CN115131943A - Acousto-optic linkage early warning method based on deep learning - Google Patents
Acousto-optic linkage early warning method based on deep learning Download PDFInfo
- Publication number
- CN115131943A CN115131943A CN202210801438.4A CN202210801438A CN115131943A CN 115131943 A CN115131943 A CN 115131943A CN 202210801438 A CN202210801438 A CN 202210801438A CN 115131943 A CN115131943 A CN 115131943A
- Authority
- CN
- China
- Prior art keywords
- acousto
- pulse
- optic
- predicted value
- accident
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 30
- 238000013135 deep learning Methods 0.000 title claims abstract description 15
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 claims abstract description 24
- 230000007704 transition Effects 0.000 claims abstract description 23
- 238000003062 neural network model Methods 0.000 claims abstract description 21
- 239000011159 matrix material Substances 0.000 claims abstract description 17
- 238000012795 verification Methods 0.000 claims abstract description 13
- 230000003287 optical effect Effects 0.000 claims description 27
- 238000012549 training Methods 0.000 claims description 23
- 238000013528 artificial neural network Methods 0.000 claims description 16
- 238000012360 testing method Methods 0.000 claims description 14
- 230000008859 change Effects 0.000 claims description 13
- 230000002068 genetic effect Effects 0.000 claims description 10
- 238000007689 inspection Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 2
- 238000012545 processing Methods 0.000 claims description 2
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000005236 sound signal Effects 0.000 description 3
- 239000000463 material Substances 0.000 description 2
- 238000001208 nuclear magnetic resonance pulse sequence Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/10—Alarms for ensuring the safety of persons responsive to calamitous events, e.g. tornados or earthquakes
-
- 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
- G06N3/08—Learning methods
- G06N3/086—Learning methods using evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B7/00—Signalling systems according to more than one of groups G08B3/00 - G08B6/00; Personal calling systems according to more than one of groups G08B3/00 - G08B6/00
- G08B7/06—Signalling systems according to more than one of groups G08B3/00 - G08B6/00; Personal calling systems according to more than one of groups G08B3/00 - G08B6/00 using electric transmission, e.g. involving audible and visible signalling through the use of sound and light sources
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Business, Economics & Management (AREA)
- Geology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Physiology (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Emergency Management (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Environmental & Geological Engineering (AREA)
- Photometry And Measurement Of Optical Pulse Characteristics (AREA)
Abstract
The application relates to an acousto-optic linkage early warning method based on deep learning, which comprises the following steps: obtaining historical acousto-optic pulses; constructing an ARIMA model based on historical acousto-optic pulses, and acquiring a first acousto-optic pulse prediction residual error; constructing a genetic-BP neural network model based on historical acousto-optic pulses; obtaining a predicted value of a first acousto-optic pulse prediction residual and a first acousto-optic pulse predicted value through a genetic-BP neural network model; obtaining a state transition matrix based on a predicted value of the first acousto-optic pulse prediction residual, the first acousto-optic pulse predicted value and the acousto-optic pulse verification set; and acquiring a final acousto-optic pulse prediction value through the ARIMA model, the genetic-BP neural network model and the state transition matrix, and early warning the sudden accident in the accident early warning area. Based on the method, graded acousto-optic early warning is achieved in the early warning area.
Description
Technical Field
The application belongs to the technical field of acousto-optic early warning, and particularly relates to an acousto-optic linkage early warning method based on deep learning.
Background
Because natural disasters and industrial production accidents frequently occur at present, people pay more attention to and pay attention to early warning of sudden accidents, and meanwhile, the early warning method has higher requirements on the accuracy and the predictability of accident early warning. For natural disasters and sudden accidents, good early warning can reduce casualties, equipment damage and the like to the maximum extent; specifically, for acousto-optic early warning, the early warning can be performed on the emergency by receiving the sound signal and the optical signal, and related personnel are informed in time to take related measures to process the emergency.
At present, for acousto-optic alarm, most acousto-optic alarm devices are adopted to monitor sudden accidents in real time, usually, a sound sensor and a light sensor are adopted to collect sound signals and light signals in a monitoring range in real time, whether the real-time sound signals and light signals are abnormal or not is judged based on a preset numerical range of conventional signals, and then early warning is carried out on the sudden accidents. However, the existing method can only perform early warning on sudden accidents based on the acquired real-time data; because the basis of accident judgment is only real-time acousto-optic signals, and then since an accident occurs, relevant personnel are informed to take treatment measures by sending out early warning, the treatment time left for the relevant personnel is often insufficient, and further the accident can not be effectively treated.
In view of this, the application provides an acousto-optic linkage early warning method based on deep learning, which can perform deep learning based on a real-time acousto-optic signal only by acquiring the real-time acousto-optic signal, so as to effectively early warn an emergency.
Disclosure of Invention
Technical problem to be solved
In view of the above disadvantages and shortcomings in the prior art, the present application provides an acousto-optic linkage early warning method based on deep learning.
(II) technical scheme
In order to achieve the purpose, the following technical scheme is adopted in the application:
in a first aspect, the present application provides a deep learning-based acousto-optic linkage early warning method, including:
s1, acquiring acoustic pulses corresponding to historical acoustic signals and optical pulses corresponding to historical optical signals of an accident early warning area, and preprocessing the acoustic pulses and the optical pulses;
the sound and light pulses comprise: an acousto-optic pulse training set, an acousto-optic pulse testing set and an acousto-optic pulse verification set;
s2, constructing an ARIMA model based on the acousto-optic pulse training set, and acquiring a first acousto-optic pulse prediction residual error through the ARIMA model based on the acousto-optic pulse testing set;
s3, based on the acousto-optic pulse training set, optimizing the BP neural network through a genetic algorithm to obtain a genetic-BP neural network model;
s4, obtaining a predicted value of the first acousto-optic pulse prediction residual error through a genetic-BP neural network model based on the first acousto-optic pulse prediction residual error; based on the acousto-optic pulse test set, obtaining a first acousto-optic pulse predicted value through a genetic-BP neural network model;
s5, obtaining a second acousto-optic pulse predicted value based on the first acousto-optic pulse predicted value and the predicted value of the first acousto-optic pulse predicted residual error; acquiring a state transition matrix through a Markov algorithm based on the second sound light pulse predicted value and the sound light pulse verification set;
s6, acquiring a final acousto-optic pulse prediction value through an ARIMA model, a genetic-BP neural network model and a state transition matrix based on an acousto-optic pulse training set;
and S7, early warning the sudden accident in the accident early warning area based on the preset conventional acousto-optic pulse conditions and the final acousto-optic pulse predicted value.
Optionally, the S1 includes:
acquiring an acoustic signal through an acoustic sensor, and extracting an acoustic pulse based on the acoustic signal;
an optical signal is acquired by a light sensor, and an optical pulse is extracted based on the optical signal.
Optionally, the S2 includes:
performing sequence inspection on the acousto-optic pulse training set, and fitting an ARIMA model by a least square method based on the inspected acousto-optic pulse training set;
taking the acousto-optic pulse test set as an input variable of the ARIMA model, and acquiring a third acousto-optic pulse predicted value;
and acquiring a first acousto-optic pulse prediction residual error based on the third acousto-optic pulse prediction value and the acousto-optic pulse verification set.
Optionally, the S3 includes: the establishing process of the genetic-BP neural network model specifically comprises the following steps:
s2-1, determining the number of input layers, output layers and hidden layers of the BP neural network;
s2-2, optimizing the connection weight and the threshold value in the BP neural network based on a genetic algorithm;
the connection weight includes: the connection weight from the input layer to the hidden layer and the connection weight from the hidden layer to the output layer.
Optionally, the S5 includes:
s5-1, acquiring a second sound light pulse prediction residual error based on a second sound light pulse prediction value and a sound light pulse verification set, performing threshold division based on the second sound light pulse prediction residual error, and marking the state of the second sound light pulse prediction residual error;
and S5-2, acquiring corresponding state transition probability based on the marked second photo pulse prediction residual state, and acquiring a state transition matrix corresponding to the second photo pulse prediction residual based on the state transition probability.
Optionally, the S6 includes:
taking the acousto-optic pulse training set as an input variable of an ARIMA model and a genetic-BP neural network model to obtain a fourth pulse predicted value;
and acquiring a final acousto-optic pulse predicted value through the state transition matrix based on the fourth pulse predicted value.
Optionally, the S7 includes:
and if the final acousto-optic pulse predicted value exceeds the numerical value change range of the acousto-optic pulse in the non-accident state, judging that the accident is a sudden accident.
Optionally, the S7 includes: and determining the acousto-optic pulse condition based on historical acousto-optic and optic pulses of the accident early warning area.
Optionally, the S7 includes:
if the final light pulse predicted value exceeds the numerical change range of the light pulse in the non-accident state and the final sound pulse predicted value does not exceed the numerical change range of the light pulse in the non-accident state, judging the first-stage sudden accident;
if the final sound pulse predicted value exceeds the numerical value change range of the sound pulse in the non-accident state and the final light pulse predicted value does not exceed the numerical value change range of the light pulse in the non-accident state, judging that a secondary sudden accident occurs;
and if the final predicted value of the light pulse exceeds the numerical variation range of the light pulse in the non-accident state and the final predicted value of the sound pulse exceeds the numerical variation range of the light pulse in the non-accident state, judging the three-stage sudden accident.
(III) advantageous effects
The method comprises the steps of acquiring real-time sound and light signals, and extracting sound and light pulses based on the sound and light signals; based on the extracted sound and light combined pulse, an ARIMA model is constructed, deep learning is carried out based on a BP neural network optimized by a genetic algorithm, error adjustment is carried out on the ARIMA model and the predicted sound and light signals corresponding to the BP neural network optimized by the genetic algorithm by utilizing a Markov state transition matrix obtained based on historical sound and light signals, the prediction error is further reduced, and further, based on the obtained simulated sound and light pulse sequence, grading early warning in a preset region is realized.
Drawings
The application is described with the aid of the following figures:
fig. 1 is a schematic flow diagram corresponding to the acousto-optic linkage early warning method based on deep learning.
Detailed Description
For the purpose of better explaining the present invention and to facilitate understanding, the present invention will be described in detail by way of specific embodiments with reference to the accompanying drawings. It is to be understood that the following specific examples are illustrative of the invention only and are not to be construed as limiting the invention. In addition, it should be noted that, in the case of no conflict, the embodiments and features in the embodiments in the present application may be combined with each other; for convenience of description, only portions related to the present invention are shown in the drawings.
The embodiment provides an acousto-optic linkage early warning method based on deep learning, which comprises the following specific steps as shown in fig. 1:
and S1, acquiring acoustic pulses corresponding to historical acoustic signals and optical pulses corresponding to historical optical signals of the accident early warning area, and preprocessing the acoustic pulses and the optical pulses.
In this embodiment, the acoustic and optical pulses include: the system comprises an acousto-optic pulse training set, an acousto-optic pulse testing set and an acousto-optic pulse verification set.
Based on the foregoing step S1, it should be noted that for the acquisition of the acoustic pulse and the optical pulse:
firstly, acquiring an acoustic signal through an acoustic sensor, and extracting an acoustic pulse based on the acoustic signal;
second, an optical signal is acquired by a light sensor, and an optical pulse is extracted based on the optical signal.
S2, constructing an ARIMA model based on the acousto-optic pulse training set, and acquiring a first acousto-optic pulse prediction residual error through the ARIMA model based on the acousto-optic pulse testing set.
Based on the foregoing step S2, the following is to be noted to obtain the first aod prediction residual:
firstly, performing sequence inspection on an acousto-optic pulse training set, and fitting an ARIMA model by a least square method based on the inspected acousto-optic pulse training set;
secondly, taking the acousto-optic pulse test set as an input variable of the ARIMA model to obtain a third acousto-optic pulse predicted value;
and thirdly, acquiring a first acousto-optic pulse prediction residual error based on the third acousto-optic pulse prediction value and the acousto-optic pulse verification set.
And S3, based on the acousto-optic pulse training set, carrying out optimization processing on the BP neural network through a genetic algorithm to obtain a genetic-BP neural network model.
Based on the foregoing step S3, to obtain the genetic-BP neural network model, it should be noted that:
s3-1, determining the number of input layers, output layers and hidden layers of the BP neural network;
and S3-2, optimizing the connection weight and the threshold value in the BP neural network based on a genetic algorithm.
In this embodiment, the connection weight includes: the connection weight from the input layer to the hidden layer and the connection weight from the hidden layer to the output layer.
Based on the foregoing step S3-2, the connection weights and the thresholds in the BP neural network are optimized, and it should be noted that:
s3-2-1, randomly initializing a population;
s3-2-2, training and testing each individual in the population through a BP neural network, obtaining a prediction error corresponding to the acousto-optic pulse, and constructing an objective function based on the prediction error;
s3-2-3, constructing a population fitness value based on the objective function, and constructing a sub-population based on the fitness value;
s3-2-4, randomly selecting 2 individuals in the sub population, and obtaining a new sub population by exchanging variables to be optimized corresponding to the selected 2 individuals;
s3-2-5, randomly selecting 1 individual based on the new sub-population; randomly selecting corresponding variables to be optimized based on the variation probability aiming at the selected individuals, and randomly mutating the corresponding variables to be optimized to obtain a final sub-population;
s3-2-6, training and testing through a BP neural network based on each individual in the final sub population, re-obtaining a prediction error corresponding to the acoustic pulse and a prediction error corresponding to the optical pulse, and re-constructing an objective function;
s3-2-7, acquiring a first generation optimal solution, an optimal weight and an optimal threshold value aiming at the acousto-optic signal based on the final sub population and the reconstructed objective function;
s3-2-8, based on the final sub population, repeating the steps S3-2-3 to S3-2-6 until the final optimal solution, the optimal weight and the optimal threshold value aiming at the sound and light pulse are obtained.
S4, obtaining a predicted value of the first acousto-optic pulse prediction residual error through a genetic-BP neural network model based on the first acousto-optic pulse prediction residual error; and acquiring a first acousto-optic pulse predicted value through a genetic-BP neural network model based on the acousto-optic pulse test set.
S5, obtaining a second acousto-optic pulse predicted value based on the first acousto-optic pulse predicted value and the predicted value of the first acousto-optic pulse predicted residual; and acquiring a state transition matrix through a Markov algorithm based on the second sound light pulse predicted value and the sound light pulse verification set.
Based on the foregoing step S5, for the acquisition of the state transition matrix, it should be noted that:
firstly, acquiring a second sound light pulse prediction residual based on a second sound light pulse prediction value and a sound light pulse verification set, carrying out threshold division based on the second sound light pulse prediction residual, and marking the state of the second sound light pulse prediction residual;
secondly, acquiring a corresponding state transition probability based on the marked state of the second optical pulse prediction residual, and acquiring a state transition matrix corresponding to the second optical pulse prediction residual based on the state transition probability.
And S6, acquiring a final acousto-optic pulse predicted value through an ARIMA model, a genetic-BP neural network model and a state transition matrix based on the acousto-optic pulse training set.
Based on the foregoing step S6, it should be noted that, for obtaining the final predicted value of the acousto-optic pulse:
firstly, taking an acousto-optic pulse training set as an input variable of an ARIMA model and a genetic-BP neural network model to obtain a fourth pulse predicted value;
and secondly, acquiring a final acousto-optic pulse predicted value through a state transition matrix based on the fourth pulse predicted value.
And S7, pre-warning the sudden accident of the accident pre-warning area based on the preset acousto-optic pulse conventional conditions and the final acousto-optic pulse predicted value.
Based on the foregoing step S7, for the emergency related to the acousto-optic linkage early warning, it should be noted that: and if the final acousto-optic pulse predicted value exceeds the numerical value change range of the acousto-optic pulse in the non-accident state, judging that the accident is a sudden accident.
In this embodiment, the sound and light pulse condition is determined based on historical sound and light pulses of the accident early warning area.
In this embodiment, based on the final predicted value of the acousto-optic pulse, the grade of the accident is divided, and the grade division specifically includes:
firstly, if the final predicted value of the light pulse exceeds the numerical variation range of the light pulse in the non-accident state and the final predicted value of the sound pulse does not exceed the numerical variation range of the light pulse in the non-accident state, judging as a primary sudden accident;
secondly, if the final sound pulse predicted value exceeds the numerical value change range of the sound pulse in the non-accident state and the final light pulse predicted value does not exceed the numerical value change range of the light pulse in the non-accident state, judging the secondary sudden accident;
and thirdly, if the final light pulse predicted value exceeds the numerical change range of the light pulse in the non-accident state and the final sound pulse predicted value exceeds the numerical change range of the light pulse in the non-accident state, judging the three-level accident.
In this embodiment, the higher the level of the sudden accident, the greater the severity of the corresponding accident.
Based on the acousto-optic linkage early warning method based on deep learning in the first embodiment, the acousto-optic pulse is extracted by acquiring real-time acousto-optic signals and based on the acousto-optic signals; based on the extracted sound and light combined pulse, an ARIMA model is constructed, deep learning is carried out based on a BP neural network optimized by a genetic algorithm, and error adjustment is carried out on the ARIMA model and the predicted sound and light signals corresponding to the BP neural network optimized by the genetic algorithm by utilizing a Markov state transition transfer matrix obtained based on historical sound and light signals, so that the prediction error is further reduced, and further, based on the obtained simulated sound and light pulse sequence, grading early warning is realized in a preset region.
It should be noted that in the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The word "comprising" does not exclude the presence of elements or steps not listed in a claim. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. Furthermore, it should be noted that in the description of the present specification, the description of the term "one embodiment", "some embodiments", "examples", "specific examples" or "some examples", etc., means that a specific feature, structure, material or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, the claims should be construed to include preferred embodiments and all such variations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various modifications and variations can be made in the present invention without departing from the spirit or scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention should also include such modifications and variations.
Claims (10)
1. A sound-light linkage early warning method based on deep learning is characterized by comprising the following steps:
s1, acquiring acoustic pulses corresponding to historical acoustic signals and optical pulses corresponding to historical optical signals of an accident early warning area, and preprocessing the acoustic pulses and the optical pulses;
the sound and light pulses comprise: an acousto-optic pulse training set, an acousto-optic pulse testing set and an acousto-optic pulse verification set;
s2, constructing an ARIMA model based on the acousto-optic pulse training set, and acquiring a first acousto-optic pulse prediction residual error through the ARIMA model based on the acousto-optic pulse testing set;
s3, based on the acousto-optic pulse training set, carrying out optimization processing on the BP neural network through a genetic algorithm to obtain a genetic-BP neural network model;
s4, obtaining a predicted value of the first acousto-optic pulse prediction residual error through a genetic-BP neural network model based on the first acousto-optic pulse prediction residual error; based on the acousto-optic pulse test set, obtaining a first acousto-optic pulse predicted value through a genetic-BP neural network model;
s5, obtaining a second acousto-optic pulse predicted value based on the first acousto-optic pulse predicted value and the predicted value of the first acousto-optic pulse predicted residual; acquiring a state transition matrix through a Markov algorithm based on the second sound light pulse predicted value and the sound light pulse verification set;
s6, acquiring a final acousto-optic pulse prediction value through an ARIMA model, a genetic-BP neural network model and a state transition matrix based on an acousto-optic pulse training set;
and S7, pre-warning the sudden accident of the accident pre-warning area based on the preset acousto-optic pulse conventional conditions and the final acousto-optic pulse predicted value.
2. The acousto-optic linkage early warning method according to claim 1, wherein the S1 includes:
acquiring an acoustic signal through an acoustic sensor, and extracting an acoustic pulse based on the acoustic signal;
an optical signal is acquired by a light sensor, and an optical pulse is extracted based on the optical signal.
3. The acousto-optic linkage early warning method according to claim 1, wherein the S2 includes:
performing sequence inspection on the acousto-optic pulse training set, and fitting an ARIMA model by a least square method based on the inspected acousto-optic pulse training set;
taking the acousto-optic pulse test set as an input variable of an ARIMA model to obtain a third acousto-optic pulse predicted value;
and acquiring a first acousto-optic pulse prediction residual error based on the third acousto-optic pulse prediction value and the acousto-optic pulse verification set.
4. The acousto-optic linkage early warning method according to claim 1, wherein the S3 includes: the establishing process of the genetic-BP neural network model specifically comprises the following steps:
s2-1, determining the number of input layers, output layers and hidden layers of the BP neural network;
s2-2, optimizing the connection weight and the threshold value in the BP neural network based on a genetic algorithm;
the connection weight includes: the connection weight from the input layer to the hidden layer and the connection weight from the hidden layer to the output layer.
5. The acousto-optic linkage early warning method according to claim 1, wherein the S5 includes:
s5-1, acquiring a second sound light pulse prediction residual error based on a second sound light pulse prediction value and a sound light pulse verification set, performing threshold division based on the second sound light pulse prediction residual error, and marking the state of the second sound light pulse prediction residual error;
and S5-2, acquiring corresponding state transition probability based on the marked second sound pulse prediction residual state, and acquiring a state transition matrix corresponding to the second sound pulse prediction residual based on the state transition probability.
6. The acousto-optic linkage early warning method according to claim 1, wherein the S6 includes:
taking the acousto-optic pulse training set as an input variable of an ARIMA model and a genetic-BP neural network model to obtain a fourth pulse predicted value;
and acquiring a final acousto-optic pulse predicted value through the state transition matrix based on the fourth pulse predicted value.
7. The acousto-optic linkage early warning method according to claim 1, wherein the S7 includes:
and if the final acousto-optic pulse predicted value exceeds the numerical value change range of the acousto-optic pulse in the non-accident state, judging that the accident is a sudden accident.
8. The acousto-optic linkage early warning method according to claim 1, wherein the S7 includes: and determining the acousto-optic pulse condition based on historical acousto-optic and optic pulses in the accident early warning area.
9. The acousto-optic linkage early warning method according to claim 1, wherein the S7 includes:
if the final predicted value of the optical pulse exceeds the numerical variation range of the optical pulse in the non-accident state and the final predicted value of the acoustic pulse does not exceed the numerical variation range of the optical pulse in the non-accident state, determining that a primary accident happens;
if the final sound pulse predicted value exceeds the numerical value change range of the sound pulse in the non-accident state and the final light pulse predicted value does not exceed the numerical value change range of the light pulse in the non-accident state, judging the secondary sudden accident;
and if the final predicted value of the light pulse exceeds the numerical variation range of the light pulse in the non-accident state and the final predicted value of the sound pulse exceeds the numerical variation range of the light pulse in the non-accident state, judging the three-stage sudden accident.
10. An electronic device, comprising:
one or more processors;
a storage device for storing one or more programs which, when executed by the one or more processors, cause the one or more processors to implement the deep learning based acousto-optic coordinated warning method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210801438.4A CN115131943B (en) | 2022-07-07 | 2022-07-07 | Acousto-optic linkage early warning method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210801438.4A CN115131943B (en) | 2022-07-07 | 2022-07-07 | Acousto-optic linkage early warning method based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115131943A true CN115131943A (en) | 2022-09-30 |
| CN115131943B CN115131943B (en) | 2023-10-31 |
Family
ID=83381911
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210801438.4A Active CN115131943B (en) | 2022-07-07 | 2022-07-07 | Acousto-optic linkage early warning method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115131943B (en) |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0552940A (en) * | 1991-08-22 | 1993-03-02 | Nec Corp | Discriminating device for degree of fluctuation of acoustic signal |
| JPH06149297A (en) * | 1992-10-31 | 1994-05-27 | Sony Corp | Method for calculating short-period prediction coefficient |
| JP2010073045A (en) * | 2008-09-19 | 2010-04-02 | Panasonic Electric Works Co Ltd | Fire alarm |
| CN101859146A (en) * | 2010-07-16 | 2010-10-13 | 哈尔滨工业大学 | Satellite fault prediction method based on predictive filtering and empirical mode decomposition |
| CN102129752A (en) * | 2010-01-12 | 2011-07-20 | 西安英诺视通信息技术有限公司 | System for online monitoring and alarming artificial destruction of outdoor facilities |
| CN102208028A (en) * | 2011-05-31 | 2011-10-05 | 北京航空航天大学 | Fault predicting and diagnosing method suitable for dynamic complex system |
| CN103323007A (en) * | 2013-06-17 | 2013-09-25 | 南京航空航天大学 | Robust federated filtering method based on time-variable measurement noise |
| CN104359977A (en) * | 2014-10-22 | 2015-02-18 | 北京理工大学 | Acoustic surface wave high-order nonlinear parameter representation method for bending fatigue state of metal plate |
| CN104537034A (en) * | 2014-12-22 | 2015-04-22 | 国家电网公司 | Electric-transmission-and-transformation-equipment state monitoring data cleaning method based on time series analysis |
| CN105719002A (en) * | 2016-01-18 | 2016-06-29 | 重庆大学 | Wind turbine generator state parameter abnormity identification method based on combination prediction |
| CN106767898A (en) * | 2016-11-17 | 2017-05-31 | 中国人民解放军国防科学技术大学 | A kind of method for detecting measuring system of satellite attitude small fault |
| CN107908891A (en) * | 2017-11-28 | 2018-04-13 | 河海大学 | A kind of Hydrological Time Series rejecting outliers method based on ARIMA SVR |
| CN108053054A (en) * | 2017-11-22 | 2018-05-18 | 中国农业大学 | A kind of water quality prediction of river method |
| CN109831264A (en) * | 2018-02-07 | 2019-05-31 | 中国科学院计算技术研究所 | The timing underwater acoustic channel prediction of quality algorithm and system returned based on nearest-neighbors |
| CN110297712A (en) * | 2019-05-20 | 2019-10-01 | 贵州联科卫信科技有限公司 | A kind of ARIMA combination forecasting method towards block chain node load estimation |
| CN112285570A (en) * | 2020-10-29 | 2021-01-29 | 哈尔滨工业大学（威海） | Electric vehicle fault diagnosis method based on attenuation memory filter |
| CN113742932A (en) * | 2021-09-13 | 2021-12-03 | 华能新能源股份有限公司 | Method for predicting and diagnosing faults of gearbox of wind turbine generator |
| CN114330517A (en) * | 2021-12-15 | 2022-04-12 | 南京航空航天大学 | Neural network-based aircraft engine sensor fault self-diagnosis method |
-
2022
- 2022-07-07 CN CN202210801438.4A patent/CN115131943B/en active Active
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0552940A (en) * | 1991-08-22 | 1993-03-02 | Nec Corp | Discriminating device for degree of fluctuation of acoustic signal |
| JPH06149297A (en) * | 1992-10-31 | 1994-05-27 | Sony Corp | Method for calculating short-period prediction coefficient |
| JP2010073045A (en) * | 2008-09-19 | 2010-04-02 | Panasonic Electric Works Co Ltd | Fire alarm |
| CN102129752A (en) * | 2010-01-12 | 2011-07-20 | 西安英诺视通信息技术有限公司 | System for online monitoring and alarming artificial destruction of outdoor facilities |
| CN101859146A (en) * | 2010-07-16 | 2010-10-13 | 哈尔滨工业大学 | Satellite fault prediction method based on predictive filtering and empirical mode decomposition |
| CN102208028A (en) * | 2011-05-31 | 2011-10-05 | 北京航空航天大学 | Fault predicting and diagnosing method suitable for dynamic complex system |
| CN103323007A (en) * | 2013-06-17 | 2013-09-25 | 南京航空航天大学 | Robust federated filtering method based on time-variable measurement noise |
| CN104359977A (en) * | 2014-10-22 | 2015-02-18 | 北京理工大学 | Acoustic surface wave high-order nonlinear parameter representation method for bending fatigue state of metal plate |
| CN104537034A (en) * | 2014-12-22 | 2015-04-22 | 国家电网公司 | Electric-transmission-and-transformation-equipment state monitoring data cleaning method based on time series analysis |
| CN105719002A (en) * | 2016-01-18 | 2016-06-29 | 重庆大学 | Wind turbine generator state parameter abnormity identification method based on combination prediction |
| CN106767898A (en) * | 2016-11-17 | 2017-05-31 | 中国人民解放军国防科学技术大学 | A kind of method for detecting measuring system of satellite attitude small fault |
| CN108053054A (en) * | 2017-11-22 | 2018-05-18 | 中国农业大学 | A kind of water quality prediction of river method |
| CN107908891A (en) * | 2017-11-28 | 2018-04-13 | 河海大学 | A kind of Hydrological Time Series rejecting outliers method based on ARIMA SVR |
| CN109831264A (en) * | 2018-02-07 | 2019-05-31 | 中国科学院计算技术研究所 | The timing underwater acoustic channel prediction of quality algorithm and system returned based on nearest-neighbors |
| CN110297712A (en) * | 2019-05-20 | 2019-10-01 | 贵州联科卫信科技有限公司 | A kind of ARIMA combination forecasting method towards block chain node load estimation |
| CN112285570A (en) * | 2020-10-29 | 2021-01-29 | 哈尔滨工业大学（威海） | Electric vehicle fault diagnosis method based on attenuation memory filter |
| CN113742932A (en) * | 2021-09-13 | 2021-12-03 | 华能新能源股份有限公司 | Method for predicting and diagnosing faults of gearbox of wind turbine generator |
| CN114330517A (en) * | 2021-12-15 | 2022-04-12 | 南京航空航天大学 | Neural network-based aircraft engine sensor fault self-diagnosis method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115131943B (en) | 2023-10-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107180070B (en) | Automatic risk information classification, identification and early warning method and system | |
| CN108764601A (en) | A kind of monitoring structural health conditions abnormal data diagnostic method based on computer vision and depth learning technology | |
| CN111415492B (en) | Slope landslide early warning method and system based on fuzzy comprehensive evaluation algorithm | |
| CN107332698A (en) | A kind of Security Situation Awareness Systems and method towards bright Great Wall intelligent perception system | |
| CN111651505B (en) | Equipment operation situation analysis and early warning method and system based on data driving | |
| CN108596507B (en) | Disaster damage early warning method for mariculture net cage | |
| CN109657880A (en) | A kind of Collision Accidents of Ships grade prediction technique and system based on Bayesian network | |
| CN114357594B (en) | Bridge abnormity monitoring method, system, equipment and storage medium based on SCA-GRU | |
| CN111325947B (en) | Storm surge flood early warning method and system based on long-term and short-term memory network | |
| CN111625988A (en) | Bridge health management analysis and prediction system and method based on deep learning | |
| CN113033011B (en) | Ship mechanical health state assessment method and system | |
| CN111223027A (en) | Urban earthquake disaster risk assessment method and system | |
| CN115774953B (en) | Pollution space-time risk supervision and evaluation system and method based on data processing | |
| CN116029555B (en) | Bridge risk identification early warning system based on lightweight neural network and application method | |
| CN111199313A (en) | Method and system for predicting landslide accumulated displacement trend based on neural network | |
| CN116677569A (en) | Fan tower stress monitoring and early warning method and system | |
| CN116680643A (en) | Abnormality detection method combining deep contrast learning and density clustering | |
| CN116720728A (en) | Risk assessment method, electronic device and storage medium | |
| CN115470850A (en) | Water quality abnormal event recognition early warning method based on pipe network water quality time-space data | |
| CN117574321B (en) | Dam safety assessment method based on mass monitoring and inspection information fusion model | |
| CN114841518A (en) | Service state evaluation method for inland river navigation aid sign | |
| CN111191855A (en) | Water quality abnormal event identification and early warning method based on pipe network multi-element water quality time sequence data | |
| CN115131943A (en) | Acousto-optic linkage early warning method based on deep learning | |
| JPH0392795A (en) | Diagnostic method of nuclear power plant | |
| CN115296933B (en) | Industrial production data risk level assessment method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A deep learning based sound and light linkage early warning method Granted publication date: 20231031 Pledgee: Guotou Taikang Trust Co.,Ltd. Pledgor: Hangzhou Shenhao Technology Co.,Ltd. Registration number: Y2024980011357 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |