CN116517860A - Ventilator fault early warning system based on data analysis - Google Patents
Ventilator fault early warning system based on data analysis Download PDFInfo
- Publication number
- CN116517860A CN116517860A CN202310418606.6A CN202310418606A CN116517860A CN 116517860 A CN116517860 A CN 116517860A CN 202310418606 A CN202310418606 A CN 202310418606A CN 116517860 A CN116517860 A CN 116517860A
- Authority
- CN
- China
- Prior art keywords
- time
- ventilator
- frequency
- audio signal
- audio
- 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
- 238000007405 data analysis Methods 0.000 title claims abstract description 15
- 230000005236 sound signal Effects 0.000 claims abstract description 150
- 238000003745 diagnosis Methods 0.000 claims abstract description 68
- 230000005856 abnormality Effects 0.000 claims abstract description 57
- 238000004458 analytical method Methods 0.000 claims abstract description 24
- 238000001914 filtration Methods 0.000 claims abstract description 17
- 230000008447 perception Effects 0.000 claims abstract description 8
- 238000000034 method Methods 0.000 claims description 52
- 230000002159 abnormal effect Effects 0.000 claims description 41
- 230000008569 process Effects 0.000 claims description 30
- 238000009826 distribution Methods 0.000 claims description 26
- 238000012545 processing Methods 0.000 claims description 25
- 238000010586 diagram Methods 0.000 claims description 23
- 230000006870 function Effects 0.000 claims description 22
- 238000009432 framing Methods 0.000 claims description 16
- 230000009467 reduction Effects 0.000 claims description 16
- 238000013527 convolutional neural network Methods 0.000 claims description 15
- 230000008859 change Effects 0.000 claims description 14
- 238000003672 processing method Methods 0.000 claims description 12
- 238000007781 pre-processing Methods 0.000 claims description 11
- 238000012216 screening Methods 0.000 claims description 11
- 238000009423 ventilation Methods 0.000 claims description 9
- 238000005070 sampling Methods 0.000 claims description 8
- 238000012544 monitoring process Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 7
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 238000001228 spectrum Methods 0.000 claims description 4
- XAGFODPZIPBFFR-UHFFFAOYSA-N aluminium Chemical compound [Al] XAGFODPZIPBFFR-UHFFFAOYSA-N 0.000 claims description 3
- 229910052782 aluminium Inorganic materials 0.000 claims description 3
- 238000013135 deep learning Methods 0.000 claims description 3
- 238000004806 packaging method and process Methods 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 abstract description 5
- 230000006872 improvement Effects 0.000 abstract description 3
- 238000011176 pooling Methods 0.000 description 13
- 238000004364 calculation method Methods 0.000 description 6
- 238000001514 detection method Methods 0.000 description 5
- 230000004913 activation Effects 0.000 description 4
- 239000003245 coal Substances 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 238000005096 rolling process Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 239000002817 coal dust Substances 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- 239000000779 smoke Substances 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 239000007789 gas Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 231100000331 toxic Toxicity 0.000 description 1
- 230000002588 toxic effect Effects 0.000 description 1
- 239000002341 toxic gas Substances 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04D—NON-POSITIVE-DISPLACEMENT PUMPS
- F04D27/00—Control, e.g. regulation, of pumps, pumping installations or pumping systems specially adapted for elastic fluids
- F04D27/008—Stop safety or alarm devices, e.g. stop-and-go control; Disposition of check-valves
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21F—SAFETY DEVICES, TRANSPORT, FILLING-UP, RESCUE, VENTILATION, OR DRAINING IN OR OF MINES OR TUNNELS
- E21F1/00—Ventilation of mines or tunnels; Distribution of ventilating currents
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21F—SAFETY DEVICES, TRANSPORT, FILLING-UP, RESCUE, VENTILATION, OR DRAINING IN OR OF MINES OR TUNNELS
- E21F17/00—Methods or devices for use in mines or tunnels, not covered elsewhere
- E21F17/18—Special adaptations of signalling or alarm devices
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04D—NON-POSITIVE-DISPLACEMENT PUMPS
- F04D27/00—Control, e.g. regulation, of pumps, pumping installations or pumping systems specially adapted for elastic fluids
- F04D27/001—Testing thereof; Determination or simulation of flow characteristics; Stall or surge detection, e.g. condition monitoring
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2131—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on a transform domain processing, e.g. wavelet transform
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/72—Wind turbines with rotation axis in wind direction
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Mining & Mineral Resources (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Geology (AREA)
- Geochemistry & Mineralogy (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Mechanical Engineering (AREA)
- Biophysics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Probability & Statistics with Applications (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention discloses a ventilator fault early warning system based on data analysis, relates to the technical field of fault early warning, and is used for solving the problem that the existing ventilator fault early warning system has a large improvement space for the fault diagnosis of a mine ventilator; the source data perception module is used for acquiring and perceiving the operation audio signals of the ventilator and the surrounding environment data; the invention is based on audio signal analysis, combines the ventilator operation parameters to carry out omnibearing sensing on the operation state of the ventilator, carries an audio signal self-adaptive filtering model, a characteristic extraction and an abnormality diagnosis model, and ensures the high-efficiency and stable operation of the system.
Description
Technical Field
The invention relates to the technical field of fault early warning, in particular to a ventilator fault early warning system based on data analysis.
Background
Coal resources are one of the most important basic energy sources and are important factors for guaranteeing national energy safety. Along with the increase of the exploitation quantity of coal enterprises, the high-speed coal cutting improves the production yield, and the great floating of coal dust particles in the air in a mine tunnel is brought along with the increase, so that ventilation equipment is required to timely disperse toxic and harmful gases and coal dust in the mine, and the flow of fresh air in the mine is ensured.
Since the ventilation operation is required to be performed synchronously with the coal mining operation, the long-time stable operation of the ventilator must be ensured, which has high requirements on the health status of each key device in the ventilator. After long-time operation, the mine ventilator is easy to generate various faults, so that toxic gas which cannot be discharged underground seriously threatens the life safety of miners, and professional technicians are required to timely find hidden abnormality in ventilator equipment in the daily maintenance process.
However, the mine ventilator is always in the dynamic ventilation process, and the positions and the time of various faults in the operation process are all uncertain, so that the difficulty of fault detection is increased;
the common ventilator state fault monitoring system generally completes the operation state parameter sensing of the ventilator system by setting sensors such as vibration, pressure, temperature, current and the like, but neglects the detection of acoustic signals generated by all parts of equipment in the operation process of the ventilator, once the equipment has fault symptoms, the vibration and the temperature signals can not obviously change in a short time, the acoustic signals are abnormally fluctuated at first, the detection and the analysis of the audio signals are absent, the comprehensive analysis of the non-contact detection and the contact detection data of the ventilator cannot be realized, and the fault analysis result of the ventilator is inaccurate;
and at present, a great room for improvement exists for the fault diagnosis technology of the mine ventilator, and the main aspects are as follows: the fault diagnosis scheme is to be optimized, the traditional ventilator fault diagnosis monitors the operation parameters of different parts through various types of sensors, and judges whether the ventilator is in fault or not through comparison with a threshold value, so that the fault diagnosis scheme is low in fault diagnosis accuracy, consumes a large amount of manpower and lacks intelligence;
in order to solve the above-mentioned defect, a technical scheme is provided.
Disclosure of Invention
The invention aims to solve the technical problem that the existing ventilator fault early warning system has a large improvement space for mine ventilator fault diagnosis, and provides a ventilator fault early warning system based on data analysis.
The aim of the invention can be achieved by the following technical scheme:
a ventilator fault early warning system based on data analysis, comprising:
the multisource data perception module is used for acquiring and perceiving the operation audio signals of the ventilator and the surrounding environment data;
the data joint module is used for carrying out preliminary processing and preprocessing on the audio signal and the surrounding environment data; performing digital sampling on the audio signal, and performing preliminary processing on the obtained audio signal by using an intelligent filtering noise reduction model; the audio signal after noise reduction and filtering is composed of audio signal pre-emphasis, framing and windowing by adopting a preprocessing method;
the abnormality diagnosis module is used for dividing the obtained ventilator audio signal and the real-time operation parameter, and comparing the obtained ventilator audio signal and the real-time operation parameter to diagnose an abnormality signal; the process of diagnosis by means of an audio signal comprises the following steps:
s1: dividing an audio signal subjected to noise reduction filtering and framing and windowing into a plurality of frames, wherein each frame is a signal with linear continuous change;
s2: then, using kurtosis values in characteristic parameters of a time domain analysis method as real-time monitoring indexes of the audio signals, setting different kurtosis thresholds, calculating the audio signals in real time to obtain audio data streams, calculating the kurtosis values of the audio data streams, and comparing the calculated kurtosis values with the different kurtosis thresholds so as to monitor early abnormality of the ventilator in real time;
s3: judging whether abnormality and the type of the abnormality occur when the abnormal value of the kurtosis occurs, storing the audio signals in a period of time before and after the abnormal moment of the kurtosis value, and further processing the audio signals in the period of time by utilizing a frequency domain processing and time-frequency processing method of the audio signals to obtain the frequency distribution and the occurrence time point of the abnormal frequency in the period of time;
the abnormality diagnosis module is also used for applying the CNN network in deep learning to abnormality diagnosis of the ventilator and constructing a fault diagnosis model;
furthermore, the multisource data perception module perceives the running audio signals of the ventilator to be composed of a plurality of directional and aluminum belt type sound pick-up devices, the sound pick-up devices respectively point to the safe running parts of the fan impeller, the non-driving side bearing and the driving side motor of the ventilator, and the audio signals in the running process of the ventilator are collected in real time;
the pickup is annular around the ventilation blower and distributes, and the pickup all is located on the straight line that uses the ventilation blower bearing to be parallel, and every pickup is the equidistance and distributes, and the key operation position of the directional ventilation blower operation of pickup self simultaneously.
Further, the data junction module adopts preprocessing specific operation steps for the noise-reduced and filtered audio signals as follows:
the audio signal after noise reduction and filtering is composed of audio signal pre-emphasis, framing and windowing by adopting a preprocessing method;
the audio signal pre-emphasis highlights the high-frequency signal, frames and windows the pre-emphasized signal, after the audio signal frames, the audio signal is matched for windowing, the original audio signal is multiplied by a window function, then the windowed signal is analyzed and researched, and the Hamming window time domain function is utilized for voice recognition.
Further, the specific operation steps of the time domain analysis method in the diagnosis process of the abnormality diagnosis module through the audio signal are as follows:
the audio signal after noise reduction filtering and framing and windowing is divided into a plurality of frames, then the kurtosis value of each frame signal is calculated, the audio acquisition frequency is set to 16000, the audio data stream in 1s is divided into 30 frames according to the framing principle, and the number of the kurtosis values calculated in one second is 30.
Further, the specific operation steps of the frequency domain processing method in the diagnosis process of the abnormality diagnosis module through the audio signal are as follows:
obtaining the frequency characteristic information of the signal by analyzing the amplitude-frequency characteristic and the phase-frequency characteristic of the audio signal; the frequency domain analysis completes the Fourier transform of the audio signal, analyzes the frequency characteristic of the signal, adopts a fast Fourier transform algorithm, and adopts the fast Fourier transform as the specific process of the frequency domain processing method of the audio signal as follows:
b1: calculating the kurtosis value of each frame of audio signal, and comparing the kurtosis value with a preset kurtosis threshold value;
b2: when the kurtosis is larger than a preset kurtosis threshold, recording the time point of the abnormal moment of the kurtosis value, and storing the audio signals in a period of time before and after the time point of the abnormal moment of the kurtosis value;
b3: b2, reading the audio signal obtained in the step, and analyzing the audio signal by adopting a fast Fourier transform method to obtain the frequency distribution of the audio signal in the period of time;
b4: observing the frequency distribution characteristics of the audio signals, comparing the frequency distribution characteristics with the audio signals when the ventilator normally operates, and if the frequency distribution ranges obtained by the two times are the same, obviously abnormality does not occur to the ventilator; if obvious difference occurs in frequency, the ventilator is abnormal, and abnormal positions of the ventilator are judged according to the energy of the frequency.
Further, the specific operation steps of the time-frequency analysis method in the diagnosis process of the abnormality diagnosis module through the audio signal are as follows:
processing the audio signal by a time-frequency analysis method to obtain a rule of frequency change along with time and the occurrence time of abnormal frequency, and selecting a short-time Fourier transform method as a time-frequency analysis method;
introducing a local frequency spectrum concept in short-time Fourier transform, intercepting a section of an original non-stationary signal by using a window function with a short time length, performing Fourier transform on the section of stationary signal, and then sliding the window function along a time axis to obtain an image instant frequency chart of the frequency change rule of the whole non-stationary signal along with time;
the frequency components contained in the audio signal and the energy of each frequency in the obtained time-frequency diagram, the brighter the color of the frequency band in the diagram, the larger the energy of the frequency band, the larger the proportion of the frequency band in the whole energy, and the corresponding frequency change rules at different moments are obtained.
Further, the specific operation steps of the fault diagnosis implementation process in the abnormality diagnosis module are as follows:
firstly, transplanting an audio signal acquisition program in a microprocessor in a pickup to acquire audio signals of different components in real time when the ventilator works; then, carrying out real-time processing on the data stream formed by the obtained audio signals, wherein the part of algorithm program is deployed after the audio acquisition program and is an audio signal time-frequency diagram based on short-time Fourier transform, and judging whether the ventilator fails or not according to the time-frequency diagram; and simultaneously, offline training is carried out on time-frequency diagrams under different working conditions by utilizing a neural network, and then the trained network model is used for automatically judging whether the ventilator is abnormal or not and the type of the abnormality on line.
Further, the specific operation steps of the warning module for sending the warning information group to the collection terminal of the manager are as follows:
packaging the ventilator abnormality diagnosis result, the occurrence time, the specific position and the established warning program to form a warning information group, and sending the warning information group to a mobile phone terminal of a manager;
the mobile phone terminal is provided with a WeChat applet, when the warning program is received, the mobile phone terminal can generate vibration and sound prompt, a warning manager checks the mobile phone at the first time, and the WeChat applet can display the abnormal diagnosis result, occurrence time and specific position of the ventilator at the moment; when the sound or vibration continues for a preset time and is not stopped, sending information to a preset emergency contact person and making a call;
after the warning is finished, the influence time of ventilator abnormality diagnosis warning is recorded, the ventilator abnormality diagnosis result, the occurrence time, the specific position and the response time are recorded according to numbers and are recorded into a list form, and meanwhile, the recorded list files are classified into three screening modes of time screening, abnormality type screening and position screening for checking historical files.
Compared with the prior art, the invention has the beneficial effects that:
(1) According to the invention, the multisource data sensing module is constructed, multisource and multiscale real-time sensing is carried out on the running state of the mine ventilator by using the multisensor, the multisource and multiscale real-time sensing is carried out on the running state of the mine ventilator by using the distributed pickup and environmental data acquisition and distributed around key running positions of the ventilator, and the real-time running state of the ventilator is captured in real time;
(2) The invention can utilize the signal processing method to extract the inherent characteristics of the audio signal after noise reduction and preprocessing of the original audio signal, the common method for extracting the characteristics of the audio signal comprises time domain characteristic parameter processing, frequency domain characteristic parameter and time frequency analysis, and the system can simultaneously apply the three methods to the audio signal processing according to the actual field requirement and the real-time property of the system, and combines the advantages and disadvantages of different methods, and extracts the characteristics of the collected audio signal, thereby obtaining finer and more accurate audio characteristic information;
(3) According to the invention, the warning module is utilized to send the warning information packet to the mobile phone terminal of the manager at the first time after the ventilator fails, and the attention of the manager to the mobile phone is improved through the sound and vibration of the mobile phone, so that the fault condition of the ventilator can be known at the first time, and further maintenance or overhaul is performed.
Drawings
For the convenience of those skilled in the art, the present invention will be further described with reference to the accompanying drawings;
FIG. 1 is a general block diagram of a system of the present invention;
FIG. 2 is a flowchart of kurtosis calculation of an audio signal according to the present invention;
FIG. 3 is a diagram showing a fault diagnosis process of the fault diagnosis model according to the present invention;
fig. 4 is a structural diagram of a CNN network according to the present invention.
Detailed Description
The technical solutions of the present invention will be clearly and completely described in connection with the embodiments, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
It should be understood that the terms "comprises" and "comprising," when used in this specification and the claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the present disclosure is for the purpose of describing particular embodiments only, and is not intended to be limiting of the disclosure. As used in the specification and claims of this disclosure, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should be further understood that the term "and/or" as used in the present disclosure and claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
1-4, a ventilator fault early warning system based on data analysis comprises a multi-source data sensing module, a data junction module, an abnormality diagnosis module and a warning module;
the multisource data perception module is used for acquiring and perceiving the operation audio signals of the ventilator and the surrounding environment data;
the sensing of the running audio signals of the ventilator consists of a plurality of directional and aluminum belt type sound pick-up devices, wherein the sound pick-up devices respectively point to the safety operation key parts of the fan impeller, the non-driving side bearing and the driving side motor of the ventilator, and collect the audio signals in the running process of the ventilator in real time; the sound pick-up is distributed annularly around the ventilator, the sound pick-up is positioned on a straight line which is parallel with the ventilator bearing, each sound pick-up is distributed at equal intervals, and meanwhile, the sound pick-up points to a key operation part of the ventilator operation, so that the omnibearing perception of the ventilator audio signal is realized;
the sensing of the surrounding environment data consists of a temperature sensor and a smoke sensor, and senses the temperature and smoke parameters around the ventilator;
the data junction module is used for carrying out preliminary processing and preprocessing on the real-time ventilator operation audio signals and the surrounding environment data uploaded by the multi-source data perception module;
performing digital sampling on the audio signal according to the required signal sampling rate, and performing preliminary processing on the acquired digital audio signal by using an intelligent filtering noise reduction model to improve the signal-to-noise ratio of the audio signal;
the PCM1808 audio processing chip is specifically adopted to finish accurate sampling of audio signals, and the process is to convert continuous analog signals into discrete digital signals through three steps of signal sampling, quantization and encoding:
and (3) signal sampling: taking a time continuously variable analog signal out of a plurality of representative sample values to represent the continuously variable analog signal, and calibrating a function of the analog audio signal which is continuous in time and amplitude as x (t), wherein the sampling process is a process of discretizing the function x (t) in time;
the quantization adopts a limited amplitude approximation to represent the original amplitude value which continuously changes in time, and the continuous amplitude of the analog signal is changed into a limited number of discrete values with a certain time interval;
the coding is to represent the quantized discrete value by binary number according to a certain rule. In order to ensure high fidelity of the ventilator field audio signals, a PCM coding format is generally adopted;
the intelligent filtering noise reduction model is used for improving the signal-to-noise ratio of an audio signal, the feature extraction accuracy of the signal and the operation processing speed, and realizing a self-adaptive, self-learning, efficient and stable filtering process aiming at background noise under different environments;
the audio signal after noise reduction and filtering is composed of audio signal pre-emphasis, framing and windowing by adopting a preprocessing method; the audio signal pre-emphasis can highlight the high-frequency signal, reduce the interference of the low-frequency signal to the high-frequency signal, and the formula of the pre-emphasis digital filter before digital conversion is as follows:
H(z)＝1-αz -1
the better the value is between 0 and 1 and is close to 1, the value is 0.98, and the amplitude-frequency characteristic is as follows:
when the signal is high frequency, the amplitude value |H (w) | approaches to 1+alpha, and when the signal is low frequency, the amplitude value|H (w) | approaches to 1-alpha, and the high frequency part can be highlighted after pre-emphasis treatment;
after framing and windowing are performed on the pre-emphasized signal, frequency domain information omission occurs after framing the audio signal, in order to reduce the influence caused by spectrum loss, windowing is performed in a matching way, the windowing is that an original sound signal is multiplied by a Window function, then analysis and research are performed on the windowed signal, and sound recognition utilizes a Hamming Window (Hamming Window) time domain function, and the function form can be expressed as:
where k=1, 2, …, N, and the frequency domain characteristics of the hamming window are expressed as:
the method comprises the steps that wired and wireless heterogeneous communication networks are adopted for receiving surrounding environment data of a ventilator;
the abnormality diagnosis module divides the obtained audio signal and the real-time operation parameters of the ventilator, calculates the operation characteristics of the ventilator, and compares the abnormal signals to diagnose;
wherein the process of diagnosing by audio signals comprises the steps of:
s1: dividing an audio signal subjected to noise reduction filtering and framing and windowing into a plurality of frames, wherein each frame is a signal with linear continuous change;
s2: then, using kurtosis in characteristic parameters of a time domain analysis method as a real-time monitoring index of the audio signal, setting different kurtosis thresholds, calculating the audio signal in real time to obtain an audio data stream, calculating a kurtosis value of the audio data stream, and comparing the kurtosis value with the different kurtosis thresholds to realize real-time monitoring of early abnormality of the ventilator;
s3: judging whether abnormality and the type of the abnormality occur when the abnormal value of the kurtosis occurs, storing the audio signals in a period of time before and after the abnormal moment of the kurtosis value, and further processing the audio signals in the period of time by utilizing a frequency domain processing and time-frequency processing method of the audio signals to obtain the frequency distribution and the occurrence time point of the abnormal frequency in the period of time;
the time domain analysis method in the step S2 is the most accurate, rapid and direct method in signal processing, analyzes the abnormal condition of the electromechanical equipment by calculating the time domain statistical characteristics of the audio signals, judges the running state of the electromechanical equipment, and can divide the health condition grade of the ventilator according to historical monitoring data and maintenance records. Common time domain indexes include skewness, kurtosis, variance, maximum value, minimum value, peak value, square root amplitude, average amplitude, root mean square amplitude, absolute average value, waveform index, amplitude index, pulse index, margin index and kurtosis index, and are based on audio signalsThe kurtosis parameter in the time domain processing method is selected as the time domain parameter to process the significance represented by the characteristics and different time domain parameters; the kurtosis reflects the distribution characteristic of a signal, the kurtosis coefficient is larger along with the increase of the proportion of abnormal pulses with amplitude values in the signal, the kurtosis value is used as a mathematical statistical variable, certain regular distribution exists, when the ventilator is in normal operation, the integral change of an audio signal is relatively uniform, no abnormal impact signal exists, the probability density distribution of different amplitude signals can be similar to normal distribution, the kurtosis coefficient Ku of normal distribution swings near a value 3 and cannot generate larger abnormal values, when a motor or a certain part of a rolling bearing is scratched or damaged, the audio signal generated in the operation process generates larger impact, namely the sound signal with abnormal amplitude values, the probability density of the abnormal pulses with amplitude values in the signal increases along with the aggravation of faults, the probability density distribution of different amplitude signals gradually deviates from the normal distribution, the more serious faults, the kurtosis coefficient of the audio signal is larger, when the kurtosis coefficient exceeds a threshold value, the probability density distribution of different amplitude signals is similar to normal distribution, the probability density distribution of the abnormal impact signal swings, the kurtosis coefficient swings near the value 3, the kurtosis coefficient swings, and the kurtosis coefficient can be used as an important basis for early fault diagnosis of the ventilator, and the kurtosis is calculated by the expression:where x (t) is a zero-mean signal and p (x) is a distribution function of the probability density of x (t); the kurtosis index value is irrelevant to the rotating speed, the size and the load of a driving motor or a rolling bearing of the ventilator, and is mainly used for detecting early abnormal forms of the rolling bearing and the driving motor;
specific: dividing an audio signal subjected to noise reduction filtering and framing and windowing into a plurality of frames, then calculating the kurtosis value of each frame of signal, setting the audio acquisition frequency to 16000, dividing an audio data stream in 1s into 30 frames according to the framing principle, and calculating 30 kurtosis values in one second so as to meet the requirement of real-time monitoring of a ventilator; the kurtosis calculation flow of the audio signal can be shown with reference to fig. 2;
the frequency domain processing method in S3 is to obtain the frequency characteristic information of the signal by analyzing the amplitude-frequency characteristic and the phase-frequency characteristic of the audio signal; the frequency domain analysis completes the Fourier transform of the audio signal, analyzes the frequency characteristic of the signal, adopts a fast FFT algorithm, the Fourier transform is a common means for analyzing the frequency characteristic of the signal, in the field of digital signal processing, the Fast Fourier Transform (FFT) is based on Discrete Fourier Transform (DFT), the calculation method is improved, the time complexity of operation is reduced, the development of digital signal processing is greatly accelerated, and the frequency domain processing method adopting the Fast Fourier Transform (FFT) as the audio signal has the following specific procedures:
b1: calculating the kurtosis value of each frame of audio signal, and comparing the kurtosis value with a preset kurtosis threshold value;
b2: if the kurtosis is larger than a preset kurtosis threshold, recording a time point of abnormal time of the kurtosis value, and storing audio signals in a period of time before and after the time point to a local appointed path;
b3: reading the audio signal, and analyzing the audio signal by adopting an FFT method to obtain the frequency distribution of the audio signal in the period of time;
b4: observing the frequency distribution characteristics of the audio signals, comparing the frequency distribution characteristics with the audio signals when the ventilator normally operates, and if the frequency distribution ranges obtained by the two times are the same, obviously abnormality does not occur to the ventilator; otherwise, if obvious difference occurs in frequency, the ventilator is abnormal, and abnormal positions of the ventilator are judged according to the energy of the frequency;
in the step S3, a time-frequency processing method is adopted, namely a time-frequency analysis method is adopted to process the audio signal to obtain a rule of frequency change along with time and the occurrence time of abnormal frequency, and a short-time Fourier transform method is selected as a time-frequency analysis method; and introducing a local frequency spectrum concept in short-time Fourier transform, intercepting a section of the original non-stationary signal by using a window function with a short time length, performing Fourier transform on the section of stationary signal, and then sliding the window function along a time axis to obtain an image of the frequency change rule of the whole non-stationary signal along with time. For the continuous signal s (t), its continuous short-time fourier transform (STFT) is defined as:
the inverse transformation is as follows:
wherein: g (t) is a window function with a very short time length; * Representing complex conjugates. When g (t) =1, the short-time fourier transform is converted into a conventional fourier transform.
From the obtained time-frequency diagram, not only can the frequency components contained in the audio signal and the energy of each frequency be clearly seen, the brighter frequency band in the diagram shows that the larger the energy of the frequency band is, the larger the proportion of the frequency band in the whole energy is, meanwhile, the frequency change rules corresponding to different moments can be obtained, the moment when the abnormal frequency of the ventilator appears can be conveniently searched, and the method has important significance for abnormality diagnosis of the ventilator;
the feature extraction of ventilator abnormality diagnosis adopts a common method in signal processing, combines the advantages and disadvantages of different methods, and extracts features of the acquired audio signals so as to obtain finer and more accurate audio feature information;
the abnormality diagnosis module is also used for applying the CNN network in deep learning to abnormality diagnosis of the ventilator to construct a fault diagnosis model, and the fault diagnosis process of the model is shown by referring to FIG. 3;
the fault diagnosis of the diagnosis model is realized as follows:
firstly, an audio signal acquisition program is transplanted into a microprocessor in a pickup and used for acquiring audio signals of different components when the ventilator works in real time, then the acquired audio data stream is processed in real time, the part of algorithm program is deployed after the audio acquisition program and mainly is based on an audio signal time-frequency diagram of short-time Fourier transform, and as the time-frequency diagrams of the ventilator under different working conditions have certain differences, whether the ventilator fails or not can be judged according to the time-frequency diagram; performing offline training on the time-frequency diagrams under different working conditions by using a neural network, and then automatically judging whether the ventilator is abnormal or not and the type of the abnormality on line by using the trained network model; the key of realizing the part is the feature extraction of the audio signal, the selection of the neural network and the parameter setting of the model;
for the type of the neural network and the parameter setting thereof, a common Convolutional Neural Network (CNN) is selected and used for automatically identifying different types of time-frequency diagrams to form a ventilator abnormality automatic diagnosis model combining short-time Fourier transform and CNN, and the working flow of the model is as follows:
firstly, kurtosis processing is adopted for multi-source signals, FFT and short-time Fourier transform are adopted for further analysis for audio signals in adjacent periods of abnormal time kurtosis values, and as the result obtained by the FFT is unfavorable for distinguishing, a time-frequency diagram obtained by the short-time Fourier transform is selected as the input of a CNN model, deeper characteristic information is obtained through convolution, pooling and full connection processes, and finally, the output result is classified by using a classification function, so that abnormal ventilator categories are obtained;
the accuracy of the fault diagnosis model is related to the parameter design of the CNN network in addition to the size of the input data, and for the parameter design of the CNN network:
input layer: the size of the input data is related to the operation time and fault diagnosis accuracy of the model, the frequency distribution in the short-time Fourier spectrogram of the audio signal under different working conditions of the ventilator is mainly concentrated in a low-frequency area, and the input data is segmented and compressed to obtain an image with the size of 32 multiplied by 32 pixels;
convolution layer: the convolution layer is a core part of a CNN network, the layer carries out convolution operation with input data through a group of convolution kernels with the same size to obtain a local receptive field of the input data, when the characteristics of the input data are extracted, one convolution kernel intelligently extracts a part of the characteristics of the data, so that a plurality of convolution kernels are needed for obtaining all the characteristics of the input data, according to different dimensions of the input data, the dimensions of the convolution kernels correspond to the two-dimensional data, the two-dimensional convolution kernels are selected for carrying out convolution calculation with the input data to extract the characteristics, the selection of the convolution kernels is carried out on original audio signals before the characteristic extraction, the influence of noise of convolution calculation can be ignored, so that the convolution kernels with the size of 5 multiplied by 5 are selected for the convolution kernels of the first layer, and the pooling layer can lead to partial characteristic information loss, so that the convolution kernels with the size of 3 multiplied number of more are selected for the second convolution layer;
pooling layer: the method is used for collecting partial characteristics of a convolution layer and reducing data dimension, the working principle is that data of the convolution layer are divided into different areas according to preset size, maximum pooling and average pooling are carried out according to selected pooling types, the maximum pooling is to select the maximum value in the area as a characteristic value, the average pooling is to calculate the average value of the area as characteristic information, the depth of a characteristic diagram is not increased because the pooling layer does not have convolution operation, the dimension of the characteristic diagram is greatly reduced through pooling operation, the calculation speed of a network model is accelerated, and the network overfitting phenomenon is avoided; adopting a maximum pooling mode for the design of the CNN pooling layer, wherein the size is 2 multiplied by 2, the step length is 2, and selecting a data filling mode of the same;
activation function: the activation function is a nonlinear mapping function which is used together with a convolution layer to enhance the expression capacity of the model, the original multi-layer perceptron model does not introduce the feature of learning data which cannot be well achieved by the activation function, the CNN network introduces the activation function to increase the nonlinearity of the model, so that the gradient descent speed of the network training is accelerated, and the method is used for solving a plurality of nonlinear data problems;
full tie layer: the data processed by the convolution layer and the pooling layer are finally transmitted to the full-connection layer, and in order to avoid the overfitting phenomenon, a Dropout regularization mechanism is introduced into the full-connection layer, so that a part of neurons can be effectively restrained during network training, and only the rest neurons are updated;
the working mechanism is as follows: the CNN network training adopts a back propagation mode, the principle is that an output result generated by forward training is compared with an actual value, if the output deviation is larger, parameters such as a convolution kernel size, a learning rate, a data set proportion and the like are required to be adjusted, so that the most suitable parameter value is obtained, the working mode of adjusting the network super-parameters based on the output result is called back propagation, and the working mode is used in a designed fault diagnosis model, so that the model output accuracy can be improved; the designed CNN structure is referred to fig. 4;
the warning module is used for receiving the ventilator abnormality diagnosis result output by the abnormality diagnosis module and acquiring the occurrence time and the specific position of the ventilator abnormality diagnosis result;
transmitting the abnormality diagnosis result, the occurrence time and the specific position of the ventilator to a mobile phone terminal of a manager, and establishing an alarm program for packaging and transmitting;
the mobile phone terminal is provided with a WeChat applet, when the warning program is received, the mobile phone terminal can generate vibration and sound prompt, a warning manager checks the mobile phone at the first time, and the WeChat applet can display the abnormal diagnosis result, occurrence time and specific position of the ventilator at the moment;
when the sound or vibration continues for a preset time and is not stopped, sending information to a preset emergency contact person and making a call, so that a manager is prevented from not checking the mobile phone;
after the warning is finished, the influence time of ventilator abnormality diagnosis warning is recorded, the ventilator abnormality diagnosis result, occurrence time, specific position and response time are recorded according to numbers and are recorded into a list form, and simultaneously, recorded list files can be classified into time screening, abnormality type screening and position screening for quick historical checking.
The preferred embodiments of the invention disclosed above are intended only to assist in the explanation of the invention. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise form disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best understand and utilize the invention. The invention is limited only by the claims and the full scope and equivalents thereof.
Claims (8)
1. A ventilator fault early warning system based on data analysis, comprising:
the multisource data perception module is used for acquiring and perceiving the operation audio signals of the ventilator and the surrounding environment data;
the data joint module is used for carrying out preliminary processing and preprocessing on the audio signal and the surrounding environment data; performing digital sampling on the audio signal, and performing preliminary processing on the obtained audio signal by using an intelligent filtering noise reduction model; the audio signal after noise reduction and filtering is composed of audio signal pre-emphasis, framing and windowing by adopting a preprocessing method;
the abnormality diagnosis module is used for dividing the obtained ventilator audio signal and the real-time operation parameter, and comparing the obtained ventilator audio signal and the real-time operation parameter to diagnose an abnormality signal; the process of diagnosis by means of an audio signal comprises the following steps:
s1: dividing an audio signal subjected to noise reduction filtering and framing and windowing into a plurality of frames, wherein each frame is a signal with linear continuous change;
s2: then, using kurtosis values in characteristic parameters of a time domain analysis method as real-time monitoring indexes of the audio signals, setting different kurtosis thresholds, calculating the audio signals in real time to obtain audio data streams, calculating the kurtosis values of the audio data streams, and comparing the calculated kurtosis values with the different kurtosis thresholds so as to monitor early abnormality of the ventilator in real time;
s3: judging whether abnormality and the type of the abnormality occur when the abnormal value of the kurtosis occurs, storing the audio signals in a period of time before and after the abnormal moment of the kurtosis value, and further processing the audio signals in the period of time by utilizing a frequency domain processing and time-frequency processing method of the audio signals to obtain the frequency distribution and the occurrence time point of the abnormal frequency in the period of time;
the abnormality diagnosis module is also used for applying the CNN network in deep learning to abnormality diagnosis of the ventilator and constructing a fault diagnosis model.
The alarm module is used for receiving the ventilator abnormality diagnosis result output by the abnormality diagnosis module and sending an alarm information group to the collection terminal of the manager.
2. The ventilator fault early warning system based on data analysis according to claim 1, wherein the multisource data perception module perceives the ventilator operation audio signals and consists of a plurality of directional and aluminum belt type sound pick-ups, the sound pick-ups respectively point to a ventilator impeller, a non-driving side bearing and a driving side motor safe operation part, and audio signals in the ventilator operation process are collected in real time;
the pickup is annular around the ventilation blower and distributes, and the pickup all is located on the straight line that uses the ventilation blower bearing to be parallel, and every pickup is the equidistance and distributes, and the key operation position of the directional ventilation blower operation of pickup self simultaneously.
3. The ventilator fault early warning system based on data analysis according to claim 1, wherein the data junction module performs preprocessing on the noise-reduced and filtered audio signal as follows:
the audio signal after noise reduction and filtering is composed of audio signal pre-emphasis, framing and windowing by adopting a preprocessing method;
the audio signal pre-emphasis highlights the high-frequency signal, frames and windows the pre-emphasized signal, after the audio signal frames, the audio signal is matched for windowing, the original audio signal is multiplied by a window function, then the windowed signal is analyzed and researched, and the Hamming window time domain function is utilized for voice recognition.
4. The ventilator fault early warning system based on data analysis according to claim 1, wherein the abnormality diagnosis module performs time domain analysis method in the diagnosis process through audio signals as follows:
the audio signal after noise reduction filtering and framing and windowing is divided into a plurality of frames, then the kurtosis value of each frame signal is calculated, the audio acquisition frequency is set to 16000, the audio data stream in 1s is divided into 30 frames according to the framing principle, and the number of the kurtosis values calculated in one second is 30.
5. The ventilator fault early warning system based on data analysis according to claim 1, wherein the specific operation steps of the frequency domain processing method in the diagnosis process of the abnormality diagnosis module through the audio signal are as follows:
obtaining the frequency characteristic information of the signal by analyzing the amplitude-frequency characteristic and the phase-frequency characteristic of the audio signal; the frequency domain analysis completes the Fourier transform of the audio signal, analyzes the frequency characteristic of the signal, adopts a fast Fourier transform algorithm, and adopts the fast Fourier transform as the specific process of the frequency domain processing method of the audio signal as follows:
b1: calculating the kurtosis value of each frame of audio signal, and comparing the kurtosis value with a preset kurtosis threshold value;
b2: when the kurtosis is larger than a preset kurtosis threshold, recording the time point of the abnormal moment of the kurtosis value, and storing the audio signals in a period of time before and after the time point of the abnormal moment of the kurtosis value;
b3: b2, reading the audio signal obtained in the step, and analyzing the audio signal by adopting a fast Fourier transform method to obtain the frequency distribution of the audio signal in the period of time;
b4: observing the frequency distribution characteristics of the audio signals, comparing the frequency distribution characteristics with the audio signals when the ventilator normally operates, and if the frequency distribution ranges obtained by the two times are the same, obviously abnormality does not occur to the ventilator; if obvious difference occurs in frequency, the ventilator is abnormal, and abnormal positions of the ventilator are judged according to the energy of the frequency.
6. The ventilator fault early warning system based on data analysis according to claim 1, wherein the specific operation steps of the time-frequency analysis method in the diagnosis process of the abnormality diagnosis module through the audio signal are as follows:
processing the audio signal by a time-frequency analysis method to obtain a rule of frequency change along with time and the occurrence time of abnormal frequency, and selecting a short-time Fourier transform method as a time-frequency analysis method;
introducing a local frequency spectrum concept in short-time Fourier transform, intercepting a section of an original non-stationary signal by using a window function with a short time length, performing Fourier transform on the section of stationary signal, and then sliding the window function along a time axis to obtain an image instant frequency chart of the frequency change rule of the whole non-stationary signal along with time;
the frequency components contained in the audio signal and the energy of each frequency in the obtained time-frequency diagram, the brighter the color of the frequency band in the diagram, the larger the energy of the frequency band, the larger the proportion of the frequency band in the whole energy, and the corresponding frequency change rules at different moments are obtained.
7. The ventilator fault early warning system based on data analysis according to claim 1, wherein the specific operation steps of the fault diagnosis implementation process in the abnormality diagnosis module are as follows:
firstly, transplanting an audio signal acquisition program in a microprocessor in a pickup to acquire audio signals of different components in real time when the ventilator works; then, carrying out real-time processing on the data stream formed by the obtained audio signals, wherein the part of algorithm program is deployed after the audio acquisition program and is an audio signal time-frequency diagram based on short-time Fourier transform, and judging whether the ventilator fails or not according to the time-frequency diagram; and simultaneously, offline training is carried out on time-frequency diagrams under different working conditions by utilizing a neural network, and then the trained network model is used for automatically judging whether the ventilator is abnormal or not and the type of the abnormality on line.
8. The ventilator fault early warning system based on data analysis according to claim 1, wherein the specific operation steps of the warning module for sending the warning information group to the collection terminal of the manager are as follows:
packaging the ventilator abnormality diagnosis result, the occurrence time, the specific position and the established warning program to form a warning information group, and sending the warning information group to a mobile phone terminal of a manager;
when the warning program is received, the mobile phone terminal generates vibration and sound prompt, and displays the abnormal diagnosis result, occurrence time and specific position of the ventilator; when the sound or vibration continues for a preset time and is not stopped, sending information to a preset emergency contact person and making a call;
after the warning is finished, the influence time of ventilator abnormality diagnosis warning is recorded, the ventilator abnormality diagnosis result, the occurrence time, the specific position and the response time are recorded according to numbers and are recorded into a list form, and meanwhile, the recorded list files are classified into three screening modes of time screening, abnormality type screening and position screening for checking historical files.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310418606.6A CN116517860A (en) | 2023-04-19 | 2023-04-19 | Ventilator fault early warning system based on data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310418606.6A CN116517860A (en) | 2023-04-19 | 2023-04-19 | Ventilator fault early warning system based on data analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116517860A true CN116517860A (en) | 2023-08-01 |
Family
ID=87396881
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310418606.6A Pending CN116517860A (en) | 2023-04-19 | 2023-04-19 | Ventilator fault early warning system based on data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116517860A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116754234A (en) * | 2023-08-17 | 2023-09-15 | 山东经典印务有限责任公司 | Automatic printing production equipment running state detection method |
| CN117092578A (en) * | 2023-10-18 | 2023-11-21 | 青岛悠进电装有限公司 | Wire harness conduction intelligent detection system based on data acquisition and processing |
| CN117150269A (en) * | 2023-09-04 | 2023-12-01 | 淮阴师范学院 | Electrical equipment operation abnormality diagnosis system based on data analysis |
| CN117422888A (en) * | 2023-09-13 | 2024-01-19 | 长龙(杭州)航空维修工程有限公司 | Aircraft performance evaluation method and system |
| CN118152864A (en) * | 2024-05-11 | 2024-06-07 | 常州全一智能科技有限公司 | Motor detection method and system based on multi-source data fusion |
-
2023
- 2023-04-19 CN CN202310418606.6A patent/CN116517860A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116754234A (en) * | 2023-08-17 | 2023-09-15 | 山东经典印务有限责任公司 | Automatic printing production equipment running state detection method |
| CN116754234B (en) * | 2023-08-17 | 2023-11-21 | 山东经典印务有限责任公司 | Automatic printing production equipment running state detection method |
| CN117150269A (en) * | 2023-09-04 | 2023-12-01 | 淮阴师范学院 | Electrical equipment operation abnormality diagnosis system based on data analysis |
| CN117150269B (en) * | 2023-09-04 | 2024-06-07 | 淮阴师范学院 | Electrical equipment operation abnormality diagnosis system based on data analysis |
| CN117422888A (en) * | 2023-09-13 | 2024-01-19 | 长龙(杭州)航空维修工程有限公司 | Aircraft performance evaluation method and system |
| CN117422888B (en) * | 2023-09-13 | 2024-05-10 | 长龙(杭州)航空维修工程有限公司 | Aircraft performance evaluation method and system |
| CN117092578A (en) * | 2023-10-18 | 2023-11-21 | 青岛悠进电装有限公司 | Wire harness conduction intelligent detection system based on data acquisition and processing |
| CN117092578B (en) * | 2023-10-18 | 2024-01-16 | 青岛悠进电装有限公司 | Wire harness conduction intelligent detection system based on data acquisition and processing |
| CN118152864A (en) * | 2024-05-11 | 2024-06-07 | 常州全一智能科技有限公司 | Motor detection method and system based on multi-source data fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116517860A (en) | Ventilator fault early warning system based on data analysis | |
| CN110867196B (en) | Machine equipment state monitoring system based on deep learning and voice recognition | |
| CN110940539B (en) | Machine equipment fault diagnosis method based on artificial experience and voice recognition | |
| KR102324776B1 (en) | Method for diagnosing noise cause of vehicle | |
| KR102272622B1 (en) | Deep learning based abnormal symptom detection system using mechanical sound of mechanical equipment | |
| CN105841961A (en) | Bearing fault diagnosis method based on Morlet wavelet transformation and convolutional neural network | |
| CN113566948A (en) | Fault audio recognition and diagnosis method for robot coal pulverizer | |
| CN110398647B (en) | Transformer state monitoring method | |
| CN114155879B (en) | Abnormal sound detection method for compensating abnormal perception and stability by using time-frequency fusion | |
| CN113298134B (en) | System and method for remotely and non-contact health monitoring of fan blade based on BPNN | |
| CN113192532A (en) | Mine hoist fault acoustic analysis method based on MFCC-CNN | |
| CN111951824A (en) | Detection method for distinguishing depression based on sound | |
| CN111695413A (en) | Signal first arrival pickup method and device combining U-Net and Temporal encoding | |
| CN112052712B (en) | Power equipment state monitoring and fault identification method and system | |
| CN113345399A (en) | Method for monitoring sound of machine equipment in strong noise environment | |
| CN113707175B (en) | Acoustic event detection system based on feature decomposition classifier and adaptive post-processing | |
| CN115687894A (en) | Tumble detection system and method based on small sample learning | |
| CN117576632B (en) | Multi-mode AI large model-based power grid monitoring fire early warning system and method | |
| CN112801033A (en) | AlexNet network-based construction disturbance and leakage identification method along long oil and gas pipeline | |
| CN112033656A (en) | Mechanical system fault detection method based on broadband spectrum processing | |
| CN116312635A (en) | Sound target detection method integrating machine recognition and auditory brain-computer interface | |
| CN110579657A (en) | Transformer fault detection system based on vibration signal analysis and detection method thereof | |
| CN115481689A (en) | System and method for simultaneously recognizing user gesture and identity based on ultrasonic waves | |
| CN113343887A (en) | Multi-sensor mixed fault signal blind separation method based on edge calculation and machine learning | |
| CN114093385A (en) | Unmanned aerial vehicle detection 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 |