Intelligent sliding rail door abnormal sound detection method and system

Abstract

The invention relates to the technical field of abnormal data recognition, and provides an intelligent sliding rail door abnormal sound detection method and system, and the method comprises the steps: collecting the vibration data of an intelligent sliding rail door and the time spent in opening and closing the door, obtaining a vibration data sequence and abnormal vibration data, and marking a standard vibration data sequence; determining the abnormal possibility of the vibration data sequence; the time anomaly correlation and the door opening and closing anomaly degree of the vibration data sequence are obtained, weighted summation is conducted on the anomaly probability and the door opening and closing anomaly degree of the vibration data sequence according to the time anomaly correlation of the vibration data sequence, and an abnormal sound evaluation value of the vibration data sequence is obtained; and according to the abnormal sound evaluation value of the vibration data sequence, judging whether the to-be-detected intelligent sliding rail door has abnormal sound or not. The fault identification accuracy of the intelligent sliding rail door can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of abnormal data recognition, and in particular to a method and system for detecting abnormal noise of an intelligent sliding door. Background Art

[0002] Due to reasons such as increased usage time, improper installation or insufficient maintenance, the smart sliding door may have abnormal noise problems during use, which not only affects the user's comfort when using the smart sliding door, but may also be a precursor to a failure of the smart sliding door. It is necessary to identify the failure of the smart sliding door in time and carry out targeted repairs in time to avoid greater losses.

[0003] In the process of identifying the fault of the smart sliding door based on the abnormal noise generated during the use of the smart sliding door, the vibration data of the smart sliding door during the use is generally collected, and the fault of the smart sliding door is identified based on the vibration data. However, the vibration data collected during the use of the smart sliding door is easily affected by external interference, which makes the fault identification result of the smart sliding door inaccurate. Summary of the invention

[0004] The present invention provides a method and system for detecting abnormal noise of an intelligent sliding door, so as to solve the problem that the fault identification result directly obtained based on the abnormal noise generated during the use of the intelligent sliding door is inaccurate. The technical solution adopted is as follows: In a first aspect, an embodiment of the present invention provides a method for detecting abnormal noise of an intelligent sliding door, the method comprising the following steps: Collect vibration data of the intelligent sliding door in the door opening and closing process in historical time, vibration data of the intelligent sliding door to be detected in the door opening and closing process, and the time spent on each opening and closing of the intelligent sliding door, obtain a first preset threshold number of vibration data sequences of the door opening and closing process in historical time, and a second preset threshold number of vibration data sequences of the intelligent sliding door to be detected in the door opening and closing process, obtain abnormal vibration data in each vibration data sequence according to the vibration data sequence of the door opening and closing process in historical time, and mark the standard vibration data sequence; Any vibration data sequence of the intelligent sliding door to be detected in the process of opening and closing the door is recorded as the first target vibration data sequence, and the abnormal possibility of the first target vibration data sequence is determined according to the difference between the first target vibration data sequence before and after denoising, and the difference between the first target vibration data sequence and the standard vibration data sequence, so as to obtain the abnormal possibility of the vibration data sequence in the process of opening and closing the door in the historical time; According to the time spent on door opening and closing corresponding to the vibration data sequence containing abnormal vibration data in the vibration data sequence of the door opening and closing process in the historical time, the time anomaly correlation of the vibration data sequence of the door opening and closing process in the historical time is obtained, and the time anomaly correlation of the first target vibration data sequence is obtained. According to the time spent on door opening and closing corresponding to all vibration data sequences of the door opening and closing process in the historical time and all vibration data sequences of the intelligent sliding door to be detected in the door opening and closing process, the door opening and closing abnormality of the vibration data sequence is determined, and the abnormal possibility and the door opening and closing abnormality of the vibration data sequence are weightedly summed according to the time anomaly correlation of the vibration data sequence to obtain the abnormal sound evaluation value of the vibration data sequence; According to the abnormal sound evaluation value of the vibration data sequence, it is determined whether the intelligent sliding door to be tested has abnormal sound.

[0005] Furthermore, the vibration data sequence of the door opening and closing process in the historical time and the vibration data sequence of the intelligent sliding door to be detected in the door opening and closing process are obtained by: Arrange all vibration data collected during the period from when the smart sliding door receives the door opening signal to when the smart sliding door completes the door closing in the historical time in the order of collection time to obtain a vibration data sequence of the door opening and closing process in the historical time; All vibration data collected from the time when the intelligent sliding door to be detected receives the door opening signal until the intelligent sliding door completes the door closing process are arranged in the order of collection time to obtain the vibration data sequence of the intelligent sliding door to be detected.

[0006] Further, the method for obtaining the abnormal possibility of the first target vibration data sequence is: De-noising the first target vibration data sequence using a mean filter algorithm to obtain a de-noised first target vibration data sequence; The accumulated sum of the absolute values ​​of the differences between the first target vibration data sequence and the corresponding vibration data in the denoised first target vibration data sequence is recorded as the noise influence degree of the first target vibration data sequence; The DTW distance between the first target vibration data sequence and the standard vibration data sequence is recorded as the standard difference of the first target vibration data sequence; The possibility of abnormality of the first target vibration data sequence is determined according to the noise influence degree and the standard difference of the first target vibration data sequence.

[0007] Further, the method of determining the possibility of abnormality of the first target vibration data sequence according to the noise influence degree and the standard difference of the first target vibration data sequence includes the following specific methods: A normalized value of the product of the noise influence degree and the standard deviation of the first target vibration data sequence is recorded as the abnormal possibility of the first target vibration data sequence.

[0008] Furthermore, the calculation formula for the time anomaly correlation of the vibration data sequence of the door opening and closing process in the historical time is: In the formula, Represents the time anomaly correlation of the vibration data series of the door opening and closing process in the historical time; represents the normalization function; Indicates the number of vibration data sequences containing abnormal vibration data in the vibration data sequences of the door opening and closing process in the historical time; Indicates the number of abnormal vibration data in the vibration data sequence of the door opening and closing process in the historical time. The time it takes to open the door corresponding to each vibration data sequence; Represents the mean time spent on door opening corresponding to the vibration data sequence of all door opening and closing processes in the historical time; Indicates the number of abnormal vibration data in the vibration data sequence of the door opening and closing process in the historical time. The time it takes to close the door corresponding to each vibration data sequence; Represents the mean of the time taken to close the door corresponding to the vibration data sequence of all door opening and closing processes in the historical time; The standard deviation of the time it takes to open the door corresponding to all vibration data sequences of the door opening and closing process in the historical time; Represents the standard deviation of the time it takes to close the door corresponding to all vibration data sequences of the door opening and closing process in the historical time.

[0009] Further, the door opening and closing abnormality of the vibration data sequence is determined according to all vibration data sequences of the door opening and closing process in the historical time and the door opening and closing time corresponding to all vibration data sequences of the intelligent sliding door to be detected in the door opening and closing process, including the specific method of: The LOF anomaly detection algorithm is used for all vibration data sequences of the door opening and closing process in the historical time and the door opening time corresponding to all vibration data sequences of the intelligent sliding door to be detected in the door opening and closing process, and the LOF value corresponding to the door opening time corresponding to each vibration data sequence is obtained; The LOF anomaly detection algorithm is used for all vibration data sequences of the door opening and closing process in the historical time and the door closing time corresponding to all vibration data sequences of the intelligent sliding door to be detected in the door opening and closing process, and the LOF value corresponding to the door closing time corresponding to each vibration data sequence is obtained; The door opening and closing abnormality of the vibration data sequence is determined according to the LOF value corresponding to the time spent on door opening and closing corresponding to the vibration data sequence.

[0010] Further, the door opening and closing abnormality of the vibration data sequence is determined according to the LOF value corresponding to the time spent on opening and closing the door corresponding to the vibration data sequence, including the specific method of: The average of the LOF values ​​corresponding to the time spent on door opening and door closing corresponding to the vibration data sequence is recorded as the door opening and closing abnormality degree of the vibration data sequence.

[0011] Furthermore, the method for obtaining the abnormal sound evaluation value of the vibration data sequence is: in, Indicates the abnormal sound evaluation value of the vibration data sequence; Represents the time anomaly correlation of vibration data series; Indicates the degree of abnormality of the door opening and closing of the vibration data sequence; Indicates the likelihood that the vibration data series is abnormal.

[0012] Further, judging whether the intelligent sliding door to be detected has abnormal sound according to the abnormal sound evaluation value of the vibration data sequence includes the following specific methods: The abnormal sound evaluation values ​​of all vibration data sequences are processed using the maximum inter-class variance method to obtain an adaptive division threshold, and the vibration data sequences with abnormal sound evaluation values ​​greater than the adaptive division threshold are recorded as abnormal vibration data sequences; If the vibration data sequence of the intelligent sliding door to be detected during the door opening and closing process does not contain an abnormal vibration data sequence, it is determined that the intelligent sliding door to be detected does not have an abnormal sound; If a normal vibration data sequence appears in the vibration data sequence of the intelligent sliding door to be detected during the door opening and closing process, it is determined that the intelligent sliding door to be detected has an abnormal sound.

[0013] In a second aspect, an embodiment of the present invention further provides an intelligent sliding door abnormal sound detection system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above methods when executing the computer program.

[0014] The beneficial effects of the present invention are: The present application is based on the fact that the vibration data collected when the door opening and closing operations are performed at different times when the smart sliding door automatically opens and closes without any abnormality is similar, and the smart sliding door with an abnormality will introduce more noise to the collected vibration data, and the vibration data collected when the door opening and closing operations are performed at different times are quite different. The possibility of abnormality of the smart sliding door in the collection time period corresponding to the vibration data sequence is evaluated, and the abnormal possibility of the vibration data sequence is obtained; further, the correlation characteristics between the smart sliding door with an abnormality and the door opening time and the door closing time are evaluated, and the vibration data sequence of the door opening and closing process in all historical time and the vibration data sequence of the smart sliding door to be detected in the door opening and closing process are obtained. The time anomaly correlation of the series is calculated, and the door opening and closing abnormality of the vibration data sequence is determined according to the time spent on opening and closing the door corresponding to all vibration data sequences of the door opening and closing process in the historical time and all vibration data sequences of the smart sliding door to be detected in the door opening and closing process, and the possibility of abnormal noise occurring in the smart sliding door corresponding to the vibration data sequence during the collection time period corresponding to the vibration data sequence is further evaluated to obtain the abnormal noise evaluation value of the vibration data sequence; finally, according to the abnormal noise evaluation value of the vibration data sequence, it is judged whether the smart sliding door to be detected has abnormal noise, so as to solve the problem of inaccurate fault identification results directly obtained according to the abnormal noise occurring during the use of the smart sliding door, and improve the accuracy of abnormal noise detection of the smart sliding door. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0016] Figure 1 A schematic flow chart of an abnormal noise detection method for an intelligent sliding door provided by an embodiment of the present invention; Figure 2 A flowchart for obtaining abnormal possibility provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] See also Figure 1, which shows a flow chart of an abnormal sound detection method for an intelligent sliding door provided by an embodiment of the present invention, the method comprising the following steps: Step S001, collecting vibration data of the smart sliding door in the door opening and closing process in historical time, vibration data of the smart sliding door to be detected in the door opening and closing process, and the time spent on each opening and closing of the smart sliding door, obtaining a first preset threshold vibration data sequence of the door opening and closing process in historical time, and a second preset threshold vibration data sequence of the smart sliding door to be detected in the door opening and closing process, according to the vibration data sequence of the door opening and closing process in historical time, obtaining abnormal vibration data in each vibration data sequence, and marking the standard vibration data sequence.

[0019] The smart sliding door opens and closes automatically. When the smart sliding door receives an open signal, the vibration sensor is used to collect vibration data of the smart sliding door at preset time intervals until the smart sliding door receives a close signal and completes closing, and then the vibration sensor data collection stops. At the same time, the time sensor is used to collect the time it takes to open and close the door from the time the smart sliding door receives an open signal to the time the smart sliding door completes closing.

[0020] Preferably, in one embodiment of the present application, when collecting vibration data, the data sampling frequency in this embodiment is 10 Hz. In actual application, as other implementation methods, the implementer can decide the sampling frequency according to the actual situation, and this application does not impose any special restrictions.

[0021] All vibration data collected from the time when the smart sliding door receives the door opening signal until the smart sliding door completes the door closing process are arranged in the order of collection time to obtain the vibration data sequence of the smart sliding door's door opening and closing process.

[0022] It should be noted that the total The vibration data sequence of the same model of smart sliding door during the door opening and closing process in the historical time. When the smart sliding door is tested for abnormal noise, a total of A vibration data sequence of a smart sliding door during the door opening and closing process. represents the first preset threshold, represents the second preset threshold. In this embodiment, the values ​​of the first preset threshold and the second preset threshold are 1000 and 20 respectively.

[0023] Technical experts in this field From the vibration data sequences of the door opening and closing processes of the same model of smart sliding doors in the historical time, a standard vibration data sequence and abnormal vibration data in each vibration data sequence are selected.

[0024] The standard vibration data sequence is a vibration data sequence of a smart sliding door without any abnormality during the door opening and closing process, and the abnormal vibration data in the vibration data sequence is the abnormal vibration data collected during the door opening and closing process of a smart sliding door with an abnormality.

[0025] At this point, the vibration data sequence of the door opening and closing process within the first preset threshold historical time, the vibration data sequence of the door opening and closing process of the second preset threshold intelligent sliding door to be detected, the abnormal vibration data and the standard vibration data sequence are obtained.

[0026] Step S002, record any vibration data sequence of the intelligent sliding door to be detected in the door opening and closing process as the first target vibration data sequence, determine the abnormal possibility of the first target vibration data sequence based on the difference between the first target vibration data sequence before and after denoising, and the difference between the first target vibration data sequence and the standard vibration data sequence, and obtain the abnormal possibility of the vibration data sequence of the door opening and closing process in the historical time.

[0027] When there is no abnormality in the smart sliding door, since the smart sliding door opens and closes automatically, without the action of external force, the vibration data collected when the smart sliding door opens and closes at different times are similar. However, an abnormal smart sliding door will introduce more noise to the collected vibration data, causing a difference between the vibration data collected from the abnormal smart sliding door and the vibration data collected from the normal smart sliding door.

[0028] Any vibration data sequence of the intelligent sliding door to be detected during the door opening and closing process is recorded as the first target vibration data sequence.

[0029] The possibility of abnormality of the first target vibration data sequence is determined based on the difference between the first target vibration data sequence before and after denoising, and the difference between the first target vibration data sequence and the standard vibration data sequence.

[0030] The first target vibration data sequence is denoised using a mean filter algorithm to obtain a denoised first target vibration data sequence. The sum of the absolute values ​​of the differences between the first target vibration data sequence and the corresponding vibration data in the denoised first target vibration data sequence is recorded as the noise influence of the first target vibration data sequence; the DTW distance between the first target vibration data sequence and the standard vibration data sequence is recorded as the standard deviation of the first target vibration data sequence; and the normalized value of the product of the noise influence of the first target vibration data sequence and the standard deviation is recorded as the abnormal possibility of the first target vibration data sequence.

[0031] Among them, using the mean filter algorithm to denoise the first target vibration data sequence is a well-known technology and will not be repeated here. Calculating the DTW distance between the first target vibration data sequence and the standard vibration data sequence is a well-known technology and will not be repeated here. It should be noted that this embodiment uses the Z-Score standard normalization method to calculate the normalized value. In actual application, the implementer can use other methods of the prior art such as the maximum and minimum normalization method, the sigmoid function, etc. to calculate the normalized value, which is not limited here.

[0032] It can be understood that the greater the difference between the first target vibration data sequence before and after denoising, the greater the possibility that the vibration data contained in the first target vibration data sequence introduces more noise, and the greater the possibility that the first target vibration data sequence corresponds to an abnormal smart sliding door. At this time, the noise influence of the first target vibration data sequence is greater. At the same time, when the difference between the first target vibration data sequence and the standard vibration data sequence is greater, the greater the possibility that the first target vibration data sequence corresponds to an abnormal smart sliding door. At this time, the greater the standard difference of the first target vibration data sequence, the greater the possibility of abnormality of the first target vibration data sequence.

[0033] The abnormal possibility acquisition flow chart is as follows Figure 2 shown.

[0034] The same method can be used to obtain the abnormal possibility of any vibration data sequence of the intelligent sliding door to be detected during the door opening and closing process.

[0035] At this point, the abnormal possibilities of all vibration data sequences of the intelligent sliding door to be detected during the door opening and closing process are obtained.

[0036] According to the same method, the abnormal possibility of the vibration data sequence of the door opening and closing process in the historical time is obtained. Specifically, the abnormal possibility of the vibration data sequence of the door opening and closing process in the historical time is obtained by: Any vibration data sequence of the door opening and closing process within the historical time is recorded as the second target vibration data sequence.

[0037] The possibility of abnormality of the second target vibration data sequence is determined based on the difference between the second target vibration data sequence before and after denoising, and the difference between the second target vibration data sequence and the standard vibration data sequence.

[0038] The second target vibration data sequence is denoised using a mean filter algorithm to obtain a denoised second target vibration data sequence. The sum of the absolute values ​​of the differences between the second target vibration data sequence and the corresponding vibration data in the denoised second target vibration data sequence is recorded as the noise influence of the second target vibration data sequence; the DTW distance between the second target vibration data sequence and the standard vibration data sequence is recorded as the standard deviation of the second target vibration data sequence; and the normalized value of the product of the noise influence of the second target vibration data sequence and the standard deviation is recorded as the abnormal possibility of the second target vibration data sequence.

[0039] At this point, the abnormal possibilities of all vibration data sequences of the intelligent sliding door to be detected during the door opening and closing process and the door opening and closing process within the historical time are obtained.

[0040] Step S003, based on the time spent on door opening and closing corresponding to the vibration data sequence containing abnormal vibration data in the vibration data sequence of the door opening and closing process within the historical time, obtain the time abnormality correlation of the vibration data sequence of the door opening and closing process within the historical time, obtain the time abnormality correlation of the first target vibration data sequence, determine the door opening and closing abnormality of the vibration data sequence according to the time spent on door opening and closing corresponding to all vibration data sequences of the door opening and closing process within the historical time and all vibration data sequences of the intelligent sliding door to be detected in the door opening and closing process, perform weighted summation of the abnormal possibility and the door opening and closing abnormality of the vibration data sequence according to the time abnormality correlation of the vibration data sequence, and obtain the abnormal sound evaluation value of the vibration data sequence.

[0041] According to the time spent on opening and closing the door corresponding to the vibration data sequence containing abnormal vibration data in the vibration data sequence of the door opening and closing process in the historical time, the correlation between the values ​​of the door opening time and the door closing time and the abnormality of the intelligent sliding door is evaluated, and the time abnormality correlation of the vibration data sequence of the door opening and closing process in the historical time is obtained. The expression of the time abnormality correlation is: In the formula, Represents the time anomaly correlation of the vibration data series of the door opening and closing process in the historical time; represents the normalization function; Indicates the number of vibration data sequences containing abnormal vibration data in the vibration data sequences of the door opening and closing process in the historical time; Indicates the number of abnormal vibration data in the vibration data sequence of the door opening and closing process in the historical time. The time it takes to open the door corresponding to each vibration data sequence; Represents the mean time spent on door opening corresponding to the vibration data sequence of all door opening and closing processes in the historical time; Indicates the number of abnormal vibration data in the vibration data sequence of the door opening and closing process in the historical time. The time it takes to close the door corresponding to each vibration data sequence; Represents the mean of the time taken to close the door corresponding to the vibration data sequence of all door opening and closing processes in the historical time; The standard deviation of the time it takes to open the door corresponding to all vibration data sequences of the door opening and closing process in the historical time; Represents the standard deviation of the time it takes to close the door corresponding to all vibration data sequences of the door opening and closing process in the historical time.

[0042] The smaller the difference between the door opening times and the door closing times during the door opening and closing process of the abnormal smart sliding door is, the stronger the consistency of the door opening time and the door closing time of the abnormal smart sliding door during the door opening and closing process is. At the same time, when the difference between the door opening time and the time spent on all door openings during the door opening and closing process of the abnormal smart sliding door is greater, and the difference between the door closing time and the time spent on all door closings during the door opening and closing process of the abnormal smart sliding door is greater, the correlation between the abnormal smart sliding door and the values ​​of the door opening time and the door closing time is stronger. At this time, the greater the time anomaly correlation is, and the more obvious the correlation characteristics between the abnormal smart sliding door and the door opening time and the door closing time are.

[0043] The mean value of the time anomaly correlation of the vibration data sequence of the door opening and closing process in all historical time is recorded as the time anomaly correlation of the first target vibration data sequence.

[0044] At this point, the time anomaly correlation of all vibration data sequences of the intelligent sliding door to be detected during the door opening and closing process is obtained.

[0045] The LOF anomaly detection algorithm is used for the door opening time corresponding to all vibration data sequences of the door opening and closing processes in the historical time and the door opening and closing time corresponding to all vibration data sequences of the intelligent sliding door to be detected, and the LOF value corresponding to the door opening time corresponding to each vibration data sequence is obtained; the LOF anomaly detection algorithm is used for the door closing time corresponding to all vibration data sequences of the door opening and closing processes in the historical time and the door opening and closing time corresponding to all vibration data sequences of the intelligent sliding door to be detected, and the LOF value corresponding to the door closing time corresponding to each vibration data sequence is obtained; the average of the LOF values ​​corresponding to the door opening and closing time corresponding to the vibration data sequence is recorded as the door opening and closing abnormality of the vibration data sequence.

[0046] Using the LOF anomaly detection algorithm to obtain the LOF value is a well-known technique and will not be described in detail.

[0047] According to the time anomaly correlation of the vibration data sequence, the abnormal possibility of the vibration data sequence and the degree of abnormality of the door opening and closing are weighted and summed to obtain the abnormal sound evaluation value of the vibration data sequence. The method for obtaining the abnormal sound evaluation value of the vibration data sequence is as follows: in, Indicates the abnormal sound evaluation value of the vibration data sequence; Represents the time anomaly correlation of vibration data series; Indicates the degree of abnormality of the door opening and closing of the vibration data sequence; Indicates the likelihood that the vibration data series is abnormal.

[0048] When the abnormal possibility of the vibration data sequence and the degree of abnormality of the door opening and closing are greater, the abnormal sound evaluation value of the vibration data sequence is greater, and the intelligent sliding door corresponding to the vibration data sequence is more likely to have abnormal sound during the collection time period corresponding to the vibration data sequence.

[0049] At this point, the abnormal noise evaluation value of the vibration data sequence is obtained.

[0050] Step S004, judging whether the intelligent sliding door to be detected has abnormal sound according to the abnormal sound evaluation value of the vibration data sequence.

[0051] The abnormal sound evaluation values ​​of all vibration data sequences are processed using the maximum inter-class variance method to obtain an adaptive division threshold. The vibration data sequences with abnormal sound evaluation values ​​greater than the adaptive division threshold are recorded as abnormal vibration data sequences.

[0052] It can be understood that the smart sliding door corresponding to the abnormal vibration data sequence is a smart sliding door that has produced abnormal noise.

[0053] If the vibration data sequence of the intelligent sliding door to be detected during the door opening and closing process does not include an abnormal vibration data sequence, it is determined that the intelligent sliding door to be detected does not have an abnormal sound.

[0054] If a normal vibration data sequence appears in the vibration data sequence of the intelligent sliding door to be detected during the door opening and closing process, it is determined that the intelligent sliding door to be detected has an abnormal sound, and the intelligent sliding door needs to be repaired in time.

[0055] At this point, the abnormal noise detection of intelligent sliding door is realized.

[0056] Based on the same inventive concept as the above method, an embodiment of the present invention also provides an intelligent sliding door abnormal sound detection system, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned intelligent sliding door abnormal sound detection methods are implemented.

[0057] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting abnormal noise of an intelligent sliding door, characterized in that: The method comprises the following steps: Collect vibration data of the intelligent sliding door in the door opening and closing process in historical time, vibration data of the intelligent sliding door to be detected in the door opening and closing process, and the time spent on each opening and closing of the intelligent sliding door, obtain a first preset threshold number of vibration data sequences of the door opening and closing process in historical time, and a second preset threshold number of vibration data sequences of the intelligent sliding door to be detected in the door opening and closing process, obtain abnormal vibration data in each vibration data sequence according to the vibration data sequence of the door opening and closing process in historical time, and mark the standard vibration data sequence; Any vibration data sequence of the intelligent sliding door to be detected in the process of opening and closing the door is recorded as the first target vibration data sequence, and the abnormal possibility of the first target vibration data sequence is determined according to the difference between the first target vibration data sequence before and after denoising, and the difference between the first target vibration data sequence and the standard vibration data sequence, so as to obtain the abnormal possibility of the vibration data sequence in the process of opening and closing the door in the historical time; According to the time spent on door opening and closing corresponding to the vibration data sequence containing abnormal vibration data in the vibration data sequence of the door opening and closing process in the historical time, the time anomaly correlation of the vibration data sequence of the door opening and closing process in the historical time is obtained, and the time anomaly correlation of the first target vibration data sequence is obtained. According to the time spent on door opening and closing corresponding to all vibration data sequences of the door opening and closing process in the historical time and all vibration data sequences of the intelligent sliding door to be detected in the door opening and closing process, the door opening and closing abnormality of the vibration data sequence is determined, and the abnormal possibility and the door opening and closing abnormality of the vibration data sequence are weightedly summed according to the time anomaly correlation of the vibration data sequence to obtain the abnormal sound evaluation value of the vibration data sequence; According to the abnormal sound evaluation value of the vibration data sequence, it is determined whether the intelligent sliding door to be tested has abnormal sound.

2. The method for detecting abnormal noise of an intelligent sliding door according to claim 1, characterized in that: The specific method for obtaining the vibration data sequence of the door opening and closing process in the historical time and the vibration data sequence of the intelligent sliding door to be detected in the door opening and closing process is: Arrange all vibration data collected during the period from when the smart sliding door receives the door opening signal to when the smart sliding door completes the door closing in the historical time in the order of collection time to obtain a vibration data sequence of the door opening and closing process in the historical time; All vibration data collected from the time when the intelligent sliding door to be detected receives the door opening signal until the intelligent sliding door completes the door closing process are arranged in the order of collection time to obtain the vibration data sequence of the intelligent sliding door to be detected.

3. The method for detecting abnormal noise of an intelligent sliding door according to claim 1, characterized in that: The method for obtaining the abnormal possibility of the first target vibration data sequence is: De-noising the first target vibration data sequence using a mean filter algorithm to obtain a de-noised first target vibration data sequence; The accumulated sum of the absolute values ​​of the differences between the first target vibration data sequence and the corresponding vibration data in the denoised first target vibration data sequence is recorded as the noise influence degree of the first target vibration data sequence; The DTW distance between the first target vibration data sequence and the standard vibration data sequence is recorded as the standard difference of the first target vibration data sequence; The possibility of abnormality of the first target vibration data sequence is determined according to the noise influence degree and the standard difference of the first target vibration data sequence.

4. The method for detecting abnormal noise of an intelligent sliding door according to claim 3, characterized in that: The method of determining the abnormal possibility of the first target vibration data sequence according to the noise influence degree and the standard difference of the first target vibration data sequence includes the following specific methods: A normalized value of the product of the noise influence degree and the standard deviation of the first target vibration data sequence is recorded as the abnormal possibility of the first target vibration data sequence.

5. The method for detecting abnormal noise of an intelligent sliding door according to claim 1, characterized in that: The calculation formula for the time anomaly correlation of the vibration data sequence of the door opening and closing process in the historical time is: In the formula, Represents the time anomaly correlation of the vibration data series of the door opening and closing process in the historical time; represents the normalization function; Indicates the number of vibration data sequences containing abnormal vibration data in the vibration data sequences of the door opening and closing process in the historical time; Indicates the number of abnormal vibration data in the vibration data sequence of the door opening and closing process in the historical time. The time it takes to open the door corresponding to each vibration data sequence; Represents the mean time spent on door opening corresponding to the vibration data sequence of all door opening and closing processes in the historical time; Indicates the number of abnormal vibration data in the vibration data sequence of the door opening and closing process in the historical time. The time it takes to close the door corresponding to each vibration data sequence; Represents the mean of the time taken to close the door corresponding to the vibration data sequence of all door opening and closing processes in the historical time; Represents the standard deviation of the time it takes to open the door corresponding to all vibration data sequences of the door opening and closing process in the historical time; Represents the standard deviation of the time it takes to close the door corresponding to all vibration data sequences of the door opening and closing process in the historical time.

6. The method for detecting abnormal noise of an intelligent sliding door according to claim 1, characterized in that: The method of determining the door opening and closing abnormality of the vibration data sequence according to all vibration data sequences of the door opening and closing process in the historical time and the door opening and closing time corresponding to all vibration data sequences of the intelligent sliding door to be detected in the door opening and closing process includes the following specific methods: The LOF anomaly detection algorithm is used for all vibration data sequences of the door opening and closing process in the historical time and the door opening time corresponding to all vibration data sequences of the intelligent sliding door to be detected in the door opening and closing process, and the LOF value corresponding to the door opening time corresponding to each vibration data sequence is obtained; The LOF anomaly detection algorithm is used for all vibration data sequences of the door opening and closing process in the historical time and the door closing time corresponding to all vibration data sequences of the intelligent sliding door to be detected in the door opening and closing process, and the LOF value corresponding to the door closing time corresponding to each vibration data sequence is obtained; The door opening and closing abnormality of the vibration data sequence is determined according to the LOF value corresponding to the time spent on door opening and closing corresponding to the vibration data sequence.

7. The method for detecting abnormal noise of an intelligent sliding door according to claim 6, characterized in that: The specific method of determining the door opening and closing abnormality of the vibration data sequence according to the LOF value corresponding to the time spent on door opening and closing corresponding to the vibration data sequence is as follows: The average of the LOF values ​​corresponding to the time spent on door opening and door closing corresponding to the vibration data sequence is recorded as the door opening and closing abnormality degree of the vibration data sequence.

8. The method for detecting abnormal noise of an intelligent sliding door according to claim 1, characterized in that: The method for obtaining the abnormal sound evaluation value of the vibration data sequence is as follows: in, Indicates the abnormal sound evaluation value of the vibration data sequence; Represents the time anomaly correlation of vibration data series; Indicates the degree of abnormality of the door opening and closing of the vibration data sequence; Indicates the likelihood that the vibration data series is abnormal.

9. The method for detecting abnormal noise of an intelligent sliding door according to claim 1, characterized in that: The specific method of judging whether the intelligent sliding door to be detected has abnormal sound according to the abnormal sound evaluation value of the vibration data sequence is as follows: The abnormal sound evaluation values ​​of all vibration data sequences are processed using the maximum inter-class variance method to obtain an adaptive division threshold, and the vibration data sequences with abnormal sound evaluation values ​​greater than the adaptive division threshold are recorded as abnormal vibration data sequences; If the vibration data sequence of the intelligent sliding door to be detected during the door opening and closing process does not contain an abnormal vibration data sequence, it is determined that the intelligent sliding door to be detected does not have an abnormal sound; If a normal vibration data sequence appears in the vibration data sequence of the intelligent sliding door to be detected during the door opening and closing process, it is determined that the intelligent sliding door to be detected has an abnormal sound.

10. An intelligent sliding door abnormal sound detection system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

Images