CN116312632A - Abnormal sound detection system and device of unmanned vending terminal - Google Patents
Abnormal sound detection system and device of unmanned vending terminal Download PDFInfo
- Publication number
- CN116312632A CN116312632A CN202310160612.6A CN202310160612A CN116312632A CN 116312632 A CN116312632 A CN 116312632A CN 202310160612 A CN202310160612 A CN 202310160612A CN 116312632 A CN116312632 A CN 116312632A
- Authority
- CN
- China
- Prior art keywords
- abnormal sound
- sound
- sound detection
- unmanned
- unmanned vending
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 94
- 238000001514 detection method Methods 0.000 title claims abstract description 79
- 230000009467 reduction Effects 0.000 claims abstract description 13
- 238000000034 method Methods 0.000 claims description 20
- 230000017525 heat dissipation Effects 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 4
- 238000003780 insertion Methods 0.000 claims description 4
- 230000037431 insertion Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000012423 maintenance Methods 0.000 abstract description 27
- 238000007689 inspection Methods 0.000 abstract description 5
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 19
- 238000004364 calculation method Methods 0.000 description 15
- 238000009826 distribution Methods 0.000 description 12
- 238000007726 management method Methods 0.000 description 12
- 238000010586 diagram Methods 0.000 description 11
- 230000003449 preventive effect Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 7
- 238000009434 installation Methods 0.000 description 6
- 238000012549 training Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000012384 transportation and delivery Methods 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 239000000428 dust Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000009432 framing Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000007787 long-term memory Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 230000015654 memory Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
- 238000005303 weighing Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F11/00—Coin-freed apparatus for dispensing, or the like, discrete articles
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F9/00—Details other than those peculiar to special kinds or types of apparatus
- G07F9/10—Casings or parts thereof, e.g. with means for heating or cooling
- G07F9/105—Heating or cooling means, for temperature and humidity control, for the conditioning of articles and their storage
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
- G10L25/30—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Computational Linguistics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Quality & Reliability (AREA)
- Control Of Vending Devices And Auxiliary Devices For Vending Devices (AREA)
Abstract
The invention relates to the technical field of unmanned retail, and discloses an abnormal sound detection system and device of an unmanned vending terminal. The automatic monitoring system is provided with the acquisition module, the noise reduction module, the detection module and the abnormal sound detection cloud platform, so that the timing or real-time maintenance of the unmanned vending terminal equipment is realized, the timeliness of finding faults caused by manual inspection and maintenance is avoided, the potential faults of the equipment are ensured to be processed in time, the running efficiency of the machine is improved, and the use experience of a user is improved.
Description
Technical Field
The invention relates to the technical field of unmanned retail, in particular to an abnormal sound detection system and device of an unmanned vending terminal.
Background
At present, in the field of unmanned retail, unmanned vending terminal equipment at home and abroad mainly relies on manual detection and maintenance. The workers are generally required to carry out routine inspection on key components such as a shipment port, a currency entrance and exit, a condenser and the like when carrying out shipment, and the requirements on the technical level, responsibility and the like of the workers are very high, so that the discovery of machine faults is often quite untimely. Some scale enterprises also employ preventative maintenance (Preventive Maintenance, pvM), also known as scheduled maintenance or time-based maintenance, i.e., maintenance is performed periodically on all of their end devices according to a scheduled schedule to predict potential machine failure and to timely provide early troubleshooting. Enterprises increase the efficiency of equipment by reducing malfunctions, sometimes resulting in unnecessary maintenance, thereby increasing operating costs. In most cases, the unmanned vending terminal equipment suddenly fails, such as a delivery channel is blocked, a coin inlet and outlet is abnormal, and after a user complains by telephone, relevant personnel arrive at the site to carry out rush repair and maintenance. This directly affects the normal use and sales of goods for the user, and the problem is often not handled at the first time, the user experience is poor. It can be seen that the conventional machine maintenance method and manner have failed to meet the actual maintenance requirements of the unmanned vending terminal.
With the development of artificial intelligence technologies such as machine learning, anomaly detection, voice recognition and the like, the technology of detecting abnormal sounds is increasingly mature, machine fault detection is carried out at an early stage based on historical data, and possible machine faults are pre-warned, so that maintenance personnel are reminded or assigned to carry out advanced treatment, and the technology becomes an important technical means for preventive maintenance of machines. The abnormal sound detection means that the neural network model is trained through the sound of machine operation, the trained model is used for predicting the sound of machine operation, whether the machine is normally operated is judged, the fault detection is carried out on the equipment in real time and at regular time, and the predictive maintenance is realized, so that the preventive maintenance on the equipment is enhanced.
The abnormal sound detection technology can effectively monitor whether the machine is normally transported or not, and is focused by researchers, so that the abnormal sound detection of the machine has become a current research hotspot. The abnormal sound detection of the machine is realized by collecting and predicting the sound of each part of the unmanned vending terminal equipment and supervising the running state of the machine, so that the timing or real-time maintenance of the unmanned vending terminal equipment is realized, the timeliness of finding faults caused by manual inspection and maintenance is avoided, the potential faults of the equipment are ensured to be processed in time, the running efficiency of the machine is improved, and the use experience of a user is improved.
At present, the unmanned vending terminal equipment often adopts a weighing sensor and an image sensor to collect commodity images, senses weight change, realizes commodity identification and face recognition through target detection, realizes commodity sales, and relates to few aspects of sound technology, and the preventive maintenance of the unmanned vending terminal equipment is suitable for applying an abnormal sound detection technology. Accordingly, the present invention has been made in view of the above circumstances, and an object of the present invention is to provide an abnormal sound detection system and apparatus for an unmanned vending terminal, which can achieve a more practical value.
Disclosure of Invention
In order to overcome the above-mentioned drawbacks of the prior art, the present invention provides an abnormal sound detection system and device for an unmanned vending terminal, so as to solve the above-mentioned problems in the prior art.
The invention provides the following technical scheme: the abnormal sound detection system of the unmanned vending terminal comprises an acquisition module, a noise reduction module, a detection module and an abnormal sound detection cloud platform, wherein the acquisition module acquires sound data of key components in the unmanned vending terminal equipment during working and processes and classifies the data, the noise reduction module uses a noise reduction algorithm to reduce noise before feature extraction and preprocesses the noise, the detection module uses an outlier method and LSTM, an abnormal sound detection model is trained under the condition that the abnormal sound is not used, the abnormal sound detection model is used, the network structure and the algorithm are changed under the condition that the abnormal sound is used, the equipment management function in the abnormal sound detection cloud platform checks the number, the positions, the running states and the like of sound sensors in the equipment, the starting time, the working plan and the like of the sound sensors in the equipment are adjusted, and authorized advanced users can manage the sound acquired by the equipment through a file management function and adjust and configure the detection model of the sensors through a model management function.
The utility model provides an unusual sound detecting system of unmanned terminal of selling goods, includes the device main part, the device main part includes data processing platform, unmanned sales counter body and sound sensor body, be provided with shipment passageway body, coin return passageway and goods shelves operation passageway on the unmanned sales counter body, shipment passageway body, coin return passageway and goods shelves operation passageway's inside all is provided with sound sensor body, sound sensor body includes the circuit board, be provided with radio receiver, potentiometre, transmission end and pilot lamp on the circuit board.
Furthermore, the top end of the self-service cabinet body is provided with a heat dissipation channel, and the heat dissipation channel is detachably installed.
Further, the front of unmanned sales counter body is provided with the spirit level, the outer wall bottom of unmanned sales counter body is provided with the leveling base, just the quantity of leveling base is four, all be connected with the lead screw body through the thread groove screw on the leveling base, the top of lead screw body is all fixed mounting has the hand wheel, the bottom of lead screw body is all connected with the support base through the bearing, the bottom of support base is all fixed mounting has the anti-skidding base.
Further, the dust screen is detachably arranged on one side of the radio, and in actual use, the service life of the sound sensor body can be effectively prolonged through the arrangement of the dust screen.
Further, the sound sensor body still includes the mounting base, the quantity of mounting base is two, the mounting base set up in the bottom of circuit board, the bottom fixed mounting of circuit board has two to insert the installation piece, the bottom of inserting the installation piece is all fixed mounting has the elasticity block, just elasticity block with mounting base looks adaptation, in the in-service use, through foretell setting, be convenient for follow-up maintenance work to the sound sensor body.
The invention has the technical effects and advantages that: the automatic monitoring system is provided with the acquisition module, the noise reduction module, the detection module and the abnormal sound detection cloud platform, so that the timing or real-time maintenance of the unmanned vending terminal equipment is realized, the timeliness of finding faults caused by manual inspection and maintenance is avoided, the potential faults of the equipment are ensured to be processed in time, the running efficiency of the machine is improved, and the use experience of a user is improved.
Drawings
FIG. 1 is a schematic diagram of the overall research content architecture of the present invention.
FIG. 2 is a schematic diagram of the general technical route of the present invention.
FIG. 3 is a schematic diagram of an LSTM cell of the present invention.
FIG. 4 is a schematic diagram of the LSTM and self-encoder fusion model of the present invention.
FIG. 5 is a schematic diagram of outlier data definition according to the present invention.
Fig. 6 is a schematic diagram of an unsupervised abnormal sound detection learning technique route according to the present invention.
Fig. 7 is a schematic diagram of an improved model structure considering abnormal sound distribution according to the present invention.
Fig. 8 is a schematic diagram of an anomaly score calculation process according to the present invention.
Fig. 9 is a schematic diagram of an unsupervised abnormal sound detection method involving abnormal sound according to the present invention.
Fig. 10 is a schematic diagram of a development route of the abnormal sound detection cloud platform system according to the present invention.
Fig. 11 is a schematic diagram of a sound collection and abnormal sound detection cloud platform scheme according to the present invention.
Fig. 12 is a schematic perspective view of a first structure of the acoustic sensor body of the present invention.
FIG. 13 is a schematic perspective view of a first embodiment of the self-service vending cabinet body of the present invention.
FIG. 14 is a schematic perspective view of a second embodiment of the self-service vending cabinet body of the present invention.
Fig. 15 is a schematic perspective view of a second structure of the acoustic sensor body of the present invention.
Fig. 16 is a schematic perspective view of a third structure of the acoustic sensor body of the present invention.
The reference numerals are: 100. a device body; 110. an unmanned sales counter body; 111. a shipment channel body; 112. coin-feed passage; 113. a shelf running channel; 114. a sound sensor body; 115. a circuit board; 116. a radio; 117. a potentiometer; 118. a transmission end; 119. an indicator light; 120. a mounting base; 121. inserting a mounting block; 122. an elastic clamping block; 123. a dust-proof gauze; 124. a level gauge; 125. leveling the base; 126. a screw rod body; 127. a support base pad; 128. and a heat dissipation channel.
Detailed Description
The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings.
Embodiment one: the invention provides an abnormal sound detection system of an unmanned vending terminal and a device thereof, the abnormal sound detection system of the unmanned vending terminal comprises an acquisition module, a noise reduction module, a detection module and an abnormal sound detection cloud platform, wherein the acquisition module acquires sound data of key components in the unmanned vending terminal during working and processes and classifies the data, the noise reduction module uses a noise reduction algorithm to reduce noise before feature extraction and preprocesses the sound, the detection module uses an outlier method and LSTM, trains an abnormal sound detection model under the condition of no abnormal sound, carries out abnormal sound detection by using the model, changes a network structure and the algorithm under the condition of abnormal sound, checks the number, the position, the running state and the like of sound sensors in equipment in the abnormal sound detection cloud platform, adjusts the starting time, the working schedule and the like of the sound sensors in the equipment, and an authorized advanced user can manage the sound acquired by the equipment through a file management function and adjusts and configures the detection model of the sensors through the model management function.
The overall framework of the project is shown in figure 1. And training normal operation sound by using unsupervised learning in the initial stage of equipment quality inspection and equipment deployment, and performing preventive maintenance on the equipment by using an unsupervised abnormal sound detection technology. When abnormality is detected, the abnormal sound is uploaded to the platform to serve as a sample of the abnormal sound, and the collected abnormal sound sample is used for optimizing a network model and an algorithm in the later period, so that the accuracy of abnormal sound detection is improved.
The specific research and development content of the project is as follows:
(1) In the initial stage of the deployment of the unmanned vending terminal equipment, because the machine is deployed in different occasions and lacks the audio data of the abnormal sound, an abnormal sound detection algorithm based on the unsupervised learning is constructed for the initial stage of the deployment of the equipment. Since the processes of delivering and inserting coins are fixed, an LSTM network is introduced into the model to learn the sound characteristics on the time sequence, and the detection accuracy of the sound events with the sequence is improved. Aiming at different machine deployment places, a noise reduction method is used for improving the extraction accuracy of the sound features.
(2) After a certain abnormal sound data is collected at the later stage of the deployment of the unmanned vending terminal equipment, a detection network model and an algorithm are changed, and the collected abnormal sound data also participates in model training. A mixed gaussian distribution can be used to distinguish between normal and abnormal distributions and to add probabilities to the loss function. And introducing a similarity calculation function, performing similarity calculation on the sound to be detected and the collected abnormal sound, and adding the similarity calculation into the abnormal score calculation. The method can increase the distinguishing degree of the abnormal sound and the normal sound, so that the abnormal sound detection is more effective.
(3) The cloud platform system for detecting the abnormal sound is designed and developed, so that preventive maintenance is conveniently carried out on the accessed unmanned vending terminal equipment. The number, the position, the running state and the like of the sound sensors in the equipment are checked through the equipment management function of the cloud platform, and the starting time, the work plan and the like of the sound sensors in the equipment are adjusted. The authorized advanced user can manage the sound collected by the equipment through the file management function, and adjust and configure the detection model of the sensor through the model management function.
The general technical route adopted by project study is as follows: 1. and collecting sound data of key components in the unmanned vending terminal equipment during working, and processing and classifying the data. 2. Noise is reduced and pre-processed using a noise reduction algorithm prior to feature extraction. 3. Using the outlier method and LSTM, an abnormal sound detection model is trained without using abnormal sounds, and abnormal sound detection is performed using the model. 4. In the case of using an abnormal sound, the network structure and algorithm are changed. The overall route of the project study is shown in figure 2.
(4) A depth self-encoder is used as a basis model for the neural network.
Based on LSTM, time sequence characteristics in the learning sound characteristics are enhanced, and the expressive power of the model is enhanced. The LSTM cell structure is shown in FIG. 3, and its core structure is the red frame inside, called the cell state.
Wherein C is t The calculation method is as follows:
wherein f t Is a forgetful door, and represents C t-1 Features for calculating C t The weights of (2) are generally calculated by a sigmoid activation function in the following manner:
f t ＝σ(W f ·[h t-1 ,x t ]+b
i t ＝σ(W i ·[h t-1 ,x t ]+b i )
finally leading to h of the next layer t From the output gate o t And cell state C t The calculation method is as follows:
o t ＝σ(W o ·[h t-1 ,x t ]+b o )
h t ＝o t *tanh(C t )
thus, each layer can trade off the learned features by calculation. Firstly, unnecessary memory is removed through a forgetting gate, then learning features needing to be supplemented are determined through an input gate, and state updating values are used for selective learning, and finally, the selectivity is determined through an output gate and the transferred features are selected, so that correction and transfer of long-term memory are realized.
The network structure combining LSTM with self-encoder is shown in fig. 4.
The encoder reads the audio sequence where the input may be a feature of each frame after framing in the MFCC features and the reconstructed output is an inverted sequence of inputs in order to make the optimization easier.
The outlier is defined as another type of sound independent of the operation sound of the unmanned vending terminal, and is represented as shown in fig. 5. The outliers may help determine a threshold for abnormal sound detection.
The depth self-encoder is adjusted and compared, and the comparison comprises evaluation indexes AUC, pAUC, training time, test time and the like.
Non-supervision abnormal sound detection method for abnormal sound participation
Calculating the distribution of abnormal sound, training the distribution of normal sound, eliminating the abnormal sound distribution from the normal sound distribution, and taking the posterior distribution as one of the loss functions during training. The modified structure is shown in fig. 7.
And adding a similarity calculation function, and adding the similarity between the sample to be detected and the abnormal sample into calculation when the abnormal score calculation is performed. The difference between the score of the normal sound and the score of the abnormal sound is increased. The anomaly score calculation process is shown in fig. 8.
The overall process route is shown in figure 9.
Abnormal sound detection cloud platform system
The equipment management function module comprises the configuration of the number and the position of sound sensors installed in specific equipment, and the feedback of the starting time of the sensors, the management of a work plan and the state.
And the sound file management function module is used for uniformly managing the sound files according to the places where the equipment is deployed and the equipment types, and machines with the same types share sound data under similar scenes.
The model management module uses the LSTM self-encoder network model for devices initially deployed in the same class of scene and modifies the threshold by providing outlier data. And the learning method is changed in the later period, so that the collected abnormal sound participates in the learning. Advanced users can flexibly configure more suitable network models.
The abnormal sound detection platform development route is shown in fig. 10.
The new generation of unmanned retail terminal equipment supporting preventive maintenance based on abnormal sound detection is manufactured through the project, and a specific sound collection and abnormal sound detection cloud platform scheme is shown in fig. 11.
Aiming at the characteristic that the sound of a key component working event in unmanned vending terminal equipment has time sequence, the LSTM structure and the encoder structure are fused, so that the time sequence characteristic of the sound is better learned, and meanwhile, the reverse sequence with output as input is reconstructed, so that model optimization is facilitated, and the accuracy and adaptability of abnormal sound detection are improved.
Aiming at the cold start problem of initial detection threshold selection of unmanned vending terminal equipment deployment, a novel method for confirming the threshold by calculating the anomaly score by using an outlier as a negative sample is provided. The method effectively solves the problems that the selected threshold value has poor correlation with the detection model and limited effect when no or a small amount of abnormal sound data exists in the initial equipment deployment stage. And unlike the past, the new method does not change training into two-classification problem, so that the method is still unsupervised learning.
For model modification in the later deployment period of unmanned vending terminal equipment, the acquired abnormal sound is used for intervening learning, gaussian distribution of abnormal sound samples is calculated, the distribution of normal sound of the abnormal sound sample distribution is trained and removed, and posterior distribution is used as one of loss functions. Besides optimizing the loss function, the similarity between the sample to be detected and the abnormal sample is calculated in detection and added to the calculation of the abnormal score, so that the abnormal score of the abnormal sample is increased, the difference of the abnormal scores between the positive sample and the negative sample is increased, and the detection accuracy is improved. At the same time, the abnormal sounds are not used as input to train the model and therefore remain unsupervised learning.
The abnormal sound detection cloud platform system which is flexibly configured is convenient for preventive maintenance of the accessed unmanned vending terminal equipment. The unmanned terminal equipment for vending needs a lot of parts of maintenance, selects reasonable sound sensor mounted position, installation quantity, and it is very important to confirm sound sensor's on-time, work plan etc.. Flexible configuration is realized through software definition, such as a sound sensor is arranged at the middle position of a delivery channel, a starting sensor is arranged at the middle position of a currency channel, and the like during delivery, coin insertion, banknote insertion and change. The collected sound is managed in the abnormal sound detection cloud platform system, equipment in similar scenes shares sound data, in addition, models used for abnormal sound detection can be managed, and sound sensors at different positions adopt different detection models, so that accurate detection is achieved.
Embodiment two:
the second embodiment differs from the first embodiment in that: an abnormal sound detection system of an unmanned vending terminal comprises a device main body 100, wherein the device main body 100 comprises a data processing platform, an unmanned vending cabinet body 110 and a sound sensor body 114, a goods outlet channel body 111, a coin-feed back channel 112 and a goods shelf operation channel 113 are arranged on the unmanned vending cabinet body 110, sound sensor bodies 114 are arranged in the goods outlet channel body 111, the coin-feed back channel 112 and the goods shelf operation channel 113, the sound sensor body 114 comprises a circuit board 115, a radio 116, a potentiometer 117, a transmission end 118 and an indicator 119 are arranged on the circuit board 115, a heat dissipation channel 128 is arranged at the top end of the unmanned vending cabinet body 110, the heat dissipation channel 128 is detachably arranged, a level meter 124 is arranged on the front surface of the unmanned vending cabinet body 110, the outer wall bottom of unmanned sales counter body 110 is provided with leveling base 125, and the quantity of leveling base 125 is four, all have lead screw body 126 through thread groove threaded connection on the leveling base 125, the top of lead screw body 126 is all fixed mounting has the hand wheel, the bottom of lead screw body 126 is all connected with support collet 127 through the bearing, the bottom of support collet 127 is all fixed mounting has anti-skidding collet, one side demountable installation of radio 116 has dustproof gauze 123, sound sensor body 114 still includes mounting base 120, the quantity of mounting base 120 is two, mounting base 120 sets up in the bottom of circuit board 115, the bottom fixed mounting of circuit board 115 has two inserts installation piece 121, the bottom of inserting installation piece 121 is all fixed mounting has elasticity block 122, and elasticity block 122 and mounting base 120 looks adaptation.
Claims (6)
1. The abnormal sound detection system of the unmanned vending terminal comprises a collection module, a noise reduction module, a detection module and an abnormal sound detection cloud platform and is characterized in that the collection module collects sound data of key components in the unmanned vending terminal equipment during operation, processes and classifies the data, the noise reduction module uses a noise reduction algorithm to reduce noise before feature extraction and preprocesses the noise, the detection module uses an outlier method and an LSTM, trains an abnormal sound detection model under the condition of not using the abnormal sound and uses the model to detect the abnormal sound, changes a network structure and the algorithm under the condition of using the abnormal sound, a device management function in the abnormal sound detection cloud platform checks the number, the position, the running state and the like of sound sensors in the device, adjusts the starting time, the working plan and the like of the sound sensors in the device, and an authorized advanced user can manage the sound collected by the device through a file management function and adjusts and configures the detection model of the sensors through a model management function.
2. The utility model provides an unusual sound detection device of unmanned terminal that sells goods, includes device main part (100), its characterized in that, device main part (100) include data processing platform, unmanned sales counter body (110) and sound sensor body (114), be provided with shipment passageway body (111), coin return passageway (112) and goods shelves operation passageway (113) on unmanned sales counter body (110), the inside of shipment passageway body (111), coin return passageway (112) and goods shelves operation passageway (113) all is provided with sound sensor body (114), sound sensor body (114) include circuit board (115), be provided with radio receiver (116), potentiometer (117), transmission end (118) and pilot lamp (119) on circuit board (115).
3. The abnormal sound detection apparatus of an unmanned vending terminal according to claim 2, wherein: the top of the self-service cabinet body (110) is provided with a heat dissipation channel (128), and the heat dissipation channel (128) is detachably installed.
4. The abnormal sound detection apparatus of an unmanned vending terminal according to claim 2, wherein: the utility model discloses a self-service counter, including self-service counter body (110), outer wall bottom of self-service counter body (110) is provided with spirit level (124), just the outer wall bottom of self-service counter body (110) is provided with leveling base (125), just the quantity of leveling base (125) is four, all have lead screw body (126) through screw thread groove threaded connection on leveling base (125), all fixed mounting has the hand wheel on the top of lead screw body (126), the bottom of lead screw body (126) all is connected with support base pad (127) through the bearing, the bottom of support base pad (127) all fixed mounting has anti-skidding base pad.
5. The abnormal sound detection apparatus of an unmanned vending terminal according to claim 2, wherein: a dustproof gauze (123) is detachably arranged on one side of the radio (116).
6. The abnormal sound detection apparatus of an unmanned vending terminal according to claim 2, wherein: the sound sensor body (114) further comprises two mounting bases (120), the number of the mounting bases (120) is two, the mounting bases (120) are arranged at the bottom end of the circuit board (115), two insertion mounting blocks (121) are fixedly mounted at the bottom end of the circuit board (115), elastic clamping blocks (122) are fixedly mounted at the bottom ends of the insertion mounting blocks (121), and the elastic clamping blocks (122) are matched with the mounting bases (120).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310160612.6A CN116312632A (en) | 2023-02-22 | 2023-02-22 | Abnormal sound detection system and device of unmanned vending terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310160612.6A CN116312632A (en) | 2023-02-22 | 2023-02-22 | Abnormal sound detection system and device of unmanned vending terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116312632A true CN116312632A (en) | 2023-06-23 |
Family
ID=86802358
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310160612.6A Pending CN116312632A (en) | 2023-02-22 | 2023-02-22 | Abnormal sound detection system and device of unmanned vending terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116312632A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117765670A (en) * | 2023-08-16 | 2024-03-26 | 广州市生基科技有限公司 | Control system of water oxygen bath machine |
-
2023
- 2023-02-22 CN CN202310160612.6A patent/CN116312632A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117765670A (en) * | 2023-08-16 | 2024-03-26 | 广州市生基科技有限公司 | Control system of water oxygen bath machine |
| CN117765670B (en) * | 2023-08-16 | 2024-05-17 | 广州市生基科技有限公司 | Control system of water oxygen bath machine |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110929918B (en) | 10kV feeder fault prediction method based on CNN and LightGBM | |
| CN100472711C (en) | Device for and method of creating a model for determining relationship between process and quality | |
| CN106204330A (en) | A kind of power distribution network intelligent diagnosis system | |
| CN109657718B (en) | Data-driven SPI defect type intelligent identification method on SMT production line | |
| CN110535159B (en) | Method and system for early warning of faults of operation units of large-scale energy storage power station | |
| CN112684301B (en) | Method and device for detecting power grid faults | |
| CN117277958B (en) | Intelligent operation and maintenance management method and system for photovoltaic power station based on big data | |
| CN110555477A (en) | municipal facility fault prediction method and device | |
| CN111412579A (en) | Air conditioning unit fault type diagnosis method and system based on big data | |
| CN116312632A (en) | Abnormal sound detection system and device of unmanned vending terminal | |
| CN116933197B (en) | Fault discrimination method and system for electricity consumption information acquisition system based on big data | |
| CN116028887B (en) | Analysis method of continuous industrial production data | |
| CN112346893A (en) | Fault prediction method, device, terminal and storage medium | |
| CN116956189A (en) | Current abnormality detection system, method, electronic equipment and medium | |
| CN107480824B (en) | Short-time passenger flow prediction system and method for urban rail transit station | |
| CN113448808A (en) | Method, system and storage medium for predicting single task time in batch processing task | |
| CN112926756A (en) | Application method of AI (AI) maintenance knowledge base of central air-conditioning equipment | |
| CN117728379A (en) | Intelligent operation scheduling method for regional power grid | |
| CN115361424A (en) | Medical equipment monitoring method and system based on data analysis of Internet of things | |
| CN113807462B (en) | Network equipment fault cause positioning method and system based on AI | |
| CN112801815A (en) | Power communication network fault early warning method based on federal learning | |
| CN109492913B (en) | Modular risk prediction method and device for power distribution terminal and storable medium | |
| CN115598459A (en) | Power failure prediction method for 10kV feeder line fault of power distribution network | |
| CN117808157B (en) | Intelligent identification-based unreported outage behavior prediction analysis system | |
| CN110569277A (en) | Method and system for automatically identifying and classifying configuration data information |
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 |