CN111627181A - Comprehensive pipe rack fire early warning method fusing multi-source parameters and gradient information thereof - Google Patents
Comprehensive pipe rack fire early warning method fusing multi-source parameters and gradient information thereof Download PDFInfo
- Publication number
- CN111627181A CN111627181A CN202010597058.4A CN202010597058A CN111627181A CN 111627181 A CN111627181 A CN 111627181A CN 202010597058 A CN202010597058 A CN 202010597058A CN 111627181 A CN111627181 A CN 111627181A
- Authority
- CN
- China
- Prior art keywords
- fire
- carbon monoxide
- concentration
- temperature
- layer
- 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
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- 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
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Emergency Management (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Educational Administration (AREA)
- Fire Alarms (AREA)
Abstract
The invention discloses a comprehensive pipe gallery fire early warning method fusing multi-source parameters and gradient information thereof, relating to the field of fire early warning, and comprising the steps of S1 establishing a fire risk prediction model; s2, collecting temperature, smoke concentration and carbon monoxide concentration, importing the temperature, smoke concentration and carbon monoxide concentration into a fire risk prediction model, and determining gradient information; s3, determining the fire type and obtaining the fire probability through the gradient information and the fire risk prediction model, and carrying out fire early warning; the method comprises the steps of establishing a fire risk prediction model based on temperature, CO concentration and smoke concentration, performing auxiliary judgment based on gradient information of temperature characteristic parameters and continuous states of CO and smoke concentration change rates, synthesizing environmental change conditions to obtain fire decision values of parameter gradients, weighting and fusing multi-source parameter decision probabilities and the fire decision values to obtain final fire probability values, judging the current fire in a proposed decision mode, and improving the reliability and the real-time performance of fire judgment by taking the output value of a single model as a judgment basis of the fire.
Description
Technical Field
The invention relates to the field of fire early warning, in particular to a comprehensive pipe rack fire early warning method fusing multi-source parameters and gradient information thereof.
Background
Utility tunnel is about to electrical conduit, gas pipeline, for the utility tunnel's operation safety, the inside early warning system that needs to set up multiple calamity of pipe gallery that fuses underground city utility tunnel corridor as an organic whole such as water supply and drainage pipeline, communication, heat supply pipeline. As one of the main monitoring objects of the environmental safety of the comprehensive pipe rack, a fire early warning system with early warning capability and high accuracy needs to be established. Most of the traditional fire detectors take single fire characteristic parameters such as gas concentration, smoke concentration and temperature as fire judgment bases, and are very easy to be interfered by the outside to cause the problems of misinformation and untimely early warning. For example, a fire alarm using temperature monitoring may send an alarm signal only when the temperature exceeds a threshold, and may not monitor a rapid temperature rise in the initial stage of a fire, and may not have a fire alarm capability. And if only a single fire characteristic parameter is monitored, the fire characteristic parameter cannot be associated with other parameters, and the problems of false alarm and missed alarm are easily caused.
Disclosure of Invention
The invention aims to solve the problems and designs a comprehensive pipe rack fire early warning method fusing multi-source parameters and gradient information thereof.
The invention realizes the purpose through the following technical scheme:
the utility model discloses a comprehensive pipe gallery fire early warning method of fusion multisource parameter and its gradient information, includes the following step:
s1, building a fire risk prediction model;
s2, collecting the temperature, the smoke concentration and the carbon monoxide concentration in the comprehensive pipe gallery, introducing the temperature, the smoke concentration and the carbon monoxide concentration into a fire risk prediction model, and respectively determining gradient information of the temperature, the smoke concentration and the carbon monoxide concentration;
and S3, determining the fire type and obtaining the fire probability through the gradient information and the fire risk prediction model, and carrying out fire early warning.
Further, the fire risk prediction model comprises an input layer, a linear change layer, a fuzzification layer, a fuzzy rule layer, a normalization layer, an anti-fuzzification layer and an output layer, the input layer is used for leading in collected temperature, smoke concentration and carbon monoxide concentration in the comprehensive pipe gallery, the linear change layer is used for carrying out data processing on the input data according to the temperature and the smoke concentration, the fuzzification layer is used for determining gradient information of the temperature, the smoke concentration and the carbon monoxide concentration, the fuzzy rule layer is used for determining a fire type, the normalization layer is used for respectively carrying out normalization processing on the temperature, the smoke concentration and the carbon monoxide concentration after the fire type is determined, the anti-fuzzification layer is used for converting a fuzzy rule processing result into a multi, the output layer is used for outputting the fire probability value and carrying out related fire early warning.
Further, in S2, the temperature, the smoke concentration, and the carbon monoxide concentration are determined by membership functions of the linear variable layer and the fuzzy variable layer, respectively, each of the parameters includes four kinds of gradient information, and the membership functions are expressed as
W1And W2Indicating a connectionAnd (6) weighting.
Further, in S3, the method includes:
s31, determining fire types in a fuzzy rule layer according to gradient information of temperature, smoke concentration and carbon monoxide concentration, wherein the fire types comprise no fire, smoldering fire and open fire;
s32, the normalization layer respectively normalizes the temperature, the smoke concentration and the carbon monoxide concentration,
wherein XiRepresenting current input data, XminIs the minimum value, X, in the entire set of datamaxIs the maximum value in the whole group of data, and X' is the normalized value;
s33, determining the temperature, the carbon monoxide concentration and the change rate of the smoke concentration at a certain moment;
s34, carbon monoxide concentration from the time, carbon monoxide concentration at the previous time, and threshold value M for carbon monoxide concentrationCODetermining the continuous state R of carbon monoxide changesCOIs expressed as RCO＝(ΔCOi＞MCO)·(ΔCOi-1＞MCO) If the condition is satisfied, (. cndot.) 1 means that the carbon monoxide concentration is obviously increased, otherwise, 0;
s35, smoke density according to the time, smoke density at the previous moment and smoke density threshold MSDetermining the state of persistence R of the change in smoke concentrationSIs expressed as RS＝(ΔSi＞MS)·(ΔSi-1＞MS) If the condition is satisfied, (. cndot.) is 1, that is, the carbon monoxide concentration is obviously increased, otherwise, the carbon monoxide concentration is 0;
s36, the change rate of temperature, the change rate of carbon monoxide concentration, the change rate of smoke concentration, and the continuous state R of carbon monoxide change at that timeCOAnd smoke concentration sustained state RSDetermining a fire decision value q at that momenti；
S37, according to the fire decision value qiAnd obtaining the fire probability value Q and carrying out corresponding fire early warning.
Further, in S36, the method includes:
s361, judging whether the temperature change rate at the moment is more than M1Wherein M is10.4; if so, qi＝q1Go to S37, 3.3; if not, the process goes to S362;
s362, judging whether the temperature change rate at the moment is larger than M2Wherein M is20.3; if not, entering S363; if yes, judging whether the sum of the temperature change rate at the previous moment and the temperature change rate at the moment is more than M3Wherein M is3If it is 0.2, q isi＝q21.3, go to S37; if not, qi＝q3Go to S37, 0.8;
s363, determining the temperature change rate at the previous moment and the temperature change rates at the previous two moments, and judging whether the sum of the temperature change rate at the previous moment, the temperature change rate at the previous two moments and the temperature change rate at the moment is more than M3If so, qi＝q3Go to S37, 0.8; if not, the process goes to S364;
s364, judging the continuous state R of the carbon monoxide changeCOIf (·) is 1, then q isi＝q4Go to S37, 0.4; if (·) ═ 0, proceed to S365;
s364, judging the continuous state R of the smoke concentrationSIf (·) is 1, then q isi＝q4Go to S37, 0.4; if (·) is 0, it indicates that the current carbon monoxide and smoke concentration is not in fire.
Further, in S37, the fire probability value Q: q ═ λ1p+λ2qj，
Wherein λ1And λ2Respectively is the weight of the multi-source parameter decision probability and the fire decision value, and p is the multi-source parameter decision probability; judging whether Q is more than f.S, wherein f is 2.5, and S is a steady-state flameless probability mean value; if the fire is established, the fire is indicated, and if the fire is not established, the possibility of the fire is low.
Further, in S37, λ1＝0.8，λ2＝0.2。
The invention has the beneficial effects that: acquiring multi-source parameter decision probability of monitoring data by establishing a fire risk prediction model with temperature, CO concentration and smoke concentration as fire characteristic parameters; the method comprises the steps of establishing a gradient model of temperature characteristic parameters, carrying out auxiliary judgment by using a continuous state function of CO and smoke concentration change rate, comprehensively monitoring the change condition of the pipe gallery environment to obtain a fire decision value of parameter gradient, weighting and fusing multi-source parameter decision probability and the fire decision value to obtain a final fire probability value, judging the current fire in a proposed decision mode, and improving the reliability and the real-time performance of fire judgment without using the output value of a single model as a judgment basis of fire.
Drawings
FIG. 1 is a model schematic diagram of a comprehensive pipe gallery fire early warning method fusing multi-source parameters and gradient information thereof;
FIG. 2 is a fuzzy rule curve diagram of the utility tunnel fire early warning method of the present invention which integrates multi-source parameters and gradient information thereof;
FIG. 3 is a flow chart of determining a fire decision value in the utility tunnel fire early warning method of the present invention incorporating multi-source parameters and gradient information thereof;
FIG. 4 is a graph of temperature change with time of fire development in the utility tunnel fire early warning method of the present invention incorporating multi-source parameters and gradient information thereof;
FIG. 5 is a graph of the variation of carbon monoxide concentration with the time of fire development in the utility tunnel fire early warning method of the present invention incorporating multi-source parameters and gradient information thereof;
FIG. 6 is a graph of the change of smoke concentration with the time of fire development in the utility tunnel fire early warning method of the present invention incorporating multi-source parameters and gradient information thereof;
FIG. 7 is a fire prediction probability map of a first data set in the utility tunnel fire early warning method of the present invention incorporating multi-source parameters and gradient information thereof;
FIG. 8 is a graph of the fire prediction probability of the second data set in the utility tunnel fire early warning method of the present invention incorporating multi-source parameters and gradient information thereof;
FIG. 9 is a fire prediction probability map of a third data set in the utility tunnel fire early warning method of the present invention incorporating multi-source parameters and gradient information thereof;
fig. 10 shows the fire judgment accuracy under different weights in the utility tunnel fire early warning method with the fusion of multi-source parameters and gradient information thereof.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. 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: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures.
In the description of the present invention, it is to be understood that the terms "upper", "lower", "inside", "outside", "left", "right", and the like, indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, or the orientations or positional relationships that the products of the present invention are conventionally placed in use, or the orientations or positional relationships that are conventionally understood by those skilled in the art, and are used for convenience of describing the present invention and simplifying the description, but do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention.
Furthermore, the terms "first," "second," and the like are used merely to distinguish one description from another, and are not to be construed as indicating or implying relative importance.
In the description of the present invention, it is also to be noted that, unless otherwise explicitly stated or limited, the terms "disposed" and "connected" are to be interpreted broadly, and for example, "connected" may be a fixed connection, a detachable connection, or an integral connection; can be mechanically or electrically connected; the connection may be direct or indirect via an intermediate medium, and may be a communication between the 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.
The following detailed description of embodiments of the invention refers to the accompanying drawings.
As shown in fig. 1, S1, establishing a fire risk prediction model; the fire risk prediction model comprises an input layer, a linear change layer, a fuzzification layer, a fuzzy rule layer, a normalization layer, an anti-fuzzification layer and an output layer, the fire risk prediction model comprises the input layer, the linear change layer, the fuzzification layer, the fuzzy rule layer, the normalization layer, the anti-fuzzification layer and the output layer, the input layer is used for leading in collected temperature, smoke concentration and carbon monoxide concentration in the comprehensive pipe gallery, the linear change layer is used for carrying out data processing on the input data through the temperature and the smoke concentration, the fuzzification layer is used for determining gradient information of the temperature, the smoke concentration and the carbon monoxide concentration, the fuzzy rule layer is used for determining the fire type, the normalization layer is used for respectively carrying out normalization processing on the temperature, the smoke concentration and the carbon monoxide concentration after the fire type is determined, the anti-fuzzification layer is used for converting a fuzzy rule processing result into a, the output layer is used for outputting the fire probability value and carrying out related fire early warning.
S2, collecting the temperature, the smoke concentration and the carbon monoxide concentration in the comprehensive pipe gallery, introducing the temperature, the smoke concentration and the carbon monoxide concentration into a fire risk prediction model, and respectively determining gradient information of the temperature, the smoke concentration and the carbon monoxide concentration; temperature, smoke concentration anddetermining gradient information L1-L4 according to the concentration of the carbon monoxide by membership functions of the linear variable layer and the fuzzy variable layer respectively; the membership function of the parameters from low to high, namely L1, L2, L3, L4, L1, L2, L3 and L4, is a Gaussian function, the output range of the function is 0-1, and the membership function is expressed as
W1And W2Representing the connection weight.
S3, determining the fire type and obtaining the fire probability through the gradient information and the fire risk prediction model, and carrying out fire early warning;
s31, as shown in figure 2, determining the fire type in a fuzzy rule layer according to the gradient information of temperature, smoke concentration and carbon monoxide concentration, wherein the fire type comprises no fire, smoldering fire and open fire; the fuzzy rules of the fuzzy rule layer are as follows:
wherein m1, m2 and m3 respectively represent fire type no fire, smoldering fire and open fire.
S32, the normalization layer performs normalization processing on the gradient information output by the fuzzy layer, and the normalization formula is as follows:
wherein: xiRepresenting current input data, XminIs the minimum value, X, in the entire set of datamaxIs the maximum value in the whole group of data, and X' is the normalized value;
s33, determining the temperature, the carbon monoxide concentration and the change rate of the smoke concentration at a certain moment; temperature gradient:
carbon monoxide gradient:
smoke concentration gradient:
where Δ T represents the rate of change in temperature at time i, Δ CO represents the rate of change in CO concentration at time i, Δ S represents the rate of change in smoke concentration at time i, and T represents the rate of change in smoke concentration at time iiTemperature value at time i, COiCO concentration value at time i, SiIs the smoke concentration value at the moment i; the change speed of the parameter gradient reaction parameters can capture fire information through the change speed of each parameter;
s34, carbon monoxide concentration from the time, carbon monoxide concentration at the previous time, and threshold value M for carbon monoxide concentrationCODetermining the continuous state R of carbon monoxide changesCOIs expressed as RCO＝(ΔCOi＞MCO)·(ΔCOi-1＞MCO) If the condition is satisfied, (. cndot.) 1 means that the carbon monoxide concentration is obviously increased, otherwise, 0;
s35, smoke density according to the time, smoke density at the previous moment and smoke density threshold MSDetermining the state of persistence R of the change in smoke concentrationSIs expressed as RS＝(ΔSi＞MS)·(ΔSi-1＞MS) If the condition is satisfied, (. cndot.) is 1, that is, the carbon monoxide concentration is obviously increased, otherwise, the carbon monoxide concentration is 0;
s36, as shown in FIG. 3, according to the change rate of the temperature, the change rate of the carbon monoxide concentration, the change rate of the smoke concentration and the continuous state R of the carbon monoxide change at the momentCOAnd smoke concentration sustained state RSDetermining a fire decision value q at that momentiThe threshold value of the temperature change rate Delta T is MiM is respectively from large to small1、M2、M3The fire disaster data curve in the pipe gallery is simulated by referring to fire disaster simulation software FDS, and the specific values are M respectively according to the change condition of characteristic parameter concentration along with time before and after the fire disaster happens1＝0.4、M2＝0.3、M3＝0.2、MCO＝0.15、MS＝0.15；qiHas a value of q1、q2、q3、q4，q1＝3.3、q2＝1.3、q3＝0.8、q4＝0.4；
S361, judging whether the temperature change rate at the moment is more than M1Wherein M is10.4; if so, qi＝q1Go to S37, 3.3; if not, the process goes to S362;
s362, judging whether the temperature change rate at the moment is larger than M2Wherein M is20.3; if not, entering S363; if yes, judging whether the sum of the temperature change rate at the previous moment and the temperature change rate at the moment is more than M3Wherein M is3If it is 0.2, q isi＝q21.3, go to S37; if not, qi＝q3Go to S37, 0.8;
s363, determining the temperature change rate at the previous moment and the temperature change rates at the previous two moments, and judging whether the sum of the temperature change rate at the previous moment, the temperature change rate at the previous two moments and the temperature change rate at the moment is more than M3If so, qi＝q3Go to S37, 0.8; if not, the process goes to S364;
s364, judging the continuous state R of the carbon monoxide changeCOIf (·) is 1, then q isi＝q4Go to S37, 0.4; if (·) ═ 0, proceed to S365;
s364, judging the smoke concentration continuous state RSIf (·) is 1, then q isi＝q4Go to S37, 0.4; if (·) is 0, it indicates that the current concentration of carbon monoxide and smoke does not support the conclusion of fire.
S37, according to the fire decision value qiObtaining a fire probability value Q, and carrying out corresponding fire early warning; fire probability value Q: q ═ λ1p+λ2qj，
Wherein λ1And λ2Respectively is the weight of the multi-source parameter decision probability and the fire decision value, and p is the multi-source parameter decision probability; judging whether Q is more than f.S, wherein f is 2.5, and S is stableIf the mean value of the probability of fire is satisfied, the occurrence of fire is indicated, if the mean value of the probability of fire is not satisfied, the probability of fire is low, and lambda is1＝0.8，λ2＝0.2。
Since the temperature keeps good characteristics throughout the development of the fire, and the fire situation can be reflected most, the temperature gradient DeltaT is considered preferentiallyi. The parameter decision is combined with the historical value of the temperature gradient, the change condition of the parameter along with the time is monitored, the fire probability is predicted in a change rate mode, and meanwhile, the carbon monoxide concentration and the smoke concentration are obviously changed in the initial stage of the fire, so that the carbon monoxide concentration and the smoke concentration are in a continuous state RCO、RSThe system is assisted in judgment, the accuracy of the system in the initial fire judgment is guaranteed, the gradient of each parameter is calculated by collecting time sequence data monitored by a temperature sensor, the change condition of the pipe gallery environment is comprehensively monitored, and if the temperature change rate is too large and reaches a threshold value under the open fire condition, the finally output fire probability is improved through the parameter decision probability; if the temperature change rate is smaller and lower than the open fire condition of the fire, the judgment can be carried out by increasing the time span, and the trend of smoldering fire in the environment can be found in time.
As shown in fig. 4, 5, and 6, by analyzing the temperature variation curve in the fire data, the variation of the medium-term temperature before and after the occurrence of multiple fires is combined, and meanwhile, the probability of fire is high when the temperature variation is severe in the process of decision according to the parameter gradient; when the temperature changes slowly and the smoke and carbon monoxide concentration continuously increases, the characteristic of smoldering fire is obvious, and therefore, the empirical decision value q is obtained in conclusion1＝3.3、q2＝1.3、q3＝0.8、q4＝0.4。
By way of example of some of the experimental data, the experimental data are shown in the following table:
experimental data sheet
| Serial number | | S | T | |
| 1 | 2.94 | 0.06 | 45 | |
| 2 | 2..93 | 0.059 | 55 | |
| 3 | 3.71 | 0.148 | 196 |
Calculating the gradient of each parameter to obtain Delta T12.56, Δ CO 0.27, Δ S1.50, and the gradient decision q can be output according to the process1＝3.3。
Fusion with fuzzy neural networks:
fire probability value Q: q ═ λ1p+λ2qj，
Wherein: lambda [ alpha ]1And λ2Respectively taking the weights of the multi-source parameter decision probability and the parameter gradient probability; q is the final output fire probability value.
The decision-making is as follows: q > fs, wherein f is 2.5; and S is a steady state flameless probability mean value which is obtained through 300 groups of training sets of the fuzzy neural network. As shown in the following formula:
wherein n is the number of the fire-free groups in the 300 training sets; x is the number ofwThe probability of fire in the w-th fireless group.
λiValue taking experiment
As shown in fig. 10, to determine the weight λ1And λ2The value of (A) is selected in the invention1For the subjects, lambda was added1Fire judgment experiments were conducted with the settings of 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9 and 1, and it can be seen in fig. 10 that the current weight value λ is1Smaller instant weight value lambda2When the size is larger, the TSM mistakenly reads the large environment of the pipe gallery, so that the false alarm rate is higher; when the weight value is lambda1Over-large weight value lambda2When the weight is too small, the TSM plays a role in the early warning process and is weakened, so that the fire early warning capability is reduced, and particularly, when the weight lambda is used1When the value of (a) is 0.8, the highest fire judgment accuracy is obtained, which is 98.67%. In summary, the weight λ is selected1＝0.8，λ2＝0.2。
In contrast to conventional fuzzy neural networks
The experimental results of 60 groups of test data are selected for displaying, as shown in fig. 7 and 9, in the flameless stage, the predicted value of the fire probability is basically consistent with the predicted value of the traditional fuzzy neural network, namely, the good prediction performance of the traditional fuzzy neural network on the flameless characteristic parameters is maintained; the fire probability prediction value of the same time sequence data is obviously higher than that of the fuzzy neural network when no fire enters a fire stage, the parameter gradient can monitor the large environment of fire characteristic parameters, and the prediction value of the fire probability is reasonably adjusted by monitoring the change rate of the temperature, so that the algorithm is more sensitive to the perception of the stage from no fire to fire, and the real-time performance of fire judgment can be realized.
As can be seen from fig. 8, false alarm caused by the fuzzy neural network can be corrected by fusing the parameter gradient, and the accuracy of fire early warning is improved. The fire probability is subjected to auxiliary judgment by introducing CO and a continuous state function of the smoke concentration change rate due to parameter gradient decision, so that the false perception of a fuzzy neural network on the data fire probability is overcome, and the false alarm in the flameless stage is better solved.
The technical solution of the present invention is not limited to the limitations of the above specific embodiments, and all technical modifications made according to the technical solution of the present invention fall within the protection scope of the present invention.
Claims (7)
1. The utility model discloses a comprehensive pipe gallery fire early warning method of fusion multisource parameter and gradient information thereof which characterized in that includes the following steps:
s1, building a fire risk prediction model;
s2, collecting the temperature, the smoke concentration and the carbon monoxide concentration in the comprehensive pipe gallery, introducing the temperature, the smoke concentration and the carbon monoxide concentration into a fire risk prediction model, and respectively determining gradient information of the temperature, the smoke concentration and the carbon monoxide concentration;
and S3, determining the fire type through the gradient information and the fire risk prediction model, obtaining the fire probability value, and performing fire early warning.
2. The utility tunnel fire early warning method fusing multi-source parameters and gradient information thereof according to claim 1, wherein the fire risk prediction model comprises an input layer, a linear transformation layer, a fuzzy rule layer, a normalization layer, an anti-fuzzy layer and an output layer, the fire risk prediction model comprises an input layer, a linear transformation layer, a fuzzy rule layer, a normalization layer, an anti-fuzzy layer and an output layer, the input layer is used for importing the collected temperature, smoke concentration and carbon monoxide concentration in the utility tunnel, the linear transformation layer is used for performing data processing of the temperature and smoke concentration on the input data, the fuzzy layer is used for determining the gradient information of the temperature, smoke concentration and carbon monoxide concentration, the fuzzy rule layer is used for determining the fire type, the normalization layer is used for respectively performing normalization processing on the temperature, smoke concentration and carbon monoxide concentration after determining the fire type, the anti-fuzzy layer is used for converting fuzzy rule processing results into multi-source parameter decision probability p, and the output layer is used for outputting fire probability values and carrying out related fire early warning.
3. The method for pre-warning the fire of the comprehensive pipe rack fusing the multi-source parameters and the gradient information thereof according to claim 2, wherein in S2, the temperature, the smoke concentration and the carbon monoxide concentration respectively determine the gradient information through membership functions of a linear variable layer and a fuzzy variable layer, each parameter comprises four kinds of gradient information, and the membership functions are expressed as
W1And W2Representing the connection weight.
4. The utility tunnel fire early warning method fusing multi-source parameters and gradient information thereof according to claim 3, wherein in S3 comprises:
s31, determining fire types in a fuzzy rule layer according to gradient information of temperature, smoke concentration and carbon monoxide concentration, wherein the fire types comprise no fire, smoldering fire and open fire;
s32, the normalization layer respectively normalizes the temperature, the smoke concentration and the carbon monoxide concentration,
wherein XiRepresenting current input data, XminIs the minimum value, X, in the entire set of datamaxIs the maximum value in the whole group of data, and X' is the normalized value;
s33, determining the temperature, the carbon monoxide concentration and the change rate of the smoke concentration at a certain moment;
s34, carbon monoxide concentration from the time, carbon monoxide concentration at the previous time, and threshold value M for carbon monoxide concentrationCODetermining the continuous state R of carbon monoxide changesCOIs expressed as RCO＝(ΔCOi＞MCO)·(ΔCOi-1＞MCO) If the condition is satisfied, (. cndot.) 1 means that the carbon monoxide concentration is obviously increased, otherwise, 0;
s35, smoke density according to the time, smoke density at the previous moment and smoke density threshold valueMSDetermining the state of persistence R of the change in smoke concentrationSIs expressed as RS＝(ΔSi＞MS)·(ΔSi-1＞MS) If the condition is satisfied, (. cndot.) is 1, that is, the carbon monoxide concentration is obviously increased, otherwise, the carbon monoxide concentration is 0;
s36, the change rate of temperature, the change rate of carbon monoxide concentration, the change rate of smoke concentration, and the continuous state R of carbon monoxide change at that timeCOAnd smoke concentration sustained state RSDetermining a fire decision value q at that momenti；
S37, according to the fire decision value qiAnd obtaining the fire probability value Q and carrying out corresponding fire early warning.
5. The utility tunnel fire early warning method fusing multi-source parameters and gradient information thereof according to claim 4, wherein in S36 comprises:
s361, judging whether the temperature change rate at the moment is more than M1Wherein M is10.4; if so, qi＝q1Go to S37, 3.3; if not, the process goes to S362;
s362, judging whether the temperature change rate at the moment is larger than M2Wherein M is20.3; if not, entering S363; if yes, judging whether the sum of the temperature change rate at the previous moment and the temperature change rate at the moment is more than M3Wherein M is3If it is 0.2, q isi＝q21.3, go to S37; if not, qi＝q3Go to S37, 0.8;
s363, determining the temperature change rate at the previous moment and the temperature change rates at the previous two moments, and judging whether the sum of the temperature change rate at the previous moment, the temperature change rate at the previous two moments and the temperature change rate at the moment is more than M3If so, qi＝q3Go to S37, 0.8; if not, the process goes to S364;
s364, judging the continuous state R of the carbon monoxide changeCOIf (·) is 1, then q isi＝q4Go to S37, 0.4; if (·) is 0, thenEntering S365;
s364, judging the continuous state R of the smoke concentrationSIf (·) is 1, then q isi＝q4Go to S37, 0.4; if (·) is 0, it indicates that the current carbon monoxide and smoke concentration is not in fire.
6. The utility tunnel fire early warning method fusing multi-source parameters and gradient information thereof according to claim 5, wherein in S37, the fire probability value Q: q ═ λ1p+λ2qj，
Wherein λ1And λ2Respectively is the weight of the multi-source parameter decision probability and the fire decision value, and p is the multi-source parameter decision probability; judging whether Q is more than f.S, wherein f is 2.5, and S is a steady-state flameless probability mean value; if the fire is established, the fire is indicated, and if the fire is not established, the possibility of the fire is low.
7. The comprehensive pipe gallery fire early warning method fusing multi-source parameters and gradient information thereof according to claim 6, wherein in S37, λ1＝0.8，λ2＝0.2。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010597058.4A CN111627181B (en) | 2020-06-28 | 2020-06-28 | Comprehensive pipe rack fire early warning method fusing multi-source parameters and gradient information thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010597058.4A CN111627181B (en) | 2020-06-28 | 2020-06-28 | Comprehensive pipe rack fire early warning method fusing multi-source parameters and gradient information thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111627181A true CN111627181A (en) | 2020-09-04 |
| CN111627181B CN111627181B (en) | 2021-11-23 |
Family
ID=72272406
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010597058.4A Active CN111627181B (en) | 2020-06-28 | 2020-06-28 | Comprehensive pipe rack fire early warning method fusing multi-source parameters and gradient information thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111627181B (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112418281A (en) * | 2020-11-11 | 2021-02-26 | 国网福建省电力有限公司电力科学研究院 | Fire detection sensor data anomaly detection method and system |
| CN112419650A (en) * | 2020-11-11 | 2021-02-26 | 国网福建省电力有限公司电力科学研究院 | Fire detection method and system based on neural network and image recognition technology |
| CN112946368A (en) * | 2021-02-02 | 2021-06-11 | 长春工程学院 | Tower grounding resistance synchronous detection method based on multi-source parameter fusion analysis |
| CN113255717A (en) * | 2021-03-25 | 2021-08-13 | 中冶赛迪重庆信息技术有限公司 | Piping lane fire detection method and system |
| CN113379781A (en) * | 2021-07-29 | 2021-09-10 | 浙江大华技术股份有限公司 | Image processing method and apparatus, storage medium, and electronic device |
| CN113470331A (en) * | 2021-07-08 | 2021-10-01 | 哲弗智能系统（上海）有限公司 | Temperature difference detection method and device for temperature-sensitive detector, temperature-sensitive detector and medium |
| CN113505925A (en) * | 2021-07-09 | 2021-10-15 | 重庆邮电大学 | ANFIS-based laboratory dangerous chemical abnormal information early warning method |
| CN113516837A (en) * | 2021-07-21 | 2021-10-19 | 重庆大学 | Urban fire judgment method and system based on multi-source information fusion and storage medium thereof |
| CN114049738A (en) * | 2021-10-19 | 2022-02-15 | 国网湖北省电力有限公司电力科学研究院 | Building electrical fire identification method and system based on smoke, temperature and electrical quantity |
| 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 |
| CN114387755A (en) * | 2021-12-13 | 2022-04-22 | 煤炭科学技术研究院有限公司 | Mine smoke detection method, device, processor and system |
| CN114582083A (en) * | 2022-01-14 | 2022-06-03 | 西安理工大学 | Tunnel monitoring multi-sensor data fusion fire early warning method |
| CN114743335A (en) * | 2021-11-29 | 2022-07-12 | 国家电网有限公司 | Cable channel fire early warning method based on fuzzy neural network |
| CN115171318A (en) * | 2022-05-13 | 2022-10-11 | 国网浙江省电力有限公司金华供电公司 | Fire detection method for lithium battery in mobile energy storage bin of energy storage power station |
| CN115206047A (en) * | 2022-05-24 | 2022-10-18 | 华能宁夏中卫光伏发电有限公司 | Photovoltaic module area fire prevention alarm method and system |
| CN116311751A (en) * | 2023-05-16 | 2023-06-23 | 陕西开来机电设备制造有限公司 | Underground coal mine use fire prevention and extinguishment electric control system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104076233A (en) * | 2014-07-18 | 2014-10-01 | 彭浩明 | Ageing degree detecting method and device |
| CN104933841A (en) * | 2015-04-30 | 2015-09-23 | 重庆三峡学院 | Fire prediction method based on self-organizing neural network |
| CN105185022A (en) * | 2015-10-21 | 2015-12-23 | 国家电网公司 | Transformer substation fire detection system based on multi-sensor information combination and detection information combination method |
| CN109255921A (en) * | 2018-11-13 | 2019-01-22 | 福州大学 | A kind of multisensor fire detection method based on hierarchical fuzzy fusion |
| CN109272037A (en) * | 2018-09-17 | 2019-01-25 | 江南大学 | A kind of self-organizing TS pattern paste network modeling method applied to infra red flame identification |
| US20190371147A1 (en) * | 2018-05-31 | 2019-12-05 | Boe Technology Group Co., Ltd. | Fire alarming method and device |
-
2020
- 2020-06-28 CN CN202010597058.4A patent/CN111627181B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104076233A (en) * | 2014-07-18 | 2014-10-01 | 彭浩明 | Ageing degree detecting method and device |
| CN104933841A (en) * | 2015-04-30 | 2015-09-23 | 重庆三峡学院 | Fire prediction method based on self-organizing neural network |
| CN105185022A (en) * | 2015-10-21 | 2015-12-23 | 国家电网公司 | Transformer substation fire detection system based on multi-sensor information combination and detection information combination method |
| US20190371147A1 (en) * | 2018-05-31 | 2019-12-05 | Boe Technology Group Co., Ltd. | Fire alarming method and device |
| CN109272037A (en) * | 2018-09-17 | 2019-01-25 | 江南大学 | A kind of self-organizing TS pattern paste network modeling method applied to infra red flame identification |
| CN109255921A (en) * | 2018-11-13 | 2019-01-22 | 福州大学 | A kind of multisensor fire detection method based on hierarchical fuzzy fusion |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112418281A (en) * | 2020-11-11 | 2021-02-26 | 国网福建省电力有限公司电力科学研究院 | Fire detection sensor data anomaly detection method and system |
| CN112419650A (en) * | 2020-11-11 | 2021-02-26 | 国网福建省电力有限公司电力科学研究院 | Fire detection method and system based on neural network and image recognition technology |
| CN112946368A (en) * | 2021-02-02 | 2021-06-11 | 长春工程学院 | Tower grounding resistance synchronous detection method based on multi-source parameter fusion analysis |
| CN113255717A (en) * | 2021-03-25 | 2021-08-13 | 中冶赛迪重庆信息技术有限公司 | Piping lane fire detection method and system |
| CN113470331B (en) * | 2021-07-08 | 2023-05-02 | 哲弗智能系统（上海）有限公司 | Differential temperature detection method and device of temperature sensing detector, temperature sensing detector and medium |
| CN113470331A (en) * | 2021-07-08 | 2021-10-01 | 哲弗智能系统（上海）有限公司 | Temperature difference detection method and device for temperature-sensitive detector, temperature-sensitive detector and medium |
| CN113505925A (en) * | 2021-07-09 | 2021-10-15 | 重庆邮电大学 | ANFIS-based laboratory dangerous chemical abnormal information early warning method |
| CN113505925B (en) * | 2021-07-09 | 2022-07-15 | 重庆邮电大学 | Laboratory hazardous chemical anomaly information early warning method based on ANFIS |
| CN113516837A (en) * | 2021-07-21 | 2021-10-19 | 重庆大学 | Urban fire judgment method and system based on multi-source information fusion and storage medium thereof |
| CN113379781A (en) * | 2021-07-29 | 2021-09-10 | 浙江大华技术股份有限公司 | Image processing method and apparatus, storage medium, and electronic device |
| CN113379781B (en) * | 2021-07-29 | 2023-04-07 | 浙江大华技术股份有限公司 | Image-based fire monitoring method and device, storage medium and electronic equipment |
| CN114049738B (en) * | 2021-10-19 | 2023-02-24 | 国网湖北省电力有限公司电力科学研究院 | Building electrical fire identification method and system based on smoke, temperature and electrical quantity |
| CN114049738A (en) * | 2021-10-19 | 2022-02-15 | 国网湖北省电力有限公司电力科学研究院 | Building electrical fire identification method and system based on smoke, temperature and electrical quantity |
| 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 |
| CN114743335A (en) * | 2021-11-29 | 2022-07-12 | 国家电网有限公司 | Cable channel fire early warning method based on fuzzy neural network |
| CN114387755A (en) * | 2021-12-13 | 2022-04-22 | 煤炭科学技术研究院有限公司 | Mine smoke detection method, device, processor and system |
| CN114582083A (en) * | 2022-01-14 | 2022-06-03 | 西安理工大学 | Tunnel monitoring multi-sensor data fusion fire early warning method |
| CN114582083B (en) * | 2022-01-14 | 2023-06-30 | 西安理工大学 | Tunnel monitoring multi-sensor data fusion fire disaster early warning method |
| CN115171318A (en) * | 2022-05-13 | 2022-10-11 | 国网浙江省电力有限公司金华供电公司 | Fire detection method for lithium battery in mobile energy storage bin of energy storage power station |
| CN115171318B (en) * | 2022-05-13 | 2024-03-12 | 国网浙江省电力有限公司金华供电公司 | Fire detection method for lithium battery in movable energy storage bin of energy storage power station |
| CN115206047A (en) * | 2022-05-24 | 2022-10-18 | 华能宁夏中卫光伏发电有限公司 | Photovoltaic module area fire prevention alarm method and system |
| CN115206047B (en) * | 2022-05-24 | 2024-04-30 | 华能宁夏中卫光伏发电有限公司 | Fireproof alarm method and system for photovoltaic module area |
| CN116311751B (en) * | 2023-05-16 | 2023-12-08 | 陕西开来机电设备制造有限公司 | Underground coal mine use fire prevention and extinguishment electric control system |
| CN116311751A (en) * | 2023-05-16 | 2023-06-23 | 陕西开来机电设备制造有限公司 | Underground coal mine use fire prevention and extinguishment electric control system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111627181B (en) | 2021-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111627181B (en) | Comprehensive pipe rack fire early warning method fusing multi-source parameters and gradient information thereof | |
| CN104933841B (en) | A kind of fire prediction method based on self organizing neural network | |
| Cheng et al. | One fire detection method using neural networks | |
| JP2019079446A (en) | Fire monitoring system | |
| CN115169479A (en) | Remote monitoring method, system and storage medium for sewage treatment process | |
| CN102013148B (en) | Multi-information fusion fire hazard detection method | |
| CN108648403A (en) | A kind of self study security against fire method for early warning and system | |
| CN108897354B (en) | Aluminum smelting process hearth temperature prediction method based on deep belief network | |
| CN105975991B (en) | A kind of improved extreme learning machine fire kind recognition methods | |
| CN114399719B (en) | Transformer substation fire video monitoring method | |
| CN113223264A (en) | QPSO-BP neural network-based intelligent fire early warning system and method | |
| CN112419650A (en) | Fire detection method and system based on neural network and image recognition technology | |
| CN115358647A (en) | Hydrogen energy industry chain risk monitoring system and monitoring method based on big data | |
| CN113990017B (en) | Forest and grassland fire early warning system and method based on PNN neural network | |
| CN114255562A (en) | Wisdom fire control early warning system based on thing networking | |
| CN112418281A (en) | Fire detection sensor data anomaly detection method and system | |
| Wang et al. | Forest fire detection system based on Fuzzy Kalman filter | |
| CN111508182A (en) | Novel monitoring alarm system for community fire fighting through Internet of things | |
| CN116311739A (en) | Multi-sensor fire detection method based on long-short-term memory network and environment information fusion | |
| CN111242278A (en) | Composite smoke sense low-false-alarm method based on intelligent algorithm | |
| CN115962428A (en) | Real-time online intelligent interpretability monitoring and tracing method for gas pipe network leakage | |
| CN115951241A (en) | Vehicle-mounted power lithium battery thermal runaway early warning protection method and system | |
| CN115905998A (en) | Cable tunnel fire extinguishing decision method combining multi-sensor data fusion with FWA-BP neural network algorithm | |
| CN113822771A (en) | Low false detection rate electricity stealing detection method based on deep learning | |
| CN115620474A (en) | Energy storage prefabricated cabin fire alarm optimization method and system based on AHP |
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 |