CN113435105A - Fire early warning system and method based on smoke detection - Google Patents
Fire early warning system and method based on smoke detection Download PDFInfo
- Publication number
- CN113435105A CN113435105A CN202110573226.0A CN202110573226A CN113435105A CN 113435105 A CN113435105 A CN 113435105A CN 202110573226 A CN202110573226 A CN 202110573226A CN 113435105 A CN113435105 A CN 113435105A
- Authority
- CN
- China
- Prior art keywords
- early warning
- fire
- value
- smoke
- sensor
- 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
- 239000000779 smoke Substances 0.000 title claims abstract description 80
- 238000001514 detection method Methods 0.000 title claims abstract description 30
- 238000000034 method Methods 0.000 title claims abstract description 30
- 230000004927 fusion Effects 0.000 claims abstract description 45
- 238000011156 evaluation Methods 0.000 claims abstract description 23
- 238000013528 artificial neural network Methods 0.000 claims abstract description 21
- 238000005259 measurement Methods 0.000 claims abstract description 14
- 230000006870 function Effects 0.000 claims description 14
- 210000002569 neuron Anatomy 0.000 claims description 11
- 239000013598 vector Substances 0.000 claims description 11
- 238000005507 spraying Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000012544 monitoring process Methods 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 4
- 238000007500 overflow downdraw method Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 239000000126 substance Substances 0.000 claims description 4
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 230000005284 excitation Effects 0.000 claims description 2
- 230000004044 response Effects 0.000 abstract description 4
- 238000012549 training Methods 0.000 description 11
- 230000008901 benefit Effects 0.000 description 8
- 241000269350 Anura Species 0.000 description 6
- 210000004027 cell Anatomy 0.000 description 3
- 230000035945 sensitivity Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/10—Actuation by presence of smoke or gases, e.g. automatic alarm devices for analysing flowing fluid materials by the use of optical means
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/08—Thermal analysis or thermal optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Chemical & Material Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Analytical Chemistry (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Health & Medical Sciences (AREA)
- Geometry (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Fire Alarms (AREA)
Abstract
The invention discloses a fire early warning method based on smoke detection, which comprises the steps of obtaining smoke concentration data of a smoke sensor at the current moment; if the smoke concentration exceeds a set threshold value, acquiring the position and building information of the current smoke sensor; remote control is carried out, and all measurement data of the smoke sensor and the temperature and humidity sensor at the current moment of the building are obtained; after carrying out homogeneous fusion on the measured data, inputting the measured data into a BP neural network for heterogeneous fusion to obtain a smoke concentration evaluation value and a temperature and humidity evaluation value, and then calculating to obtain an early warning information fusion value; and determining an early warning grade based on the early warning information fusion value, and executing a fire early warning strategy according to the early warning grade. According to the invention, the whole building smoke concentration, temperature and humidity are fused and modeled to determine the early warning level, and the occurrence probability of the fire is pre-judged, so that the timely early warning and the efficient response to the fire are realized.
Description
Technical Field
The invention relates to the technical field of fire early warning, in particular to a fire early warning system and method based on smoke detection.
Background
Under the current economic and social conditions, high-rise buildings become the mainstream of urban construction, and although the high-rise buildings have many benefits, a plurality of problems also occur, one of which is the fire problem. Compared with the prior high-rise building, the height, the shape, the material, the structure and the function of the prior high-rise building are greatly changed, and the complexity and the difficulty of fire prevention and extinguishment are increased. Because of high floor height, large scale, complete living facilities and more combustible materials, the high-rise building has the disadvantages of rapid fire spread, difficult fire fighting and evacuation and easy huge loss in case of fire. With the development of network technology and intelligent control technology, the traditional fire early warning management is also developing towards intellectualization, so that it is necessary to provide a fire early warning system and method based on smoke detection.
Disclosure of Invention
The invention provides a fire early warning method based on smoke detection, which is characterized in that when the smoke concentration of a smoke sensor exceeds a threshold value, the smoke concentration, temperature and humidity data of the whole building are acquired, the fire early warning level is determined through data fusion modeling, the occurrence probability of a fire is pre-judged, and the timely early warning and efficient response to the fire are realized.
The invention also aims to provide a fire early warning system based on smoke detection, which can be applied to various fire early warning scenes and has the advantages of large early warning range and high sensitivity.
The technical scheme of the invention is as follows:
a fire early warning method based on smoke detection comprises the following steps:
step one, obtaining smoke concentration data of a smoke sensor at the current moment;
step two, if the smoke concentration exceeds a set threshold value, acquiring the position and building information of the current smoke sensor;
step three, remote control is carried out, and all measurement data of the smoke sensor and the temperature and humidity sensor at the current moment of the building are obtained;
step four, inputting the measured data after homogeneous fusion into a BP neural network for heterogeneous fusion to obtain a smoke concentration evaluation value and a temperature and humidity evaluation value, and then calculating to obtain an early warning information fusion value;
and fifthly, determining an early warning grade based on the early warning information fusion value, and executing a fire early warning strategy according to the early warning grade.
Preferably, the step four specifically includes the following steps:
threshold segmentation is carried out on the smoke concentration data and the temperature and humidity data, interference data are deleted, and then normalization processing is carried out;
calculating the weight of the sensor based on a self-adaptive weighted fusion method to obtain a state value of the measured data;
establishing a prediction model based on a BP neural network algorithm to obtain an evaluation value of the measured data;
and calculating an early warning information fusion value according to the evaluation value.
Preferably, the adaptive weighted fusion specifically includes the following steps:
calculating the variance and the total mean square error of the data to obtain a function equation of the total mean square error relative to the weight of the sensor;
the state model obtained by analyzing the function equation is:
wherein the content of the first and second substances,representing a sensor state value, aiRepresenting sensor measurements, piRepresenting sensor weight, i representing sensor, S2Representing the total mean square error and n representing the number of sensors.
Preferably, the deep neural network algorithm includes:
establishing a three-layer BP neural network;
determining input layer neuron vector x ═ (x)1,x2,x3)TWherein x is1Representing the state value of smoke concentration, x2Representing the value of the temperature state, x3Represents a humidity state value;
mapping input layer vectors to hidden layers, wherein the number of neurons of the hidden layers is m;
obtaining the output layer neuron vector o ═ (o)1,o2,o3)TWherein o is1Indicates the evaluation value of smoke density, o2Indicates the temperature evaluation value, o3Indicating the humidity evaluation value.
Preferably, the BP neural network algorithm further comprises parameter optimization of the number of hidden layer neurons of the BP neural network and the learning rate of the reverse fine tuning algorithm based on the mixed frog-leaping algorithm.
Preferably, the excitation functions of the hidden layer and the output layer adopt S-shaped functions
Preferably, the calculation formula of the early warning information fusion value is as follows:
λ denotes the fusion value, λ0Denotes the standard fusion value, ξ denotes the correction factor, e denotes the base of the natural logarithm.
Preferably, the early warning level includes:
when lambda is more than 0.45 lambda0Then, the early warning level is extreme risk;
when 0.35 lambda0≤λ≤0.45λ0Then, the early warning grade is severe risk;
when 0.15 lambda0≤λ<0.35λ0Then, the early warning level is a high risk;
when the value is 0.05 lambda0≤λ<0.15λ0Then, the early warning grade is moderate risk;
when lambda is less than 0.05 lambda0The early warning level is low risk.
Preferably, the early warning strategy comprises:
when the early warning level is extreme risk and severe risk, immediately starting fire-extinguishing spraying equipment, sending early warning information to a user, whistling by an alarm, and alarming to a fire center;
when the early warning level is high risk, immediately starting the fire extinguishing spraying equipment, sending early warning information to a user, and whistling by an alarm;
when the early warning level is moderate risk, immediately starting the fire extinguishing spraying equipment and sending early warning information to a user;
and when the early warning level is low risk, immediately sending early warning information to the user.
A fire early warning system based on smoke detection comprises:
the smoke sensor is arranged in a building in a fire monitoring area and used for collecting space smoke concentration data;
the temperature and humidity sensor is arranged in a building in a fire monitoring area and used for acquiring space temperature and humidity data;
the cloud computing platform is used for fusing data to obtain an early warning information fusion value;
a controller for performing policy configuration according to the fusion value;
the short message sending module is used for sending early warning information to a user;
and the user management module is used for recording and storing the user operation log.
The invention has the beneficial effects that:
1. according to the fire early warning method based on smoke detection provided by the invention, when the smoke concentration of a smoke sensor exceeds a threshold value, the smoke concentration, temperature and humidity data of the whole building are obtained, the fire early warning grade is determined through data fusion modeling, the occurrence probability of fire is pre-judged, and timely early warning and efficient response to the fire are realized.
2. According to the fire early warning method based on smoke detection, the sensor state model is constructed based on the self-adaptive weighted fusion method for homogenous fusion, the influence of sensor measurement difference on consistency measurement is eliminated, and the influence of interference data on the state value is effectively reduced.
3. According to the fire early warning method based on smoke detection, provided by the invention, a BP neural network model is established for heterogeneous fusion, and the model is optimized based on a mixed frog-leaping algorithm, so that the neural network algorithm has the advantages of small error, high calculation precision and higher convergence speed.
4. The invention also provides a fire early warning system based on smoke detection, which can be applied to various fire early warning scenes and has the advantages of large early warning range and high sensitivity.
Drawings
Fig. 1 is a flowchart of a fire early warning method based on smoke detection according to the present invention.
Fig. 2 is a flowchart of a method for obtaining an early warning data fusion value according to an embodiment of the present invention.
Fig. 3 is a frame diagram of a fire early warning system based on smoke detection according to the present invention.
Detailed Description
The present invention is described in terms of particular embodiments, other advantages and features of the invention will become apparent to those skilled in the art from the following disclosure, and it is to be understood that the described embodiments are merely exemplary of the invention and that it is not intended to limit the invention to the particular embodiments disclosed. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that in the description of the present invention, the terms "in", "upper", "lower", "lateral", "inner", etc. indicate directions or positional relationships based on those shown in the drawings, which are merely for convenience of description, and do not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
Furthermore, it should be noted that, in the description of the present invention, unless otherwise explicitly specified or limited, the terms "disposed," "mounted," "connected," and "connected" are to be construed broadly and may be, for example, fixedly connected, detachably connected, or integrally connected; may be a mechanical connection; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
As shown in fig. 1, the fire early warning method based on smoke detection includes:
and S110, acquiring smoke concentration data of the smoke sensor at the current moment.
And S120, if the smoke concentration exceeds a set threshold value, acquiring the position and building information of the current smoke sensor.
S130, remotely controlling to obtain the measurement data of all smoke sensors and temperature and humidity sensors at the current moment of the building;
s140, carrying out homogeneous fusion on the measured data, inputting the measured data into a BP neural network for heterogeneous fusion to obtain a smoke concentration evaluation value and a temperature and humidity evaluation value, and then calculating to obtain an early warning information fusion value;
s150, determining an early warning grade based on the early warning information fusion value, and executing a fire early warning strategy according to the early warning grade.
Further, step S140 specifically includes:
and S141, performing threshold segmentation on the smoke concentration data and the temperature and humidity data, deleting interference data, and performing normalization processing.
The method has the advantages that the measurement data of all smoke sensors and greenhouse sensors of the building are obtained, the number of the measurement data is large, the calculation complexity and the feature extraction difficulty are greatly increased, invalid interference data can be deleted through threshold segmentation, then normalization processing is carried out, and the influences of dimension and magnitude among variables are eliminated.
Wherein the content of the first and second substances,denotes the normalized value, xiRepresenting sampled values, xminRepresenting the minimum sample value, xmaxRepresenting the maximum sample value.
And S142, calculating the weight of the sensor based on the self-adaptive weighted fusion method to obtain the state value of the measured data.
Calculating the variance and the total mean square error of the data to obtain a function equation of the total mean square error relative to the weight of the sensor;
The total mean square error is then:
since the measured values are independent of each other and are unbiased estimates of a, an unbiased estimate of a can be obtained
The total mean square error can be obtained:
according to the theory of extreme value of multivariate function, the current value can be obtainedAnd then, establishing a fusion model by using the corresponding minimum mean square error:
wherein the content of the first and second substances,representing a sensor state value, aiRepresenting sensor measurements, piRepresenting sensor weight, i representing sensor, S2Representing the total mean square error, n representing the number of sensors,having the advantage of viewing state estimation, delta2The variance is indicated.
S143, establishing a prediction model based on the BP neural network algorithm, and obtaining an evaluation value of the measured data.
The BP neural network system structure adopted by the invention comprises three layers, wherein the first layer is an input layer and has m nodes corresponding to m detection signals representing submarine cables, the second layer is a hidden layer and has k nodes, and the number of the nodes is determined by the training process of the network in a self-adaptive mode; the third layer is an output layer and has l nodes.
The mathematical model of the network is:
inputting a vector: x ═ x1,x2,…xm)T;
Intermediate layer vector: y ═ y1,y2,…yk)T;
Outputting a vector: o ═ o (o)1,o2,…ol)T;
The input layer 3 parameters are respectively expressed as: x is the number of1Representing the state value of smoke concentration, x2Representing the value of the temperature state, x3Represents a humidity state value;
the output layer has 3 parameters expressed as: o1Indicates the evaluation value of smoke density, o2Indicates the temperature evaluation value, o3Indicating the humidity evaluation value.
And (5) training the BP neural network.
After the BP neural network node model is established, the training of the BP neural network can be carried out. And obtaining a training sample according to historical experience data, and giving a connection weight between the input node i and the hidden layer node j and a connection weight between the hidden layer node j and the output layer node k.
(1) Training method
Each subnet adopts a separate training method; when training, firstly providing a group of training samples, wherein each sample consists of an input sample and an ideal output pair, and when all actual outputs of the network are consistent with the ideal outputs of the network, the training is finished; otherwise, the ideal output of the network is consistent with the actual output by correcting the weight.
(2) Training algorithm
The BP network is trained by using a back Propagation (Backward Propagation) algorithm, and the steps can be summarized as follows:
the first step is as follows: and selecting a network with a reasonable structure, and setting initial values of all node thresholds and connection weights.
The second step is that: for each input sample, the following calculations are made:
(a) forward calculation: for j unit of l layer
In the formula (I), the compound is shown in the specification,represents the weighted sum of the j unit information of the l layer at the nth calculation,represents the connection weight between the j cell of the l layer and the cell i of the previous layer (i.e. the l-1 layer),the working signal sent by the unit i of the previous layer (i.e. the layer l-1, the node number is nl-1); when i is 0, order Is the threshold of the j cell of the l layer.
If the activation function of the unit j is a sigmoid function, then
And is
(b) And (3) calculating the error reversely:
for the output unit;
for hidden unit
(c) Correcting the weight value:
The third step: inputting a new sample or a new period sample until the network converges, and randomly re-ordering the input sequence of the samples in each period during training.
The BP algorithm adopts a gradient descent method to solve the extreme value of a nonlinear function, and has the problems of local minimum, low convergence speed and the like. A more effective algorithm is a Levenberg-Marquardt optimization algorithm, which enables the network learning time to be shorter and can effectively inhibit the network from being locally minimum. The weight adjustment rate is
Δω=(JTJ+μX)-1JTβ;
Wherein J represents a Jacobian (Jacobian) matrix of error to weight differentiation, X represents an input vector, beta represents an error vector, and the variable mu is a scalar quantity which is self-adaptive and adjusted and is used for determining whether the learning is completed according to a Newton method or a gradient method.
When the system is designed, the system model is a network which is only initialized, the weight needs to be learned and adjusted according to data samples obtained in the using process, and therefore the parameter optimization is designed on the basis of the mixed frog-leaping algorithm on the number of hidden layer neurons of the BP neural network and the learning rate of the reverse fine-tuning algorithm.
The mixed frog leaping algorithm is a heuristic algorithm based on population, so the quality of the initial population is extremely important to the searching performance of the algorithm. When the initial population is unevenly distributed in the feasible domain, the search range of the algorithm is limited to a certain extent, so that the global search capability of the algorithm is reduced. Therefore, the invention adopts the orthogonal design theory to initialize the frog population. Group mpSmaller, not only results in reduced communication between groups, but also slows the algorithm's solution speed, so mpThe amount should not be too small. Number n of frogs in the grouppThis increase in frog population will result in a larger number of frog populations, with more frogs approaching the optimal solution, but in population mpIn the case of definite, npThe increase of (2) does not improve the solving precision but increases the calculation cost. Number of local iterations N1If the value is too small, the meaning of local depth search is lost, and if the value is too large, the efficiency of the algorithm is greatly reduced, and the selected algorithm parameters are shown in table 1.
In order to ensure the diversity of frog clan groups and prevent from falling into a local optimal solution, when q frogs are randomly selected to construct a sub-group, the frogs with larger fitness value are endowed with larger weight, the frogs with smaller fitness value are endowed with smaller weight, and the weight distribution formula is as follows:
wherein p isjRepresenting the optimal weight, npRepresenting the number of frogs in the group.
TABLE 1 Combined frog-leaping algorithm number
And S144, calculating an early warning information fusion value according to the evaluation value.
The calculation formula of the early warning information fusion value is as follows:
wherein, lambda represents the early warning information fusion value, lambda0Denotes the standard fusion value, ξ denotes the correction factor, e denotes the base of the natural logarithm.
Standard fusion value lambda0The correction coefficient xi is finely adjusted by multiple times of fusion iteration of the collected data set.
When lambda is more than 0.45 lambda0And when the early warning level is extreme risk, the fire extinguishing spraying equipment is immediately started, early warning information is sent to a user, and the alarm whistles and gives an alarm to a fire center.
When 0.35 lambda0≤λ≤0.45λ0And when the early warning level is severe risk, immediately starting the fire-extinguishing spraying equipment, sending early warning information to a user, and alarming by an alarm to a fire center.
When 0.15 lambda0≤λ<0.35λ0When the early warning level is high risk, the fire extinguishing spraying equipment is immediately startedAnd sending early warning information to the user, and the alarm whistling.
When the value is 0.05 lambda0≤λ<0.15λ0And when the early warning level is moderate risk, immediately starting the fire extinguishing spraying equipment and sending early warning information to the user.
When lambda is less than 0.05 lambda0And when the early warning level is low risk, immediately sending early warning information to the user.
According to the fire early warning method based on smoke detection provided by the invention, when the smoke concentration of a smoke sensor exceeds a threshold value, the smoke concentration, temperature and humidity data of the whole building are obtained, the fire early warning grade is determined through data fusion modeling, the occurrence probability of fire is pre-judged, and timely early warning and efficient response to the fire are realized.
As shown in fig. 3, the fire early warning system based on smoke detection includes a smoke sensor, a temperature and humidity sensor, a cloud computing platform, a controller, a short message sending module, and a user management module.
The smoke sensor is arranged in a building in a fire monitoring area and used for collecting space smoke concentration data;
the temperature and humidity sensor is arranged in a building in a fire monitoring area and used for acquiring space temperature and humidity data;
the cloud computing platform is used for fusing data to obtain an early warning information fusion value;
a controller for performing policy configuration according to the fusion value;
the short message sending module is used for sending early warning information to a user;
and the user management module is used for recording and storing the user operation log.
The fire early warning system based on smoke detection provided by the invention can be applied to various fire early warning scenes, and has the advantages of large early warning range and high sensitivity.
The above descriptions are only examples of the present invention, and common general knowledge of known specific structures, characteristics, and the like in the schemes is not described herein too much, and it is easily understood by those skilled in the art that the scope of the present invention is obviously not limited to these specific embodiments. Without departing from the invention, several changes and modifications can be made, which should also be regarded as the protection scope of the invention, and these will not affect the effect of the invention and the practicality of the patent.
Claims (10)
1. A fire early warning method based on smoke detection is characterized by comprising the following steps:
step one, obtaining smoke concentration data of a smoke sensor at the current moment;
step two, if the smoke concentration exceeds a set threshold value, acquiring the position and building information of the current smoke sensor;
step three, remote control is carried out, and measurement data of all smoke sensors and temperature and humidity sensors at the current moment of the building are obtained;
step four, inputting the measurement data after homogeneous fusion into a BP neural network for heterogeneous fusion to obtain a smoke concentration evaluation value and a temperature and humidity evaluation value, and then calculating to obtain an early warning information fusion value;
and fifthly, determining an early warning grade based on the early warning information fusion value, and executing a fire early warning strategy according to the early warning grade.
2. The fire early warning method based on smoke detection as claimed in claim 1, wherein the fourth step specifically comprises the following steps:
threshold segmentation is carried out on the smoke concentration data and the temperature and humidity data, interference data are deleted, and then normalization processing is carried out;
calculating the weight of the sensor based on a self-adaptive weighted fusion method to obtain a state value of the measurement data;
establishing a prediction model based on a BP neural network algorithm to obtain an evaluation value of the measurement data;
and calculating an early warning information fusion value according to the evaluation value.
3. The fire early warning method based on smoke detection as claimed in claim 2, wherein the adaptive weighted fusion specifically comprises the following steps:
calculating the variance and the total mean square error of the data to obtain a function equation of the total mean square error relative to the weight of the sensor;
analyzing the function equation to obtain a state model as follows:
4. The fire early warning method based on smoke detection as claimed in claim 3, wherein the deep neural network algorithm comprises:
establishing a three-layer BP neural network;
determining input layer neuron vector x ═ (x)1,x2,x3)TWherein x is1Representing the state value of smoke concentration, x2Representing the value of the temperature state, x3Represents a humidity state value;
the input layer vector is mapped to a hidden layer, and the number of neurons of the hidden layer is m;
obtaining the output layer neuron vector o ═ (o)1,o2,o3)TWherein o is1Indicates the evaluation value of smoke density, o2Indicates the temperature evaluation value, o3Indicating the humidity evaluation value.
5. The fire early warning method based on smoke detection according to claim 4, wherein the BP neural network algorithm further comprises parameter optimization of the number of hidden layer neurons of the BP neural network and the learning rate of a reverse fine tuning algorithm based on a mixed frog-leap algorithm.
7. The fire early warning method based on smoke detection as claimed in claim 6, wherein the calculation formula of the early warning information fusion value is:
wherein, lambda represents the early warning information fusion value, lambda0Denotes the standard fusion value, ξ denotes the correction factor, e denotes the base of the natural logarithm.
8. A fire alerting method based on smoke detection as claimed in claim 7, wherein the alerting levels comprise:
when lambda is more than 0.45 lambda0Then, the early warning level is extreme risk;
when 0.35 lambda0≤λ≤0.45λ0Then, the early warning grade is severe risk;
when 0.15 lambda0≤λ<0.35λ0Then, the early warning level is a high risk;
when the value is 0.05 lambda0≤λ<0.15λ0Then, the early warning grade is moderate risk;
when lambda is less than 0.05 lambda0The early warning level is low risk.
9. The fire early warning method based on smoke detection as claimed in claim 8, wherein the early warning strategy comprises:
when the early warning level is extreme risk and severe risk, immediately starting fire-extinguishing spraying equipment, sending early warning information to a user, whistling by an alarm, and giving an alarm to a fire center;
when the early warning level is high risk, immediately starting fire-extinguishing spraying equipment, sending early warning information to a user, and whistling by an alarm;
when the early warning level is moderate risk, immediately starting fire-extinguishing spraying equipment and sending early warning information to a user;
and when the early warning level is low risk, immediately sending early warning information to the user.
10. A fire early warning system based on smoke detection, comprising:
the smoke sensor is arranged in a building in a fire monitoring area and used for collecting space smoke concentration data;
the temperature and humidity sensor is arranged in a building in a fire monitoring area and used for acquiring space temperature and humidity data;
the cloud computing platform is used for fusing the data to obtain an early warning information fusion value;
a controller for performing policy configuration according to the fusion value;
the short message sending module is used for sending early warning information to a user;
and the user management module is used for recording and storing the user operation log.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110573226.0A CN113435105A (en) | 2021-05-25 | 2021-05-25 | Fire early warning system and method based on smoke detection |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110573226.0A CN113435105A (en) | 2021-05-25 | 2021-05-25 | Fire early warning system and method based on smoke detection |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113435105A true CN113435105A (en) | 2021-09-24 |
Family
ID=77803421
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110573226.0A Pending CN113435105A (en) | 2021-05-25 | 2021-05-25 | Fire early warning system and method based on smoke detection |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113435105A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113841593A (en) * | 2021-10-29 | 2021-12-28 | 山东润浩水利科技有限公司 | Intelligent farmland irrigation system and irrigation method based on Internet of things |
CN114064628A (en) * | 2021-11-25 | 2022-02-18 | 北京中海兴达建设有限公司 | Data processing system for fire early warning of construction site |
CN115049988A (en) * | 2022-08-17 | 2022-09-13 | 南方电网数字电网研究院有限公司 | Edge calculation method and device for power distribution network monitoring and prejudging |
CN115376266A (en) * | 2022-08-19 | 2022-11-22 | 广州市万保职业安全事务有限公司 | AIOT-based fire safety risk early warning method and system |
CN115578832A (en) * | 2022-11-25 | 2023-01-06 | 天津新亚精诚科技有限公司 | Wireless monitoring alarm system applied to fire-fighting Internet of things |
CN116631136A (en) * | 2023-07-26 | 2023-08-22 | 邹城市美安电子科技有限公司 | Intelligent fire alarm system of building floor |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06325270A (en) * | 1993-05-18 | 1994-11-25 | Hochiki Corp | Fire judging device and learning method for neural network |
CN112116776A (en) * | 2020-10-21 | 2020-12-22 | 国网陕西省电力公司电力科学研究院 | System and method for predicting and monitoring cable fire in transformer substation |
-
2021
- 2021-05-25 CN CN202110573226.0A patent/CN113435105A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH06325270A (en) * | 1993-05-18 | 1994-11-25 | Hochiki Corp | Fire judging device and learning method for neural network |
CN112116776A (en) * | 2020-10-21 | 2020-12-22 | 国网陕西省电力公司电力科学研究院 | System and method for predicting and monitoring cable fire in transformer substation |
Non-Patent Citations (1)
Title |
---|
王玉成等: "多传感器火灾检测***的设计", 《智慧工厂》, no. 01, 15 January 2018 (2018-01-15), pages 72 - 75 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113841593A (en) * | 2021-10-29 | 2021-12-28 | 山东润浩水利科技有限公司 | Intelligent farmland irrigation system and irrigation method based on Internet of things |
CN114064628A (en) * | 2021-11-25 | 2022-02-18 | 北京中海兴达建设有限公司 | Data processing system for fire early warning of construction site |
CN115049988A (en) * | 2022-08-17 | 2022-09-13 | 南方电网数字电网研究院有限公司 | Edge calculation method and device for power distribution network monitoring and prejudging |
CN115376266A (en) * | 2022-08-19 | 2022-11-22 | 广州市万保职业安全事务有限公司 | AIOT-based fire safety risk early warning method and system |
CN115578832A (en) * | 2022-11-25 | 2023-01-06 | 天津新亚精诚科技有限公司 | Wireless monitoring alarm system applied to fire-fighting Internet of things |
CN115578832B (en) * | 2022-11-25 | 2023-03-10 | 天津新亚精诚科技有限公司 | Wireless monitoring alarm system applied to fire-fighting Internet of things |
CN116631136A (en) * | 2023-07-26 | 2023-08-22 | 邹城市美安电子科技有限公司 | Intelligent fire alarm system of building floor |
CN116631136B (en) * | 2023-07-26 | 2023-10-03 | 邹城市美安电子科技有限公司 | Intelligent fire alarm system of building floor |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN113435105A (en) | Fire early warning system and method based on smoke detection | |
CN104933841B (en) | A kind of fire prediction method based on self organizing neural network | |
CN105488528B (en) | Neural network image classification method based on improving expert inquiry method | |
CN110473592B (en) | Multi-view human synthetic lethal gene prediction method | |
CN112036540B (en) | Sensor number optimization method based on double-population hybrid artificial bee colony algorithm | |
CN106815782A (en) | A kind of real estate estimation method and system based on neutral net statistical models | |
Jough et al. | Prediction of seismic collapse risk of steel moment frame mid-rise structures by meta-heuristic algorithms | |
CN112488401B (en) | Fire escape route guiding method and system | |
CN111079860B (en) | Method for constructing multi-granularity response model of emergency based on scene similarity | |
CN111988786B (en) | Sensor network covering method and system based on high-dimensional multi-target decomposition algorithm | |
CN108416458A (en) | A kind of tunnel rich water rock mass Synthetic Geological Prediction Ahead of Construction method based on BP neural network | |
CN108594793A (en) | A kind of improved RBF flight control systems fault diagnosis network training method | |
CN106815635A (en) | A kind of forecasting system and method for insulator surface equivalent salt deposit density | |
CN113722980A (en) | Ocean wave height prediction method, system, computer equipment, storage medium and terminal | |
CN116069045B (en) | Radiation environment detection method and system based on mobile robot | |
CN110941902A (en) | Lightning stroke fault early warning method and system for power transmission line | |
CN111445087A (en) | Flood prediction method based on extreme learning machine | |
CN114599069A (en) | Underwater wireless sensor network routing method based on energy self-collection | |
CN109857988A (en) | A kind of safety monitoring method of cold ground assembled Modern Wood Construction | |
CN110910263A (en) | Grain condition monitoring method based on GA-BP neural network | |
CN115357568A (en) | Method, device, processor and storage medium for realizing social information discovery based on time-space domain credible graph neural network calculation | |
CN115510753A (en) | Data collection method based on matrix completion and reinforcement learning in crowd-sourcing network | |
CN113177675A (en) | Air conditioner cold load prediction method based on optimization neural network of longicorn group algorithm | |
CN113889282A (en) | Individual mobile intervention infectious disease prevention and control method and system | |
CN108171333B (en) | Simulation intelligent rescue communication decision optimization method under low communication environment |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210924 |