CN109580134B - Industrial environment liquid leakage detection method and device based on double RFID tags - Google Patents
Industrial environment liquid leakage detection method and device based on double RFID tags Download PDFInfo
- Publication number
- CN109580134B CN109580134B CN201811391257.9A CN201811391257A CN109580134B CN 109580134 B CN109580134 B CN 109580134B CN 201811391257 A CN201811391257 A CN 201811391257A CN 109580134 B CN109580134 B CN 109580134B
- Authority
- CN
- China
- Prior art keywords
- window
- signal
- rssi
- leakage
- phase
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M3/00—Investigating fluid-tightness of structures
- G01M3/40—Investigating fluid-tightness of structures by using electric means, e.g. by observing electric discharges
Abstract
The embodiment of the invention provides an industrial environment leakage detection method and device based on double RFID tags, and the method comprises the following steps: preprocessing the acquired tag signal to obtain a signal window, wherein the tag signal is related to the mutual inductive coupling effect between the two RFID tags; acquiring classification features according to each signal window, identifying the state of each signal window according to a pre-trained decision tree based on the classification features, and determining a first preset number of continuous signal windows detected as liquid leakage state windows, wherein the classification features are used for classifying the state of each signal window into normal, interference or liquid leakage; and extracting a signal section corresponding to the leakage state window, and obtaining a starting point and an ending point of leakage according to the signal section. According to the industrial environment leakage detection method and device based on the double RFID tags, provided by the embodiment of the invention, the purpose of leakage detection is achieved by analyzing the signal characteristics of two adjacent RFID tags, and the detection process is timely and accurate and is not easily influenced by the external environment.
Description
Technical Field
The invention relates to the field of industrial detection, in particular to a method and a device for detecting industrial environment liquid leakage based on double RFID tags.
Background
In modern industry, auxiliary machinery for water cooling, lubricant oil circulation and liquid purification is indispensable. Auxiliary machines often require flanges to connect the pump and the tank, and the flanges are often not welded for easy maintenance and replacement, which may lead to leakage problems. Leakage is a very frequent occurrence and can be a safety hazard (e.g., fire hazard from combustible leakage) and cause numerous economic losses (e.g., machine damage from cooling water or lubricant leakage). Therefore, leakage detection is an extremely important task in modern industrial scenarios.
Conventional weep detection still relies on a large number of high delays and inaccurate manual detections. Many efforts have been made to achieve automatic leakage detection, but existing solutions are still not suitable for complex large-scale industrial scenarios. For example, vision-based methods cannot work in dark environments or in non-line-of-sight (NLOS) scenarios common in industrial environments. The use of specialized sensors for detection appears to be an effective alternative, but most sensors are costly to deploy and maintain, consume high power, and are difficult to deploy on machines.
Studies have shown that the presence of liquid may alter the properties of the Radio Frequency (RF) signal, which may be an indicator of liquid leakage. However, unlike existing work, the problem of leak detection in an industrial environment presents the following key requirements and challenges:
and (5) the requirement of timeliness. To reduce damage from weeping, the weeping event must be detected in a timely manner, however, small amounts of liquid may have too little effect on the signal to be detectable by commercial RFID devices. Therefore, the existing liquid identification method based on the obvious signal change caused by more than 100ml of liquid is not applicable to the leakage detection scene.
High accuracy requirements. The goal is to accurately detect a weep event occurring at a potential weep point (e.g., a flange). However, many other confounding events (such as changes in indoor humidity, or the dynamics of the surrounding environment) affect the signal propagation between the tag and the reader. These events can also change the characteristics of the signal, resulting in false positives. Excessive false positives are also a similarly serious problem, as they can result in additional labor costs to handle the false positives, cause unnecessary machine downtime, and even result in improper operation.
Disclosure of Invention
In order to solve the technical defects, the embodiment of the invention provides an industrial environment leakage detection method and device based on double RFID tags.
In a first aspect, an embodiment of the present invention provides an industrial environment leakage detection method based on a dual RFID tag, including:
preprocessing the acquired tag signal to obtain a signal window, wherein the tag signal is related to the mutual inductive coupling effect between the two RFID tags;
acquiring a classification characteristic according to each signal window, identifying the state of each signal window according to a pre-trained decision tree based on the classification characteristic, and determining a first preset number of continuous signal windows detected as liquid leakage state windows, wherein the classification characteristic is used for classifying the state of each signal window into normal, interference or liquid leakage;
and extracting a signal section corresponding to the leakage state window, and obtaining a starting point and an ending point of leakage according to the signal section.
In a second aspect, an embodiment of the present invention provides an industrial environment liquid leakage detection apparatus based on a dual RFID tag, including:
the system comprises an acquisition module, a signal window and a signal processing module, wherein the acquisition module is used for preprocessing acquired tag signals to obtain a signal window, and the tag signals are related to the mutual inductive coupling effect between two RFID tags;
the state identification module is used for acquiring classification features according to each signal window, identifying the state of each signal window according to a pre-trained decision tree based on the classification features, and determining a first preset number of continuous signal windows which are detected as liquid leakage state windows, wherein the classification features are used for classifying the state of each signal window into normal, interference or liquid leakage;
and the signal extraction module is used for extracting a signal section corresponding to the leakage state window and obtaining a starting point and an ending point of leakage according to the signal section.
In a third aspect, an embodiment of the present invention provides an electronic device, including a memory and a processor, where the processor and the memory complete communication with each other through a bus; the memory stores program instructions executable by the processor, and the processor calls the program instructions to execute the method for detecting leakage of industrial environment based on the dual RFID tag according to the first aspect.
In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the dual RFID tag-based industrial environment liquid leakage detection method according to the first aspect.
According to the industrial environment leakage detection method and device based on the double RFID tags, the novel real-time leakage detection technology based on the RFID is provided, the double tags are used as sensors for leakage detection according to the fact that coupling effect of two adjacent RFID tags can be influenced by liquid, the purpose of leakage detection is achieved by analyzing signal characteristics of the two tags, even a small amount of leakage can be detected in time, the leakage is not easily influenced by the external environment, the starting point and the ending point of the leakage can be accurately identified, and the occurrence of false alarm is reduced.
Drawings
Fig. 1 is a schematic flow chart of an industrial environment leakage detection method based on a dual RFID tag according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of an industrial environment leakage detection device based on a dual RFID tag according to an embodiment of the present invention;
fig. 3 is a schematic physical structure diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
Fig. 1 is a schematic flow chart of an industrial environment leakage detection method based on a dual RFID tag according to an embodiment of the present invention, as shown in fig. 1, including:
step 11, preprocessing the acquired tag signals to obtain a signal window, wherein the tag signals are related to the mutual inductive coupling effect between the two RFID tags;
and step 13, extracting a signal section corresponding to the liquid leakage state window, and obtaining a starting point and an ending point of liquid leakage according to the signal section.
Radio Frequency Identification (RFID) is a communication technology that can identify a specific target and read and write related data through Radio signals without establishing mechanical or optical contact between an Identification system and the specific target. The RFID tags are composed of coupling elements and chips, each RFID tag has a unique electronic code and is attached to an object to identify a target object.
In the industrial environment liquid leakage detection, the Received Signal Strength Indication (RSSI) and the phase value of the RFID tag are kept stable before liquid leakage; after the leakage begins, the RSSI slightly rises, the RSSI begins to fall along with the increase of the leakage, and the cotton is recovered to be stable after being saturated, and meanwhile, the phase value monotonically falls.
Before a leak occurs, there is a difference between the RSSI of the two RFID tags even though they are in the same relative position to the reader. The RSSI separation of two RFID tags is caused by the mutual inductive coupling effect between the two RFID tags. After liquid leakage begins, the RSSI gap of the two RFID tags is rapidly reduced, because the coupling effect between the two RFID tags is weakened by liquid, therefore, the reduction of the RSSI gap can be used as a characteristic of liquid leakage, the disappearance of the coupling effect causes the rapid change of tag signals, and the liquid leakage can be detected in a short time. Since two RFID tags are in close proximity to each other, which means that other confounding events will cause similar interference to them, signal changes caused by other confounding events can be filtered out by observing the relationship (e.g., gaps) between the tag signals.
The mutual inductive coupling effect between the two RFID tags can influence the tag signals, the RSSI interval of the two RFID tags is caused by the mutual inductive coupling effect between the two RFID tags, and the reduction of the RSSI interval can be used as a characteristic of liquid leakage, so that whether liquid leakage occurs or not can be reflected through the tag signals. Firstly, obtaining a tag signal, preprocessing the obtained tag signal, wherein the purpose of preprocessing is to perform data smoothing and data alignment, obtaining the tag signal by adopting a time window, then performing denoising processing, and performing data alignment according to a time stamp to obtain signal windows, wherein the data volume of each signal window is the same, and the time is aligned.
Next, status recognition is performed. The goal of state identification is to label the signal in each signal window as a certain state, a normal state, an interference state and a leakage state, wherein the interference state refers to the signal change caused by environmental interference. The principle of state recognition is that the signal pattern in the liquid leakage state, i.e., the mutual inductive coupling between the RFID tags and the superposition of impedance changes caused by the leaked liquid, is different from other states. Thus, in the state classification phase, first features that can distinguish different states, i.e. classification features, are extracted from each signal window, and then the states are identified from these classification features. The classification features selected include signal features of a single tag and features related to the relationship between two tag signals for accurate classification. And after the classification features are extracted, the state of each signal window is identified by using a pre-trained decision tree. However, once a signal window of the leakage state is detected, the occurrence of leakage cannot be directly judged, and the method inevitably leads to a high misjudgment rate. In the design of the double label, after a first preset number of continuous signal windows for detecting leakage is identified (which means that the possibility of leakage is high), the leakage is alarmed to occur, and the first preset number of continuous signal windows for detecting leakage is determined as a leakage state window. For example, if k consecutive signal windows determined as leakage states are identified, an alarm is issued, where the first preset number k is determined according to actual conditions before detection, and is not specifically limited herein, and the tradeoff between detection delay and accuracy is realized by empirically setting k to 3.
And finally, extracting a signal section corresponding to the leakage state window, and further analyzing. For the signal segment, firstly, the states of the front window and the rear window of each signal window are judged according to the signal windows contained in the signal segment, so as to further determine the signal windows with liquid leakage, finish the rough segmentation of the signal, then, the RSSI and the phase of each signal window are obtained according to the signal segments after the rough segmentation, and the starting point and the ending point of the liquid leakage are obtained according to the RSSI and the phase.
According to the industrial environment leakage detection method based on the double RFID tags, the novel real-time leakage detection technology based on the RFID is provided, the double tags are used as sensors for leakage detection according to the fact that coupling effect of two adjacent RFID tags can be influenced by liquid, the purpose of leakage detection is achieved by analyzing signal characteristics of the two tags, even a small amount of leakage can be detected in time, the leakage detection method is not easily influenced by the external environment, the starting point and the ending point of leakage can be accurately identified, and the occurrence of false alarm is reduced.
On the basis of the above embodiment, the preprocessing the acquired tag signal specifically includes:
acquiring the tag signal according to a time window, and acquiring the RSSI, the phase and the timestamp of the tag signal;
aligning the data of the tag signals according to the time stamps, and acquiring an RSSI average value and a phase average value in a preset time period;
and carrying out interpolation based on an interpolation function according to the RSSI average value and the phase average value in the preset time period to obtain the signal windows, wherein the interpolation function is used for ensuring that the signal quantity of each time window is the same.
The preprocessing process includes three steps, data acquisition, data smoothing, and data alignment, respectively.
The first is data acquisition, which requires data acquisition and preprocessing prior to state recognition. And acquiring the tag signal by using a time window, and storing the RSSI, the phase and the timestamp of the tag data reported each time.
And secondly, data smoothing is carried out, because the acquired tag signal has noise, the data needs to be smoothed, a low-pass filter can be used for filtering the tag signal, and the phase of the tag signal is unwrapped.
And finally, data alignment is carried out, and the tag signals need to be aligned according to the time stamps because the reporting interval is random. And selecting an M value according to the reporting frequency of the label signal, averaging the RSSI and the phase within every M milliseconds, and interpolating by using an interpolation function to ensure that the data volume of each time window is the same and the time is aligned to obtain the signal window.
And carrying out subsequent state identification and signal extraction processes according to the obtained signal window.
On the basis of the above embodiment, the classification features include:
the relative change amount of a signal, the direct deviation from the stable value of the signal, the slope of the linear fit of the signal, the mean square error of the linear fit of the signal, the variance of the signal, and the mean of the signal, wherein the signal comprises the RSSI of the two RFID tags and the phase of the two RFID tags.
Extracting features capable of distinguishing different states, i.e. classification features, from each signal window, and then identifying the states according to the classification features, wherein the classification features comprise:
the relative change amount of the signal, the direct deviation from the stable value of the signal, the slope of the linear fitting of the signal, the mean square error of the linear fitting of the signal, the variance of the signal and the mean value of the signal, wherein the signal comprises the RSSI of the two RFID tags and the phase of the two RFID tags.
Selected features in embodiments of the invention include signal features of a single tag and features related to the relationship between two tag signals for accurate classification.
According to the industrial environment leakage detection method based on the double RFID tags, the novel real-time leakage detection technology based on the RFID is provided, the double tags are used as sensors for leakage detection according to the fact that coupling effect of two adjacent RFID tags can be influenced by liquid, the purpose of leakage detection is achieved by analyzing signal characteristics of the two tags, even a small amount of leakage can be detected in time, the leakage detection method is not easily influenced by the external environment, the starting point and the ending point of leakage can be accurately identified, and the occurrence of false alarm is reduced.
On the basis of the above embodiment, the extracting a signal segment corresponding to the liquid leakage state window, and obtaining a starting point and an ending point of liquid leakage according to the signal segment specifically include:
calculating the context window proportion of each signal window of the signal section marked as a liquid leakage state according to the number of preset continuous signal windows, and obtaining a rough segmentation window of the liquid leakage state window according to the context window proportion and a second preset number;
acquiring the RSSI and the phase of a first window and a last window in the coarse segmentation of the coarse segmentation window, and acquiring the RSSI and the phase of a front window and a rear window, wherein the front window is a signal window close to the first window in the coarse segmentation window, and the rear window is a signal window close to the last window in the coarse segmentation window;
and determining a starting point of liquid leakage according to the RSSI and the phase of the first window and the RSSI and the phase of the previous window, and determining an ending point of liquid leakage according to the RSSI and the phase of the last window and the RSSI and the phase of the next window.
The calculating the context window proportion of each signal window of the signal segment marked as a liquid leakage state specifically comprises:
wherein the content of the first and second substances,
for any one of the windows of the signal,
and N is the preset continuous signal window number, and is the context window proportion of any signal window marked as a liquid leakage state.
The determining a starting point of liquid leakage according to the RSSI and the phase of the first window and the RSSI and the phase of the preceding window, and determining an ending point of liquid leakage according to the RSSI and the phase of the last window and the RSSI and the phase of the succeeding window specifically include:
acquiring a corresponding first fitted straight line according to the RSSI and the phase of the first window, and acquiring a corresponding second fitted straight line according to the RSSI and the phase of the previous window, wherein the intersection point of the first fitted straight line and the second fitted straight line is the starting point of the leakage;
and acquiring a corresponding third fitted straight line according to the RSSI and the phase of the last window, and acquiring a corresponding fourth fitted straight line according to the RSSI and the phase of the subsequent window, wherein the intersection point of the third fitted straight line and the fourth fitted straight line is the end point of the liquid leakage.
After a leakage is detected, the next step is to extract the signal segment containing the complete leakage process for further analysis. However, the starting and ending points of the weep process must be accurately identified, otherwise the subsequent analysis will be error-prone. One simple approach is to use the results of the state recognition process-simply identifying the first window of the leakage state window as the starting point and the last window as the end point. However, this is very fragile because the recognition accuracy of each window is very low, and for this problem, a two-step segmentation method is proposed, which first provides a coarse segmentation result according to the state recognition result, and then locates the precise start point and end point on the coarse segmentation segment to obtain a refined segmentation result.
In the coarse segmentation stage, continuous leakage state windows are firstly identified, and signals contained in the windows are spliced into a coarse signal segmentation result. The problem encountered here is how to deal with errors in the state recognition phase. The method adopted by the embodiment of the invention is that each signal window considers the states of the windows before and after the signal window. This is because when more contextual windows are identified as being in a weep state, the windows are more likely to be weep state windows.
In particular, a window of a mark signal is given
Sequence, calculating a parameter for each signal window
Any signal window of the parametric representation signal segment is identified asThe calculation formula of the contextual window proportion of the liquid leakage state is as follows:
wherein the content of the first and second substances,
for any one of the windows of the signal,
and N is the preset continuous signal window number, and is the context window proportion of any signal window marked as a liquid leakage state. Obtaining the roughly divided windows of the liquid leakage state windows according to the context window proportion and a second preset number
Then extracting
And taking the continuous signal windows with the length larger than the second preset number M as rough segmentation windows of the liquid leakage state windows. In the system of the embodiment of the present invention, N is set to 3 and M is set to 4 empirically.
Roughly dividing and outputting 4 signal streams, namely RSSI and phase data of two labels, which respectively represent Ri、PiI is e {1,2 }. The goal of this step is to determine the starting and ending points of the weep process from the four signal streams.
In this section, we mainly discuss how to find a signal stream of determined start and end points (in sp)[s]And ep[s]Denotes S ∈ S ═ R1,R2,P1,P2}) which are key to the calculation of the segmentation bleed-off signal. As mentioned above, the RSSI value remains stable under normal conditions, but begins to change when a leak occurs, eventually returning to stability when the cotton is saturated with liquid. Thus, translating the segmentation problem into finding two critical inflection points in the RSSI readings, a search directly from noisy raw data is clearly error-prone. To solve this problem, a linear regression is used (which can filter out the noise), according to a first window
Obtaining a corresponding first fitting straight line LS by the RSSI and the phase1According to the preceding window
Obtaining a corresponding second fitted straight line LB for the RSSI and the phase1First fitted straight line LS1And a second fitted straight line LB1The intersection point of (a) is the starting point of the leakage. According to the last window
Obtaining a corresponding third fitting straight line LS by the RSSI and the phase2According to the rear window
Obtaining a corresponding fourth fitted straight line LA by the RSSI and the phase1Third fitted straight line LS2And a fourth fitted straight line LA1The intersection point of (a) is the end point of the leakage.
According to the industrial environment leakage detection method based on the double RFID tags, the novel real-time leakage detection technology based on the RFID is provided, the double tags are used as sensors for leakage detection according to the fact that coupling effect of two adjacent RFID tags can be influenced by liquid, the purpose of leakage detection is achieved by analyzing signal characteristics of the two tags, even a small amount of leakage can be detected in time, the leakage detection method is not easily influenced by the external environment, the starting point and the ending point of leakage can be accurately identified, and the occurrence of false alarm is reduced.
On the basis of the above embodiment, the method further includes:
and acquiring the corresponding leakage amount and rate based on a preset signal template according to the starting point of the leakage and the ending point of the leakage.
After obtaining the signal section corresponding to the liquid leakage process, characteristics of the liquid leakage process, such as liquid leakage amount and liquid leakage rate, need to be estimated, which is also a concern in the industrial field.
To estimate the amount and rate of leakage, the progress of the leakage is first tracked. Therefore, firstly, simple and low-overhead calibration is carried out to obtain a complete and accurate signal template in the liquid leakage process. The entire process of leakage is tracked by matching the previously extracted signal segments with the signal template.
Tag calibration is performed only once prior to tag deployment. In the calibration process, the leakage process of the dual-label sensor is triggered manually, the leakage rate is stable, and the leakage liquid amount is controlled to be 20 mL. The above process is repeated 30 times at different leakage rates on different sensors, during which the reader continuously reads the tag, resulting in RSSI and phase. Finally, a template is generated from the collected data using an algorithm closest to the center.
The liquid leakage process is estimated by adopting a Dynamic Time Warping (DTW) based search algorithm tracking process. The goal of the DTW algorithm is at SaAnd SbFind a matching function f betweenMThereby minimizing
Although DTW is widely used, it cannot be directly used to track a weep process. Because DTW is mainly used to compare two complete data streams, the two data streams are expected to take on similar shapes. And the weeping process may be stopped before the cotton is saturated. Thus, the resulting signal segment associated with a leak may be only a portion of the template. Using the DTW algorithm to directly match the two streams results in large errors, so we improve the algorithm.
In a given template sequence SaAnd a detected leakage sequence SbIn the case of (c), some key observations are summarized: (1) r (S)b) Determining a template S without changing shapeaThe time scale of (a); (2) v (S)b) Without increasing or decreasing SaBut rather to spatially truncate it by entering a steady state ahead of time. Thus, based on these observations, r (S) can be estimated by inferring the temporal scale factor and the spatial truncation ratiob) And v (S)b). Two alignment sequences SaAnd Sb′＝Sb[fM]Is cut offPoint t maximizes subsequence S by reversinga[1:t]And Sb′[1:t]Is determined. Then, the average matching distance is estimated
Rate of liquid leakageVolume of leakage
Is fMInverse mapping of (3).
According to the industrial environment leakage detection method based on the double RFID tags, the novel real-time leakage detection technology based on the RFID is provided, the double tags are used as sensors for leakage detection according to the fact that coupling effect of two adjacent RFID tags can be influenced by liquid, the purpose of leakage detection is achieved by analyzing signal characteristics of the two tags, even a small amount of leakage can be detected in time, the leakage detection method is not easily influenced by the external environment, the starting point and the ending point of leakage can be accurately identified, and the occurrence of false alarm is reduced. Meanwhile, besides accurate detection, quantitative characteristics (such as liquid leakage speed and volume) of a liquid leakage process are provided, and the liquid leakage detection method can be used as an important decision basis when liquid leakage occurs.
Fig. 2 is a schematic structural diagram of an industrial environment liquid leakage detection device based on a dual RFID tag according to an embodiment of the present invention, as shown in fig. 2, including an obtaining module 21, a state identification module 22, and a signal extraction module 23, where:
the acquisition module 21 is configured to pre-process the acquired tag signal to obtain a signal window, where the tag signal is related to a mutual inductive coupling effect between two RFID tags;
the state identification module 22 is configured to obtain a classification feature according to each signal window, identify a state of each signal window according to a pre-trained decision tree based on the classification feature, and determine a first preset number of consecutive signal windows detected as liquid leaks as liquid leak state windows, where the classification feature is used to classify the state of each signal window as normal, interference, or liquid leak;
the signal extraction module 23 is configured to extract a signal segment corresponding to the leakage state window, and obtain a starting point and an ending point of leakage according to the signal segment.
Before a leak occurs, there is a difference between the RSSI of the two RFID tags even though they are in the same relative position to the reader. The RSSI separation of two RFID tags is caused by the mutual inductive coupling effect between the two RFID tags. After liquid leakage begins, the RSSI gap of the two RFID tags is rapidly reduced, because the coupling effect between the two RFID tags is weakened by liquid, therefore, the reduction of the RSSI gap can be used as a characteristic of liquid leakage, the disappearance of the coupling effect causes the rapid change of tag signals, and the liquid leakage can be detected in a short time. Since two RFID tags are in close proximity to each other, which means that other confounding events will cause similar interference to them, signal changes caused by other confounding events can be filtered out by observing the relationship (e.g., gaps) between the tag signals.
The mutual inductive coupling effect between the two RFID tags can influence the tag signals, the RSSI interval of the two RFID tags is caused by the mutual inductive coupling effect between the two RFID tags, and the reduction of the RSSI interval can be used as a characteristic of liquid leakage, so that whether liquid leakage occurs or not can be reflected through the tag signals. Firstly, the obtaining module 21 obtains a tag signal, and preprocesses the obtained tag signal, the purpose of the preprocessing is to perform data smoothing and data alignment, a time window is adopted to obtain the tag signal, then denoising processing is performed, and data alignment is performed according to a time stamp to obtain signal windows, the data volume of each signal window is the same, and the time is aligned with the signal window.
Next, the state recognition module 22 performs state recognition. The goal of state identification is to label the signal in each signal window as a certain state, a normal state, an interference state and a leakage state, wherein the interference state refers to the signal change caused by environmental interference. The principle of state recognition is that the signal pattern in the liquid leakage state, i.e., the mutual inductive coupling between the RFID tags and the superposition of impedance changes caused by the leaked liquid, is different from other states. Thus, in the state classification phase, the state identification module 22 first extracts features from each signal window that are capable of distinguishing between different states, i.e., classification features, and then identifies the states according to these classification features. The classification features selected include signal features of a single tag and features related to the relationship between two tag signals for accurate classification. And after the classification features are extracted, the state of each signal window is identified by using a pre-trained decision tree. However, once a signal window of the leakage state is detected, the occurrence of leakage cannot be directly judged, and the method inevitably leads to a high misjudgment rate. In the design of the double label, after a first preset number of continuous signal windows for detecting leakage is identified (which means that the possibility of leakage is high), the leakage is alarmed to occur, and the first preset number of continuous signal windows for detecting leakage is determined as a leakage state window. For example, if k consecutive signal windows determined as leakage states are identified, an alarm is issued, where the first preset number k is determined according to actual conditions before detection, and is not specifically limited herein, and the tradeoff between detection delay and accuracy is realized by empirically setting k to 3.
Finally, the signal extraction module 23 extracts a signal segment corresponding to the liquid leakage state window for further analysis. For this signal segment, the signal extraction module 23 first determines the states of the front and rear windows of each signal window to further determine the signal window with liquid leakage, completes the rough division of the signal, then obtains the RSSI and the phase of each signal window according to the roughly divided signal segment, and obtains the starting point and the ending point of the liquid leakage according to the RSSI and the phase.
The apparatus provided in the embodiment of the present invention is used for executing the above method embodiments, and for detailed descriptions and specific processes, reference is made to the above method embodiments, which are not described herein again.
According to the industrial environment leakage detection device based on the double RFID tags, the novel real-time leakage detection technology based on the RFID is provided, the double tags are used as sensors for leakage detection according to the fact that coupling effect of two adjacent RFID tags can be influenced by liquid, the purpose of leakage detection is achieved by analyzing signal characteristics of the two tags, even a small amount of leakage can be detected in time, the leakage detection device is not easily influenced by the external environment, the starting point and the ending point of the leakage can be accurately identified, and the occurrence of false alarm is reduced.
Fig. 3 is a schematic entity structure diagram of an electronic device according to an embodiment of the present invention, and as shown in fig. 3, the electronic device may include: a processor (processor)310, a communication Interface (communication Interface)320, a memory (memory)330 and a bus 340, wherein the processor 310, the communication Interface 320 and the memory 330 complete communication with each other through the bus 340. Bus 340 may be used for information transfer between the electronic device and the sensor. The processor 310 may call logic instructions in the memory 330 to perform the following method: preprocessing the acquired tag signal to obtain a signal window, wherein the tag signal is related to the mutual inductive coupling effect between the two RFID tags; acquiring a classification characteristic according to each signal window, identifying the state of each signal window according to a pre-trained decision tree based on the classification characteristic, and determining a first preset number of continuous signal windows detected as liquid leakage state windows, wherein the classification characteristic is used for classifying the state of each signal window into normal, interference or liquid leakage; and extracting a signal section corresponding to the leakage state window, and obtaining a starting point and an ending point of leakage according to the signal section.
In addition, the logic instructions in the memory 330 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
An embodiment of the present invention provides a non-transitory computer-readable storage medium, which stores computer instructions, where the computer instructions cause a computer to execute a method for detecting leakage in an industrial environment based on a dual RFID tag, which is provided in the foregoing embodiments, for example, the method includes: preprocessing the acquired tag signal to obtain a signal window, wherein the tag signal is related to the mutual inductive coupling effect between the two RFID tags; acquiring a classification characteristic according to each signal window, identifying the state of each signal window according to a pre-trained decision tree based on the classification characteristic, and determining a first preset number of continuous signal windows detected as liquid leakage state windows, wherein the classification characteristic is used for classifying the state of each signal window into normal, interference or liquid leakage; and extracting a signal section corresponding to the leakage state window, and obtaining a starting point and an ending point of leakage according to the signal section.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Various modifications and additions may be made to the described embodiments by those skilled in the art without departing from the spirit of the invention or exceeding the scope as defined in the appended claims.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the invention has been described in detail 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 technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (8)
1. A double-RFID-tag-based industrial environment leakage detection method is characterized by comprising the following steps:
preprocessing the acquired tag signal to obtain a signal window, wherein the tag signal is related to the mutual inductive coupling effect between the two RFID tags;
acquiring a classification characteristic according to each signal window, identifying the state of each signal window according to a pre-trained decision tree based on the classification characteristic, and determining a first preset number of continuous signal windows detected as liquid leakage state windows, wherein the classification characteristic is used for classifying the state of each signal window into normal, interference or liquid leakage;
extracting a signal section corresponding to the leakage state window, and obtaining a starting point and an ending point of leakage according to the signal section;
the extracting of the signal section corresponding to the leakage state window and the obtaining of the starting point and the ending point of leakage according to the signal section specifically comprise:
calculating the context window proportion of each signal window of the signal section marked as a liquid leakage state according to the number of preset continuous signal windows, and obtaining a rough segmentation window of the liquid leakage state window according to the context window proportion and a second preset number;
acquiring the RSSI and the phase of a first window and a last window in the coarse segmentation of the coarse segmentation window, and acquiring the RSSI and the phase of a front window and a rear window, wherein the front window is a signal window close to the first window in the coarse segmentation window, and the rear window is a signal window close to the last window in the coarse segmentation window;
and determining a starting point of liquid leakage according to the RSSI and the phase of the first window and the RSSI and the phase of the previous window, and determining an ending point of liquid leakage according to the RSSI and the phase of the last window and the RSSI and the phase of the next window.
2. The method according to claim 1, wherein the preprocessing the acquired tag signal specifically comprises:
acquiring the tag signal according to a time window, and acquiring the RSSI, the phase and the timestamp of the tag signal;
aligning the data of the tag signals according to the time stamps, and acquiring an RSSI average value and a phase average value in a preset time period;
and carrying out interpolation based on an interpolation function according to the RSSI average value and the phase average value in the preset time period to obtain the signal windows, wherein the interpolation function is used for ensuring that the signal quantity of each time window is the same.
3. The method of claim 1, wherein the classifying features comprise:
the relative change amount of a signal, the direct deviation from the stable value of the signal, the slope of the linear fit of the signal, the mean square error of the linear fit of the signal, the variance of the signal, and the mean of the signal, wherein the signal comprises the RSSI of the two RFID tags and the phase of the two RFID tags.
4. The method according to claim 3, wherein the determining a starting point of the liquid leakage according to the RSSI and the phase of the first window and the RSSI and the phase of the preceding window, and the determining an ending point of the liquid leakage according to the RSSI and the phase of the last window and the RSSI and the phase of the following window specifically comprises:
acquiring a corresponding first fitted straight line according to the RSSI and the phase of the first window, and acquiring a corresponding second fitted straight line according to the RSSI and the phase of the previous window, wherein the intersection point of the first fitted straight line and the second fitted straight line is the starting point of the leakage;
and acquiring a corresponding third fitted straight line according to the RSSI and the phase of the last window, and acquiring a corresponding fourth fitted straight line according to the RSSI and the phase of the subsequent window, wherein the intersection point of the third fitted straight line and the fourth fitted straight line is the end point of the liquid leakage.
5. The method of claim 4, further comprising:
and acquiring the corresponding leakage amount and rate based on a preset signal template according to the starting point of the leakage and the ending point of the leakage.
6. An industrial environment weeping detection device based on two RFID labels, its characterized in that includes:
the system comprises an acquisition module, a signal window and a signal processing module, wherein the acquisition module is used for preprocessing acquired tag signals to obtain a signal window, and the tag signals are related to the mutual inductive coupling effect between two RFID tags;
the state identification module is used for acquiring classification features according to each signal window, identifying the state of each signal window according to a pre-trained decision tree based on the classification features, and determining a first preset number of continuous signal windows which are detected as liquid leakage state windows, wherein the classification features are used for classifying the state of each signal window into normal, interference or liquid leakage;
the signal extraction module is used for extracting a signal section corresponding to the liquid leakage state window and obtaining a starting point and an end point of liquid leakage according to the signal section;
the signal extraction module is further configured to:
calculating the context window proportion of each signal window of the signal section marked as a liquid leakage state according to the number of preset continuous signal windows, and obtaining a rough segmentation window of the liquid leakage state window according to the context window proportion and a second preset number;
acquiring the RSSI and the phase of a first window and a last window in the coarse segmentation of the coarse segmentation window, and acquiring the RSSI and the phase of a front window and a rear window, wherein the front window is a signal window close to the first window in the coarse segmentation window, and the rear window is a signal window close to the last window in the coarse segmentation window;
and determining a starting point of liquid leakage according to the RSSI and the phase of the first window and the RSSI and the phase of the previous window, and determining an ending point of liquid leakage according to the RSSI and the phase of the last window and the RSSI and the phase of the next window.
7. An electronic device, comprising a memory and a processor, wherein the processor and the memory communicate with each other via a bus; the memory stores program instructions executable by the processor, the processor calling the program instructions being capable of executing a dual RFID tag based industrial environment leakage detection method according to any of claims 1 to 5.
8. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements a dual RFID tag based industrial environment leakage detection method according to any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811391257.9A CN109580134B (en) | 2018-11-21 | 2018-11-21 | Industrial environment liquid leakage detection method and device based on double RFID tags |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811391257.9A CN109580134B (en) | 2018-11-21 | 2018-11-21 | Industrial environment liquid leakage detection method and device based on double RFID tags |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109580134A CN109580134A (en) | 2019-04-05 |
| CN109580134B true CN109580134B (en) | 2020-01-14 |
Family
ID=65923581
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811391257.9A Active CN109580134B (en) | 2018-11-21 | 2018-11-21 | Industrial environment liquid leakage detection method and device based on double RFID tags |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109580134B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108169248A (en) * | 2016-12-07 | 2018-06-15 | 成都理工大学 | A kind of adaptive S parameter matching process for liquid hazardous material detection |
| CN108443714A (en) * | 2018-04-09 | 2018-08-24 | 山东省科学院激光研究所 | Determine the method and device of pipe leakage |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1293366A (en) * | 2000-12-19 | 2001-05-02 | 石油大学(北京)机电工程学院 | Equipment and method for detecting leakage of pipeline |
| CN103136587B (en) * | 2013-03-07 | 2015-12-09 | 武汉大学 | Based on the power distribution network running status classifying identification method of support vector machine |
| US9274020B2 (en) * | 2013-10-14 | 2016-03-01 | General Electric Company | System and a method for detecting liquid water |
| CN204496030U (en) * | 2015-04-15 | 2015-07-22 | 北京瑞芯谷科技有限公司 | A kind of pipeline detection system |
| CN106567997B (en) * | 2016-04-24 | 2020-01-21 | 内蒙古科技大学 | Oil and gas pipeline remote real-time health monitoring system based on Internet of things |
| CN206496202U (en) * | 2017-01-19 | 2017-09-15 | 常州黄河机电工程有限公司 | Intelligent conduit pipeline water leakage detection warning protection device |
-
2018
- 2018-11-21 CN CN201811391257.9A patent/CN109580134B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108169248A (en) * | 2016-12-07 | 2018-06-15 | 成都理工大学 | A kind of adaptive S parameter matching process for liquid hazardous material detection |
| CN108443714A (en) * | 2018-04-09 | 2018-08-24 | 山东省科学院激光研究所 | Determine the method and device of pipe leakage |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109580134A (en) | 2019-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102799867B (en) | Meter pointer angle identification method based on image processing | |
| Song et al. | A noise robust method based on completed local binary patterns for hot-rolled steel strip surface defects | |
| CN109115800B (en) | Method for rapidly detecting burrs of product and accurately measuring length | |
| CN109740609B (en) | Track gauge detection method and device | |
| CN104647893B (en) | A kind of printing chromatography error detection method based on cross hairs | |
| CN110706247A (en) | Target tracking method, device and system | |
| CN110866515A (en) | Method and device for identifying object behaviors in plant and electronic equipment | |
| KR20160118830A (en) | Method for recognizing object using pressure sensor | |
| WO2021017000A1 (en) | Method and apparatus for acquiring meter reading, and memory, processor and terminal | |
| CN111811567B (en) | Equipment detection method based on curve inflection point comparison and related device | |
| CN111259718A (en) | Escalator retention detection method and system based on Gaussian mixture model | |
| CN109580134B (en) | Industrial environment liquid leakage detection method and device based on double RFID tags | |
| CN104992175A (en) | Over-merged steel billet character segmentation method based on tree type relationship | |
| CN114445398A (en) | Method and device for monitoring state of side protection plate of hydraulic support of coal mining machine | |
| CN109543498A (en) | A kind of method for detecting lane lines based on multitask network | |
| Choi et al. | Localizing slab identification numbers in factory scene images | |
| CN117400066A (en) | Numerical control machine tool wear identification method and system | |
| CN102855473A (en) | Image multi-target detecting method based on similarity measurement | |
| CN107680104A (en) | Hand Gesture Segmentation method based on kalman filtering | |
| CN110059610A (en) | A kind of pointer instrument method for early warning based on image recognition | |
| CN105488448A (en) | Barcode and two-dimensional code distinguishing method | |
| Mol et al. | Text recognition using poisson filtering and edge enhanced maximally stable extremal regions | |
| CN113160220A (en) | Door handle homing and bending detection method based on deep learning | |
| KR101492338B1 (en) | Method and apparatus for smart managing of product using rf tag and vison sensor | |
| CN112329770A (en) | Instrument scale identification method and device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |