Fire bus type alarm controller

Abstract

The invention discloses a fire bus type alarm controller, which relates to the field of fire alarm and comprises an alarm control module used for acquiring start-stop control authority of each functional module and receiving and submitting alarm information; the detection and acquisition module is used for acquiring environmental factor data of smoke, temperature, humidity, airflow direction and airflow intensity in real time and working states and performance indexes of all detection devices to form analysis data real-time streams, and transmitting the analysis data real-time streams to the alarm control module through a bus; by acquiring environmental factors and equipment performance data in real time and analyzing data streams, potential fire risks are quickly judged, timely alarm giving is facilitated through a quick response mechanism, the response speed of fire response is increased, mutual influence between the environmental factors and the equipment performance is more accurately identified through a machine learning algorithm, and the fire risk can be quickly judged. Therefore, the prediction accuracy is effectively improved, and false alarms are reduced.

Description

Technical Field

[0001] The invention relates to the technical field of fire alarms, in particular to a fire bus type alarm controller. Background Art

[0002] With the development of smart buildings and the Internet of Things, alarm systems that integrate multiple sensors and smart devices need to be able to analyze big data in real time and respond quickly. Alarm controllers can access the system to share and use various data. In traditional alarm systems, multiple devices need to be wired independently, which makes the construction complex and costly. The bus-type system connects multiple devices through a data bus, which can significantly reduce wiring projects and improve construction efficiency, thereby achieving more accurate fire alarms. It is widely used in large commercial buildings, industrial plants or public places.

[0003] Existing fire alarm systems may generate false alarms or missed alarms in environmental changes such as high temperature and smoke. Under different environmental conditions, the fire alarm system's response to various influencing factors is prone to inconsistency, making it difficult to conduct effective evaluation. In addition, the alarm standards are often static and difficult to dynamically adjust according to actual conditions.

[0004] In addition, current fire alarm controllers have the problem of information processing lag and are unable to respond quickly to changing environmental conditions. This can easily lead to the fire alarm controller being unable to issue an alarm in time at a critical moment, affecting the efficiency of fire rescue. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In view of the above-mentioned shortcomings of the prior art, the present invention provides a fire bus type alarm controller, which can effectively solve the problems of the prior art.

[0007] (II) Technical solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0009] The present invention discloses a fire bus alarm controller, comprising:

[0010] The alarm control module is used to obtain the start and stop control authority of each functional module and receive and submit alarm information;

[0011] The detection and acquisition module is used to obtain real-time environmental factor data such as smoke, temperature, humidity, airflow direction and airflow intensity, as well as the working status and performance indicators of each detection equipment, to form a real-time analysis data stream, which is transmitted to the alarm control module through the bus;

[0012] The performance analysis module is used to obtain the detection equipment status data in the real-time stream of the current periodic analysis data, calculate the performance coefficient of each device, and generate the periodic performance coefficient by combining the historical performance data of the device;

[0013] The environmental analysis module is used to obtain the environmental factor data in the real-time stream of the current periodic analysis data, calculate the impact coefficient of each environmental factor, and generate the periodic environmental impact coefficient;

[0014] The model building module is used to build an interference model through a machine learning algorithm based on historical environmental impact data and performance data as training samples, identify the interference factors of environmental impact data and performance data on the detection and acquisition module, obtain the weight coefficients of several interference factors, and output the interference feature set;

[0015] A model prediction module is used to output the predicted performance coefficient and environmental coefficient of the next cycle based on the interference feature set of the current cycle;

[0016] Standard equipment module, used to verify the accuracy of predicted performance coefficient and environmental coefficient, and use the performance coefficient and environmental coefficient that meet the accurate threshold as a reference for setting alarm standards, supporting user-defined adjustments;

[0017] The data correction module is used to receive the predicted performance coefficient and environmental coefficient of the current cycle, and correct the performance coefficient and environmental coefficient currently collected in real time in combination with the current alarm information, and output the performance data and environmental coefficient after interference correction;

[0018] The judgment module is used to judge in real time whether to trigger an alarm based on the interference-corrected data and the preset alarm standard.

[0019] Furthermore, the performance analysis module obtains the current working status, sensitivity, response time and failure rate indicators of the detection equipment, cleans the collected data, removes outliers and noise, standardizes different types of status data, and combines historical performance data to determine the performance of each device in the past cycle, evaluates its important coefficients, allocates weights using the weighted average method, regularly performs calculations and updates the performance coefficients according to a preset time period, and outputs the calculation results to the model building module.

[0020] Furthermore, the calculation formula for the weighted average method is:

[0021] P=∑(S i ×W i ) / ∑W i ;

[0022] Where P represents the periodic performance coefficient, S i represents the standardized state coefficient of the i-th detection device, W iRepresents the weight of the i-th device.

[0023] Furthermore, the environmental analysis module analyzes the environmental factor data of the current period through the random forest algorithm, and calculates the environmental impact coefficient in combination with the weather forecast data. The specific process is as follows:

[0024] Step 1: Obtain environmental factors such as smoke density, temperature, humidity, airflow direction and airflow intensity, collect historical environmental factor data and corresponding fire event records, and obtain weather forecast data for the relevant area from the Internet;

[0025] Step 2: Remove missing values ​​and outliers from the data obtained in step 1, standardize the data, select feature data with a high correlation with fire risk that meets the preset threshold, and construct a data set;

[0026] Step 3: Divide the data set into training set and test set, import the random forest model, and perform training and testing;

[0027] Step 4: Analyze the impact coefficient of each environmental factor on fire risk through the built-in feature importance score of the random forest model to form a periodic environmental impact report.

[0028] Furthermore, the random forest model obtains the feature importance by calculating the impact score of each feature on the fire risk, and calculates the periodic environmental impact coefficient, and the calculation formula is:

[0029]

[0030] In the formula, E represents the periodic environmental impact coefficient, N represents the number of samples, and W j represents the weight of feature j, k represents the number of selected features, and N T represents the number of decision trees, I(D) represents the complexity of the original data set D, and I(Q) represents the subset after feature j is divided.

[0031] Furthermore, the periodic environmental impact report is formed by collecting importance scores of all trees in various features, comparing them with several actual fire incidents, and outputting a comparison and adjustment reference.

[0032] Furthermore, the data correction module performs weighted correction on the performance coefficient and environmental coefficient currently collected in real time, and merges the corrected performance coefficient and environmental coefficient with the currently collected data to form the final interference-corrected performance data and environmental coefficient.

[0033] Furthermore, the standard equipment module is interactively connected to an interactive module via a wireless network, and the interactive module is used to provide an interactive interface to the user, provide access and operation rights for custom settings, set alarm parameters, adjust alarm standards, and configure interference correction coefficients.

[0034] Furthermore, if the determination module determines that an alarm is triggered, the interference-corrected analysis data stream is delivered in the next preset period after the real-time analysis data stream is delivered.

[0035] Furthermore, the alarm control module is interactively connected to the detection and acquisition module through an electrical medium, the detection and acquisition module is interactively connected to the performance analysis module and the environmental analysis module through a wireless network, the model construction module is interactively connected to the performance analysis module, the environmental analysis module, the model prediction module, the standard equipment module and the data correction module through a wireless network, the model construction module is interactively connected to the model prediction module through a wireless network, the data correction module is interactively connected to the judgment module through a wireless network, and the judgment module is interactively connected to the alarm control module through a wireless network.

[0036] (III) Beneficial effects

[0037] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0038] 1. By acquiring environmental factors and equipment performance data in real time and analyzing data streams to quickly determine potential fire risks, this rapid response mechanism helps to issue alarms in a timely manner, improve the speed of fire response, and more accurately identify the mutual influence between environmental factors and equipment performance through machine learning algorithms, thereby more effectively improving the accuracy of predictions and reducing false alarms.

[0039] 2. By simultaneously monitoring environmental factors and the actual working status of the detection equipment, a more comprehensive safety monitoring solution is provided. By comprehensively considering the data from both aspects, it can better reflect whether the equipment can work effectively under the current environmental conditions.

[0040] 3. By customizing alarm parameters and standards according to specific needs, the system can be flexibly adjusted according to different environments and application scenarios, which increases the adaptability of the system and user satisfaction. The performance coefficient and environmental impact coefficient of the equipment are calculated regularly, so that fire risk prediction can be based on more comprehensive and accurate historical data and current conditions, providing continuous monitoring capabilities, which helps to detect potential problems in advance. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0042] Figure 1 It is a schematic diagram of the framework of the present invention.

[0043] The numbers in the figure represent: 1. Alarm control module; 2. Detection and acquisition module; 3. Performance analysis module; 4. Environmental analysis module; 5. Model building module; 6. Model prediction module; 7. Standard equipment module; 8. Data correction module; 9. Judgment module; 10. Interaction module. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0045] The present invention will be further described below in conjunction with the embodiments.

[0046] Example 1

[0047] A fire bus alarm controller of this embodiment, such as Figure 1 As shown, including:

[0048] Alarm control module 1, used to obtain the start and stop control authority of each functional module, and receive and submit alarm information;

[0049] The detection and acquisition module 2 is used to obtain the environmental factor data of smoke, temperature, humidity, airflow direction and airflow intensity and the working status and performance indicators of each detection equipment in real time, form a real-time analysis data stream, and transmit it to the alarm control module 1 through the bus; it provides a basis for subsequent data analysis and enhances the sensitivity and adaptability of the system;

[0050] The performance analysis module 3 is used to obtain the detection equipment status data in the real-time stream of the current cycle analysis data, calculate the performance coefficient of each device, and generate a periodic performance coefficient in combination with the historical performance data of the device. The performance analysis module 3 obtains the current working status, sensitivity, response time and failure rate indicators of the detection equipment, cleans the collected data, removes outliers and noise, standardizes different types of status data, and determines the performance of each device in the past cycle in combination with historical performance data. The important coefficients are evaluated and weighted average method is used to assign weights. The calculation is performed regularly according to the preset time period and the performance coefficient is updated, and the calculation results are output to the model construction module 5; the performance of the detection equipment is periodically evaluated, and the performance coefficient under special conditions is calculated to ensure that the equipment is always in the best working condition, so as to improve the accuracy and response speed of fire detection;

[0051] The calculation formula for weighted average allocation weight is:

[0052] P=∑(S i ×W i ) / ∑W i ;

[0053] Where P represents the periodic performance coefficient, S i represents the standardized state coefficient of the i-th detection device, W i Represents the weight of the i-th device;

[0054] Environmental analysis module 4, used to obtain environmental factor data in the real-time stream of current periodic analysis data, calculate the impact coefficient of each environmental factor, and generate a periodic environmental impact coefficient;

[0055] Model building module 5, used to build an interference model through a machine learning algorithm based on historical environmental impact data and performance data as training samples, identify the interference factors of environmental impact data and performance data on the detection and acquisition module 2, obtain weight coefficients of several interference factors, and output an interference feature set;

[0056] Model prediction module 6 is used to output the predicted performance coefficient and environmental coefficient of the next cycle based on the interference feature set of the current cycle, providing a forward-looking judgment basis for fire warning, thereby improving the effectiveness of system response;

[0057] The standard equipment module 7 is used to verify the accuracy of the predicted performance coefficient and environmental coefficient, and use the performance coefficient and environmental coefficient that meet the accurate threshold as a reference for the formulation of the alarm standard, and supports user-defined adjustment. The standard equipment module 7 is interactively connected to the interactive module 10 through a wireless network. The interactive module 10 is used to provide the user with an interactive interface, provide access and operation rights for customized settings, set alarm parameters, adjust alarm standards, and configure interference correction coefficients; verify the accuracy of the prediction, and adjust the alarm standards according to user needs, so that the system has strong flexibility and customizability to meet the security needs of different users;

[0058] The data correction module 8 is used to receive the predicted performance coefficient and environmental coefficient of the current cycle, and correct the performance coefficient and environmental coefficient currently collected in real time in combination with the current alarm information, and output the performance data and environmental coefficient after interference correction. The data correction module 8 weightedly corrects the performance coefficient and environmental coefficient currently collected in real time, and merges the corrected performance coefficient and environmental coefficient with the current collected data to form the final interference-corrected performance data and environmental coefficient; dynamic correction is performed in combination with real-time data to improve flexibility, ensure the accuracy and timeliness of alarm data, and provide users with more reliable early warning information;

[0059] The judgment module 9 is used to judge in real time whether to trigger an alarm based on the interference-corrected data and the preset alarm standard. If the judgment module 9 judges that an alarm is triggered, the interference-corrected analysis data stream will be submitted in the next preset period after the real-time analysis data stream is submitted, so as to ensure timely response when danger occurs and enhance fire safety protection measures.

[0060] As a preferred implementation in this embodiment, Figure 1 As shown, the alarm control module 1 is interactively connected with the detection and acquisition module 2 through an electrical medium, the detection and acquisition module 2 is interactively connected with the performance analysis module 3 and the environment analysis module 4 through a wireless network, the model construction module 5 is interactively connected with the performance analysis module 3, the environment analysis module 4, the model prediction module 6, the standard equipment module 7 and the data correction module 8 through a wireless network, the model construction module 5 is interactively connected with the model prediction module 6 through a wireless network, the data correction module 8 is interactively connected with the judgment module 9 through a wireless network, and the judgment module 9 is interactively connected with the alarm control module 1 through a wireless network.

[0061] Compared with existing technologies, it can obtain and analyze changes in environmental factors such as smoke, temperature, and humidity in real time, improve the early warning capability of fire, calculate the performance coefficient and periodic performance coefficient of the equipment in real time, ensure that the status of each detection equipment under different conditions is continuously monitored, and can detect equipment failures in time to ensure the effectiveness of the system;

[0062] The interference model built using machine learning algorithms can identify and analyze the relationship between environmental impacts and performance data, helping to adjust alarm conditions more accurately, thereby reducing false alarm rates. By weighted correction of real-time collected data, it combines current data with historical data to provide more accurate performance coefficients and environmental coefficients, thereby improving the robustness of the alarm.

[0063] Example 2

[0064] In other aspects, this embodiment also provides a method for calculating the environmental impact coefficient by using a random forest algorithm. The random forest algorithm analyzes the environmental factor data of the current period and calculates the environmental impact coefficient in combination with the weather forecast data. The specific process is as follows:

[0065] Step 1: Obtain environmental factors such as smoke density, temperature, humidity, airflow direction and airflow intensity, collect historical environmental factor data and corresponding fire event records, and obtain weather forecast data for the relevant area from the Internet;

[0066] Step 2: Remove missing values ​​and outliers from the data obtained in step 1, standardize the data, select feature data with a high correlation with fire risk that meets the preset threshold, and construct a data set;

[0067] Step 3: Divide the data set into training set and test set, import the random forest model, and perform training and testing;

[0068] Step 4: Analyze the impact coefficient of each environmental factor on fire risk through the built-in feature importance score of the random forest model to form a periodic environmental impact report;

[0069] The random forest model obtains the feature importance by calculating the impact score of each feature on the fire risk and calculates the periodic environmental impact coefficient. The calculation formula is:

[0070]

[0071] In the formula, E represents the periodic environmental impact coefficient, N represents the number of samples, and W j represents the weight of feature j, k represents the number of selected features, and N T represents the number of decision trees, I(D) represents the complexity of the original data set D, and I(Q) represents the subset after feature j is divided;

[0072] Periodic environmental impact reports are formed by collecting the importance scores of all trees in various characteristics, comparing them with several actual fire incidents, and outputting comparison and adjustment references.

[0073] Compared with existing technologies, by combining multiple environmental factors such as smoke concentration, temperature, humidity, airflow direction and intensity, and weather forecast data, it can more comprehensively reflect the comprehensive impact of the environment on fire risk, process high-dimensional data, and automatically select features that are highly correlated with fire risk, reducing the subjectivity and complexity of manual feature selection. Through the feature importance scoring function, it can clearly show the contribution of each environmental factor to fire risk, enabling decision makers to conduct more targeted environmental management and risk control;

[0074] By comparing with historical fire event records, the effectiveness of the model can be verified, promoting scientific decision-making based on data, and improving the efficiency and accuracy of fire prevention and control.

[0075] Working principle: The present invention performs bus control through the alarm control module 1 to feedback the alarm signal, obtains sensor data through the detection and acquisition module 2, obtains the operation setting data of the acquisition equipment associated with the performance analysis module 3 through the performance analysis module 3, and collects the environmental data of the detection area in real time through the environment analysis module 4. The model construction module 5 constructs an interference model based on the environmental data and the equipment operation equipment data, and then the model prediction module 6 predicts and outputs the predicted data before being interfered by the environmental data and the equipment operation equipment data. The setting standard of the model construction module 5 is set through the standard equipment module 7, and the user-defined standard is accessed through the interaction module 10. The current alarm data is corrected for interference through the data correction module 8. The judgment module 9 judges whether the alarm is triggered based on the interference-corrected data and the preset alarm standard. If it is judged to be yes, the correct alarm signal is output, otherwise the current alarm signal is interrupted;

[0076] The present invention combines the current alarm information and corrects the performance coefficient and the environmental coefficient in real time, thereby reducing the probability of false alarms and missed alarms. It realizes efficient data bus transmission and real-time analysis between the functional modules, can respond quickly to environmental changes, and improve the reaction speed of the system. It integrates environmental factors and equipment performance data, and can comprehensively evaluate fire risks from multiple aspects. This comprehensive analysis improves the intelligence level of the system, continuously optimizes performance and environmental analysis based on historical data, and improves the self-learning and self-adjustment capabilities of the system.

[0077] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fire bus alarm controller, characterized in that: include: An alarm control module (1) is used to obtain the start and stop control authority of each functional module and receive and submit alarm information; The detection and acquisition module (2) is used to obtain in real time environmental factor data such as smoke, temperature, humidity, airflow direction and airflow intensity, and the working status and performance indicators of each detection device, form a real-time analysis data stream, and transmit it to the alarm control module (1) through the bus; The performance analysis module (3) is used to obtain the detection device status data in the real-time stream of the current periodic analysis data, calculate the performance coefficient of each device, and generate a periodic performance coefficient by combining the performance history data of the device; The environmental analysis module (4) is used to obtain the environmental factor data in the real-time stream of the current periodic analysis data, calculate the influence coefficient of each environmental factor, and generate a periodic environmental influence coefficient; A model building module (5) is used to build an interference model through a machine learning algorithm based on historical environmental impact data and performance data as training samples, identify interference factors of environmental impact data and performance data on the detection and acquisition module (2), obtain weight coefficients of several interference factors, and output an interference feature set; A model prediction module (6), used for outputting a predicted performance coefficient and an environmental coefficient for the next cycle based on the interference feature set of the current cycle; The standard equipment module (7) is used to verify the accuracy of the predicted performance coefficient and environmental coefficient, and use the performance coefficient and environmental coefficient that meet the accurate threshold as a reference for setting the alarm standard, supporting user-defined adjustment; A data correction module (8) is used to receive the predicted performance coefficient and environmental coefficient of the current cycle, and correct the performance coefficient and environmental coefficient currently collected in real time in combination with the current alarm information, and output the performance data and environmental coefficient after interference correction; The judgment module (9) is used to judge in real time whether to trigger an alarm based on the interference-corrected data and a preset alarm standard.

2. A fire bus alarm controller according to claim 1, characterized in that: The performance analysis module (3) obtains the current working status, sensitivity, response time and failure rate indicators of the detection equipment, cleans the collected data, removes outliers and noise, standardizes different types of status data, combines historical performance data, determines the performance of each device in the past cycle, evaluates its important coefficient, allocates weights using the weighted average method, regularly performs calculations and updates the performance coefficients according to a preset time period, and outputs the calculation results to the model construction module (5).

3. A fire bus alarm controller according to claim 2, characterized in that: The calculation formula for the weighted average method is: P=∑(S i ×W i ) / ∑W i ; Where P represents the periodic performance coefficient, S i represents the standardized state coefficient of the i-th detection device, W i Represents the weight of the i-th device.

4. A fire bus alarm controller according to claim 1, characterized in that: The environmental analysis module (4) analyzes the environmental factor data of the current period by using a random forest algorithm, and calculates the environmental impact coefficient in combination with weather forecast data. The specific process is as follows: Step 1: Obtain environmental factors such as smoke density, temperature, humidity, airflow direction and airflow intensity, collect historical environmental factor data and corresponding fire event records, and obtain weather forecast data for the relevant area from the Internet; Step 2: Remove missing values ​​and outliers from the data obtained in step 1, standardize the data, select feature data with a high correlation with fire risk that meets the preset threshold, and construct a data set; Step 3: Divide the data set into training set and test set, import the random forest model, and perform training and testing; Step 4: Analyze the impact coefficient of each environmental factor on fire risk through the built-in feature importance score of the random forest model to form a periodic environmental impact report.

5. A fire bus alarm controller according to claim 4, characterized in that: The random forest model obtains the feature importance by calculating the impact score of each feature on the fire risk, and calculates the periodic environmental impact coefficient, and the calculation formula is: In the formula, E represents the periodic environmental impact coefficient, N represents the number of samples, and W j represents the weight of feature j, k represents the number of selected features, and N T represents the number of decision trees, I(D) represents the complexity of the original data set D, and I(Q) represents the subset after feature j is divided.

6. A fire bus alarm controller according to claim 4, characterized in that: The periodic environmental impact report is formed by collecting the importance scores of all trees in various characteristics, comparing them with several actual fire events, and outputting a comparison and adjustment reference.

7. A fire bus alarm controller according to claim 1, characterized in that: The data correction module (8) performs weighted correction on the performance coefficient and environmental coefficient currently collected in real time, and fuses the corrected performance coefficient and environmental coefficient with the currently collected data to form the final interference-corrected performance data and environmental coefficient.

8. A fire bus alarm controller according to claim 1, characterized in that: The standard equipment module (7) is interactively connected to an interactive module (10) via a wireless network, and the interactive module (10) is used to provide an interactive interface to the user, provide access and operation rights for custom settings, set alarm parameters, adjust alarm standards, and configure interference correction coefficients.

9. A fire bus alarm controller according to claim 1, characterized in that: If the judgment module (9) judges that an alarm is triggered, it submits the interference-corrected analysis data stream in the next preset period after the real-time analysis data stream is submitted.

10. A fire bus alarm controller according to claim 1, characterized in that: The alarm control module (1) is interactively connected to the detection and acquisition module (2) via an electrical medium; the detection and acquisition module (2) is interactively connected to the performance analysis module (3) and the environment analysis module (4) via a wireless network; the model construction module (5) is interactively connected to the performance analysis module (3), the environment analysis module (4), the model prediction module (6), the standard equipment module (7) and the data correction module (8) via a wireless network; the model construction module (5) is interactively connected to the model prediction module (6) via a wireless network; the data correction module (8) is interactively connected to the judgment module (9) via a wireless network; and the judgment module (9) is interactively connected to the alarm control module (1) via a wireless network.

Images