Fire prevention monitoring system for garment production

Abstract

The invention discloses a fire prevention and monitoring system for garment production, and relates to the technical field of fire prevention, and the key points of the technical scheme are that the system comprises a detection end used for collecting gas and temperature conditions in a workshop; the control end is used for receiving the data collected by the detection end, processing and analyzing the data and judging the fire risk according to the data; the interaction end is connected with the control end, comprises a display for displaying environmental data in a workshop and facilitating man-machine interaction, and further comprises an alarm unit arranged in the workshop, and can accurately capture subtle changes of smoke, temperature and combustible gas concentration at the initial stage of a fire, monitor fire hazards in a production environment in real time and give an alarm. And guarantee is provided for production safety.

Description

Technical Field

[0001] The present invention relates to the technical field of fire prevention, and more particularly to a fire prevention monitoring system for garment production. Background Art

[0002] In the field of clothing production, ensuring the safety of the production environment is of vital importance, and the fire prevention monitoring system is a key link in ensuring safety. However, the fire detection systems currently on the market still need to be improved in some aspects. Specifically, the existing systems have certain limitations in terms of sensitivity, linkage and intelligence.

[0003] First, in terms of sensitivity, traditional fire detection systems are sometimes unable to identify potential fire hazards in a timely and accurate manner. This may result in the inability to quickly take control measures in the early stages of a fire, thereby increasing the risk of fire spreading.

[0004] Secondly, in terms of linkage, when a fire occurs, the existing system is often difficult to immediately and effectively coordinate with other fire-fighting equipment such as sprinkler systems and smoke exhaust systems. This delay may miss the best time to extinguish the fire, further increasing the losses.

[0005] Furthermore, the lack of intelligence is also a significant problem of the current system. Most existing systems still use relatively simple data processing methods, such as threshold-based alarm triggering mechanisms, but lack more intelligent risk assessment and early warning models.

[0006] Specifically, existing fire detection systems rely on a single detection method, such as monitoring only through smoke or temperature sensors. This method is limited in both detection range and accuracy. For example, at the beginning of a fire when smoke concentration is low, a single smoke detector may not be able to issue a warning in time. Similarly, temperature sensors may also produce false alarms or missed alarms due to natural fluctuations in ambient temperature.

[0007] Therefore, in order to solve the above technical problems, the present invention proposes a fire prevention monitoring system for clothing production. Summary of the invention

[0008] In view of the deficiencies in the prior art, in a first aspect, the present invention provides a fire prevention monitoring system for clothing production.

[0009] To achieve the above-mentioned object, the present invention provides the following technical solutions: a fire prevention monitoring system for clothing production, the detection end is used to collect gas and temperature conditions in the workshop;

[0010] The control end is used to receive the data collected by the detection end, process and analyze the data, and determine the fire risk based on the data;

[0011] The interactive end is connected to the control end, and includes a display for displaying environmental data in the workshop and facilitating human-computer interaction, and also includes an alarm unit set up in the workshop.

[0012] Furthermore, the detection end includes a photoelectric smoke detector, a temperature sensor and a combustible gas detector.

[0013] Furthermore, the control end includes a data receiving unit and a data processing unit, wherein the data receiving unit is respectively connected to the photoelectric smoke detector, the temperature sensor and the combustible gas detector in the detection end.

[0014] Furthermore, a fire risk assessment model is constructed in the data processing unit through a decision tree model, and the fire risk level is determined through the fire risk assessment model.

[0015] Furthermore, the alarm unit includes a sound alarm module and a light source alarm module.

[0016] Furthermore, the decision tree model in the data processing unit can train and verify the fire risk assessment model to improve the prediction accuracy of the fire risk assessment model.

[0017] In a second aspect, the present invention provides a data processing method for fire prevention monitoring, wherein

[0018] The data processing unit includes the following data processing steps:

[0019] A. Preprocess the collected data, including cleaning the data to remove noise and outliers;

[0020] B. Extracting the key features of the fire risk assessment model, where the key features are set in advance through the interactive terminal;

[0021] C. Import the data into the fire risk assessment model to determine the fire risk level.

[0022] Furthermore, the fire risk level includes a low risk level, a medium risk level, and a high risk level, wherein when the fire risk reaches a medium risk level or a high risk level, an alarm unit will be triggered.

[0023] Furthermore, when preprocessing the data in step A, the weighted moving average and Z-score algorithms are used to clean the data to remove noise and outliers.

[0024] Furthermore, the interactive terminal is linked with the fire protection system in the workshop, and when it is detected that the fire risk level is a high risk level, the fire protection system can be controlled to perform fire extinguishing work.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] 1. In the present invention, an efficient and reliable fire detection module is constructed. By comprehensively using photoelectric smoke detectors, temperature sensors and combustible gas detectors, the sensitivity and accuracy of fire detection are greatly improved. This multi-sensor collaborative approach enables the system to accurately capture subtle changes in smoke, temperature and combustible gas concentrations at the early stage of a fire, monitor fire hazards in the production environment in real time, and provide protection for production safety.

[0027] 2. In the present invention, the control end can efficiently and accurately analyze the monitoring data in real time, promptly judge the fire risk, and provide strong support for the safety management of the production environment.

[0028] 3. In the present invention, a fire risk assessment model based on a decision tree is constructed to judge the fire risk in real time and provide a scientific basis for safety management. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0030] Figure 1 It is a schematic diagram of the overall structure of the present invention;

[0031] Figure 2 It is a schematic diagram of the data processing flow in the present invention. DETAILED DESCRIPTION

[0032] like Figure 1-2 As shown, the present invention provides a fire prevention monitoring system for clothing production, including a detection end for collecting gas and temperature conditions in a workshop;

[0033] The detection end consists of three main sensors: photoelectric smoke detectors, temperature sensors and combustible gas detectors.

[0034] The photoelectric smoke detector is the core component of the module and is mainly used to detect the presence of smoke particles in the air. It should be installed on the ceiling of the production workshop to ensure that the detector has an unobstructed view and can effectively capture the generation of smoke. The working principle of this detector is based on the photoelectric effect. It monitors smoke by emitting and receiving infrared beams. When smoke particles appear in the air, the smoke scatters the beam, causing the signal of the receiver to change, thus achieving real-time detection of smoke.

[0035] Temperature sensors are used to monitor ambient temperature and identify abnormally high temperatures. They should be installed around clothing production equipment and inside workshops, especially near high-temperature equipment such as irons and dryers. The working principle of the sensor can use components such as thermocouples or thermistors to collect temperature data of the surrounding environment in real time and transmit this data to the control end for subsequent analysis.

[0036] The function of the combustible gas detector is to detect the presence of combustible gases in the air, such as methane and propane, etc. It is installed in gas storage areas and fuel use areas to prevent fires caused by gas leaks. The working principle of the detector is based on semiconductor sensors or catalytic combustion sensors. When the concentration of combustible gas in the air exceeds the set threshold, it will immediately sound an alarm to remind the staff to take corresponding measures.

[0037] Installation and layout of equipment: Photoelectric smoke detectors should be installed every certain area (usually 100 square meters, depending on the size and layout of the workshop). Especially in high-risk areas such as warehouses and ironing rooms, the number of detectors can be increased to enhance the coverage of monitoring. MQ-2 smoke sensor can be selected. When gas molecules come into contact with the surface of the sensor, the gas will change the conductivity of the semiconductor, thereby affecting the output signal of the sensor. It has high sensitivity (200-1000ppm), fast response (10-20s), and real-time detection of smoke concentration in the air. The working voltage is set to 5V DC. Temperature sensors should be arranged around equipment that is prone to high temperatures, such as irons and dryers. In addition, temperature sensors are also required in the storage area of ​​flammable items in the warehouse. NTC thermistor sensors are selected. Its nominal resistance at 25°C is 10kΩ. When the temperature rises, the carrier concentration in the material increases, thereby reducing the resistance value. The mathematical expression of the relationship between the resistance and temperature of the NTC thermistor is described by the Steinhart-Hart equation. The temperature detection range of NTC thermistor sensor is -40℃~+125℃, and it responds between a few hundred milliseconds and a few seconds. The layout of combustible gas detectors should pay special attention to gas storage areas and equipment use areas to ensure that there are 3 to 5 detectors covering the area. MQ-2 smoke sensor can also detect liquefied petroleum gas (LPG), natural gas, smoke, propane, and hydrogen, which will not be repeated here.

[0038] Steinhart-Hart equation:

[0039]

[0040] Where T is the absolute temperature (unit: Kelvin, K); R is the thermistor value at temperature T (unit: ohm, Ω); A, B, C are coefficients obtained by fitting experimental data using the least squares method

[0041] The interactive end is connected to the control end, including a display for displaying environmental data in the workshop and facilitating human-machine interaction, and also includes an alarm unit set in the workshop;

[0042] The control end is used to receive the data collected by the detection end, process and analyze the data, and determine the fire risk based on the data;

[0043] The control end is the core of the fire monitoring system. It is responsible for receiving real-time data from various sensors (smoke detectors, temperature sensors and combustible gas detectors) and analyzing and judging potential fire risks. The main components of the system include data receiving unit, data processing engine, user interface and alarm system. The data receiving unit is connected to different types of sensors with twisted pair cables and supports RS-485 communication protocol.

[0044] The data processing engine is the core of the system. It uses a decision tree algorithm to analyze monitoring data in real time, integrates information from various sensors, and improves the accuracy of fire risk judgment. The user interface provides an intuitive operating platform, and system administrators can view monitoring data, alarm status, and historical records in real time to enhance data visualization. The control end is linked with other fire protection systems (sprinkler systems and smoke exhaust systems). When a fire risk is detected, the system can automatically start the sprinkler and smoke exhaust devices to control the spread of the fire as much as possible. At the same time, the alarm notifies relevant personnel to take timely measures. The system has an emergency plan management function, and provides corresponding emergency treatment plans and evacuation instructions according to different fire risk levels (low risk, medium risk, and high risk) to ensure the safety of employees.

[0045] The data processing engine needs to pre-process the collected monitoring data, clean the data to remove noise and outliers (weighted moving average and Z-score algorithms); extract key features (temperature change rate, smoke concentration change) to provide necessary information for fire risk assessment; the system will build a fire risk assessment model, and use algorithms such as decision tree for training and verification to improve the model's prediction accuracy; through real-time monitoring, the system can respond quickly to any abnormal situation. Once it is determined that the risk exceeds the set threshold, the alarm system will be immediately activated and relevant data will be recorded for subsequent analysis.

[0046] Decision tree algorithm to build fire risk assessment model

[0047] 1. Decision Tree Algorithm:

[0048] (1) Basic concept: A decision tree consists of nodes, edges, and leaf nodes

[0049] Root node: represents the starting state of the data set, which is divided according to a certain feature set (temperature change rate, smoke concentration change rate, and combustible gas concentration change rate).

[0050] Internal nodes: represent tests on feature sets, and edges between nodes represent different results.

[0051] Leaf node: represents the final decision result (divided into low risk, medium risk and high risk), that is, the output of the decision tree.

[0052] (2) Construction process:

[0053] (a) Select the best features for segmentation:

[0054] Information Gain: Select features that maximize information gain. Information gain refers to the amount of entropy reduction in a sample by selecting a feature set for partitioning. Entropy is a measure of uncertainty.

[0055] Information gain ratio: Considering the diversity of feature sets, information gain ratio is used to select features. Information gain ratio is the ratio of information gain to the inherent information of feature sets.

[0056] Gini index: used to measure the purity of a sample. The lower the Gini index, the purer the sample. The decision tree algorithm can select a feature set by minimizing the Gini index.

[0057] (b) Dividing the data set: Divide the data set into different subsets according to the selected feature set, each subset corresponds to a certain value of the feature set (temperature, smoke concentration and combustible gas concentration change rate).

[0058] (c) Recursively build a decision tree: For each subset, repeat steps (a) and (b) until the stopping condition is met.

[0059] (d) Generate tree structure: combine all nodes into a tree structure to form the final decision tree.

[0060] (3) Stopping conditions: preventing overfitting

[0061] Maximum depth: Limit the maximum depth of the tree and control the complexity of the model. The maximum depth of the model used in this system is set to 5.

[0062] Minimum number of samples: Stop dividing when the number of samples in a node is less than a certain threshold.

[0063] Purity threshold: Stop when the purity of the node reaches a certain standard.

[0064] Fire risk assessment model:

[0065] (1) Data preparation:

[0066] Feature columns: include the extracted temperature change rate, smoke concentration change, and combustible gas concentration change.

[0067] Label column: marks whether a fire has occurred, using multivariate classification (fire level low, medium, high).

[0068] (2) Data partitioning: The data set is divided into a training set and a test set, with 70% used for training and 30% for testing.

[0069] (3) Decision tree model construction:

[0070] Use the machine learning library (Scikit-learn) in Python to build a decision tree model. train_test_split is a function used to divide the dataset into training and test sets, and DecisionTreeClassifier is a class for building a decision tree model; X is a feature dataset containing temperature, smoke concentration, and rate of change of combustible gas concentration, y is a label dataset, multivariate classification (indicating low, medium, and high fire levels), test_size = 0.3 means that 30% of the data will be used for testing and 70% for training. random_state = 42 is a random seed to ensure consistent results each time it is run; a decision tree classifier object clf is created. criterion = 'gini' means using Gini impurity to evaluate the quality of the split, and max_depth = 5 limits the maximum depth of the tree to prevent overfitting. "clf.fit(X_train,y_train);y_pred=clf.predict(X_test)" is to fit the decision tree model using the training set X_train and the corresponding label y_train. In this process, the model will learn how to judge whether there is a risk of fire based on the features; use the test set X_test to make predictions, get the prediction result y_pred, and get an array containing the model's predicted labels for each test sample. Finally, the classification report is generated and printed through the classification_report function. The evaluation indicators are accuracy, precision, recall, and F1-score.

[0071] (4) Model Optimization

[0072] Adjust hyperparameters: Adjust the hyperparameters of the decision tree (maximum depth, minimum number of sample splits) through grid search or random search.

[0073] Ensemble methods: Ensemble learning methods such as random forest and gradient boosting can improve the stability and accuracy of the model.

[0074] Fire risk level (threshold) and emergency response plan:

[0075] Low risk level

[0076] Definition: The probability of fire is low, there are no flammable materials in the area, and the consequences of fire are minor.

[0077] Characteristics: The temperature is lower than the threshold of 57°C (the temperature change rate is lower than 6.4°C / s), the smoke concentration is lower than the threshold of 5%OBS / M (the smoke concentration change rate is lower than 0.5%OBS / Ms), and the combustible gas concentration is lower than the threshold of 25LEL (the combustible gas concentration change rate is lower than 2.5LEL / s).

[0078] Low-risk emergency response plan:

[0079] Daily management: Conduct fire safety inspections regularly to ensure that fire-fighting facilities are in good condition.

[0080] Training and Drills: Provide fire protection knowledge training and escape drills to employees on a regular basis.

[0081] Emergency response: Once a fire is discovered, call the fire alarm immediately and carry out initial fire extinguishing (fire extinguisher, open sprinkler, smoke exhaust device).

[0082] Evacuation and rescue: Take the green safety channel to ensure that people can evacuate to a safe area safely and orderly.

[0083] Medium risk level

[0084] Definition: The probability of fire is moderate and there is a certain amount of flammable materials in the area.

[0085] Characteristics: The temperature is close to or slightly higher than the threshold value of 57°C (the temperature change rate is close to or slightly higher than 6.4°C / s), the smoke concentration is close to or slightly higher than the threshold value of 5%OBS / M (the smoke concentration change rate is close to or slightly higher than 0.5%OBS / Ms), and the combustible gas concentration is close to or slightly higher than the threshold value of 25LEL (the combustible gas concentration change rate is close to or slightly higher than 2.5LEL / s).

[0086] Emergency response plan for medium risk level:

[0087] Strengthen management: Establish a sound fire safety management system and clearly identify those responsible.

[0088] Equipment maintenance: Regularly inspect and maintain fire-fighting facilities to ensure they function properly.

[0089] Training and Drills: Carry out regular fire training and emergency drills to improve employees' ability to deal with fires.

[0090] Emergency Response:

[0091] (a) When a fire is discovered, call the police immediately and carry out initial fire extinguishing (fire extinguisher, sprinkler and smoke exhaust device).

[0092] (b) Activate the evacuation plan, take the green safety channel, and ensure that personnel evacuate quickly according to the predetermined route.

[0093] (c) Isolate the combustible gas storage area and equipment area in the workshop to reduce the intensity of the fire.

[0094] High risk level

[0095] Definition: The probability of fire is high and there are a large number of flammable and explosive materials in the area.

[0096] Characteristics: The temperature is much higher than the threshold of 57°C (the temperature change rate is much higher than 6.4°C / s), the smoke concentration is much higher than the threshold of 5%OBS / M (the smoke concentration change rate is much higher than 0.5%OBS / Ms), and the combustible gas concentration is much higher than the threshold of 25LEL (the combustible gas concentration change rate is much higher than 2.5LEL / s).

[0097] Emergency response plan for medium risk level:

[0098] Comprehensive management: Establish and improve the fire safety management system and equip professional fire management personnel.

[0099] Risk Assessment: Conduct fire risk assessments regularly to identify potential hazards and develop rectification plans.

[0100] Equipment maintenance: Regularly inspect and maintain fire-fighting facilities to ensure they function properly.

[0101] Training and Drills: Conduct professional firefighting training and emergency drills, including equipment use, evacuation routes, and emergency handling procedures.

[0102] Emergency Response:

[0103] (a) When a fire is discovered, call the fire department immediately and turn on the sprinkler and smoke exhaust systems.

[0104] (b) Organize professional rescue teams to evacuate personnel and put out fires.

[0105] (c) Isolate the combustible gas storage area and equipment area in the workshop to reduce the intensity of the fire.

[0106] (d) After the fire is extinguished, conduct a detailed fire cause investigation and damage assessment to improve future fire prevention efforts.

[0107] In summary, the present invention proposes a fully optimized fire prevention monitoring system. The system realizes full coverage monitoring and intelligent response of the target area by deeply integrating multiple functional modules such as fire detection, data acquisition, central control, alarm and emergency response. The present invention uses a variety of sensors (including photoelectric smoke detectors, temperature sensors and combustible gas detectors) to significantly improve the sensitivity and accuracy of fire detection. At the same time, the collected data is analyzed in real time with the help of the central control system, and a risk assessment model is constructed using an advanced decision tree algorithm, thereby realizing accurate prediction of fire risks. In addition, the system can also effectively link with various fire-fighting equipment to ensure that emergency response measures can be quickly and accurately initiated when a fire occurs, thereby minimizing the losses caused by the fire.

[0108] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any ordinary technician in the industry can smoothly implement the present invention as shown in the drawings and described above. However, any equivalent changes, modifications and evolutions made by technicians familiar with the profession without departing from the scope of the technical solution of the present invention using the technical content disclosed above are all equivalent embodiments of the present invention. At the same time, any equivalent changes, modifications and evolutions made to the above embodiments based on the essential technology of the present invention are still within the protection scope of the technical solution of the present invention.

Claims

1. A fire prevention monitoring system for clothing production, characterized in that: include: The detection end is used to collect gas and temperature conditions in the workshop; The control end is used to receive the data collected by the detection end, process and analyze the data, and determine the fire risk based on the data; The interactive end is connected to the control end, and includes a display for displaying environmental data in the workshop and facilitating human-computer interaction, and also includes an alarm unit set up in the workshop.

2. A fire prevention monitoring system for clothing production according to claim 1, characterized in that: The detection end includes a photoelectric smoke detector, a temperature sensor and a combustible gas detector.

3. A fire prevention monitoring system for clothing production according to claim 2, characterized in that: The control end comprises a data receiving unit and a data processing unit, wherein the data receiving unit is respectively connected with the photoelectric smoke detector, the temperature sensor and the combustible gas detector in the detection end.

4. A fire prevention monitoring system for clothing production according to claim 3, characterized in that: A fire risk assessment model is constructed in the data processing unit through a decision tree model, and the fire risk level is determined through the fire risk assessment model.

5. A fire prevention monitoring system for clothing production according to claim 4, characterized in that: The alarm unit comprises a sound alarm module and a light source alarm module.

6. A fire prevention monitoring system for clothing production according to claim 5, characterized in that: The decision tree model in the data processing unit trains and verifies the fire risk assessment model to improve the prediction accuracy of the fire risk assessment model.

7. A data processing method for fire prevention monitoring, applied to a fire prevention monitoring system for clothing production as claimed in any one of claims 1 to 6, characterized in that: The data processing unit includes the following data processing steps: A. Preprocess the collected data, including cleaning the data to remove noise and outliers; B. Extracting the key features of the fire risk assessment model, where the key features are set in advance through the interactive terminal; C. Import the data into the fire risk assessment model to determine the fire risk level.

8. A data processing method for fire prevention monitoring according to claim 7, characterized in that: The fire risk level includes a low risk level, a medium risk level, and a high risk level, wherein when the fire risk reaches a medium risk level or a high risk level, an alarm unit will be triggered.

9. A data processing method for fire prevention monitoring according to claim 8, characterized in that: When preprocessing the data in step A, the weighted moving average and Z-score algorithms are used to clean the data to remove noise and outliers.

10. A data processing method for fire prevention monitoring according to claim 9, characterized in that: The interactive terminal is linked with the fire protection system in the workshop, and when it is detected that the fire risk level is a high risk level, the fire protection system is controlled to perform fire extinguishing work.

Images