CN116295583A - Sensor abnormal data correction method, device, equipment and readable storage medium - Google Patents
Sensor abnormal data correction method, device, equipment and readable storage medium Download PDFInfo
- Publication number
- CN116295583A CN116295583A CN202310195638.4A CN202310195638A CN116295583A CN 116295583 A CN116295583 A CN 116295583A CN 202310195638 A CN202310195638 A CN 202310195638A CN 116295583 A CN116295583 A CN 116295583A
- Authority
- CN
- China
- Prior art keywords
- sensor
- detected
- data
- abnormal
- detection data
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/049—Temporal neural networks, e.g. delay elements, oscillating neurons or pulsed inputs
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D18/00—Testing or calibrating apparatus or arrangements provided for in groups G01D1/00 - G01D15/00
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Testing Or Calibration Of Command Recording Devices (AREA)
Abstract
The application discloses a sensor abnormal data correction method, device and equipment and a readable storage medium, and belongs to the field of Internet of things. The method comprises the steps of obtaining detection data acquired by a sensor to be detected; based on a generalized extreme student bias test, judging whether the detection data acquired by the sensor to be detected is abnormal data or not; if yes, outputting the correction value of the sensor to be detected based on a nonlinear autoregressive exogenous input neural network model, wherein the nonlinear autoregressive exogenous input neural network model outputs the correction value after being calculated by combining the historical detection data of the sensor to be detected and other sensors in the same sensor group with a calculation mode set by the model. Not only can the abnormal detection data be accurately judged, but also the correction value can be output.
Description
Technical Field
The application relates to the field of internet of things, in particular to a sensor abnormal data correction method, device and equipment and a readable storage medium.
Background
Along with the continuous development of modern technology, the Internet of things of the emerging industry is developed, the sensor is the basis of the Internet of things, the intelligent sensor is the industry of the basis of the Internet of things, and the technical level of the intelligent sensor is also a key factor affecting the popularization of the application of the Internet of things.
In a general internet of things system, a working environment of a sensor node is often severe, a networking structure is complex, and a sampling interval is often longer because the sensor node is used for guaranteeing long-time operation of equipment. The time period represented by each group of data is longer, the performance of the processor of the node is insufficient, complex data verification, receiving and transmitting confirmation and other works are difficult to realize, and high requirements are put on the reliability of the data.
However, once an abnormality occurs in a group of data, the abnormality may cause data errors or voids for a long time, which may cause difficulty in subsequent data processing and analysis, and there is currently no effective method for detecting data abnormality, so that the abnormal data does not affect the subsequent data processing and analysis.
The foregoing is merely provided to facilitate an understanding of the principles of the present application and is not admitted to be prior art.
Disclosure of Invention
The main purpose of the present application is to provide a method, a device and a readable storage medium for correcting abnormal data of a sensor, which aims to solve the technical problem that an effective method for detecting abnormal data processing is lacking at present, so that the abnormal data does not influence the subsequent data processing analysis.
In order to achieve the above object, the present application provides a sensor abnormal data correction method, including the steps of:
acquiring detection data acquired by a sensor to be detected;
based on a generalized extreme student bias test, judging whether the detection data acquired by the sensor to be detected is abnormal data or not;
if yes, outputting the correction value of the sensor to be detected based on a nonlinear autoregressive exogenous input neural network model, wherein the nonlinear autoregressive exogenous input neural network model outputs the correction value after being calculated by combining the historical detection data of the sensor to be detected and other sensors in the same sensor group with a calculation mode set by the model.
Optionally, the step of determining whether the detection data collected by the sensor to be detected is abnormal data based on the generalized extreme student bias test includes:
inputting detection data acquired by the to-be-detected sensor into a statistical model, wherein the statistical model is composed of historical detection data of the to-be-detected sensor;
judging whether the detection data acquired by the to-be-detected sensor follow the statistical model or not, wherein the detection data acquired by the to-be-detected sensor is abnormal data if the detection data does not follow the statistical model.
Optionally, the step of determining whether the detection data collected by the sensor to be detected is abnormal data based on the generalized extreme student bias test includes:
if not, updating the statistical model and the nonlinear autoregressive exogenous input neural network model based on the detection data acquired by the sensor to be detected.
Optionally, if so, outputting the correction value of the sensor to be detected based on a nonlinear autoregressive exogenous input neural network model, where the step of outputting the correction value after the nonlinear autoregressive exogenous input neural network model calculates by combining the historical detection data of the sensor to be detected and other sensors in the same sensor group with a calculation mode set by the model includes:
counting the abnormal times of the sensor to be detected;
judging whether the abnormal times reach a first preset times or not;
if the abnormal times reach the first preset times, feeding back the abnormal information of the sensor to be detected to an operation page.
Optionally, the step of acquiring detection data acquired by the sensor to be detected includes:
receiving an environmental parameter data packet acquired by a sensor group;
and analyzing the environmental parameter data packet to obtain detection data of the actual sensor.
Optionally, the step of parsing the environmental parameter data packet to obtain detection data of the actual sensor includes:
detecting whether the actual sensor contains detection data of all preset sensors;
if not, marking the missing sensor, and recording the missing times and adding one.
Optionally, if not, marking the missing sensor, and recording the number of missing times plus one, which comprises:
judging whether the missing times reach a second preset times or not;
if yes, feeding back the missing sensor to the operation page.
In addition, in order to achieve the above object, the present application further provides a sensor abnormal data correction device, the device including:
the acquisition module is used for acquiring detection data acquired by the sensor to be detected;
the judging module is used for judging whether the detection data acquired by the sensor to be detected is abnormal data or not based on a generalized extreme student bias test;
and the output module is used for outputting the correction value of the sensor to be detected based on a nonlinear autoregressive exogenous input neural network model if the sensor to be detected is detected, wherein the nonlinear autoregressive exogenous input neural network model outputs the correction value after being calculated by combining the historical detection data of the sensor to be detected and other sensors in the same sensor group with a calculation mode set by the model.
In addition, to achieve the above object, the present application further provides a sensor abnormal data correction apparatus, the apparatus including: the sensor abnormality data correction device comprises a memory, a processor and a sensor abnormality data correction program stored on the memory and capable of running on the processor, wherein the sensor abnormality data correction program is configured to realize the steps of the sensor abnormality data correction method.
In addition, in order to achieve the above object, the present application further provides a readable storage medium having stored thereon a sensor abnormal data correction program that, when executed by a processor, implements the steps of the above-described sensor abnormal data correction method.
In order to solve the technical problem of data error detection of the sensors, the method and the device classify a plurality of different environment sensors into a group, acquire detection data acquired by the sensors to be detected, judge whether the detection data acquired by the sensors to be detected are abnormal data based on a generalized extreme chemical biochemical deviation test, if so, input a neural network model based on a nonlinear autoregressive external source, and output a correction value of the sensors to be detected. Not only can the abnormal detection data be accurately judged, but also the correction value of the sensor to be detected can be output.
Drawings
FIG. 1 is a schematic diagram of a sensor abnormal data correction device of a hardware operating environment according to an embodiment of the present application;
FIG. 2 is a flowchart of a first embodiment of a sensor abnormal data correction method according to the present application;
FIG. 3 is a schematic diagram of a first embodiment of an abnormal sensor data correction apparatus according to the present application;
fig. 4 is a schematic functional block diagram of a first embodiment of the sensor abnormal data correction device of the present application.
The realization, functional characteristics and advantages of the present application will be further described with reference to the embodiments, referring to the attached drawings.
Detailed Description
It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
Referring to fig. 1, fig. 1 is a schematic structural diagram of a sensor abnormal data correction device of a hardware running environment according to an embodiment of the present application.
As shown in fig. 1, the sensor abnormal data correction apparatus may include: a processor 1001, such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, a memory 1005. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a WIreless interface (e.g., a WIreless-FIdelity (WI-FI) interface). The Memory 1005 may be a high-speed random access Memory (Random Access Memory, RAM) Memory or a stable nonvolatile Memory (NVM), such as a disk Memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.
Those skilled in the art will appreciate that the configuration shown in fig. 1 does not constitute a limitation of the sensor abnormality data correction device, and may include more or fewer components than shown, or certain components may be combined, or a different arrangement of components.
As shown in fig. 1, an operating system, a data storage module, a network communication module, a user interface module, and a sensor abnormal data correction program may be included in the memory 1005 as one type of storage medium.
In the sensor abnormal data correction device shown in fig. 1, the network interface 1004 is mainly used for data communication with other devices; the user interface 1003 is mainly used for data interaction with a user; the processor 1001 and the memory 1005 in the sensor abnormal data correction apparatus of the present application may be provided in the sensor abnormal data correction apparatus, and the sensor abnormal data correction apparatus calls a sensor abnormal data correction program stored in the memory 1005 through the processor 1001 and executes the sensor abnormal data correction method provided in the embodiment of the present application.
An embodiment of the present application provides a method for correcting abnormal sensor data, and referring to fig. 2, fig. 2 is a schematic flow chart of a first embodiment of the method for correcting abnormal sensor data.
In this embodiment, the method for correcting abnormal sensor data includes:
step S10: acquiring detection data acquired by a sensor to be detected;
it should be understood that, as an emerging industry, the internet of things is an important ring for integrating the physical world and the information world by enabling an article to speak and release information through a perception recognition technology, and is the most unique part of the internet of things, which is different from other networks. The tentacle of the Internet of things is a large amount of information generating equipment positioned on a perception and identification layer, and comprises RFID, a sensor network, a positioning system and the like. The data perceived by the sensor network is one of the important sources of mass information of the Internet of things.
The sensor network is a distributed intelligent network system which is formed by a large number of tiny sensor nodes which are deployed in an action area and have wireless communication and calculation capabilities and can autonomously complete specified tasks according to environments in an ad hoc mode. The distance between nodes of the sensing network is very short, and a multi-hop (multi-hop) wireless communication mode is generally adopted for communication. The sensor network can operate in an independent environment, and can also be connected to the Internet through a gateway, so that a user can access the sensor network remotely.
The sensor network integrates sensor technology, embedded computing technology, modern network and wireless communication technology, distributed information processing technology and the like, can cooperatively monitor, sense and collect information of various environments or monitored objects in real time through various integrated micro sensors, processes the information through an embedded system, and transmits the sensed information to a user terminal in a multi-hop relay mode through a random self-organizing wireless communication network. Thus truly realizing the concept of 'ubiquitous computing'.
The sensor network node comprises four basic units, namely a sensing unit (comprising a sensor and an analog-to-digital conversion functional module), a processing unit (comprising an embedded system, including a CPU, a memory, an embedded operating system and the like), a communication unit (comprising a wireless communication module) and a power supply part. Further, other functional units that may be selected include positioning systems, motion systems, power generation devices, and the like.
In a sensor network, nodes are deployed in large numbers inside or near a perceived object in various ways. The nodes form a wireless network in a self-organizing mode, sense, collect and process specific information in a network coverage area in a cooperative mode, and can collect, process and analyze information at any place and at any time. A typical sensor network architecture includes distributed sensor nodes (clusters), sink nodes, the internet, user interfaces, and the like.
The sensing nodes can communicate with each other, are self-organized to form a network and are connected to a Sink (base station node) in a multi-hop mode, and after receiving data, the Sink node completes connection with a public Internet network through a Gateway (Gateway). The whole system manages and controls this system through a task manager. The characteristics of the sensor network lead the sensor network to have very wide application prospect, and the ubiquitous characteristics lead the sensor network to be an indispensable part of our lives in the near future.
However, in a general internet of things system, since the sensor node is to ensure that the device operates for a long time, the sampling interval is often long. The time period represented by each group of data is longer, and because the working environment of the sensor node in the Internet of things system is often bad, the networking structure is complex, the performance of the processor of the node is insufficient, the complex data verification, the receiving and transmitting confirmation and other works are difficult to realize, the probability of data abnormality is increased, and the higher requirements are put forward on the reliability of the data. And once an abnormality occurs in a group of data, the abnormality can cause data errors or blanks in a longer time, which causes difficulties in subsequent data processing and analysis.
For this reason, in the present embodiment, referring to fig. 3, a plurality of sensor nodes are arranged in a relatively closed and stable space, and the sensors are grouped into one group. After the sensor parameters are uploaded to the background server, the background analyzes the sensors in the same group to generate various parameters of each sensor and the relation (algorithm) between the various parameters and other sensors. And updates the algorithms as the data is continuously uploaded. The algorithm is a generalized extreme chemical biochemical deviation test algorithm and a nonlinear autoregressive exogenous input neural network model algorithm, and is operated on a server to judge detection data received from a sensor node, if the detection data is found to be abnormal data, the data is marked and corrected, and the original data and the corrected data are simultaneously stored. For further analysis and processing.
Specifically, a plurality of sensors are included in the sensor group, and detection data such as air temperature, air humidity, carbon dioxide concentration, etc. of the plurality of sensors are collected. The detection data of the plurality of sensors are packed into one data packet, i.e., an environmental parameter data packet in the present embodiment. The server typically uses a network as a medium, and can provide services to the outside through an intranet or the internet. The server has the greatest characteristics of strong operation capability or a computer with a large amount of disk storage space, so that the server can complete a large amount of operation work and a large amount of data storage in a short time. If the server receives the environmental parameter data packet collected by the sensor group, the environmental parameter data packet is analyzed to obtain detection data of various sensors.
Step S20: based on a generalized extreme student bias test, judging whether the detection data acquired by the sensor to be detected is abnormal data or not;
further, the step S20 includes:
step S21: inputting detection data acquired by the to-be-detected sensor into a statistical model, wherein the statistical model is composed of historical detection data of the to-be-detected sensor;
step S22: judging whether the detection data acquired by the to-be-detected sensor follow the statistical model or not, wherein the detection data acquired by the to-be-detected sensor is abnormal data if the detection data does not follow the statistical model.
In particular, a variety of techniques based on different schemes or methodologies may be used to identify anomalies. For example, a method of graphics (box graph, scatter graph); distance-based schemes (nearest neighbor algorithm, clustering algorithm); statistical methods (GESD, quartile based techniques) and the like. Each solution has its advantages and disadvantages, the effect of which depends on the actual use case. In this embodiment, a generalized extreme student bias test (generalized extreme studentized deviate test), hereinafter referred to as GESD, is used to determine whether the data is anomalous. GESD is a simple statistical method for detecting one or more outliers in a univariate dataset that follow an approximately normal distribution. The GESD method assumes that normal data follows some statistical model (or distribution), while data that does not follow the model (or distribution) is outliers. The discriminant formula may be written as a function,
Y(t)＝f1(y(t),y(t–1),...,y(t–d))
y is a binary value, and represents whether the latest data Y (t) is abnormal data, and Y (t-d) is Y-type data before the d moment.
The GESD detects whether the latest input data is abnormal data or not according to the historical data, wherein when the historical data does not exist in the first operation of the algorithm, preset default data are called to generate a statistical model corresponding to the sensor to be detected. If the data is normal, the latest data is transmitted to the algorithm updating module for updating the historical data in the statistical model. And outputting the data, and if the data is judged to be abnormal, transmitting the data to a data correction module.
Step S30: if yes, outputting the correction value of the sensor to be detected based on a nonlinear autoregressive exogenous input neural network model, wherein the nonlinear autoregressive exogenous input neural network model outputs the correction value after being calculated by combining the historical detection data of the sensor to be detected and other sensors in the same sensor group with a calculation mode set by the model.
Specifically, in this embodiment, the correction algorithm included in the correction module is a nonlinear autoregressive exogenous input neural network model algorithm (nonlinear autoregressive neural network with external input, referred to as NARX for short). NARX is a model used to describe nonlinear discrete systems. The NARX neural network structure comprises an input layer, an hidden layer and an output layer. The number of input layer nodes is set according to the number of input values, and the number of output layer nodes is set according to the number of predicted values. And constructing the NARX neural network by reasonably setting the hidden layer number and the node number. The NARX neural network is similar to the training method of the back propagation neural network. The NARX neural network is characterized by comprising: the NARX neural network adds a delay and feedback mechanism, so that the memory capacity of historical data is enhanced, and the NARX neural network is a dynamic neural network. And NARX is suitable for time-series prediction and is applied to solve nonlinear-series prediction problems in various fields.
In this embodiment, because there is an objectively related correlation between the sensors collected between the sensors monitoring different environmental parameters in the same environment, a nonlinear autoregressive exogenous input neural network model is set according to the sensors in the specifically set sensor group, and based on the historical detection data of all the sensors. Really achieves the aim of adapting to local conditions and can effectively improve the accuracy of detecting and judging abnormal data.
The formula of the NARX model can be written as a function:
y(t)＝f2(y(t–1),...,y(t–d),x(t–1),z(t–1)...,z(t–d))
wherein y (t) is the output value of the algorithm, namely the correction value of the current abnormal data. y (t-d) is the value of the y sensor before the d moment, and x and z are the values of the relevant sensors in the same group. And outputting the corrected data and then entering a data storage module.
Further, the step S30 further includes;
step S31: counting the abnormal times of the sensor to be detected;
specifically, when a certain sensor frequently or continuously uploads abnormal data, there must be a certain fault that is not checked. For this reason, each sensor determined to be corresponding to the abnormal data is recorded.
Step S32: judging whether the abnormal times reach a first preset times or not;
specifically, the first preset number of times is used as a threshold value for determining whether the sensor is continuously or frequently abnormal in the present embodiment. Judging whether the counted abnormal times of a certain sensor reach the first preset times or not by judging whether the abnormal times of the certain sensor reach the first preset times or not.
Step S33: the sensor abnormality information is fed back to the operation page.
Specifically, if it is detected that the counted abnormal times of a certain sensor have reached the first preset times, it is indicated that the sensor may have a fault that is difficult to remove, and abnormal state information of the sensor is fed back to an operation page for an operation user to check, so that measures related to the fault removal are taken.
In this embodiment, in order to solve the technical problem of data error of sensor detection, the present application classifies a plurality of different environmental sensors into a group, obtains detection data of a sensor to be detected, determines whether the detection data of the sensor to be detected is abnormal data based on a generalized extreme chemical biochemical deviation test, if yes, inputs a neural network model based on a nonlinear autoregressive external source, and outputs a correction value of the sensor to be detected. Not only can the abnormal detection data be accurately judged, but also the correction value can be output. The sensor node can operate in a common sensor node, and can be widely applied to the fields of coal mine gas monitoring, coalbed methane industry, geological disaster monitoring, environmental pollution monitoring, safety production monitoring and the like.
Based on the foregoing embodiment, a second embodiment of the present application is provided, further, the step S20 includes:
step S24: if not, updating the statistical model and the nonlinear autoregressive exogenous input neural network model based on the detection data acquired by the sensor to be detected.
Specifically, if the detected data is not abnormal data, i.e. normal data, the data is stored in the storage module and enters the algorithm updating module at the same time, and related data such as x, Y, z and the like in the data judging module function Y (t) and the data correcting module function Y (t) are updated. The memory module in this embodiment stores data including time, related data (including normal data and corrected data), original data, and the like.
Therefore, the embodiment constructs a logic for continuously updating the algorithm model according to the detected data and continuously improving the detection and correction accuracy.
In this embodiment, if the detected data is normal data, the detected data is used as a statistical model and historical data in a nonlinear autoregressive exogenous input neural network model, so as to update an algorithm. And further, the accuracy of detecting and correcting the abnormal data is continuously improved.
Based on the foregoing embodiments, a third embodiment of the present application is provided, further, before step S10, including:
step S00: receiving an environmental parameter data packet acquired by a sensor group;
step S01: and analyzing the environmental parameter data packet to obtain detection data of the actual sensor.
Specifically, a plurality of sensors are included in the sensor group, and detection data such as air temperature, air humidity, carbon dioxide concentration, and the like of the plurality of sensors are collected. The detection data of the plurality of sensors are packed into one data packet, i.e., an environmental parameter data packet in the present embodiment. The server typically uses a network as a medium, and can provide services to the outside through an intranet or the internet. The server has the greatest characteristics of strong operation capability or a computer with a large amount of disk storage space, so that the server can complete a large amount of operation work and a large amount of data storage in a short time. If the server receives the environmental parameter data packet collected by the sensor group, the environmental parameter data packet is analyzed to obtain the detection data of the sensor in practice.
Further, the step S01 includes:
step S02: detecting whether the actual sensor comprises all preset sensors or not;
specifically, the server knows the sensors arranged in the sensor group, and after analyzing the environmental parameter data packet to obtain multiple types of detection data, the server detects whether the sensor corresponds to the preset sensor one by one or not, and whether the missing sensor does not upload data or not.
Step S03: if not, marking the missing sensor, and recording the missing times and adding one.
Specifically, if the sensor corresponding to the detection data included in the environmental parameter data packet is not in one-to-one correspondence with the preset sensor, it indicates that the sensor does not upload the data successfully, the sensor which is not corresponding to the sensor is marked as a missing sensor, and the marking times are recorded, and the times are increased by one.
Further, the step S03 includes:
step S04: judging whether the missing times reach a second preset times or not;
specifically, it is determined whether the number of times the sensor is missing, labeled as "missing sensor", reaches a second preset number of times, which is to verify that the sensor has failed such that the ability to collect and upload data is not available at all.
Step S05: if yes, feeding back the missing sensor to the operation page.
Specifically, if the number of times the sensor is missing, labeled as "missing sensor", reaches a second preset number of times, it is indicated that the sensor may have a difficult to troubleshoot fault, e.g., loss, damage, energy shortage, etc. And feeding the missing state information of the sensor back to an operation page for the operation user to check, so as to take the measures of relevant discharge faults.
In this embodiment, by counting the sensors corresponding to the actually received detection data, it is possible to determine whether or not the detection data of a certain sensor is missing, and to timely find an abnormality, and by calculating the sensor where the abnormal data exceeds a certain limit, it is possible to timely find that the sensor has a certain fault.
In addition, the embodiment of the application further provides a sensor abnormal data correction device, referring to fig. 4, fig. 4 is a schematic functional block diagram of a first embodiment of the sensor abnormal data correction device of the application. The sensor abnormal data correction device includes:
the acquiring module 10 is configured to parse the environmental parameter data packet to obtain detection data of an actual sensor;
the judging module 20 is configured to judge whether the detection data collected by the sensor to be detected is abnormal data based on a generalized extreme student bias test;
and the output module 30 is used for outputting the correction value of the sensor to be detected based on the nonlinear autoregressive exogenous input neural network model if yes.
The specific embodiment executed by the sensor abnormal data correction device is basically the same as each embodiment of the sensor abnormal data correction method, and is not described herein.
In addition, the embodiment of the application also provides a readable storage medium.
The readable storage medium of the present application stores thereon a sensor abnormal data correction program which, when executed by a processor, implements the steps of the sensor abnormal data correction method described above.
The specific embodiment of the sensor abnormal data correction program stored in the readable storage medium is basically the same as each embodiment of the sensor abnormal data correction method described above, and is not described herein.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one sensor anomaly data correction" does not exclude the presence of additional identical elements in a process, method, article, or system that comprises the element.
The foregoing embodiment numbers of the present application are merely for describing, and do not represent advantages or disadvantages of the embodiments.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) as described above, including several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method described in the embodiments of the present application.
The foregoing description is only of the preferred embodiments of the present application, and is not intended to limit the scope of the claims, and all equivalent structures or equivalent processes using the descriptions and drawings of the present application, or direct or indirect application in other related technical fields are included in the scope of the claims of the present application.
Claims (10)
1. The sensor abnormal data correction method is characterized in that the sensor abnormal data correction method is applied to a sensor group, wherein the sensor group comprises at least two sensors for measuring environmental parameters, and the sensor group comprises the following steps of:
acquiring detection data acquired by a sensor to be detected;
based on a generalized extreme student bias test, judging whether the detection data acquired by the sensor to be detected is abnormal data or not;
if yes, outputting the correction value of the sensor to be detected based on a nonlinear autoregressive exogenous input neural network model, wherein the nonlinear autoregressive exogenous input neural network model outputs the correction value after being calculated by combining the historical detection data of the sensor to be detected and other sensors in the same sensor group with a calculation mode set by the model.
2. The sensor abnormal data correction method according to claim 1, wherein the step of judging whether the detected data collected by the sensor to be detected is abnormal data based on a generalized extreme student bias test comprises:
inputting detection data acquired by the to-be-detected sensor into a statistical model, wherein the statistical model is composed of historical detection data of the to-be-detected sensor;
judging whether the detection data acquired by the to-be-detected sensor follow the statistical model or not, wherein the detection data acquired by the to-be-detected sensor is abnormal data if the detection data does not follow the statistical model.
3. The sensor abnormal data correction method according to claim 1, wherein the step of determining whether the detected data collected by the sensor to be detected is abnormal data based on a generalized extreme student bias test comprises:
if not, updating the statistical model and the nonlinear autoregressive exogenous input neural network model based on the detection data acquired by the sensor to be detected.
4. The method for correcting abnormal sensor data according to claim 1, wherein if yes, outputting the correction value of the sensor to be detected based on a nonlinear autoregressive exogenous input neural network model, and the step of calculating the output correction value by the nonlinear autoregressive exogenous input neural network model by combining the historical detection data of the sensor to be detected and other sensors in the same sensor group with a calculation mode set by the model includes:
counting the abnormal times of the sensor to be detected;
judging whether the abnormal times reach a first preset times or not;
if the abnormal times reach the first preset times, feeding back the abnormal information of the sensor to be detected to an operation page.
5. The sensor abnormality data correction method according to claim 1, characterized in that the step of acquiring the detection data acquired by the sensor to be detected includes, before:
receiving an environmental parameter data packet acquired by a sensor group;
and analyzing the environmental parameter data packet to obtain detection data of the actual sensor.
6. The sensor anomaly data correction method of claim 5, wherein the step of parsing the environmental parameter data packet to obtain the detection data of the actual sensor comprises:
detecting whether the actual sensor comprises all preset sensors or not;
if not, marking the missing sensor, and recording the missing times and adding one.
7. The method for correcting abnormal sensor data according to claim 1, wherein the step of marking the missing sensor and recording the number of deletions plus one if not comprises:
judging whether the missing times reach a second preset times or not;
if yes, feeding back the missing sensor to the operation page.
8. A sensor abnormality data correction device, characterized by comprising:
the acquisition module is used for acquiring detection data acquired by the sensor to be detected;
the judging module is used for judging whether the detection data acquired by the sensor to be detected is abnormal data or not based on a generalized extreme student bias test;
and the output module is used for outputting the correction value of the sensor to be detected based on a nonlinear autoregressive exogenous input neural network model if the sensor to be detected is detected, wherein the nonlinear autoregressive exogenous input neural network model outputs the correction value after being calculated by combining the historical detection data of the sensor to be detected and other sensors in the same sensor group with a calculation mode set by the model.
9. A sensor abnormal data correction apparatus, characterized by comprising: a memory, a processor, and a sensor abnormality data correction program stored on the memory and executable on the processor, the sensor abnormality data correction program configured to implement the steps of the sensor abnormality data correction method according to any one of claims 1 to 7.
10. A readable storage medium, wherein a sensor abnormal data correction program is stored on the readable storage medium, and the sensor abnormal data correction program, when executed by a processor, implements the steps of the sensor abnormal data correction method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310195638.4A CN116295583A (en) | 2023-02-21 | 2023-02-21 | Sensor abnormal data correction method, device, equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310195638.4A CN116295583A (en) | 2023-02-21 | 2023-02-21 | Sensor abnormal data correction method, device, equipment and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116295583A true CN116295583A (en) | 2023-06-23 |
Family
ID=86788181
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310195638.4A Pending CN116295583A (en) | 2023-02-21 | 2023-02-21 | Sensor abnormal data correction method, device, equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116295583A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117129236A (en) * | 2023-09-11 | 2023-11-28 | 深邦智能科技集团(青岛)有限公司 | Remote control-based motor vehicle equipment calibration detection method and system |
| CN118311379A (en) * | 2024-06-11 | 2024-07-09 | 深圳聚创致远科技有限公司 | High-voltage power cable online monitoring fault positioning method and system |
-
2023
- 2023-02-21 CN CN202310195638.4A patent/CN116295583A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117129236A (en) * | 2023-09-11 | 2023-11-28 | 深邦智能科技集团(青岛)有限公司 | Remote control-based motor vehicle equipment calibration detection method and system |
| CN117129236B (en) * | 2023-09-11 | 2024-03-26 | 深邦智能科技集团(青岛)有限公司 | Remote control-based motor vehicle equipment calibration detection method and system |
| CN118311379A (en) * | 2024-06-11 | 2024-07-09 | 深圳聚创致远科技有限公司 | High-voltage power cable online monitoring fault positioning method and system |
| CN118311379B (en) * | 2024-06-11 | 2024-08-23 | 深圳聚创致远科技有限公司 | High-voltage power cable online monitoring fault positioning method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116295583A (en) | Sensor abnormal data correction method, device, equipment and readable storage medium | |
| CN112202736B (en) | Communication network anomaly classification method based on statistical learning and deep learning | |
| Miao et al. | Agnostic diagnosis: Discovering silent failures in wireless sensor networks | |
| US11269752B1 (en) | System and method for unsupervised anomaly prediction | |
| CN111047082A (en) | Early warning method and device for equipment, storage medium and electronic device | |
| EP2680636A1 (en) | Reporting method, detection method, device and system for network state | |
| US20200272923A1 (en) | Identifying locations and causes of network faults | |
| CN112611936A (en) | Distribution network transformer fault dynamic detection and classification system based on edge calculation | |
| US11544557B2 (en) | IoT-based network architecture for detecting faults using vibration measurement data | |
| Gao et al. | On threshold-free error detection for industrial wireless sensor networks | |
| CN118075791A (en) | Wireless communication network fault prediction method and system | |
| KR102296568B1 (en) | Swithchboard Management And Control System based on Machine Learning | |
| KR20230032286A (en) | System for detecting abnormal value with periodicity using time-series data | |
| CN117977536A (en) | Smart power grid load identification method | |
| Niggemann et al. | A learning anomaly detection algorithm for hybrid manufacturing systems | |
| Nardelli et al. | Framework for the identification of rare events via machine learning and iot networks | |
| Syed et al. | A novel adaptive neuro-fuzzy inference system-differential evolution (Anfis-DE) assisted software fault-tolerance methodology in wireless sensor network (WSN) | |
| Chu et al. | The research of Wireless Edge Computing Gateway with Anomaly Detection | |
| Ahmadi Livani et al. | Outlier detection in wireless sensor networks using distributed principal component analysis | |
| Farruggia et al. | Detecting faulty wireless sensor nodes through stochastic classification | |
| Soller et al. | Predicting machine errors based on adaptive sensor data drifts in a real world industrial setup | |
| Scheffel et al. | WSN data confidence attribution using predictors | |
| US20200351170A1 (en) | Method for Detecting and Determining a Failure Probability of a Radio Network and Central Computer | |
| Neuhaus et al. | Autoencoder-based characterisation of passive IEEE 802.11 link level measurements | |
| Livani et al. | An energy-efficient anomaly detection approach for wireless sensor networks |
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 |