An internet-based intelligent fire alarm method and system

Abstract

The application belongs to the technical field of fire alarm, and discloses a smart fire alarm method and system based on the Internet, which comprises the following steps: by analyzing temperature, humidity and electrical equipment state, using the Internet for real-time data transmission, executing data classification, and identifying fire risk indicators, generating environment risk analysis results; in the application, by analyzing temperature, humidity and electrical equipment state, potential fire risks are identified in time, early warning can be performed in the initial stage of fire formation, potential damage of fire is significantly reduced, the response speed and prevention ability of fire alarm are significantly improved, real-time data transmission ensures that information can be quickly transmitted to the fire center in the initial stage of fire occurrence, allows emergency response units to prepare in advance, effectively reduces personnel casualties and property losses, enhances the adaptability to different environmental conditions, provides important data for post-fire analysis, and helps to optimize and adjust future strategies.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of fire alarm, specifically relates to an intelligent fire alarm method and system based on the Internet. BACKGROUND

[0002] The technical field of fire alarm mainly involves the design, implementation and management of systems for detecting early signs of fire, such as smoke, flame and abnormal temperature, from basic smoke detectors and fire alarms to integrated safety solutions that incorporate advanced sensing technology, automatic control systems and real-time data analysis; modern fire alarm systems can include a variety of sensors, such as photoelectric sensors, ion sensors, thermal sensors and gas detectors, which can be connected and monitored and analyzed in real time through a central monitoring system; not only can provide rapid response in case of fire, but also can help prevent the occurrence of fire through data prediction and trend analysis.

[0003] Among them, the intelligent fire alarm method based on the Internet refers to using Internet technology to enhance the function and efficiency of the traditional fire alarm system, by connecting the fire alarm system to the Internet, realizing remote monitoring, data sharing and intelligent analysis; the main purpose is to realize the real-time monitoring of the fire situation in the building, quickly transmit the alarm information to the fire center and the associated emergency response units, reduce personnel casualties and property losses; the system based on the Internet can collect and analyze a large amount of data to help optimize fire prevention strategies and improve the effectiveness of preventive measures; through intelligent algorithms, it can identify potential fire risks and issue early warnings, effectively improving the intelligent level of fire safety management.

[0004] Although the existing fire alarm technology contains a variety of sensors and real-time monitoring systems, it still has deficiencies in processing complex data and conducting efficient risk assessment, and the data analysis relies on post-processing rather than real-time analysis, limiting the system's response speed in emergency situations; the risk assessment in the prior art does not consider posterior probability, resulting in a lack of flexibility in responding to environmental changes, and the alarm system cannot be accurately adjusted to respond to specific risk levels; static risk assessment mode leads to false alarms or missed alarms under changing environmental conditions, affecting the accuracy and reliability of fire alarm; for example, in the case where the change in specific environmental conditions does not reach the preset threshold, the existing system cannot trigger an alarm in time, to some extent, increasing the risk of fire. SUMMARY

[0005] The technical problem to be solved by the present application is to provide an intelligent fire alarm method and system based on the Internet, which can significantly improve the response speed and prevention ability of fire alarm through the comprehensive application of real-time monitoring and intelligent analysis.

[0006] In order to solve the above technical problems, the present application provides the following technical scheme,

[0007] The application discloses an Internet-based intelligent fire alarm method, which comprises the following steps:

[0008] S1: by analyzing temperature, humidity and electrical equipment state, using the Internet for real-time data transmission, executing data classification, and identifying fire risk indicators, generating environment risk analysis results;

[0009] S2: using the environment risk analysis results, constructing a probability updating model, calculating the posterior probability under different environmental conditions, analyzing the risk level of fire using the posterior probability, and obtaining the posterior probability updating results;

[0010] S3: through the posterior probability updating results, through the Internet, the fire risk is evaluated, the risk weight of environmental factors is calculated, the risk level of the environment is analyzed through the risk weight calculation results, and the risk level determination results are obtained;

[0011] S4: according to the risk level determination results, adjusting the trigger parameters of intelligent fire alarm, setting the trigger threshold dynamically according to the risk level, optimizing the timeliness of alarm triggering, and generating alarm threshold adjustment results;

[0012] S5: implementing the alarm threshold adjustment results, real-time monitoring environmental data and comparing with preset threshold, real-time risk assessment of fire, updating the response state of intelligent fire alarm, and generating real-time risk assessment results;

[0013] S6: based on the real-time risk assessment results, using the Internet, simulating the intelligent fire emergency response test, evaluating the implementation effect of the response strategy, and generating the intelligent fire alarm log.

[0014] The application further optimizes the above technical solutions as follows:

[0015] The environment risk analysis results include key temperature indicators, key humidity indicators and fire equipment operation indicators, the posterior probability updating results include temperature adjustment probability, humidity adjustment probability and environmental safety probability, the risk level determination results include temperature risk level, humidity risk level and fire risk level, the alarm threshold adjustment results include temperature trigger threshold, humidity trigger threshold and electrical abnormal trigger threshold, the real-time risk assessment results include real-time data of temperature, real-time data of humidity and electrical data, and the intelligent fire alarm log includes response time record, simulation test evaluation and strategy effect analysis results.

[0016] Further optimization: the step of generating environment risk analysis results by analyzing temperature, humidity and electrical equipment state, using the Internet for real-time data transmission, executing data classification, and identifying fire risk indicators is as follows:

[0017] S101: By analyzing temperature, humidity and electrical equipment state, uploading data to the central server through the Internet, marking differentiated data as temperature data, humidity data and equipment state data, and performing data formatting processing, real-time environment data is obtained;

[0018] S102: Based on the real-time environment data, data classification processing is performed, data is filtered, data in normal range and abnormal range is distinguished, data in abnormal range is marked with iteration state, potential abnormal state is identified, and classification state data is obtained;

[0019] S103: According to the classification state data, fire risk assessment is performed, the risk value of differentiated data is calculated according to the preset risk assessment threshold, the overall fire risk level of the environment is evaluated, and the environment risk analysis result is obtained.

[0020] Further optimization: using the environment risk analysis result, constructing a probability updating model, calculating the posterior probability under differentiated environment conditions, using the posterior probability to analyze the risk level of fire, and obtaining the posterior probability updating result, the steps are as follows:

[0021] S201: Using the environment risk analysis result, initializing the environment parameter and historical environment data comparison model, through data normalization processing, checking whether the compared data is on the target scale, identifying the deviation of real-time environment parameter and historical data, and obtaining the environment change parameter;

[0022] S202: Based on the environment change parameter, updating the posterior probability model using parameter adjustment strategy, gradually adjusting the model parameter to match the new environment change information, and performing model verification after parameter updating, verifying the applicability of parameter adjustment, and obtaining the adjusted posterior probability;

[0023] S203: According to the adjusted posterior probability, classifying and quantitatively analyzing the fire risk level of differentiated environment, comparing the analysis result with the preset risk level threshold, and obtaining the posterior probability updating result.

[0024] Further optimization: through the posterior probability updating result, through the Internet, the fire risk is evaluated, the risk weight of environmental factors is calculated, the risk level of the environment is analyzed through the risk weight calculation result, and the risk level determination result is obtained, the steps are as follows:

[0025] S301: Based on the posterior probability updating result, through the Internet, synchronizing the current fire risk data, checking the integrity of the data, and analyzing the causes of the fire risk, obtaining the synchronous risk data;

[0026] S302: According to the synchronization risk data, the weight distribution of environmental factors is performed, the differentiated weight values are set according to the risk contribution degree of each factor, the fire risk value of the environmental factor is calculated, and the weighted risk score is obtained;

[0027] S303: According to the weighted risk score, the differentiated risk level is assigned to the environmental factor, each risk level is associated with the risk value of the target range, the real-time of the fire risk assessment is checked, and the risk level determination result is obtained.

[0028] Further optimization: According to the risk level determination result, the trigger parameter of intelligent fire alarm is adjusted, the trigger threshold is dynamically set according to the risk level, the timeliness of alarm triggering is optimized, and the step of generating alarm threshold adjustment result is specifically:

[0029] S401: Based on the risk level determination result, the trigger parameter of intelligent fire alarm is queried according to the Internet, the old trigger threshold associated with the differentiated level risk is recorded, and the real-time threshold record is obtained;

[0030] S402: According to the real-time threshold record, the intelligent fire alarm trigger threshold is adjusted according to the new risk level, it is checked whether the threshold matches the real-time risk level, the intelligent fire alarm parameter is updated, and the adjusted trigger parameter is obtained;

[0031] S403: The configuration file of intelligent fire alarm is updated through the adjusted trigger parameter, the effect of implementing the new threshold evaluation is reloaded, the adjusted response time is tested, and the alarm threshold adjustment result is obtained.

[0032] Further optimization: The alarm threshold adjustment result is implemented, the environmental data is monitored in real time and compared with the preset threshold, the real-time risk assessment of fire is performed, the response state of intelligent fire alarm is updated, and the step of generating real-time risk assessment result is specifically:

[0033] S501: According to the alarm threshold adjustment result, real-time monitoring is performed through the Internet, real-time collection of environmental temperature, humidity and smoke concentration data in differentiated areas is performed, data is cyclically monitored and recorded, and real-time monitoring data is obtained;

[0034] S502: Based on the real-time monitoring data, the new trigger threshold is compared, it is analyzed whether the parameter exceeds the safe range, the potential fire risk area is identified, and the risk assessment is performed, and the early warning risk level is obtained;

[0035] S503: The early warning risk level is used to update the alarm response state, if the evaluation result exceeds the target risk level, the associated alarm signal is triggered, the time and area of alarm activation are recorded, and the real-time risk assessment result is obtained.

[0036] Further optimization: based on the real-time risk assessment result, using the Internet, carrying out intelligent fire emergency response simulation test, evaluating the implementation effect of the response strategy, and generating the intelligent fire alarm log, the steps are specifically as follows:

[0037] S601: using the real-time risk assessment result, simulating the intelligent fire alarm through the Internet, simulating the fire response under different environmental conditions through network connection, checking the activation state of the intelligent fire alarm and response equipment, verifying the reaction time of the equipment under each simulation scenario, and obtaining simulation test records;

[0038] S602: according to the simulation test records, using a decision tree algorithm to analyze the response time and effect of the intelligent fire alarm under different test scenarios, and obtaining a response strategy evaluation result;

[0039] S603: based on the response strategy evaluation result, recording the date, time, response situation and adjustment measures of each simulation test through the Internet, and recording and tracking the time and area of the intelligent fire alarm, and obtaining an intelligent fire alarm log.

[0040] Further optimization: the formula of the decision tree algorithm is as follows:

[0041] ;

[0042] wherein, is the weighted Gini impurity, is the proportion of samples under the classification to the total samples, is the critical weight coefficient of the classification , is the total number of categories .

[0043] An intelligent fire alarm system based on the Internet, which is used to execute the above-mentioned intelligent fire alarm method based on the Internet, the system comprises:

[0044] The data collection module transmits data through the Internet according to temperature, humidity and electrical equipment state, classifies as environmental data, and analyzes and identifies fire risk indicators to generate environmental risk analysis results;

[0045] The risk analysis module establishes a probability updating model according to the environmental risk analysis results, calculates the posterior probability under different environmental conditions, evaluates the fire risk level by using probability, and obtains the posterior probability updating result;

[0046] The risk identification module updates the posterior probability, calculates the weight of the environmental risk level through the Internet, analyzes the risk level of the differentiated environmental factors, and obtains a risk level determination result.

[0047] The threshold optimization module adjusts the trigger parameters of the intelligent fire alarm, dynamically sets the trigger threshold according to the differentiated risk level, optimizes the timeliness of the alarm trigger, and generates an alarm threshold adjustment result.

[0048] The real-time monitoring module applies the alarm threshold adjustment result, monitors the environmental data in real time, compares with the preset threshold, evaluates the real-time risk of the fire, updates the response state of the intelligent fire alarm, generates the intelligent fire alarm log through the simulation test of the intelligent fire alarm response.

[0049] The above technical scheme has the following beneficial effects:

[0050] In the present application, through the comprehensive application of real-time monitoring and intelligent analysis, the reaction speed and prevention ability of the fire alarm are significantly improved, the temperature, humidity and electrical equipment state are analyzed, the potential fire risk is identified in time, the early warning can be performed in the early stage of fire formation, and the potential damage of fire is significantly reduced; real-time data transmission ensures that information can be quickly transmitted to the fire center in the early stage of fire occurrence, allows the emergency response unit to prepare in advance, effectively reduces personnel casualties and property losses; the application of posterior probability provides more accurate data support in evaluating the fire risk level, makes the dynamic adjustment of trigger parameters more flexible, enhances the adaptability to different environmental conditions, provides important data for post-fire analysis, and helps to optimize and adjust future strategies. BRIEF DESCRIPTION OF DRAWINGS

[0051] Figure 1 The figure is a workflow schematic diagram of the embodiment of the present application.

[0052] Figure 2 The figure is a detailed flowchart of step S1 in the embodiment of the present application.

[0053] Figure 3 The figure is a detailed flowchart of step S2 in the embodiment of the present application.

[0054] Figure 4 The figure is a detailed flowchart of step S3 in the embodiment of the present application.

[0055] Figure 5 The figure is a detailed flowchart of step S4 in the embodiment of the present application.

[0056] Figure 6 The figure is a detailed flowchart of step S5 in the embodiment of the present application.

[0057] Figure 7A detailed flowchart for step S6 in the embodiment of the present application;

[0058] Figure 8 A flowchart of the system in the embodiment of the present application. DETAILED DESCRIPTION

[0059] Embodiment one:

[0060] Please refer to Figure 1 The present application provides a technical solution, a smart fire alarm method based on the Internet, comprising the following steps:

[0061] S1: By analyzing temperature, humidity and electrical equipment state, using the Internet for real-time data transmission, executing data classification, marking the iteration state of abnormal range data, and identifying fire risk indicators, generating environment risk analysis results;

[0062] S2: Using the environment risk analysis results, constructing a probability updating model, calculating the posterior probability under different environmental conditions, using the posterior probability to analyze the risk level of fire, classifying and quantitatively analyzing the fire risk level of different environments, and obtaining the posterior probability updating results;

[0063] S3: Through the posterior probability updating results, through the Internet, the fire risk is evaluated, the risk weight of environmental factors is calculated, the risk level of the environment is analyzed through the risk weight calculation results, and the risk level determination result is obtained;

[0064] S4: According to the risk level determination result, adjust the trigger parameter of the smart fire alarm, dynamically set the trigger threshold according to the risk level, optimize the timeliness of the alarm trigger, and generate the alarm threshold adjustment result;

[0065] S5: Implement the alarm threshold adjustment result, real-time monitor the environmental data and compare with the preset threshold, real-time risk assessment of fire, update the response state of the smart fire alarm, and generate the real-time risk assessment result;

[0066] S6: Based on the real-time risk assessment result, using the Internet, the smart fire emergency response simulation test is carried out, the date, time, response situation and adjustment measures of each simulation test are recorded, the implementation effect of the response strategy is evaluated, and the smart fire alarm log is generated.

[0067] The environmental risk analysis result includes a key temperature index, a key humidity index, and a fire equipment operation index, the posterior probability update result includes a temperature adjustment probability, a humidity adjustment probability, and an environmental safety probability, the risk level determination result includes a temperature risk level, a humidity risk level, and a fire risk level, the alarm threshold adjustment result includes a temperature trigger threshold, a humidity trigger threshold, and an electrical anomaly trigger threshold, the real-time risk assessment result includes real-time data of temperature, real-time data of humidity, and electrical data, and the intelligent fire alarm log includes a response time record, a simulation test evaluation, and a strategy effect analysis result.

[0068] Referring to Figure 2 By analyzing temperature, humidity, and electrical equipment state, using the Internet for real-time data transmission, performing data classification, and identifying fire risk indicators, the steps for generating the environmental risk analysis result are as follows:

[0069] S101: By analyzing temperature, humidity, and electrical equipment state, uploading data to a central server using the Internet, marking differentiated data as temperature data, humidity data, and equipment state data, and performing data formatting processing, the execution process of obtaining real-time environmental data is as follows:

[0070] S101 sub-step analyzes temperature, humidity, and electrical equipment state, deploys multi-point sensors to capture key parameters in real time, and transmits data to a central server through the Internet; on the server side, data is preliminarily formatted to distinguish temperature data, humidity data, and equipment state data, which can mark various data and perform necessary formatting to facilitate subsequent analysis and processing, including time stamp, data type identification, and specific measurement value, to provide standardized input for further data analysis and application, obtaining real-time environmental data.

[0071] S102: Based on real-time environmental data, perform data classification processing, filter data, distinguish normal range and abnormal range data, and mark iteration state of abnormal range data to identify potential abnormal state, the execution process of obtaining classification state data is as follows:

[0072] S102 sub-step is based on real-time environmental data, performs data classification processing, sets normal and abnormal threshold standards, and filters and distinguishes data according to the standards; all data outside the normal operating range is marked as abnormal and is marked with iteration state to identify potential abnormal state; classification processing not only helps to quickly identify abnormal range data that needs attention, but also makes the response to potential risks more rapid and effective, obtaining classification state data.

[0073] S103: According to the classification state data, fire risk assessment is carried out, and the risk value of the differentiated data is calculated according to the preset risk assessment threshold. The overall fire risk level of the environment is evaluated, and the execution process of the environment risk analysis result is as follows:

[0074] S103 substep According to the classification state data, fire risk assessment is carried out, and the risk value of the differentiated data is calculated according to the preset risk assessment threshold. The abnormal frequency, severity and influence range of various parameters are calculated. Based on the calculation, the overall fire risk level of the environment is evaluated, and the quantitative analysis of the potential fire risk in the building is provided. Risk assessment helps managers take appropriate preventive measures or emergency response to ensure the safety of buildings and residents, and obtains the environment risk analysis result, which uses the formula:

[0075] ;

[0076] Among them, is the fire risk level of the environment, is the weight of the first classification state data, is the number of the first classification state data, is the corresponding risk index.

[0077] Please refer to Figure 3 , the environment risk analysis result is used to construct a probability updating model to calculate the posterior probability under the differentiated environment condition, and the posterior probability is used to analyze the risk level of the fire, and the step of obtaining the posterior probability updating result is as follows:

[0078] S201: Use the environment risk analysis result to initialize the environment parameter and historical environment data comparison model. Through data normalization processing, check whether the compared data is on the target scale, identify the deviation between real-time environment parameters and historical data, and obtain the execution process of environment change parameters as follows:

[0079] S201 substep Use the environment risk analysis result to initialize an environment parameter and historical environment data comparison model. Normalize the real-time monitoring data and historical data to ensure that they are on the same scale when compared and analyzed, allowing accurate identification of the deviation between real-time environment parameters and historical data, and better understanding of the change trend of the environment. After normalization processing, the specific parameters of environmental change can be clearly identified, such as temperature, humidity mutation or continuous high state. The parameters will be used for subsequent model updating and risk assessment, and the posterior probability updating result is obtained, which uses the formula:

[0080] ;

[0081] Among them, denotes the normalized data, is the original data, and are the minimum and maximum values of the data, respectively.

[0082] S202: Based on the environmental change parameter, the parameter adjustment strategy is adopted to update the posterior probability model, gradually adjust the model parameters to match the new environmental change information, and perform model verification after parameter updating to verify the applicability of parameter adjustment. The execution process of the adjusted posterior probability is as follows:

[0083] S202 sub-step is based on the environmental change parameter, and the posterior probability model is updated by adopting the parameter adjustment strategy. It involves gradually adjusting the model parameters to match the new environmental change information. Through comparative analysis and feedback adjustment, the prediction ability and accuracy of the model are continuously optimized. After parameter updating, model verification is performed to confirm the applicability and effectiveness of parameter adjustment. It includes the response and prediction accuracy of the test model under new environmental conditions, ensures that the model update can reflect the changes in the actual environment, improves the reliability of environmental risk assessment, and obtains the adjusted posterior probability. The formula is:

[0084] ;

[0085] Among them, is the updated posterior probability, is the original posterior probability, is the adjustment coefficient, is the environmental change parameter.

[0086] S203: According to the adjusted posterior probability, the fire risk level of the differentiated environment is classified and quantitatively analyzed, and the analysis result is compared with the preset risk level threshold to obtain the execution process of the posterior probability update result as follows:

[0087] S203 sub-step is based on the adjusted posterior probability to classify and quantitatively analyze the fire risk level of the differentiated environment. The risk level obtained by analysis is compared with the preset risk level threshold to evaluate the fire risk of the current environment, allowing decision-makers to understand the trend of environmental risk and take appropriate preventive measures or adjust safety strategies. Through comparison and analysis, the posterior probability update result is obtained.

[0088] Please refer to Figure 4 , through the posterior probability update result, through the Internet, the fire risk is evaluated, the risk weight of environmental factors is calculated, the risk level of the environment is analyzed through the risk weight calculation result, and the steps of obtaining the risk level determination result are as follows:

[0089] S301: Based on the posterior probability update result, synchronize the current fire risk data through the Internet, check the integrity of the data, and analyze the causes of the fire risk, and the execution process of the synchronized risk data is as follows:

[0090] S301 sub-step based on posterior probability update result, synchronize the current fire risk data to the intelligent fire fighting system through the Internet, it is essential to ensure the integrity and security of the data; after data synchronization, verify the integrity of the data, ensure that all key information has been correctly transmitted and has not been damaged or tampered with, analyze the synchronized fire data in depth, identify and understand the main reasons for the risk change, including temperature anomaly, humidity change or electrical equipment failure and other factors, not only can provide the current risk state, but also can provide data support for prevention and response measures, get the synchronized risk data, the formula is:

[0091] ;

[0092] Among them, represents the correlation between each factor and fire risk, is the observation value of the current environmental factor, is the historical data related to fire risk, and are the average values of the observation value of the environmental factor and the historical data related to fire risk respectively.

[0093] S302: According to the synchronized risk data, weight distribution of environmental factors is carried out, according to the risk contribution degree of each factor, the differentiated weight value is set, the fire risk value of environmental factors is calculated, and the execution process of weighted risk score is as follows:

[0094] S302 sub-step according to the synchronized risk data, weight distribution of environmental factors is carried out; according to the contribution degree of each factor to fire risk, the differentiated weight value is set; for example, if the data shows that temperature anomaly contributes most to fire risk, higher weight is given to temperature factor; through weight distribution, the contribution of each environmental factor to the overall fire risk can be calculated more accurately, reflecting the fire risk situation under the comprehensive action of each factor in the environment, getting the weighted risk score, the formula is:

[0095] ;

[0096] Among them, is the risk value of environmental factors, is the weight, is the predicted risk index of environmental factors.

[0097] S303: Assign different risk levels to environmental factors by weighting risk scores, each risk level associated with a specific range of risk values, check the real-time nature of fire risk assessment, and obtain the execution process of risk level determination results as follows:

[0098] S303 sub-step assigns different risk levels to different environmental factors by weighting risk scores, each risk level associated with a specific range of risk values, ensuring the real-time nature and accuracy of fire risk assessment, quickly identifying and responding to potential high-risk areas or conditions, and timely taking preventive measures, effectively guiding fire departments or building managers to conduct corresponding risk management, and obtaining risk level determination results.

[0099] Please refer to Figure 5 , according to the risk level determination result, adjust the trigger parameters of intelligent fire alarm, set the trigger threshold dynamically according to the risk level, optimize the timeliness of alarm triggering, and the steps of generating alarm threshold adjustment results are as follows:

[0100] S401: Based on the risk level determination result, access the configuration interface based on the Internet, query the trigger parameters of intelligent fire alarm, record the old trigger threshold associated with the differentiated risk level, and obtain the execution process of real-time threshold record as follows:

[0101] S401 sub-step based on risk level determination result, access configuration interface through Internet; In this interface, query and record the intelligent fire alarm trigger parameters associated with the current risk level, including the old trigger threshold; Ensure real-time understanding of intelligent fire alarm system, can track existing settings and prepare for upcoming adjustments; Record the threshold is the key, because it is the basis for evaluating the current system response and making necessary adjustments, and obtain real-time threshold record.

[0102] S402: According to the real-time threshold record, adjust the intelligent fire alarm trigger threshold according to the new risk level, check whether the threshold matches the real-time risk level, update the intelligent fire alarm parameters, and obtain the execution process of the adjusted trigger parameters as follows:

[0103] S402 sub-step adjusts the intelligent fire alarm trigger threshold according to the real-time threshold record; This adjustment is based on the new risk level and the requirement for system sensitivity; By evaluating the matching of risk level and current threshold, adjust the trigger threshold appropriately to improve the prevention and response capability of the system; After completing the adjustment, update the system parameters, including updating the trigger parameters in the intelligent fire alarm system, and ensure that the new parameters match the actual risk level, because it directly affects the performance of the system in actual fire situations, and obtain the adjusted trigger parameters, using the formula:

[0104] ;

[0105] wherein, is the adjusted new trigger threshold, is the old trigger threshold, is the adjustment factor determined based on the risk level.

[0106] S403: Update the configuration file of the intelligent fire alarm system with the adjusted trigger parameters, reload the configuration and evaluate the implementation effect of the new threshold, test the response time after threshold adjustment, and obtain the execution process of the alarm threshold adjustment result as follows:

[0107] S403 sub-step: Update the alarm configuration file of the intelligent fire alarm system with the adjusted trigger parameters; reload the configuration and evaluate the implementation effect of the new threshold; including testing the response time after threshold adjustment and the overall reaction ability of the system; through testing, ensure that the newly set threshold not only reflects the latest risk assessment result, but also can provide effective alarm in emergency, get the alarm threshold adjustment result.

[0108] Please refer to Figure 6 , implement the alarm threshold adjustment result, real-time monitor environmental data and compare with preset threshold, conduct real-time risk assessment of fire, update the response state of intelligent fire alarm, and generate the steps of real-time risk assessment result as follows:

[0109] S501: According to the alarm threshold adjustment result, real-time monitoring is carried out through the Internet, real-time collection of environmental temperature, humidity and smoke concentration data in different areas is carried out, and the data is monitored and recorded in a cycle, and the execution process of real-time monitoring data is as follows:

[0110] S501 sub-step: According to the alarm threshold adjustment result, real-time monitoring of the intelligent fire alarm system is carried out through the Internet, including real-time collection of environmental temperature, humidity and smoke concentration data in different areas; using sensor network to continuously monitor key parameters, and through the data management platform of the system to monitor and record in a cycle; real-time monitoring data provides continuous update of environmental state for the system, ensures that the system can respond quickly when any abnormal situation occurs, and generates real-time risk assessment result.

[0111] S502: Based on the real-time monitoring data, compare with the set new trigger threshold, analyze whether the parameters exceed the safe range, identify the potential fire risk area, and conduct risk assessment, and obtain the execution process of the early warning risk level as follows:

[0112] S502, based on real-time monitoring data, compares the new trigger threshold set in the intelligent fire fighting system, analyzes whether the parameters exceed the safety range, and identifies potential fire risk areas according to the exceeding situation; the analysis process includes statistical evaluation of temperature, humidity and smoke data to determine whether the area is in a high-risk state; risk assessment is made according to the data to generate early warning risk level.

[0113] S503: Use the early warning risk level to update the alarm response state, if the evaluation result exceeds the target risk level, trigger the associated alarm signal, and record the time and area of alarm activation, the execution process of real-time risk assessment result is as follows;

[0114] S503, using the early warning risk level, updates the alarm response state of the intelligent fire fighting system; if the evaluation result shows that the risk level of a certain area exceeds the set target risk level, the system will automatically trigger the associated alarm signal; at the same time, record the time and specific area of alarm activation, ensure that all key information is recorded in detail, so as to facilitate post-analysis and improvement measures, get the risk assessment result.

[0115] Please refer to Figure 7 , based on real-time risk assessment results, use the Internet to simulate intelligent fire emergency response test, evaluate the implementation effect of response strategy, generate intelligent fire alarm log steps are as follows:

[0116] S601: Use real-time risk assessment results to simulate intelligent fire alarm through the Internet, simulate fire response under different environmental conditions through network connection, check the activation state of intelligent fire alarm and response equipment, verify the reaction time of equipment in each simulation scenario, and the execution process of simulation test record is as follows:

[0117] S601, using real-time risk assessment results, simulates intelligent fire fighting system through the Internet, tests fire response under different environmental conditions to ensure that the system can work effectively in various situations; through network connection, set up simulation fire scene, check the activation state of intelligent fire alarm and response equipment; record the reaction time of equipment in each scene in detail, evaluate the timeliness of alarm system and the efficiency of response equipment, get simulation test record.

[0118] S602: According to the simulation test record, use decision tree algorithm to analyze the response time and effect of intelligent fire alarm in different test scenarios, get the response strategy evaluation result execution process as follows:

[0119] S602, according to the simulation test record, the decision tree algorithm is used to analyze the response time and effect of the intelligent fire alarm system in different test scenarios; the decision tree algorithm classifies data to identify the effectiveness of the response strategy by evaluating the response time and scenario variables; helps to determine the factors that most affect the performance of the alarm system, and provides the basis for optimizing the alarm strategy, and obtains the response strategy evaluation result.

[0120] The formula of the decision tree algorithm is as follows:

[0121] ;

[0122] Among them, is the weighted Gini impurity, is the proportion of samples in the classification to the total samples, is the weight coefficient of the classification key, is the total number of categories .

[0123] The execution process is as follows:

[0124] Determine the weight coefficient of each category , the coefficient is adjusted according to the influence and sensitivity of each category in the previous historical data, calculate the occurrence probability of each category, accumulate the weighted to evaluate the weighted impurity of each split node, select the split attribute with the smallest weighted Gini impurity, determine the structure of the tree, and the determination of the weight coefficient .

[0125] S603: Based on the response strategy evaluation result, through the Internet, record the date, time, response and adjustment measures of each simulation test, and record and track the time and area of the intelligent fire alarm, and the execution process of the intelligent fire alarm log is as follows:

[0126] S603, based on the response strategy evaluation result, record the date, time, response and adjustment measures of each simulation test through the Internet, provide detailed historical data for long-term monitoring and maintenance of the system; record includes specific details of each test and corresponding adjustment and optimization measures, ensure that all important information is tracked and archived; the log is crucial for continuously improving the performance and response capability of the intelligent fire alarm system, forming the intelligent fire alarm log.

[0127] Please refer to Figure 8The application discloses an internet-based intelligent fire alarm system, which is used for executing the above-mentioned internet-based intelligent fire alarm method, and comprises the following modules.

[0128] The data collection module transmits data through the Internet according to temperature, humidity and electrical equipment states, classifies the data as environmental data, analyzes and identifies fire risk indexes, and generates an environmental risk analysis result.

[0129] The risk analysis module establishes a probability updating model according to the environmental risk analysis result, calculates a posterior probability under differentiated environmental conditions, evaluates a fire risk grade by using the probability, and obtains a posterior probability updating result.

[0130] The risk identification module calculates the weight of the environmental risk grade through the Internet by using the posterior probability updating result, analyzes the risk grade of the differentiated environmental factors, and obtains a risk grade determination result.

[0131] The threshold optimization module adjusts the triggering parameters of the intelligent fire alarm by using the risk grade determination result, dynamically sets a triggering threshold according to the differentiated risk grade, optimizes the timeliness of the alarm triggering, generates an alarm threshold adjustment result, and applies the alarm threshold adjustment result to the real-time monitoring module.

[0132] The real-time monitoring module applies the alarm threshold adjustment result to real-time monitoring of the environmental data, compares the environmental data with preset threshold values, evaluates the real-time risk of a fire, updates the response state of the intelligent fire alarm, simulates the response of the intelligent fire alarm, and generates an intelligent fire alarm log.

[0133] The above is only a preferred embodiment of the application, and does not limit the application in other forms. Any person skilled in the art can modify or change the above-mentioned technical content to obtain equivalent embodiments applied to other fields. However, any simple modification, equivalent change and modification made according to the technical essence of the application to the above-mentioned embodiments still belong to the protection scope of the application.

Claims

1. An Internet-based intelligent fire alarm method, characterized in that: The method comprises the following steps: By analyzing temperature, humidity and electrical equipment state, using the Internet for real-time data transmission, performing data classification, and identifying fire risk indicators, an environmental risk analysis result is generated; Using the environmental risk analysis result, a probability updating model is constructed, the posterior probability under the differential environmental condition is calculated, the risk level of fire is analyzed by using the posterior probability, and a posterior probability updating result is obtained; Through the posterior probability updating result, the fire risk is evaluated through the Internet, the risk weight of environmental factors is calculated, the risk level of the environment is analyzed through the risk weight calculation result, and a risk level determination result is obtained; According to the risk level determination result, the trigger parameters of intelligent fire alarm are adjusted, the trigger threshold is set according to the risk level, the timeliness of alarm triggering is optimized, and an alarm threshold adjustment result is generated; Implementing the alarm threshold adjustment result, real-time monitoring of environmental data and comparison with the preset threshold are performed, real-time risk assessment of fire is carried out, the response state of intelligent fire alarm is updated, and a real-time risk assessment result is generated; Based on the real-time risk assessment result, the Internet is used to perform intelligent fire emergency response simulation test, evaluate the implementation effect of the response strategy, and generate an intelligent fire alarm log; The step of using the environmental risk analysis result to construct a probability updating model, calculating the posterior probability under the differential environmental condition, and using the posterior probability to analyze the risk level of fire to obtain the posterior probability updating result is specifically: Using the environmental risk analysis result, an environmental parameter and historical environmental data comparison model is initialized, data normalization processing is performed, it is checked whether the compared data is on the target scale, the deviation of real-time environmental parameters and historical data is identified, and an environmental change parameter is obtained; Based on the environmental change parameter, a parameter adjustment strategy is used to update the posterior probability model, the model parameters are gradually adjusted to match the new environmental change information, and model verification is performed after parameter updating, the applicability of parameter adjustment is verified, and an adjusted posterior probability is obtained; According to the adjusted posterior probability, the fire risk level of the differential environment is classified and quantitatively analyzed, the analysis result is compared with the preset risk level threshold, and a posterior probability updating result is obtained. 2.The Internet-based intelligent fire alarm method according to claim 1, characterized in that: The environmental risk analysis result includes key temperature indicators, key humidity indicators and fire equipment operation indicators, the posterior probability updating result includes temperature adjustment probability, humidity adjustment probability and environmental safety probability, the risk level determination result includes temperature risk level, humidity risk level and fire risk level, the alarm threshold adjustment result includes temperature trigger threshold, humidity trigger threshold and electrical abnormal trigger threshold, the real-time risk assessment result includes real-time data of temperature, real-time data of humidity and electrical data, and the intelligent fire alarm log includes response time record, simulation test evaluation and strategy effect analysis result. 3.The Internet-based intelligent fire alarm method according to claim 2, characterized in that: The step of analyzing temperature, humidity and electrical equipment state, using the Internet for real-time data transmission, performing data classification, and identifying fire risk indicators to generate an environmental risk analysis result is specifically: By analyzing temperature, humidity and electrical equipment state, uploading data to a central server through the Internet, marking differentiated data as temperature data, humidity data and equipment state data, and performing data formatting processing, real-time environment data is obtained; Based on the real-time environment data, data classification processing is performed, data is filtered, data in normal and abnormal ranges is distinguished, and iteration state marking is performed on data in the abnormal range to identify potential abnormal states, obtaining classification state data; According to the classification state data, fire risk assessment is performed, and the risk value of differentiated data is calculated according to the preset risk assessment threshold, the overall fire risk level of the environment is evaluated, and the environment risk analysis result is obtained. 4.The Internet-based intelligent fire alarm method according to claim 3, characterized in that: Through the posterior probability update result, the fire risk is evaluated through the Internet, the risk weight of environmental factors is calculated, the risk level of the environment is analyzed through the risk weight calculation result, and the risk level judgment result is obtained. The steps are as follows: Based on the posterior probability update result, the current fire risk data is synchronized through the Internet, the integrity of the data is checked, and the reason for the fire risk is analyzed, and the synchronous risk data is obtained; According to the synchronous risk data, the weight distribution of environmental factors is performed, the differentiated weight value is set according to the risk contribution degree of each factor, the fire risk value of environmental factors is calculated, and the weighted risk score is obtained. Through the weighted risk score, the differentiated risk level is assigned to the environmental factors, each risk level is associated with the risk value of the target range, the real-time of fire risk assessment is checked, and the risk level judgment result is obtained. 5.The Internet-based intelligent fire alarm method according to claim 4, characterized in that: According to the risk level judgment result, the trigger parameters of intelligent fire alarm are adjusted, the trigger threshold is dynamically set according to the risk level, the timeliness of alarm triggering is optimized, and the alarm threshold adjustment result is generated. The steps are as follows: Based on the risk level judgment result, the trigger parameters of intelligent fire alarm are adjusted, the trigger threshold is dynamically set according to the risk level, the timeliness of alarm triggering is optimized, and the alarm threshold adjustment result is generated. The steps are as follows: According to the real-time threshold record, the trigger threshold of intelligent fire alarm is adjusted according to the new risk level, whether the threshold matches the real-time risk level is checked, the parameters of intelligent fire alarm are updated, and the adjusted trigger parameters are obtained. Through the adjusted trigger parameters, the configuration file of intelligent fire alarm is updated, the effect of implementing the new threshold evaluation is reloaded, the response time after adjustment is tested, and the alarm threshold adjustment result is obtained. 6.The Internet-based intelligent fire alarm method according to claim 5, characterized in that: Implement the alarm threshold adjustment result, monitor the environment data in real time and compare with the preset threshold, perform real-time risk assessment of fire, update the response state of intelligent fire alarm, and generate real-time risk assessment result. The steps are as follows: According to the alarm threshold adjustment result, real-time monitoring is performed through the Internet, real-time collection of environment temperature, humidity and smoke concentration data in differentiated areas is performed, data is cyclically monitored and recorded, and real-time monitoring data is obtained; Based on the real-time monitoring data, the parameters are compared with the set new trigger threshold to analyze whether the parameters exceed the safe range, identify potential fire risk areas, and perform risk assessment to obtain the early warning risk level. The early warning risk level is used to update an alarm response state, trigger an associated alarm signal if the evaluation result exceeds a target risk level, and record the time and area of the alarm activation to obtain a real-time risk evaluation result. 7.The Internet-based intelligent fire alarm method according to claim 6, characterized in that: Based on the real-time risk evaluation result, the intelligent fire alarm is simulated and tested through the Internet, the fire response under different environmental conditions is simulated through network connection, the activation state of the intelligent fire alarm and response equipment is checked, the reaction time of the equipment under each simulation scenario is verified, and a simulation test record is obtained. Based on the real-time risk evaluation result, the intelligent fire alarm is simulated and tested through the Internet, the fire response under different environmental conditions is simulated through network connection, the activation state of the intelligent fire alarm and response equipment is checked, the reaction time of the equipment under each simulation scenario is verified, and a simulation test record is obtained. Based on the real-time risk evaluation result, the intelligent fire alarm is simulated and tested through the Internet, the fire response under different environmental conditions is simulated through network connection, the activation state of the intelligent fire alarm and response equipment is checked, the reaction time of the equipment under each simulation scenario is verified, and a simulation test record is obtained. The formula of the decision tree algorithm is as follows: 8.The Internet-based intelligent fire alarm method according to claim 7, characterized in that: The Internet-based intelligent fire alarm method according to claim 8, wherein the system comprises: ； wherein, is the weighted Gini impurity, is the proportion of samples in the class under consideration, is the weight coefficient that is critical for the class under consideration, is the total number of classes .

9. An Internet-based intelligent fire alarm system, characterized in that: The data collection module transmits data through the Internet according to temperature, humidity and electrical equipment state, classifies the data as environmental data, analyzes and identifies fire risk indicators, and generates an environmental risk analysis result; The risk analysis module establishes a probability updating model according to the environmental risk analysis result, calculates the posterior probability under different environmental conditions, evaluates the fire risk level by probability, and obtains a posterior probability updating result; The risk identification module calculates the weight of the environmental risk level through the Internet based on the posterior probability updating result, analyzes the risk level of different environmental factors, and obtains a risk level determination result; The threshold optimization module adjusts the trigger parameters of the intelligent fire alarm using the risk level determination result, dynamically sets the trigger threshold according to the different risk levels, optimizes the timeliness of the alarm trigger, and generates an alarm threshold adjustment result; The real-time monitoring module applies the alarm threshold adjustment result to monitor the environmental data in real time and compares it with the preset threshold to evaluate the real-time risk of the fire, update the response state of the intelligent fire alarm, simulate the intelligent fire alarm response, and generate an intelligent fire alarm log; The risk analysis module establishes a probability updating model according to the environmental risk analysis result, calculates the posterior probability under different environmental conditions, evaluates the fire risk level by probability, and obtains a posterior probability updating result. Using the environmental risk analysis result, the environmental parameter and historical environmental data comparison model are initialized, the data to be compared is checked to see if it is on the target scale through data normalization processing, the deviation of the real-time environmental parameter and the historical data is identified, and the environmental change parameter is obtained. ​ Based on the environmental change parameter, a parameter adjustment strategy is adopted to update the posterior probability model, gradually adjust the model parameters to match the new environmental change information, and perform model verification after parameter updating to verify the applicability of parameter adjustment, and obtain the adjusted posterior probability; According to the adjusted posterior probability, the fire risk level of the differentiated environment is classified and quantitatively analyzed, the analysis result is compared with the preset risk level threshold, and the posterior probability updating result is obtained.

Images