;
step 2.2, traversing the SensorList matrix, reading m pieces of normal data in a specified data table as initial data, and forming a dynamic queue matrix SensorQueue, wherein m is the length of a dynamic queue and can be specified independently;
step 2.3, after the program is finished, starting data real-time acquisition by means of the data transmission service configured in the step 1, extracting a measuring point mapping code and sensing data in the data aiming at the latest acquired transmission data, taking the measuring point mapping code and the sensing data as a retrieval condition, finding an index aiming at the sensor in a SensorList, and pushing the sensing data into a queue D while keeping the queue D = SensorQueue [ index ];
step 2.4, respectively obtaining two random and unequal queue center points k according to the formulas 1-2 1 ',k 2 ', and make k 1 ,k 2 Are respectively equal to k 1 ',k 2 '；
k 1 ′＝random(min(D),max(D)) (1)
k 2 ′＝random(min(D),max(D)) (2)
Step 2.5, respectively calculating element distance k in the queue D according to the formulas 3-4 1 ,k 2 Judging the category of each element in the queue according to the absolute value of the Manhattan distance of the central point and the formula 5;
dis 1 (i)＝|k 1 -D(i)| (3)
dis 2 (i)＝|k 2 -D(i)| (4)
category(i)＝(dis 1 (i)＜dis 2 (i))+2(dis 2 (i)＜dis 1 (i)) (5)
step 2.6, respectively calculating the element distance k in the queue D according to the formulas 3 to 4 1 ,k 2 The absolute value of the manhattan distance of the center point,judging the type of each element in the queue according to the formula 5;
dis 1 (i)＝|k 1 -D(i)| (3)
dis 2 (i)＝|k 2 -D(i)| (4)
category(i)＝(dis 1 (i)＜dis 2 (i))+2(dis 2 (i)＜dis 1 (i)) (5)
step 2.7, calculating new classification central point k after two classifications according to formulas 6-7 1 ',k 2 ', and judges k 1 ',k 2 Whether or not to k 1 ,k 2 If equal, execute step 2.8, otherwise let k 1 ,k 2 Are respectively equal to k 1 ',k 2 ', and performing step 2.5;
step 2.8, calculating an abnormal state evaluation variable noise of the latest sensing data according to the formula 8, wherein in the embodiment, a monitoring data abnormal state evaluation threshold value k =2 is taken, and when the noise is less than or equal to 2, the latest sensing data is abnormal outlier data, the latest sensing data is directly marked as a deleted state and is directly removed from the queue; when the noise is larger than 2, the latest sensing data is in a normal state, the latest sensing data is marked as a normal state, and the queue is enabled to normally execute dequeue operation;
and step 3: and (4) according to the judgment result of the abnormal state of the real-time sensing data in the step (2), performing persistent storage on the data. Specifically, the current data acquisition time is acquired, so that the latest sampling time data in the sensor definition table is updated; when the monitoring data is judged to be in an abnormal state, updating the abnormal sampling number in the sensor definition table to be the original number plus one, storing the monitoring data into a data table to which the data belongs, and setting a deleted field to be 1 to indicate that the monitoring data is in a logic deletion state; and when the monitoring data is in a normal state, storing the data into a data table to which the data belongs, and setting the deleted field to be 0. The final effect of the data cleaning method is shown in fig. 5, and fig. 5 is a comparison graph of the effect before and after data cleaning according to the embodiment of the invention;
and 4, step 4: and (3) analyzing and processing the dynamically updated sensor parameters in the step (3) by adopting a dynamic time window form, analyzing the running state of the sensor, and if the sensor which works abnormally is monitored, sending alarm information to a specified user in a form of short messages and mails. The step 4 specifically comprises the following steps:
step 4.1, setting a dynamic time window to be 1h, and dynamically reading data of a responsible person, a sampling period, latest sampling time and abnormal sampling quantity in a sensor definition table in a timing task mode;
step 4.2, based on the latest sampling time and sampling period data dynamically acquired in the step 4.1, and combining the current time, calculating the interval between the last sampling time and the current time, and when the interval exceeds 3 times of the sampling period, indicating that the sensor is in a broken line state;
4.3, judging whether the abnormal sampling quantity data is larger than the abnormal sampling threshold value 10 or not based on the abnormal sampling quantity data in the step 4.1, and if so, indicating that the data acquisition of the sensor is abnormal;
step 4.4, based on the calculation results of step 4.2 and step 4.3, when the sensor is detected to be in a broken line or abnormal data acquisition state, sending short message and mail alarm information by combining the information of the responsible person obtained in step 4.1 to inform the responsible person to check and maintain the sensor, wherein fig. 6 is a sensor fault alarm diagram of the embodiment of the invention; while updating the sensor definition with an exception sample number of 0.
FIG. 7 is a schematic diagram of raw data in a classification method used in the present invention;
FIG. 8 is a diagram illustrating the processing results of the binary classification method used in the present invention.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (7)
1. The utility model provides a mine multi-element sensor data acquisition, washing and discrimination system which characterized in that: the method comprises the following steps:
the data acquisition subsystem is used for carrying out real-time acquisition and storage on information cloud configuration and mechanical response information of the rock mechanical response monitoring equipment in the mine production process;
the abnormal data cleaning subsystem of the multi-element sensor is used for constructing a dynamic queue based on the data acquisition subsystem after monitoring equipment information is configured and rock mechanics response information data in the mine production process acquired in real time, realizing queue data classification according to a binary classification principle, judging the abnormal state of the latest acquired monitoring data according to the classification result of elements at the tail of the queue and finishing data cleaning of the multi-element sensor;
and the sensor fault detection subsystem is used for dynamically updating the running state of the multi-element sensor based on the cleaning subsystem, and judging the fault state of the sensor in running by adopting a dynamic time window form and a data threshold value or data acquisition time interval mode.
2. The mine multielement sensor data collection, cleaning and discrimination system as recited in claim 1, wherein: the data acquisition subsystem includes:
the multivariate environment sensing equipment is used for sensing information such as rock mass mechanical response and the like in real time in the mine production process;
the system comprises a cloud database, a monitoring data storage position, a basic information table and a monitoring data storage position, wherein the cloud database is used for storing basic information and diversified monitoring data of a sensor in a persistent manner, and the basic information table of the sensor comprises a transmission protocol, a measuring point mapping code, a sensor name, a monitoring area, a responsible person, a sampling period, latest sampling time, abnormal sampling quantity and a monitoring data storage position;
the diversified monitoring data table is mainly used for storing sensor numbers, data acquisition time, acquired data and data states;
the cloud service management system is used for visually configuring and persistently storing parameters of a sensor such as a data transmission protocol, a measuring point mapping code, a sensor sampling period, a monitoring area and a responsible person to a sensor basic information table, and storing sensing data of a multi-element sensor to the multi-element monitoring data table in real time on the basis so as to realize real-time cloud service configuration of the monitoring data to data transmission parameters of a cloud database.
3. The mine multi-sensor data acquisition, cleaning and discrimination system according to claim 1, characterized in that: the multi-element sensor is configured through a data transmission protocol cloud end, and real-time transmission of multi-element data is achieved.
4. The mine multielement sensor data collection, cleaning and discrimination system as recited in claim 1, wherein: based on the data acquisition subsystem after monitoring equipment information configuration, rock mechanics response information data in the mine production process of real-time collection constructs dynamic queues, realizes queue data classification according to the principle of two classifications, and according to the classification structure of queue tail elements, the abnormal state of the latest collected monitoring data is distinguished, and the process of completing the data cleaning of the multi-element sensor is as follows:
step 1.1, reading a measuring point mapping code and a real-time data storage position field in a sensor definition table, and constructing a sensor definition matrix;
step 1.2, traversing a sensor definition matrix, reading m pieces of normal data in a specified data table as initial data to form a dynamic queue matrix, wherein m is the length of a dynamic queue and can be specified independently;
step 1.3, starting data real-time acquisition, extracting a measuring point mapping code and sensing data in the data aiming at the latest acquired transmission data, finding an index aiming at the sensor in a sensor definition matrix by taking the mapping code as a retrieval condition, and pushing the sensing data into a dynamic queue matrix index queue;
step 1.4, finding the maximum value and the minimum value in the dynamic queue matrix index queue, and generating two unequal random central points k between the maximum value and the minimum value 1 ,k 2 As the initial center point, calculating the distance k between each element in the queue and the center point 1 ,k 2 The Manhattan distance of the queue element is divided into two types under the condition of the minimum Manhattan distance to obtain a classification result;
step 1.5, combining the classification results and generating a new central point k according to a midpoint calculation formula 1 ',k 2 ', when k 1 ',k 2 ' and k 1 ,k 2 When they are completely equal, go to step 1.6, otherwise k is 1 ',k 2 ' values assigned to k 1 ,k 2 And here back to step 1.4;
step 1.6, calculating the number of elements of the category where the latest monitoring data are located according to the classification result, so as to judge the state of the latest acquired monitoring data, if the number of elements of the category where the tail elements of the queue are located is less than k, determining that the state is an abnormal state, removing the abnormal state from the queue matrix index queue, and otherwise, determining that the state is a normal state; dequeuing the queue head element of the queue matrix index queue, and enqueuing the monitoring data to realize dynamic update of a time window; and k is an abnormal state evaluation threshold of the monitored data, when k =1, the data is in a completely abnormal state, and the evaluation criterion of the abnormal data is lower as k is increased.
5. The mine multielement sensor data collection, cleaning and discrimination system as recited in claim 1, wherein: the value of the abnormal data is between [1,4], and the optimal value of the abnormal data is 2.
6. The mine multielement sensor data collection, cleaning and discrimination system as recited in claim 2, wherein: acquiring the current latest data acquisition time so as to update the latest sampling time data in the sensor definition table;
when the monitoring data are judged to be in an abnormal state, updating the abnormal sampling number in the sensor definition table to be the original number plus one, storing the monitoring data into a data table recorded in a queue matrix index queue, and setting a deleted field to be 1 to indicate that the monitoring data are in a logic deletion state;
and when the monitoring data is in a normal state, storing the data into a data table recorded in the queue matrix index queue, and setting a deleted field of the data to be 0.
7. The mine multivariate sensor data acquisition, cleaning and discrimination system as set forth in claim 1, characterized in that: the process of dynamically updating the running state of the multi-element sensor based on the cleaning subsystem adopts a dynamic time window form and a data threshold value or data acquisition time interval mode to realize the judgment of the fault state of the sensor in the running process as follows:
step 2.1, setting a dynamic time window, and dynamically reading data of a responsible person, a sampling period, latest sampling time and abnormal sampling quantity in a sensor definition table in a timing task mode;
and 2.2, calculating the interval between the last sampling time and the current time based on the latest sampling time and the sampling period data which are obtained dynamically and by combining the current time, and indicating that the sensor is in a broken line state when the interval exceeds n times of the sampling period. Wherein n is a user-specified parameter used to characterize the sensitivity of the sensor disconnection detection of the system;
and 2.3, judging whether the abnormal sampling quantity data is larger than an abnormal sampling threshold value k or not based on the abnormal sampling quantity data in the step 2.1, and if so, indicating that the data acquisition of the sensor is abnormal, wherein k is a user-defined parameter and is used for representing the sensitivity of abnormal data detection.
And 2.4, based on the judgment results of the step 2.2 and the step 2.3, when the sensor is detected to be in a broken line state or in a data acquisition abnormal state, sending a short message and mail alarm information by combining the information of the responsible person acquired in the step 2.1, informing the responsible person to check and maintain the sensor, and updating the abnormal sampling number in the definition of the sensor to be 0.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211338736.0A CN115567562A (en) | 2022-10-28 | 2022-10-28 | Mine multi-sensor data acquisition, cleaning and fault discrimination system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211338736.0A CN115567562A (en) | 2022-10-28 | 2022-10-28 | Mine multi-sensor data acquisition, cleaning and fault discrimination system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115567562A true CN115567562A (en) | 2023-01-03 |
Family
ID=84768263
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211338736.0A Pending CN115567562A (en) | 2022-10-28 | 2022-10-28 | Mine multi-sensor data acquisition, cleaning and fault discrimination system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115567562A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116702030A (en) * | 2023-05-31 | 2023-09-05 | 浙江大学 | Blast furnace state monitoring method and device based on sensor reliability analysis |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103324691A (en) * | 2013-06-03 | 2013-09-25 | 河海大学 | Voice frequency searching method based on M-tree |
| CN103793599A (en) * | 2014-01-17 | 2014-05-14 | 浙江远图智控系统有限公司 | Travel anomaly detection method based on hidden Markov model |
| CN110366031A (en) * | 2019-07-24 | 2019-10-22 | 长春融成智能设备制造股份有限公司 | A kind of monitoring abnormal state and method for diagnosing faults digitizing workshop MES system view-based access control model |
| CN112925279A (en) * | 2021-01-30 | 2021-06-08 | 重庆简华科技有限公司 | Fault comprehensive analysis system based on MES system |
| CN113535445A (en) * | 2021-01-06 | 2021-10-22 | 腾讯科技（深圳）有限公司 | Anomaly detection method and device and computer readable storage medium |
-
2022
- 2022-10-28 CN CN202211338736.0A patent/CN115567562A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103324691A (en) * | 2013-06-03 | 2013-09-25 | 河海大学 | Voice frequency searching method based on M-tree |
| CN103793599A (en) * | 2014-01-17 | 2014-05-14 | 浙江远图智控系统有限公司 | Travel anomaly detection method based on hidden Markov model |
| CN110366031A (en) * | 2019-07-24 | 2019-10-22 | 长春融成智能设备制造股份有限公司 | A kind of monitoring abnormal state and method for diagnosing faults digitizing workshop MES system view-based access control model |
| CN113535445A (en) * | 2021-01-06 | 2021-10-22 | 腾讯科技（深圳）有限公司 | Anomaly detection method and device and computer readable storage medium |
| CN112925279A (en) * | 2021-01-30 | 2021-06-08 | 重庆简华科技有限公司 | Fault comprehensive analysis system based on MES system |
Non-Patent Citations (1)
| Title |
|---|
| 吕云翔等编著: "机器学习原理及应用", 31 August 2021, 机械工业出版社, pages: 97 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116702030A (en) * | 2023-05-31 | 2023-09-05 | 浙江大学 | Blast furnace state monitoring method and device based on sensor reliability analysis |
| CN116702030B (en) * | 2023-05-31 | 2024-01-23 | 浙江大学 | Blast furnace state monitoring method and device based on sensor reliability analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110458195B (en) | Abnormal data discrimination method based on multi-criterion fusion | |
| WO2020134032A1 (en) | Method for detecting abnormality of service system, and apparatus therefor | |
| CN115567562A (en) | Mine multi-sensor data acquisition, cleaning and fault discrimination system | |
| EP3301586A1 (en) | Pre-processor and diagnosis device | |
| CN110334140B (en) | Method and device for processing data reported by equipment and server | |
| CN114118224A (en) | Neural network-based system-wide remote measurement parameter anomaly detection system | |
| CN110738255A (en) | device state monitoring method based on clustering algorithm | |
| CN106971058A (en) | A kind of pumping station operation monitoring data abnormal point detecting method | |
| CN109308589B (en) | Power grid automation data quality monitoring method, storage medium, terminal equipment and system | |
| CN109063885A (en) | A kind of substation's exception metric data prediction technique | |
| CN115509797A (en) | Method, device, equipment and medium for determining fault category | |
| CN117034149A (en) | Fault processing strategy determining method and device, electronic equipment and storage medium | |
| CN115034094B (en) | Prediction method and system for operation state of metal processing machine tool | |
| CN115168154B (en) | Abnormal log detection method, device and equipment based on dynamic baseline | |
| CN114331688A (en) | Method and device for detecting batch operation state of bank counter system business | |
| CN113535458A (en) | Abnormal false alarm processing method and device, storage medium and terminal | |
| CN112907111A (en) | Intelligent monitoring data acquisition and analysis method based on Internet of things technology | |
| CN111740856A (en) | Network communication equipment alarm acquisition abnormity early warning method based on abnormity detection algorithm | |
| CN115829160B (en) | Time sequence abnormality prediction method, device, equipment and storage medium | |
| CN116861204B (en) | Intelligent manufacturing equipment data management system based on digital twinning | |
| CN112163618B (en) | Equipment fault detection method and detection system | |
| CN115952085B (en) | Middleware system supporting rapid test development | |
| CN117433591B (en) | Automatic equipment monitoring system based on multi-sensor data | |
| CN111314170B (en) | Feature fuzzy P2P protocol identification method based on connection statistical rule analysis | |
| CN117870779B (en) | Remote wireless intelligent monitoring method and system for automation equipment |
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 |