Fire fighting system alarm method based on multi-sensor coupling data support utility index prediction

Abstract

The invention discloses a fire-fighting system alarm method based on multi-sensor coupling data support utility index prediction, and relates to the technical field of fire-fighting system alarm, and the method comprises the steps: synchronously collecting real-time measurement data, and forming respective time sequences; carrying out normalization processing on various measurement data by adopting a minimum-maximum normalization method to obtain a normalized time sequence; an autoregressive integrated moving average model algorithm is adopted for the normalized time sequence, and normalized values of various parameters at the next moment in the future are predicted; and calculating a multi-sensor coupling data support utility index according to the predicted normalized values of various parameters at the next moment in the future, evaluating the coupling utility of the sensors, and triggering a fire behavior judgment strategy. According to the method, comprehensive identification of early characteristics of the fire is realized, so that the system can process abnormal data in a complex environment more stably, technical support is provided for upgrading of an intelligent fire-fighting system, and the method has engineering application value and popularization significance.

Description

Technical Field

[0001] This invention relates to the field of fire alarm technology, and in particular to a fire alarm method based on multi-sensor coupled data to support utility index prediction. Background Technology

[0002] Traditional fire alarm systems primarily rely on single-type sensors, such as smoke detectors or heat detectors, to determine the fire situation by monitoring changes in a specific physical parameter in the environment. These systems have revealed significant limitations in long-term application, exhibiting uneven sensitivity to specific types of fires. For example, smoke detectors are effective against slow-burning fires but relatively slow to respond to rapidly developing open flames, while heat detectors are the opposite. A more prominent problem is the high false alarm rate; non-fire factors can easily trigger alarms, severely undermining the reliability and trustworthiness of the alarm system. Furthermore, single-parameter detection struggles to accurately identify and warn of early-stage fires, often only confirming the fire after it has developed to a certain extent, missing the optimal window for early intervention. This technology is no longer sufficient to meet the demands of modern complex building structures, diverse usage scenarios, and increasingly stringent safety requirements, making the upgrading of fire alarm technology imperative.

[0003] To overcome the inherent limitations of single sensors, multi-sensor information fusion technology has been introduced into the field of fire alarms. Its core idea is to mimic the human cognitive process of using multiple senses to judge situations. This method integrates two or more sensors based on different principles at the same detection node, such as combining smoke, temperature, light, and carbon monoxide gas sensors, to form a multi-dimensional sensing unit. These sensors simultaneously monitor the environment, capturing potential fire characteristic information from multiple physical dimensions, including smoke concentration, ambient temperature, flame spectral characteristics, and characteristic gas concentrations. However, simply stacking multiple sensors cannot fundamentally solve the problem. The key lies in how to effectively process and comprehensively analyze this massive amount of heterogeneous, asynchronous, and potentially contradictory or complementary data. The goal is to extract more essential and reliable fire characteristics from multi-source information, suppressing the randomness and uncertainty of single information sources. This improves alarm accuracy while significantly reducing the probability of false alarms and missed alarms, enabling earlier and more accurate detection of the nascent stages of a fire. Summary of the Invention

[0004] In view of the problems existing in the current fire alarm method based on utility index prediction supported by multi-sensor coupled data, this invention is proposed. Therefore, the problem to be solved by this invention is how to provide a fire alarm method based on utility index prediction supported by multi-sensor coupled data.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] In a first aspect, the present invention provides a fire alarm method for predicting utility index based on multi-sensor coupled data, which includes: synchronously collecting real-time measurement data of smoke concentration, ambient temperature, ambient humidity, CO concentration and ambient radiation intensity to form their respective time series;

[0007] The min-max normalization method is used to normalize various types of measurement data to obtain normalized values ​​and normalized time series of various types of data at each time point;

[0008] An autoregressive integrated moving average model algorithm is used to predict the normalized values ​​of various parameters at the next time step in the normalized time series.

[0009] Based on the predicted normalized values ​​of various parameters at the next moment, the multi-sensor coupled data support utility index is calculated. The sensor coupling utility is evaluated based on the multi-sensor coupled data support utility index, and the fire detection strategy is triggered.

[0010] As a preferred embodiment of the fire alarm method for predicting utility index based on multi-sensor coupled data described in this invention, the real-time measurement data includes time series of smoke concentration, time series of ambient temperature, time series of ambient humidity, time series of CO concentration, and time series of ambient radiation intensity.

[0011] As a preferred embodiment of the fire alarm method for predicting utility index based on multi-sensor coupled data described in this invention, the normalization process includes:

[0012] The min-max normalization method is used to normalize various types of measurement data, and the expression is:

[0013] ;

[0014] in: For the i-th type of measurement data, the first... The normalized value at each moment. The minimum value of all time-time measurements in the i-th type of measurement data. The maximum value of all time-time measurements in the i-th type of measurement data; For the i-th type of measurement data, the first... Measurements at each moment;

[0015] The normalized time series of smoke concentration, ambient temperature, ambient humidity, CO concentration, and ambient radiation intensity were obtained by calculation.

[0016] As a preferred embodiment of the fire alarm system method based on multi-sensor coupled data to support utility index prediction as described in this invention, the step of using an autoregressive integrated moving average model algorithm to predict the normalized values ​​of various parameters at the next future moment includes:

[0017] For each type of parameter, the normalized time series is tested for stationarity. If the series is non-stationary, differencing is performed to eliminate the trend. The differencing series is defined as follows:

[0018] ;

[0019] in: This represents a first-order difference operator, where d is the order of the difference; This represents the value of the normalized time series at time t. for The value of the stationary sequence at time t after d-order difference;

[0020] An autoregressive moving average model is constructed on the differencing stationary sequence, mathematically expressed as:

[0021] ;

[0022] in: and , respectively, are the parameters of the i-th autoregressive term and the j-th moving average term. It is a white noise sequence with zero mean; For constant terms; The order of the autoregressive term; The order of the moving average term;

[0023] The model parameters are determined by maximum likelihood estimation or least squares method, and the model is used to predict the next step to obtain the predicted value of the difference sequence.

[0024] The predicted values ​​of the difference sequence are restored to the predicted normalized values ​​of the original sequence through inverse difference operation.

[0025] As a preferred embodiment of the fire alarm method for fire protection system based on multi-sensor coupled data to support utility index prediction as described in this invention, the predicted normalized values ​​of various parameters at the next future moment are specifically the predicted normalized values ​​of smoke concentration, ambient temperature, ambient humidity, CO concentration, and ambient radiation intensity.

[0026] As a preferred embodiment of the fire alarm system method based on multi-sensor coupled data support utility index prediction according to the present invention, wherein: the formula for calculating the multi-sensor coupled data support utility index is:

[0027] ;

[0028] in, The utility index supports multi-sensor coupled data; For smoke concentration at Normalized value of the time-matter prediction; For ambient humidity Normalized value of the time-matter prediction; For the ambient temperature at Normalized value of the time-matter prediction; For environmental radiation intensity at The normalized value of the time-time prediction.

[0029] As a preferred embodiment of the fire alarm method for supporting utility index prediction based on multi-sensor coupled data as described in this invention, the step of evaluating sensor coupling utility and triggering corresponding strategies based on multi-sensor coupled data supporting utility index includes:

[0030] If the support utility index is greater than or equal to the preset threshold, the sensor coupling utility is determined to be Level 1, the fire system alarm success rate is high, and the standard alarm procedure is executed.

[0031] If the support utility index is less than a preset threshold, the sensor coupling utility is determined to be level two, the alarm success rate of the fire protection system is low, and an enhanced fire discrimination strategy is triggered. The strategy includes weighting the sensor data and performing secondary data verification.

[0032] As a preferred embodiment of the fire alarm method for predicting utility index based on multi-sensor coupled data described in this invention, the weighted adjustment of sensor data includes: calculating the Pearson correlation coefficient between each sensor observation sequence and the reference sequence as a trend consistency index; normalizing the trend consistency value to obtain sensor weights; and performing weighted fusion of multi-source sensor data according to the weights.

[0033] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of a fire alarm system method based on multi-sensor coupled data to support utility index prediction.

[0034] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements the steps of a fire alarm system method based on multi-sensor coupled data to support utility index prediction.

[0035] The beneficial effects of this invention are as follows: By establishing a multi-sensor coupled data support utility index, this invention measures the synergy and consistency of various sensors in fire detection in real time, thereby achieving comprehensive identification of early fire characteristics. This not only overcomes the limitation of single sensors being susceptible to interference but also compensates for the lack of unified quantitative criteria in traditional multi-sensor fusion, enabling the system to handle abnormal data in complex environments more robustly. Furthermore, this invention combines time series prediction methods to infer the sensor response trend at the next moment, giving the alarm strategy foresight and adaptability, thus significantly improving the accuracy, timeliness, and overall reliability of fire early warning systems. This method can provide key technical support for the upgrade of intelligent fire protection systems and has significant engineering application value and promotional significance. Attached Figure Description

[0036] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart of a fire alarm system based on multi-sensor coupled data to support utility index prediction. Detailed Implementation

[0038] To make the above-mentioned objects, features, and advantages of the present invention more readily understood, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0039] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0040] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0041] Reference Figure 1This is the first embodiment of the present invention, which provides a fire alarm method for predicting utility index based on multi-sensor coupled data, including:

[0042] Step 1: Simultaneously measure real-time data of smoke concentration, ambient temperature, ambient humidity, CO concentration, and radiation intensity.

[0043] Time series of smoke concentrations were measured Time series of ambient temperature Time series of ambient humidity Time series of CO concentration Time series of environmental radiation intensity .

[0044] Step 2: Normalize the measurement data using the min-max normalization method, mapping data of different dimensions to the (0,1) interval. The expression for the min-max normalization method is:

[0045] ;

[0046] in: For the i-th type of measurement data, the first... The normalized value at each moment. The minimum value of all time-time measurements in the i-th type of measurement data. The maximum value of all time-time measurements in the i-th type of measurement data. For the i-th type of measurement data, the first... The measurement value at each moment.

[0047] Get any time Normalized value of smoke concentration Normalized value of ambient temperature Normalized values ​​of ambient humidity Normalized values ​​of CO concentration Normalized values ​​of environmental radiation intensity .

[0048] Normalized time series of smoke concentration was obtained. Normalized time series of ambient temperature Normalized time series of ambient humidity Normalized time series of CO concentration Normalized time series of environmental radiation intensity

[0049] Step 3: For non-stationary time series with normalized parameters, the Autoregressive Integrated Moving Average (ARIMA) model algorithm is used to predict the normalized parameter values ​​at the next time step.

[0050] The specific steps of the ARIMA algorithm are as follows:

[0051] Time series The step prediction aims to obtain future moments based on historical observations. The optimal estimate. ARIMA is a type of data-driven forecasting model widely used for stationary and weakly non-stationary time series. Its forecasting process can be divided into three stages: difference preprocessing, model building and parameter estimation, and one-step forecasting and inverse difference recovery.

[0052] First, the original sequence is tested for stationarity. If the sequence exhibits a trend or non-stationary characteristics, it needs to be stabilized using a difference operator. Let the difference order be d, and the difference sequence is defined as:

[0053] ;

[0054] in: This represents a first-order difference operator; an appropriate difference order can effectively eliminate the trend of a sequence, making the differenced sequence... To meet the stability requirements, for The value of the stationary sequence at time t after d-order differencing.

[0055] Secondly, an ARMA(p,q) model is constructed on the stationary sequence after differencing, and its mathematical form is:

[0056] ;

[0057] in: and , respectively, are the parameters of the i-th autoregressive term and the j-th moving average term. It is a white noise sequence with zero mean; This is a constant term used to adjust the model mean so that the model can describe a sequence with a non-zero mean. The order of the autoregressive term; The moving average term is the order; the model order (p, q) can be determined by the partial autocorrelation function, the autocorrelation function, and the information criterion.

[0058] The parameter estimates are then obtained using maximum likelihood estimation or least squares method, and are expressed as follows:

[0059] ;

[0060] In the formula, This is an estimate of the constant term in the ARIMA model; These are the coefficient estimates for the autoregressive component; These are the coefficient estimates for the moving average component.

[0061] Thus, the fitted ARMA model is obtained.

[0062] After obtaining the model parameters, a prediction is made for the next step. According to the one-step prediction theory of the ARMA model, the conditional expectation is expressed as:

[0063] ;

[0064] in: For the ARIMA model, the difference sequence values ​​for the next time step The predicted value; Let be the conditional expectation, representing the expectation given all historical observations. to In the case of the future The mathematical expectation; The constant term is an estimate obtained through parameter estimation; and These are the estimated values ​​of the i-th autoregressive term and the j-th moving average term, respectively. For observable measurements of historical difference sequences; This is the residual estimate obtained from model fitting.

[0065] Finally, the predicted difference components Inverse difference (IR) recovers the predicted values ​​of the original sequence. For first-order difference ARIMA(p, 1, q), the IR recovery formula is:

[0066] ;

[0067] In the formula, To analyze time series at future moments The predicted value; For time series at time... The actual observed value; This represents the next-step prediction of the difference sequence obtained through the ARIMA model. , Indicates to Perform a first-order difference. This represents the true value at future moments in the time series.

[0068] Using the ARIMA algorithm to obtain Normalized value of smoke concentration prediction at time twentieth hour Normalized values ​​of ambient temperature prediction Normalized values ​​of environmental humidity prediction Normalized values ​​of predicted CO concentration Normalized value of environmental radiation intensity .

[0069] Step 4: Calculate the next future moment. The multi-sensor coupled data support utility index is calculated using the following formula:

[0070] ;

[0071] in, The utility index supports multi-sensor coupled data; For smoke concentration at Normalized value of the time-matter prediction; For ambient humidity Normalized value of the time-matter prediction; For the ambient temperature at Normalized value of the time-matter prediction; For environmental radiation intensity at The normalized value of the time-time prediction.

[0072] like This indicates that the sensor coupling effect is Level 1, and the alarm success rate of the fire protection system is high;

[0073] like This indicates that the sensor coupling effect is level two, and the alarm success rate of the fire protection system is low. At this time, it is necessary to trigger an enhanced fire discrimination strategy, adjust the weight of the sensor, and perform secondary data verification to avoid false alarms or missed alarms.

[0074] Weighted adjustment of sensors refers to using a weight allocation method based on trend consistency. This involves calculating the correlation coefficient between each sensor and a reference trend as a reliability indicator, then normalizing this coefficient to obtain the weight, which is used for subsequent multi-source data fusion. Sensors with higher trend consistency receive greater weights; sensors with significant trend deviations automatically have their weights reduced, thus enhancing the stability and anti-interference capability of fire detection.

[0075] Weighted adjustments include: constructing a reference trend sequence The first principal component is obtained by weighted averaging or principal component analysis of all sensors.

[0076] For each sensor sequence The Pearson correlation coefficient with the reference sequence is calculated and expressed as:

[0077]

[0078] in, This indicates the degree to which the sensor response trend matches the overall fire situation trend.

[0079] Furthermore, the trend consistency indicator is normalized into weights. The resulting weights are used for sensor data fusion. m represents the number of sensors. By fusing multi-source measurement data, the influence of low-reliability sensors on alarm results is reduced, and sensors with severe trend deviations and high noise are automatically downweighted, thereby improving the stability of multi-sensor fusion and the accuracy of fire detection.

[0080] This embodiment also provides a computer device applicable to a fire alarm system based on multi-sensor coupled data to support utility index prediction, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement all or part of the steps of the method described in the above embodiments of the present invention.

[0081] This embodiment also provides a storage medium storing a computer program thereon. When the computer program is executed by a processor, it performs the method in any optional implementation of the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0082] The storage medium proposed in this embodiment and the data storage method proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0083] In summary, this invention establishes a multi-sensor coupled data support utility index to measure the synergy and consistency of various sensors in fire detection in real time, thereby achieving comprehensive identification of early fire characteristics. This index not only overcomes the limitation of single sensors being susceptible to interference but also compensates for the lack of unified quantitative criteria in traditional multi-sensor fusion, enabling the system to handle abnormal data in complex environments more robustly. Furthermore, this invention combines time series prediction methods to infer the sensor response trend at the next moment, giving the alarm strategy foresight and adaptability, thus significantly improving the accuracy, timeliness, and overall reliability of fire early warning systems. This method can provide key technical support for the upgrade of intelligent fire protection systems and has significant engineering application value and promotional significance.

[0084] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A fire alarm system based on multi-sensor coupled data to support utility index prediction, characterized in that: include, Real-time measurement data of smoke concentration, ambient temperature, ambient humidity, CO concentration and ambient radiation intensity are collected simultaneously to form their respective time series; The min-max normalization method is used to normalize various types of measurement data to obtain normalized values ​​and normalized time series of various types of data at each time point; An autoregressive integrated moving average model algorithm is used to predict the normalized values ​​of various parameters at the next time step in the normalized time series. Based on the predicted normalized values ​​of various parameters at the next moment, the multi-sensor coupled data support utility index is calculated. The sensor coupling utility is evaluated based on the multi-sensor coupled data support utility index, and the fire detection strategy is triggered.

2. The fire alarm method for a fire protection system based on multi-sensor coupled data to support utility index prediction as described in claim 1, characterized in that: The real-time measurement data includes time series of smoke concentration, ambient temperature, ambient humidity, CO concentration, and ambient radiation intensity.

3. The fire alarm method for a fire protection system based on multi-sensor coupled data to support utility index prediction as described in claim 1, characterized in that: The normalization process includes: The min-max normalization method is used to normalize various types of measurement data, and the expression is: ； in: For the i-th type of measurement data, the first... The normalized value at each moment. The minimum value of all time-time measurements in the i-th type of measurement data. The maximum value of all time-time measurements in the i-th type of measurement data; For the i-th type of measurement data, the first... Measurements at each moment; The normalized time series of smoke concentration, ambient temperature, ambient humidity, CO concentration, and ambient radiation intensity were obtained by calculation.

4. The fire alarm method for a fire protection system based on multi-sensor coupled data to support utility index prediction as described in claim 1, characterized in that: The method of using an autoregressive integrated moving average model to predict the normalized values ​​of various parameters at the next future time step includes: For each type of parameter, the normalized time series is tested for stationarity. If the series is non-stationary, differencing is performed to eliminate the trend. The differencing series is defined as follows: ； in: This represents a first-order difference operator, where d is the order of the difference; This represents the value of the normalized time series at time t. for The value of the stationary sequence at time t after d-order difference; An autoregressive moving average model is constructed on the differencing stationary sequence, mathematically expressed as: ； in: and , respectively, are the parameters of the i-th autoregressive term and the j-th moving average term. It is a white noise sequence with zero mean; For constant terms; The order of the autoregressive term; The order of the moving average term; The model parameters are determined by maximum likelihood estimation or least squares method, and the model is used to predict the next step to obtain the predicted value of the difference sequence. The predicted values ​​of the difference sequence are restored to the predicted normalized values ​​of the original sequence through inverse difference operation.

5. The fire alarm method for a fire protection system based on multi-sensor coupled data to support utility index prediction as described in claim 4, characterized in that: The predicted normalized values ​​of various parameters at the next future moment are specifically the predicted normalized values ​​of smoke concentration, ambient temperature, ambient humidity, CO concentration, and ambient radiation intensity.

6. The fire alarm method for a fire protection system based on multi-sensor coupled data to support utility index prediction as described in claim 1, characterized in that: The formula for calculating the multi-sensor coupled data support utility index is as follows: ； in, The utility index supports multi-sensor coupled data; For smoke concentration at Normalized value of the time-matter prediction; For ambient humidity Normalized value of the time-matter prediction; For the ambient temperature at Normalized value of the time-matter prediction; For environmental radiation intensity at The normalized value of the time-time prediction.

7. The fire alarm method for a fire protection system based on multi-sensor coupled data to support utility index prediction as described in claim 1, characterized in that: The method of evaluating sensor coupling utility based on multi-sensor coupled data support utility index and triggering corresponding strategies includes: If the support utility index is greater than or equal to the preset threshold, the sensor coupling utility is determined to be Level 1, the fire system alarm success rate is high, and the standard alarm procedure is executed. If the support utility index is less than a preset threshold, the sensor coupling utility is determined to be level two, the alarm success rate of the fire protection system is low, and an enhanced fire discrimination strategy is triggered. The strategy includes weighting the sensor data and performing secondary data verification.

8. The weighted adjustment of sensor data as described in claim 7, characterized in that: The weighted adjustment of sensor data includes: calculating the Pearson correlation coefficient between each sensor observation sequence and the reference sequence as a trend consistency index; normalizing the trend consistency value to obtain the sensor weight; and performing weighted fusion of multi-source sensor data according to the weight.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the fire alarm system method based on multi-sensor coupled data to support utility index prediction as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the fire alarm method for predicting the utility index based on multi-sensor coupled data as described in any one of claims 1 to 8.

Images