CN111192428A - Fire-fighting alarm method and system based on big data hidden variable model - Google Patents
Fire-fighting alarm method and system based on big data hidden variable model Download PDFInfo
- Publication number
- CN111192428A CN111192428A CN202010022344.8A CN202010022344A CN111192428A CN 111192428 A CN111192428 A CN 111192428A CN 202010022344 A CN202010022344 A CN 202010022344A CN 111192428 A CN111192428 A CN 111192428A
- Authority
- CN
- China
- Prior art keywords
- fire
- combustible gas
- concentration
- scene
- variable
- 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
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/06—Electric actuation of the alarm, e.g. using a thermally-operated switch
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/17—Function evaluation by approximation methods, e.g. inter- or extrapolation, smoothing, least mean square method
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
- 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
- G08B17/117—Actuation by presence of smoke or gases, e.g. automatic alarm devices for analysing flowing fluid materials by the use of optical means by using a detection device for specific gases, e.g. combustion products, produced by the fire
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Emergency Management (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Databases & Information Systems (AREA)
- Algebra (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Fire Alarms (AREA)
Abstract
The invention provides a fire-fighting alarm method and system based on a big data hidden variable model. The fire-fighting alarm method comprises the steps of receiving temperature, smoke concentration and combustible gas concentration in real time under each fire-fighting scene, and utilizing a fire occurrence event model F: a is F ═ a1f(T)+a2g(S)+a3h (X) outputting the fire occurrence probability under the corresponding fire fighting scene; when the fire occurrence probability under the fire fighting scene is larger than a first class error threshold value of the statistical hypothesis test, performing fire alarm on the corresponding fire fighting scene; the fire occurrence probability of all fire scenes is obtainedWhen the average value of the probability of fire occurrence is larger than the second type error threshold value of the statistical hypothesis test, sending data abnormal alarm information and further judging the reason causing the data abnormal alarm information by human intervention; wherein f (T) is an implicit function of fire and temperature T univariates, g (S) is an implicit function of fire and smoke single concentration S univariates, and h (X) is an implicit function of fire and combustible gas concentration X univariates; a is1、a2And a3All are constant coefficients.
Description
Technical Field
The invention belongs to the field of fire alarm, and particularly relates to a fire alarm method and system based on a big data hidden variable model.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
In the field of fire fighting, the conventional alarm mechanism is based on the alarm of sensors, such as smoke alarm, temperature alarm, combustible gas sensor, etc., based on simple rules. For example, a boundary value of a certain alarm temperature is set in advance, and an alarm is given when the temperature judged by the temperature sensor is greater than the boundary value. This traditional method in fire fighting, although simple, is not very efficient.
The inventor finds that the traditional alarm method mainly has the following problems:
(1) the accuracy is low: because the alarm judgment is carried out according to the preset boundary value, misjudgment exists at a high probability, and the situation that the fire alarm is not formed but the fire alarm is possibly relatively more in the misjudgment. Generally speaking, if the problem is viewed from the perspective of hypothesis testing in statistics, the first type error rate (which should be alarmed, but not alarmed in time, causing loss) and the second type error rate (which should not be alarmed, but alarmed, causing waste of public resources) of the alarm mechanism based on the conventional method are relatively high and are not ideal enough.
(2) Judging according to relative segmentation: by relative segmentation of judgment basis is meant that the use of sensory data throughout the alarm mechanism is based on simple judgment logic rather than an organic combination. Usually, the logic of the whole alarm mechanism is composed of a series of connected OR gates. That is, after any sensor triggers an alarm, the whole alarm mechanism is started, and the alarm mechanism formed by the logic may cause the occurrence of the situation that the fire alarm is not formed, but the fire alarm is generated, so that public resources are wasted. And a part of alarm mechanism is formed by connecting a series of doors. That is, all sensors or a higher percentage of sensors are required to trigger an alarm before the entire alarm mechanism can be activated. Such logically configured alarm mechanisms may result in the situation that the fire alarm is configured to not be timely alarmed, causing losses.
(3) The boundary values are difficult to set: another great drawback of the conventional alarm mechanism is that the alarm limit values are difficult to set. If the alarm mechanism is logically connected by an OR gate, the boundary value cannot be set too low; if the alarm mechanism is logically connected with an AND gate, the boundary value cannot be set too high. However, for a specific scene and environment, the set boundary values are rarely referred to each other, and the usability of the boundary values sometimes needs to be obtained by a lot of expensive experiments, and the more experiments are needed to obtain boundary values suitable for the specific scene, the higher the cost is.
Disclosure of Invention
In order to solve the above problems, a first aspect of the present invention provides a fire alarm method based on a big data hidden variable model, which can improve fire alarm accuracy, increase the relevance and systematicness between judgments, and make boundary values well-documented.
In order to achieve the purpose, the invention adopts the following technical scheme:
a fire-fighting alarm method based on a big data hidden variable model comprises the following steps:
under each fire scene, receiving temperature, smoke concentration and combustible gas concentration in real time, and utilizing a fire occurrence event model F: a is F ═ a1f(T)+a2g(S)+a3h (X) outputting the fire occurrence probability under the corresponding fire fighting scene; when the fire occurrence probability under the fire fighting scene is larger than a first class error threshold value of the statistical hypothesis test, performing fire alarm on the corresponding fire fighting scene;
calculating the average value of the fire occurrence probability of all fire scenes, and when the average value of the fire occurrence probability is larger than a second type error threshold value of statistical hypothesis test, sending data abnormal alarm information and further judging the reason causing the data abnormal alarm information by human intervention;
wherein f (T) is an implicit function of fire and temperature T univariates, g (S) is an implicit function of fire and smoke single concentration S univariates, and h (X) is an implicit function of fire and combustible gas concentration X univariates; a is1、a2And a3All are constant coefficients, namely a temperature coefficient, a smoke concentration coefficient and a combustible gas concentration coefficient.
As one embodiment, the reason for causing the data abnormality warning information includes: fire occurrence model failures and sensor failures.
The technical scheme has the advantages that the second type of error threshold value is determined by the statistical hypothesis test, the threshold value is used for verifying the effectiveness of the model, when the average value of the fire occurrence probability is larger than the second type of error threshold value of the statistical hypothesis test, data abnormity alarm information is sent out and manual intervention is needed, the actual situation is inspected, and therefore whether the model fails or whether the sensor fails or not is inspected, input data abnormity is caused, and the accuracy of fire alarm is finally improved.
As one embodiment, the implicit function f (T) of the fire and temperature single variable, the implicit function g (S) of the fire and smoke single concentration single variable and the implicit function h (X) of the fire and combustible gas concentration single variable are constructed by the following steps:
and (3) calling correlation data of the fire, temperature, smoke concentration and combustible gas concentration respectively under each fire scene from a fire protection fire database, and screening a hidden function f (T) of a fire and temperature single variable, a hidden function g (S) of a fire and smoke single concentration variable and a hidden function h (X) of a fire and combustible gas concentration single variable from a hidden function set according to the function fitting degree and the function structure.
The technical scheme has the advantages that the used hidden variable model not only respectively considers the influence condition of each judgment basis, but also considers the influence condition of the whole variable on the fire, and the relevance and systematicness between the judgment bases are increased to a certain extent.
As an embodiment, the building process of the fire protection fire database is as follows:
and respectively acquiring the temperature change, the smoke concentration change and the combustible gas concentration under each experimental fire scene in real time until a fire disaster occurs, and obtaining the associated data between the fire disaster and the temperature, the smoke concentration and the combustible gas concentration under each fire scene so as to construct a fire-fighting fire disaster database.
The technical scheme has the advantages that the fire protection fire database is built, so that a data basis is provided for building the fire protection single quantity implicit function with the temperature, the smoke concentration and the combustible gas concentration and the temperature, the smoke concentration and the combustible gas concentration which are combined to form the implicit function, the efficiency and the accuracy of building the fire occurrence event model are guaranteed, and the accuracy of fire protection fire alarm is finally improved.
As one embodiment, the implicit functions in the set of implicit functions include tanh functions and logistic functions.
As an embodiment, the experimental fire scenario includes: the household indoor environment, the working environment and the machine room.
As an embodiment, the method comprises the steps of calling data related to the temperature, the smoke concentration and the combustible gas concentration of a fire under each fire fighting scene from a fire fighting fire database, and recording the temperature coefficient, the smoke concentration coefficient and the combustible gas concentration coefficient when the fire breaks out;
then carrying out maximum inspection on the fire with the temperature coefficient, the smoke concentration coefficient and the combustible gas concentration coefficient and the implicit functions of the temperature single variable, the smoke single concentration variable and the combustible gas concentration single variable respectively to obtain the temperature coefficient a1Smoke concentration coefficient a2And the coefficient of combustible gas concentration a3And further constructing a fire occurrence event model F: a is F ═ a1f(T)+a2g(S)+a3h(X)。
The technical scheme has the advantages that the hidden variable model is used, the maximum likelihood estimation method is utilized, the first type of errors and the second type of errors are effectively avoided, and the fire fighting fire alarm accuracy is improved.
In order to solve the above problems, a second aspect of the present invention provides a fire alarm system based on a big data hidden variable model, which can improve fire alarm accuracy, increase the relevance and systematicness between judgments, and make boundary values well-documented.
In order to achieve the purpose, the invention adopts the following technical scheme:
a fire alarm system based on big data hidden variable model comprises:
fire scene fire alarm module, it is used for under each fire scene, real-time receipt temperature, smog concentration and combustible gas concentration utilize conflagration emergence event model F: a is F ═ a1f(T)+a2g(S)+a3h (X) outputting the fire occurrence probability under the corresponding fire fighting scene; when the fire occurrence probability under the fire fighting scene is larger than a first class error threshold value of the statistical hypothesis test, performing fire alarm on the corresponding fire fighting scene;
the data abnormity alarm module is used for solving the average value of the fire occurrence probability of all fire scenes, and when the average value of the fire occurrence probability is larger than the second type error threshold value of the statistical hypothesis test, sending data abnormity alarm information and needing human intervention to further judge the reason causing the data abnormity alarm information;
wherein f (T) is an implicit function of fire and temperature T univariates, g (S) is an implicit function of fire and smoke single concentration S univariates, and h (X) is an implicit function of fire and combustible gas concentration X univariates; a is1、a2And a3All are constant coefficients, namely a temperature coefficient, a smoke concentration coefficient and a combustible gas concentration coefficient.
In order to solve the above-described problems, a third aspect of the present invention provides a computer-readable storage medium capable of improving fire alarm accuracy, increasing the relevance and systematicness between judgments, and making boundary values well documented.
In order to achieve the purpose, the invention adopts the following technical scheme:
a computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps in the big-data implicit variable model based fire alarm method as described above.
In order to solve the above-described problems, a fourth aspect of the present invention provides a computer device capable of improving fire alarm accuracy, increasing the relevance and systematicness between judgments, and making boundary values well-documented.
A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the steps of the fire alarm method based on big data implicit variable model as described above.
The invention has the beneficial effects that:
(1) the accuracy is low: because the hidden variable model is used and the maximum likelihood estimation method is utilized, the first kind of errors and the second kind of errors are effectively avoided
(2) Aiming at the judgment basis relative division: because the used hidden variable model not only respectively considers the influence condition of each judgment basis, but also considers the influence condition of the whole variable on the fire, the relevance and systematicness between the judgment bases are increased to a certain extent.
(3) Difficult to estimate and determine for boundary values: in the construction of the hidden variable model, parameters are scientifically estimated by using fitting of data and a statistical method, and likelihood values are also scientifically estimated, so that the obtained boundary values are well documented, and the problem that the boundary values are difficult to estimate is solved to a certain extent.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, are included to provide a further understanding of the invention, and are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and together with the description serve to explain the invention and not to limit the invention.
FIG. 1 is a flow chart of a fire alarm method based on a big data hidden variable model according to an embodiment of the invention;
FIG. 2 is an alarm schematic diagram of a fire-fighting alarm method based on a big data hidden variable model applied to an actual scene according to an embodiment of the invention;
FIG. 3 is a diagram illustrating the mapping of the function tanh (x);
FIG. 4 is a diagram of the mapping of the function locality (x);
FIG. 5 is a flow chart of the construction of the implicit function f (T) of fire and temperature single variables according to the embodiment of the present invention;
FIG. 6 is a flow chart of the construction of implicit function g (S) of fire and smoke single concentration variable according to the embodiment of the present invention;
FIG. 7 is a flow chart of the construction of a univariate implicit function h (X) of fire and combustible gas concentration according to an embodiment of the present invention;
FIG. 8 is a flow chart of a fire occurrence model F construction according to an embodiment of the present invention;
FIG. 9 is a schematic structural diagram of a fire alarm system based on a big data hidden variable model according to an embodiment of the invention.
Detailed Description
The invention is further described with reference to the following figures and examples.
It is to be understood that the following detailed description is exemplary and is intended to provide further explanation of the invention as claimed. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
(1) Hidden variable
The implicit variable is a function which is performed by using the information of the measurable variable and some characteristics of the variable of interest when the information of the measurable random variable cannot be directly measured but the information of interest cannot be directly measuredThereby solving the problem that information is not available in the above case. Mathematically, hidden variables are a mapping relationship. Variables of interest to the user are denoted by variable I, measurable variables are denoted by M, and f: (1).) Representing an implicit function. Then, I ═ f (m) can be obtained.
(2) Hidden variable model
The hidden variable model refers to a machine learning model or a statistical model constructed by using hidden variables.
For example, linear regression can be used as the simplest statistical model, with Y ═ a1X1+a2X2+…+a0For example, the hidden variable model of the model can be expressed as Y ═ a1f1(X1)+a2f2(X2)+…+a0E, if each implicit function fiThe above formula can be further simplified to satisfy the consistency property in functional science
f(Y)=a1X1+a2X2+…+a0+∈
Wherein the variable Y represents a variable of interest to the user, X1、X2…, denotes a measurable variable, f1(.)、f2(.) …, denotes an implicit function; e also represents a constant coefficient;
the implicit variable model is described above as an example of linear regression, and the artificial neural network model is essentially a linear regression implicit variable model. From this point of view, the hidden variable model can make what would otherwise appear to be a simple model very energetic.
(3) Application of hidden variable model in fire fighting field
The fire fighting field is a field which is very suitable for using an implicit variable model, because the scenes are considered, all scenes for fire fighting are suitable for introducing implicit variables, namely, the probability of fire occurrence cannot be directly measured, and only the implicit variables such as temperature, smoke density, combustible gas density and the like can be measured, and then the implicit variable model is constructed according to the implicit variables, so that the model is used for removing the phase change risk of 'measuring' the fire occurrence.
The random variable of the fire occurrence of interest is represented by variable F, the measurable variable temperature by variable T, the measurable variable smoke density by variable S, and the measurable variable combustible gas density by variable X. Assuming a linear regression model is used as the model for the hidden variables, the model for the fire occurrence event can be expressed as:
F=a1f(T)+a2g(S)+a3h(X)
wherein f (T) is an implicit function of fire and temperature single variables, g (S) is an implicit function of fire and smoke single concentration variables, and h (X) is an implicit function of fire and combustible gas concentration single variables; a is1、a2And a3All are constant coefficients, namely a temperature coefficient, a smoke concentration coefficient and a combustible gas concentration coefficient. a is1、a2And a3Is a parameter of linear regression, and can be estimated by maximum likelihood estimation.
Example one
Fig. 1 shows a flow chart of a fire alarm method based on a big data hidden variable model according to the embodiment.
Fig. 2 is an alarm schematic diagram of the fire-fighting alarm method based on the big data hidden variable model in the embodiment applied to an actual scene. In fig. 2, "scene … …" represents an omission.
As shown in fig. 1 and fig. 2, a fire alarm method based on a big data hidden variable model according to this embodiment includes:
step 1: under each fire scene, receiving temperature, smoke concentration and combustible gas concentration in real time, and utilizing a fire occurrence event model F: a is F ═ a1f(T)+a2g(S)+a3h (X) outputting the fire occurrence probability under the corresponding fire fighting scene; when the fire occurrence probability under the fire fighting scene is larger than a first class error threshold value of the statistical hypothesis test, performing fire alarm on the corresponding fire fighting scene; wherein f (T) is an implicit function of fire and temperature T univariates, g (S) is an implicit function of fire and smoke single concentration S univariates, and h (X) is an implicit function of fire and combustible gas concentration X univariates; a is1、a2And a3All are constant coefficients, namely a temperature coefficient, a smoke concentration coefficient and a combustible gas concentration coefficient.
As one embodiment, the implicit function f (T) of the fire and temperature single variable, the implicit function g (S) of the fire and smoke single concentration single variable and the implicit function h (X) of the fire and combustible gas concentration single variable are constructed by the following steps:
and (3) calling correlation data of the fire, temperature, smoke concentration and combustible gas concentration respectively under each fire scene from a fire protection fire database, and screening a hidden function f (T) of a fire and temperature single variable, a hidden function g (S) of a fire and smoke single concentration variable and a hidden function h (X) of a fire and combustible gas concentration single variable from a hidden function set according to the function fitting degree and the function structure.
Wherein, the construction process of the fire-fighting fire database is as follows:
and respectively acquiring the temperature change, the smoke concentration change and the combustible gas concentration under each experimental fire scene in real time until a fire disaster occurs, and obtaining the associated data between the fire disaster and the temperature, the smoke concentration and the combustible gas concentration under each fire scene so as to construct a fire-fighting fire disaster database.
In specific implementation, the temperature change, the smoke concentration change and the combustible gas concentration can be monitored in real time by using a temperature sensor, a smoke concentration sensor and a combustible gas concentration sensor respectively.
According to the fire-fighting fire database, a data basis is provided for building the fire, the single implicit function of the temperature, the smoke concentration and the combustible gas concentration and the joint implicit function of the temperature, the smoke concentration and the combustible gas concentration, the efficiency and the accuracy of building a fire occurrence event model are guaranteed, and the accuracy of fire-fighting fire alarm is finally improved.
Specifically, the implicit functions in the set of implicit functions include tanh functions and logistic functions.
the tan h function is defined as: for input data x, the function tanh (x) maps x to:
the mapping diagram for the function tanh (x) is shown in fig. 3.
tanh (x) is a common implicit function, and is widely applied in the field of neural network models, including the field of signal processing.
The logistic function is defined as: for input data x, the function locality (x) maps x to
For the mapping diagram of the function logic (x), as shown in FIG. 4
Here, logistic (x) is a common implicit function, and in the neural network model field under the non-deep learning framework, the function is the most used connection function.
In a specific implementation, the experimental fire scenario includes: the household indoor environment, the working environment and the machine room.
The indoor environment of the house refers to common indoor environments of the house, including bedrooms, living rooms, toilets and kitchens, and experiments can be carried out according to various different environments.
The working environment refers to various working environments including offices, factories, and the like, and experiments can be performed according to these scenes.
The computer room generally refers to an environment scene in which various devices operate, and includes a computer room, a server room and the like, and experiments can be performed according to the scenes.
The construction of the implicit function f (T) of the fire and temperature single variable, the implicit function g (S) of the fire and smoke single concentration variable and the implicit function h (X) of the fire and combustible gas concentration single variable are respectively shown in fig. 5, 6 and 7.
In fig. 5, 6 and 7, the range of implicit functions is one of the ranges of implicit functions selected, and a decision is needed according to the scene, the data, the degree of fitting of the functions, and the like. And judging whether the fitting of the data is good or not through data fitting, wherein the data fitting refers to substituting data obtained through experiments into a theoretical model, and the fitting of the data is judged to be good or not and can be evaluated through function fitting degree obtained by calculating the difference between the actual value and the theoretical value of the model.
In fig. 5, since only the temperature is increased, and a fire cannot be caused, in an experiment, the control of the comburent is performed so as to control the temperature change to the probability of the fire in the scene, but whether the fire occurs can be defined according to the difficulty of fire extinguishing (for example, the time required for extinguishing, the cost of the covering material, and the like). The implicit function set is a pre-selected possible implicit function, and the experimental fire scene set is a pre-selected scene set.
In fig. 6, since only the increase of the smoke concentration actually does not cause a fire, the control of the comburent may be performed so as to control the change of the smoke concentration during the experiment, but whether a fire occurs may be defined according to the difficulty of fire extinguishing (for example, the time required for extinguishing, the cost of the covering, etc.). The implicit function set is a pre-selected possible implicit function, and the experimental fire scene set is a pre-selected scene set.
In fig. 7, since only the increase of the combustible gas concentration actually does not result in a fire, the control of the combustibles may be performed during the experiment so as to control the change of the combustible gas concentration to the probability of the scene fire, but whether a fire occurs may be defined according to the difficulty of fire extinguishing (for example, the time required for extinguishing, the cost of the covering, etc.). The implicit function set is a pre-selected possible implicit function, and the experimental fire scene set is a pre-selected scene set.
The technical scheme has the advantages that the used hidden variable model not only respectively considers the influence condition of each judgment basis, but also considers the influence condition of the whole variable on the fire, and the relevance and systematicness between the judgment bases are increased to a certain extent.
As an embodiment, the method comprises the steps of calling data related to the temperature, the smoke concentration and the combustible gas concentration of a fire under each fire fighting scene from a fire fighting fire database, and recording the temperature coefficient, the smoke concentration coefficient and the combustible gas concentration coefficient when the fire breaks out;
then carrying out maximum inspection on the fire with the temperature coefficient, the smoke concentration coefficient and the combustible gas concentration coefficient and the implicit functions of the temperature single variable, the smoke single concentration variable and the combustible gas concentration single variable respectively to obtain the temperature coefficient a1Smoke concentration coefficient a2And the coefficient of combustible gas concentration a3And further constructing a fire occurrence event model F: a is F ═ a1f(T)+a2g(S)+a3h(X)。
Specifically, a process of constructing a fire occurrence event model F is shown in fig. 8.
In fig. 8, since only three measurable variables are improved, which cannot actually cause a fire, in an experiment, the combustibles may be controlled to control the change of the probability of the fire in the scene by the three measurable variables, but whether a fire occurs may be defined according to the difficulty of fire extinguishing (for example, the time required for extinguishing, the cost of the covering, etc.).
The technical scheme has the advantages that the hidden variable model is used, the maximum likelihood estimation method is utilized, the first type of errors and the second type of errors are effectively avoided, and the fire fighting fire alarm accuracy is improved.
Step 2: and solving the average value of the fire occurrence probability of all fire scenes, and when the average value of the fire occurrence probability is larger than the second type error threshold value of the statistical hypothesis test, sending out data abnormal alarm information and further judging the reason causing the data abnormal alarm information by human intervention.
The reason for causing the data abnormal alarm information comprises the following reasons: fire occurrence model failures and sensor failures.
The first type error threshold value of the statistical hypothesis test means that if the fire probability is considered to be greater than a certain threshold value, a timely alarm is required, and since the hidden variable modeling is used, 0.5 is used as the statistical hypothesis test first type error threshold value in the embodiment.
And when the average value of the probability of fire occurrence is greater than the second type error threshold value of the statistical hypothesis test, sending data abnormity alarm information and needing human intervention to investigate the actual situation, thereby inspecting whether the model is failed or whether the sensor is failed, causing input data abnormity, and finally improving the accuracy of fire alarm. The statistical hypothesis test in this example has a second type error threshold of 0.2.
The beneficial effects produced by the embodiment are as follows:
the accuracy is low: because the hidden variable model is used and the maximum likelihood estimation method is utilized, the first kind of errors and the second kind of errors are effectively avoided
Aiming at the judgment basis relative division: because the used hidden variable model not only respectively considers the influence condition of each judgment basis, but also considers the influence condition of the whole variable on the fire, the relevance and systematicness between the judgment bases are increased to a certain extent.
Difficult to estimate and determine for boundary values: in the construction of the hidden variable model, parameters are scientifically estimated by using fitting of data and a statistical method, and likelihood values are also scientifically estimated, so that the obtained boundary values are well documented, and the problem that the boundary values are difficult to estimate is solved to a certain extent.
Example two
As shown in fig. 9, the fire alarm system based on big data hidden variable model of the embodiment includes:
(1) fire scene fire alarm module, it is used for under each fire scene, real-time receipt temperature, smog concentration and combustible gas concentration utilize conflagration emergence event model F: a is F ═ a1f(T)+a2g(S)+a3h (X) outputting the fire occurrence probability under the corresponding fire fighting scene; when the fire occurrence probability under the fire fighting scene is larger than a first class error threshold value of the statistical hypothesis test, performing fire alarm on the corresponding fire fighting scene; wherein f (T) is an implicit function of fire and temperature T univariates, g (S) is an implicit function of fire and smoke single concentration S univariates, and h (X) is an implicit function of fire and combustible gas concentration X univariates; a is1、a2And a3All are constant coefficients, temperature coefficient and smoke respectivelyConcentration coefficient and combustible gas concentration coefficient.
As one embodiment, the implicit function f (T) of the fire and temperature single variable, the implicit function g (S) of the fire and smoke single concentration single variable and the implicit function h (X) of the fire and combustible gas concentration single variable are constructed by the following steps:
and (3) calling correlation data of the fire, temperature, smoke concentration and combustible gas concentration respectively under each fire scene from a fire protection fire database, and screening a hidden function f (T) of a fire and temperature single variable, a hidden function g (S) of a fire and smoke single concentration variable and a hidden function h (X) of a fire and combustible gas concentration single variable from a hidden function set according to the function fitting degree and the function structure.
Wherein, the construction process of the fire-fighting fire database is as follows:
and respectively acquiring the temperature change, the smoke concentration change and the combustible gas concentration under each experimental fire scene in real time until a fire disaster occurs, and obtaining the associated data between the fire disaster and the temperature, the smoke concentration and the combustible gas concentration under each fire scene so as to construct a fire-fighting fire disaster database.
According to the fire-fighting fire database, a data basis is provided for building the fire, the single implicit function of the temperature, the smoke concentration and the combustible gas concentration and the joint implicit function of the temperature, the smoke concentration and the combustible gas concentration, the efficiency and the accuracy of building a fire occurrence event model are guaranteed, and the accuracy of fire-fighting fire alarm is finally improved.
Specifically, the implicit functions in the set of implicit functions include tanh functions and logistic functions.
In a specific implementation, the experimental fire scenario includes: the household indoor environment, the working environment and the machine room.
The indoor environment of the house refers to common indoor environments of the house, including bedrooms, living rooms, toilets and kitchens, and experiments can be carried out according to various different environments.
The working environment refers to various working environments including offices, factories, and the like, and experiments can be performed according to these scenes.
The computer room generally refers to an environment scene in which various devices operate, and includes a computer room, a server room and the like, and experiments can be performed according to the scenes.
The construction of the implicit function f (T) of the fire and temperature single variable, the implicit function g (S) of the fire and smoke single concentration variable and the implicit function h (X) of the fire and combustible gas concentration single variable are respectively shown in fig. 5, 6 and 7.
In fig. 5, 6 and 7, the range of implicit functions is one of the ranges of implicit functions selected, and a decision is needed according to the scene, the data, the degree of fitting of the functions, and the like. And judging whether the fitting of the data is good or not through data fitting, wherein the data fitting refers to substituting data obtained through experiments into a theoretical model, and the fitting of the data is judged to be good or not and can be evaluated through function fitting degree obtained by calculating the difference between the actual value and the theoretical value of the model.
In fig. 5, since only the temperature is increased, and a fire cannot be caused, in an experiment, the control of the comburent is performed so as to control the temperature change to the probability of the fire in the scene, but whether the fire occurs can be defined according to the difficulty of fire extinguishing (for example, the time required for extinguishing, the cost of the covering material, and the like). The implicit function set is a pre-selected possible implicit function, and the experimental fire scene set is a pre-selected scene set.
In fig. 6, since only the increase of the smoke concentration actually does not cause a fire, the control of the comburent may be performed so as to control the change of the smoke concentration during the experiment, but whether a fire occurs may be defined according to the difficulty of fire extinguishing (for example, the time required for extinguishing, the cost of the covering, etc.). The implicit function set is a pre-selected possible implicit function, and the experimental fire scene set is a pre-selected scene set.
In fig. 7, since only the increase of the combustible gas concentration actually does not result in a fire, the control of the combustibles may be performed during the experiment so as to control the change of the combustible gas concentration to the probability of the scene fire, but whether a fire occurs may be defined according to the difficulty of fire extinguishing (for example, the time required for extinguishing, the cost of the covering, etc.). The implicit function set is a pre-selected possible implicit function, and the experimental fire scene set is a pre-selected scene set.
The technical scheme has the advantages that the used hidden variable model not only respectively considers the influence condition of each judgment basis, but also considers the influence condition of the whole variable on the fire, and the relevance and systematicness between the judgment bases are increased to a certain extent.
As an embodiment, the method comprises the steps of calling data related to the temperature, the smoke concentration and the combustible gas concentration of a fire under each fire fighting scene from a fire fighting fire database, and recording the temperature coefficient, the smoke concentration coefficient and the combustible gas concentration coefficient when the fire breaks out;
then carrying out maximum inspection on the fire with the temperature coefficient, the smoke concentration coefficient and the combustible gas concentration coefficient and the implicit functions of the temperature single variable, the smoke single concentration variable and the combustible gas concentration single variable respectively to obtain the temperature coefficient a1Smoke concentration coefficient a2And the coefficient of combustible gas concentration a3And further constructing a fire occurrence event model F: a is F ═ a1f(T)+a2g(S)+a3h(X)。
Specifically, a process of constructing a fire occurrence event model F is shown in fig. 8.
In fig. 8, since only three measurable variables are improved, which cannot actually cause a fire, in an experiment, the combustibles may be controlled to control the change of the probability of the fire in the scene by the three measurable variables, but whether a fire occurs may be defined according to the difficulty of fire extinguishing (for example, the time required for extinguishing, the cost of the covering, etc.).
The technical scheme has the advantages that the hidden variable model is used, the maximum likelihood estimation method is utilized, the first type of errors and the second type of errors are effectively avoided, and the fire fighting fire alarm accuracy is improved.
(2) And the data abnormity alarm module is used for solving the average value of the fire occurrence probability of all fire scenes, and when the average value of the fire occurrence probability is greater than the second type error threshold value of the statistical hypothesis test, sending data abnormity alarm information and needing human intervention to further judge the reason causing the data abnormity alarm information.
The reason for causing the data abnormal alarm information comprises the following reasons: fire occurrence model failures and sensor failures.
The first type error threshold value of the statistical hypothesis test means that if the fire probability is considered to be greater than a certain threshold value, a timely alarm is required, and since the hidden variable modeling is used, 0.5 is used as the statistical hypothesis test first type error threshold value in the embodiment.
And when the average value of the probability of fire occurrence is greater than the second type error threshold value of the statistical hypothesis test, sending data abnormity alarm information and needing human intervention to investigate the actual situation, thereby inspecting whether the model is failed or whether the sensor is failed, causing input data abnormity, and finally improving the accuracy of fire alarm. The statistical hypothesis test in this example has a second type error threshold of 0.2.
The beneficial effects produced by the embodiment are as follows:
the accuracy is low: because the hidden variable model is used and the maximum likelihood estimation method is utilized, the first kind of errors and the second kind of errors are effectively avoided
Aiming at the judgment basis relative division: because the used hidden variable model not only respectively considers the influence condition of each judgment basis, but also considers the influence condition of the whole variable on the fire, the relevance and systematicness between the judgment bases are increased to a certain extent.
Difficult to estimate and determine for boundary values: in the construction of the hidden variable model, parameters are scientifically estimated by using fitting of data and a statistical method, and likelihood values are also scientifically estimated, so that the obtained boundary values are well documented, and the problem that the boundary values are difficult to estimate is solved to a certain extent.
EXAMPLE III
A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the steps of the big-data hidden variable model-based fire alarm method as described in the first embodiment.
The beneficial effects produced by the embodiment are as follows:
the accuracy is low: because the hidden variable model is used and the maximum likelihood estimation method is utilized, the first kind of errors and the second kind of errors are effectively avoided
Aiming at the judgment basis relative division: because the used hidden variable model not only respectively considers the influence condition of each judgment basis, but also considers the influence condition of the whole variable on the fire, the relevance and systematicness between the judgment bases are increased to a certain extent.
Difficult to estimate and determine for boundary values: in the construction of the hidden variable model, parameters are scientifically estimated by using fitting of data and a statistical method, and likelihood values are also scientifically estimated, so that the obtained boundary values are well documented, and the problem that the boundary values are difficult to estimate is solved to a certain extent.
Example four
A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor implementing the steps in the big-data hidden variable model based fire alarm method as described in embodiment one when executing the program.
The beneficial effects produced by the embodiment are as follows:
the accuracy is low: because the hidden variable model is used and the maximum likelihood estimation method is utilized, the first kind of errors and the second kind of errors are effectively avoided
Aiming at the judgment basis relative division: because the used hidden variable model not only respectively considers the influence condition of each judgment basis, but also considers the influence condition of the whole variable on the fire, the relevance and systematicness between the judgment bases are increased to a certain extent.
Difficult to estimate and determine for boundary values: in the construction of the hidden variable model, parameters are scientifically estimated by using fitting of data and a statistical method, and likelihood values are also scientifically estimated, so that the obtained boundary values are well documented, and the problem that the boundary values are difficult to estimate is solved to a certain extent.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, and the like) having computer-usable program code embodied therein.
The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by a computer program, which can be stored in a computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), or the like.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A fire-fighting alarm method based on a big data hidden variable model is characterized by comprising the following steps:
under each fire scene, receiving temperature, smoke concentration and combustible gas concentration in real time, and utilizing a fire occurrence event model F: a is F ═ a1f(T)+a2g(S)+a3h (X) outputting the fire occurrence probability under the corresponding fire fighting scene; when the fire occurrence probability under the fire fighting scene is larger than a first class error threshold value of the statistical hypothesis test, performing fire alarm on the corresponding fire fighting scene;
calculating the average value of the fire occurrence probability of all fire scenes, and when the average value of the fire occurrence probability is larger than a second type error threshold value of statistical hypothesis test, sending data abnormal alarm information and further judging the reason causing the data abnormal alarm information by human intervention;
wherein f (T) is an implicit function of fire and temperature T univariates, g (S) is an implicit function of fire and smoke single concentration S univariates, and h (X) is an implicit function of fire and combustible gas concentration X univariates; a is1、a2And a3Are all usualThe coefficients are temperature coefficient, smoke concentration coefficient and combustible gas concentration coefficient respectively.
2. A fire alarm method based on big data hidden variable model as claimed in claim 1, wherein the cause of data abnormal alarm information includes: fire occurrence model failures and sensor failures.
3. A fire alarm method based on big data hidden variable model as claimed in claim 1, wherein the hidden function f (t) of fire and temperature single variable, the hidden function g(s) of fire and smoke single concentration variable and the hidden function h (x) of fire and combustible gas concentration single variable are constructed by:
and (3) calling correlation data of the fire, temperature, smoke concentration and combustible gas concentration respectively under each fire scene from a fire protection fire database, and screening a hidden function f (T) of a fire and temperature single variable, a hidden function g (S) of a fire and smoke single concentration variable and a hidden function h (X) of a fire and combustible gas concentration single variable from a hidden function set according to the function fitting degree and the function structure.
4. A fire alarm method based on big data hidden variable model as claimed in claim 3, wherein the construction process of the fire protection fire database is:
and respectively acquiring the temperature change, the smoke concentration change and the combustible gas concentration under each experimental fire scene in real time until a fire disaster occurs, and obtaining the associated data between the fire disaster and the temperature, the smoke concentration and the combustible gas concentration under each fire scene so as to construct a fire-fighting fire disaster database.
5. A fire alarm method based on big data implicit variable model according to claim 3, wherein the implicit functions in the implicit function set include tanh function and logistic function.
6. A fire alarm method based on big data hidden variable model as claimed in claim 3, wherein the experimental fire scene includes: the household indoor environment, the working environment and the machine room.
7. A fire alarm method based on big data hidden variable model as claimed in claim 3, characterized in that the data relating to the fire, temperature, smoke concentration and combustible gas concentration under each fire scene are retrieved from the fire-fighting fire database, and the temperature coefficient, smoke concentration coefficient and combustible gas concentration coefficient when the fire occurs are recorded;
then carrying out maximum inspection on the fire with the temperature coefficient, the smoke concentration coefficient and the combustible gas concentration coefficient and the implicit functions of the temperature single variable, the smoke single concentration variable and the combustible gas concentration single variable respectively to obtain the temperature coefficient a1Smoke concentration coefficient a2And the coefficient of combustible gas concentration a3And further constructing a fire occurrence event model F: a is F ═ a1f(T)+a2g(S)+a3h(X)。
8. The utility model provides a fire alarm system based on latent variable model under big data which characterized in that includes:
fire scene fire alarm module, it is used for under each fire scene, real-time receipt temperature, smog concentration and combustible gas concentration utilize conflagration emergence event model F: a is F ═ a1f(T)+a2g(S)+a3h (X) outputting the fire occurrence probability under the corresponding fire fighting scene; when the fire occurrence probability under the fire fighting scene is larger than a first class error threshold value of the statistical hypothesis test, performing fire alarm on the corresponding fire fighting scene;
the data abnormity alarm module is used for solving the average value of the fire occurrence probability of all fire scenes, and when the average value of the fire occurrence probability is larger than the second type error threshold value of the statistical hypothesis test, sending data abnormity alarm information and needing human intervention to further judge the reason causing the data abnormity alarm information;
wherein f (T) is an implicit function of a single variable of fire and temperature T, g (S) is an implicit function of a single variable of fire and smoke concentration S, h (X)Is an implicit function of fire and a single variable of the concentration X of the combustible gas; a is1、a2And a3All are constant coefficients, namely a temperature coefficient, a smoke concentration coefficient and a combustible gas concentration coefficient.
9. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the steps of the big-data implicit variable model based fire alarm method as claimed in any one of claims 1 to 7.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the steps in the big-data implicit variable model based fire alarm method of any of claims 1-7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010022344.8A CN111192428A (en) | 2020-01-09 | 2020-01-09 | Fire-fighting alarm method and system based on big data hidden variable model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010022344.8A CN111192428A (en) | 2020-01-09 | 2020-01-09 | Fire-fighting alarm method and system based on big data hidden variable model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111192428A true CN111192428A (en) | 2020-05-22 |
Family
ID=70708778
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010022344.8A Pending CN111192428A (en) | 2020-01-09 | 2020-01-09 | Fire-fighting alarm method and system based on big data hidden variable model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111192428A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111882806A (en) * | 2020-06-19 | 2020-11-03 | 杭州后博科技有限公司 | Indoor fire-fighting threat identification method and system |
CN112820062A (en) * | 2021-01-19 | 2021-05-18 | 武汉拓宝科技股份有限公司 | Fire occurrence probability prediction method and system |
CN113295824A (en) * | 2021-05-24 | 2021-08-24 | 江苏威尔安智能科技有限公司 | Portable intelligent gas detection alarm instrument based on Internet of things transmission data |
CN116153007A (en) * | 2022-12-29 | 2023-05-23 | 浙江佳伯尔电子科技有限公司 | Alarm control system and method based on sensor |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150379858A1 (en) * | 2011-10-25 | 2015-12-31 | Hydrotech, Inc. | Pump Monitoring Device |
CN107037423A (en) * | 2016-11-09 | 2017-08-11 | 谭顺成 | Multi-object tracking method is filtered with reference to the PHD of amplitude information |
CN108230637A (en) * | 2017-12-29 | 2018-06-29 | 深圳市泛海三江科技发展有限公司 | A kind of fire alarm method and system |
CN108769177A (en) * | 2018-05-23 | 2018-11-06 | 南京思特齐科技有限公司 | Internet of Things fire-fighting linkage system and method |
CN109920192A (en) * | 2019-03-15 | 2019-06-21 | 深圳力维智联技术有限公司 | Fire alarm method, system and computer readable storage medium |
CN110533871A (en) * | 2019-08-13 | 2019-12-03 | 浙江大华技术股份有限公司 | Smoke monitoring method, apparatus, computer equipment and storage medium |
-
2020
- 2020-01-09 CN CN202010022344.8A patent/CN111192428A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150379858A1 (en) * | 2011-10-25 | 2015-12-31 | Hydrotech, Inc. | Pump Monitoring Device |
CN107037423A (en) * | 2016-11-09 | 2017-08-11 | 谭顺成 | Multi-object tracking method is filtered with reference to the PHD of amplitude information |
CN108230637A (en) * | 2017-12-29 | 2018-06-29 | 深圳市泛海三江科技发展有限公司 | A kind of fire alarm method and system |
CN108769177A (en) * | 2018-05-23 | 2018-11-06 | 南京思特齐科技有限公司 | Internet of Things fire-fighting linkage system and method |
CN109920192A (en) * | 2019-03-15 | 2019-06-21 | 深圳力维智联技术有限公司 | Fire alarm method, system and computer readable storage medium |
CN110533871A (en) * | 2019-08-13 | 2019-12-03 | 浙江大华技术股份有限公司 | Smoke monitoring method, apparatus, computer equipment and storage medium |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111882806A (en) * | 2020-06-19 | 2020-11-03 | 杭州后博科技有限公司 | Indoor fire-fighting threat identification method and system |
CN111882806B (en) * | 2020-06-19 | 2022-09-06 | 杭州后博科技有限公司 | Indoor fire-fighting threat identification method and system |
CN112820062A (en) * | 2021-01-19 | 2021-05-18 | 武汉拓宝科技股份有限公司 | Fire occurrence probability prediction method and system |
CN112820062B (en) * | 2021-01-19 | 2022-05-03 | 武汉拓宝科技股份有限公司 | Fire occurrence probability prediction method and system |
CN113295824A (en) * | 2021-05-24 | 2021-08-24 | 江苏威尔安智能科技有限公司 | Portable intelligent gas detection alarm instrument based on Internet of things transmission data |
CN113295824B (en) * | 2021-05-24 | 2022-03-01 | 江苏威尔安智能科技有限公司 | Portable intelligent gas detection alarm instrument based on Internet of things transmission data |
CN116153007A (en) * | 2022-12-29 | 2023-05-23 | 浙江佳伯尔电子科技有限公司 | Alarm control system and method based on sensor |
CN116153007B (en) * | 2022-12-29 | 2023-09-26 | 浙江佳伯尔电子科技有限公司 | Alarm control system and method based on sensor |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111192428A (en) | Fire-fighting alarm method and system based on big data hidden variable model | |
CA2879090C (en) | Mobile device with automatic acquisition and analysis of building automation system | |
Du et al. | Sensor fault detection and its efficiency analysis in air handling unit using the combined neural networks | |
CN112270122B (en) | Inversion evaluation method for fire source parameters of building fire | |
Batov | The distinctive features of “smart” buildings | |
WO2019052241A1 (en) | Refrigerant leak detection method and device for air conditioner | |
CN116227763B (en) | Visual model-based fire emergency command method and system | |
CN1146768C (en) | Monitoring system for industrial plant | |
Sterling et al. | Model-based fault detection and diagnosis of air handling units: A comparison of methodologies | |
CN111260176A (en) | Method and system for eliminating fault conditions in a technical installation | |
Michaelides et al. | Contaminant event monitoring in multi-zone buildings using the state-space method | |
Han et al. | Fault detection and diagnosis of HVAC systems | |
Behravan et al. | Automatic model-based fault detection and diagnosis using diagnostic directed acyclic graph for a demand-controlled ventilation and heating system in Simulink | |
Mallak et al. | A graph-based sensor fault detection and diagnosis for demand-controlled ventilation systems extracted from a semantic ontology | |
CN113870522A (en) | Enhanced alerts using BACnet objects | |
Khire et al. | Model based failure mode effect analysis on whole building energy performance | |
Behravan et al. | Fault injection framework for fault diagnosis based on machine learning in heating and demand-controlled ventilation systems | |
CN116909257A (en) | Fault detection method, system, equipment and storage medium of industrial control system | |
KR102495864B1 (en) | Apparatus, method and computer program for deriving digital twin model | |
KR102408444B1 (en) | Method and System of virtually monitoring indoor-air-quality using Artificial Neural Network | |
CN114198863A (en) | Fault detection method and device for air conditioner in machine room, computer equipment and storage medium | |
Lazreg et al. | A Bayesian network model for fire assessment and prediction | |
CN109101702A (en) | A kind of fault detection method based on time-constrain rarefaction representation and Feature Dimension Reduction | |
KR20200065900A (en) | Method and apparatus for accomplishing goal of self-adaptive system | |
CN116338113B (en) | Environment intelligent perception simulation system, method, server and storage medium |
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 |
Application publication date: 20200522 |
|
RJ01 | Rejection of invention patent application after publication |