Electrical fire real-time early warning system based on Internet of Things

Abstract

The invention relates to the technical field of electrical fire early warning, in particular to an electrical fire real-time early warning system based on the Internet of Things, which comprises an environment monitoring module, an equipment state monitoring module, a fire risk assessment module, a fire source positioning module and an early warning module, wherein the environment monitoring module is used for collecting environment data around electrical equipment in real time; the equipment state monitoring module is used for monitoring operation data of the electrical equipment in real time; the fire risk assessment module assesses the fire risk and generates a fire risk level; the fire source positioning module is used for positioning the specific position of a potential fire source; and the early warning module is used for generating an early warning signal. According to the invention, through multi-dimensional data acquisition and deep learning analysis, the fire risk is accurately evaluated, the fire source is positioned in real time, and in combination with a differentiated early warning signal generation mechanism, the early warning accuracy and response efficiency of the electrical fire are effectively improved, and the fire prevention and control capability is significantly enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical fire early warning, and in particular to a real-time electrical fire early warning system based on the Internet of Things. Background Art

[0002] With the widespread use of modern electrical equipment, electrical fires have become a common and fatal safety hazard in industrial, commercial and residential areas. The occurrence of electrical fires is usually closely related to factors such as equipment overload, short circuit, insulation aging, and environmental conditions. Traditional fire detection systems mainly rely on the monitoring of environmental parameters such as smoke and temperature to achieve fire alarms. However, such systems can only trigger alarms when the fire spreads or produces obvious smoke or temperature rises. They lack effective early warning for the early stages of a fire and cannot detect potential fire sources in a timely manner. In addition, existing systems are usually unable to monitor and comprehensively evaluate multiple factors in real time, such as the status of electrical equipment, changes in environmental conditions, etc., resulting in low accuracy in fire prediction and failure to effectively avoid the occurrence of electrical fires.

[0003] In order to improve the early warning capability of electrical fires, Internet of Things technology has gradually been applied to fire monitoring systems in recent years. By collecting real-time data on the surrounding environment and equipment status of electrical equipment, combined with big data analysis and deep learning algorithms, accurate prediction of fire risks can be achieved. However, the existing Internet of Things fire early warning system still has certain technical bottlenecks. For example, how to accurately assess fire risks in complex electrical environments, how to locate potential fire sources in real time, and how to generate efficient early warning signals based on multi-dimensional data, these problems have not yet been effectively solved. Summary of the invention

[0004] Based on the above objectives, the present invention provides a real-time early warning system for electrical fires based on the Internet of Things.

[0005] The real-time early warning system for electrical fires based on the Internet of Things includes an environmental monitoring module, an equipment status monitoring module, a fire risk assessment module, a fire source location module, and an early warning module; among which: Environmental monitoring module: used to collect environmental data around electrical equipment in real time, including temperature, humidity and oxygen content, and transmit the collected environmental data to the fire risk assessment module through the Internet of Things; Equipment status monitoring module: used to monitor the operating data of electrical equipment in real time, including current, voltage and surface temperature, and transmit the monitored operating data to the fire risk assessment module; Fire risk assessment module: used to receive environmental data from the environmental monitoring module and the environmental data from the equipment status monitoring module, use the deep belief network algorithm to assess the fire risk, generate the fire risk level, and transmit the assessment results to the fire source location module and the early warning module; Fire source location module: Based on the assessment results of the fire risk assessment module and combined with the surface temperature and current data provided by the equipment status monitoring module, the specific location of the potential fire source is located and the location information is transmitted to the early warning module; Early warning module: used to generate early warning signals according to the fire risk level and fire source location information, and send the early warning information to the user terminal and emergency response center through the Internet of Things.

[0006] Furthermore, the environment monitoring module includes a sensor unit, a data processing unit and a communication unit; wherein: Sensor unit: including a temperature sensor, a humidity sensor and an oxygen concentration sensor, which are used to collect the temperature, humidity and oxygen content around the electrical equipment respectively, and convert the collected data into electrical signals; Data processing unit: used to filter, correct and convert the electrical signals collected by the sensor unit to form standardized digital environmental data; Communication unit: includes a wireless communication submodule or a wired communication submodule; the wireless communication submodule adopts wireless communication technology, including Wi-Fi, Bluetooth and Zigbee, to transmit digital environmental data to the fire risk assessment module through a wireless network; the wired communication submodule adopts wired communication technology, including Ethernet and industrial bus protocol, to transmit data to the fire risk assessment module through a local area network.

[0007] Furthermore, the device status monitoring module includes a current monitoring unit, a voltage monitoring unit and a temperature monitoring unit, wherein: Current monitoring unit: including a current sensor, used to monitor the current data of the electrical equipment in real time and convert the current signal into an electrical signal; the current sensor is a Hall effect sensor or a current transformer; Voltage monitoring unit: including a voltage sensor, used to monitor the voltage data of the electrical equipment in real time, detect the input voltage or output voltage of the electrical equipment through the voltage sensor, and convert the voltage signal into an electrical signal; the voltage sensor is a voltage transmitter or a voltage drop sensor; Temperature monitoring unit: used to collect the surface temperature of the equipment through a temperature sensor and convert the temperature signal into an electrical signal. The temperature sensor is a thermocouple or a thermistor.

[0008] Furthermore, the fire risk assessment module includes a data receiving unit, a data processing unit, a risk assessment unit and a risk level generating unit, wherein: Data receiving unit: used to receive environmental data from the environmental monitoring module and operating data from the equipment status monitoring module, and transmit the received data to the data processing unit; Data processing unit: used to pre-process the environmental data and operation data obtained by the data receiving unit, including data cleaning, denoising and standardization, to form a standardized data set; Risk Assessment Unit: Uses deep belief network algorithms to analyze and model standardized data sets, and predicts electrical fire risks based on historical and real-time data; Risk level generation unit: Based on the fire risk prediction results output by the risk assessment unit, the fire risk level is generated according to the preset risk assessment standards, including low risk, medium risk and high risk, and the assessment results are transmitted to the fire source location module and the early warning module.

[0009] Furthermore, the data processing unit includes: Data cleaning: used to remove missing values, outliers and duplicate data. Specifically, missing values ​​are filled by the mean interpolation method, and outliers are identified and removed using the Z-score method. For duplicate data, a hash value-based deduplication algorithm is used to remove redundant data. Denoising: remove noise from the signal through the mean filtering algorithm to enhance the stability of the data; Standardization: The data was processed using the Z-score standardization method, with the mean of all features adjusted to 0 and the standard deviation adjusted to 1 to eliminate the impact of different feature scales and ensure that the data was analyzed within a unified standard range.

[0010] Furthermore, the risk assessment unit includes: Data input layer construction: First, the standardized data set is divided into training data and test data; the training data includes historical data from the environmental monitoring module and the equipment status monitoring module, and the test data is the environmental data and operation data collected in real time; Pre-training of hidden layers and visible layers: by performing unsupervised learning on the training data, specifically using a greedy layer-by-layer training strategy to pre-train each hidden layer of the deep belief network; Supervised learning and error back propagation: After pre-training the hidden layers, supervised learning is used to fine-tune the deep belief network, and the error back propagation algorithm is used to adjust the weights and biases; Fire risk prediction: The trained deep belief network will analyze the test data through forward propagation to calculate the predicted fire risk value.

[0011] Furthermore, the risk level generating unit includes: Receive prediction results: Receive the fire risk prediction results from the risk assessment unit and set them as , the result is a numerical value in the range of 0 to 1; Fire risk level calculation: by comparing the predicted results Classify the fire risk level according to the preset risk assessment standards; When , it means the fire risk is low and the level is low risk; when , indicating that the fire risk is at a medium level, the level is medium risk; when , indicating that the fire risk is high and the level is high risk.

[0012] Furthermore, the fire source positioning module includes a data fusion unit, a positioning algorithm unit and a position output unit, wherein: Data fusion unit: used to receive the assessment results from the fire risk assessment module and the surface temperature and current data provided by the equipment status monitoring module, and fuse these data to form fused data; Positioning algorithm unit: Based on the fusion data provided by the data fusion unit, it uses multi-sensor data fusion and positioning algorithm to determine the specific location of the potential fire source; Position output unit: formats the fire source location information calculated by the positioning algorithm unit, and transmits the location information to the early warning module through the Internet of Things.

[0013] Furthermore, the positioning algorithm unit specifically includes: Abnormal signal calculation: According to the surface temperature, current data and fire risk prediction results in the fused data, the abnormal signal strength of each monitoring point is calculated. The calculation is performed by weighted average. The formula is: ,in, Indicates abnormal signal strength; and are weight coefficients, which respectively represent the contribution of surface temperature, current and fire risk prediction results to the abnormal signal, and ; is the surface temperature; is the current; The fire risk prediction results; Sensor position calibration: Receive the coordinates of each sensor position and the corresponding abnormal signal strength , calculate the contribution of each sensor based on its spatial position and signal strength ; Multi-sensor data fusion: The contribution from different sensors is fused through weighted averaging to ultimately determine the coordinates of potential fire sources , the formula is: and ,in, is the final location coordinate of the fire source, For the contribution of each sensor, and For the The position coordinates of the sensors, is the total number of sensors.

[0014] Furthermore, the warning module includes a warning signal generating unit and a warning information transmitting unit, wherein: Warning signal generating unit: used to generate a warning signal according to the fire risk level provided by the fire risk assessment module and the fire source location information provided by the fire source location module in accordance with the preset fire warning rules; Rule 1: When the fire risk level is high and the fire source is less than 5 meters away from the electrical equipment, an emergency warning signal is generated to prompt immediate emergency measures; Rule 2: When the fire risk level is medium risk and the distance between the fire source and the electrical equipment is between 5 meters and 20 meters, a medium priority warning signal is generated and enhanced monitoring is recommended; Rule 3: When the fire risk level is low and the fire source is more than 20 meters away from the electrical equipment, a general warning signal is generated to alert users; Warning information transmission unit: used to transmit the generated warning signal to the user terminal and emergency response center through the Internet of Things.

[0015] Beneficial effects of the present invention: The present invention can realize accurate early warning of electrical fires through multi-dimensional data collection and comprehensive analysis; the system can perform multi-level and accurate assessment of fire risks by real-time collection of environmental data and equipment status data around electrical equipment, and combining with a deep belief network algorithm; compared with the existing technology, it can not only predict fire risks in advance, but also accurately determine the location of potential fire sources through a fire source locating module, thus avoiding the problem of delayed response of traditional fire detection systems.

[0016] The present invention, through the early warning signal generating unit, can intelligently generate early warning signals of different levels in combination with the fire risk level and the fire source location information, and accurately determine whether an emergency response is required based on a preset distance threshold; this differentiated early warning method based on risk assessment results can dynamically adjust the early warning strategy according to actual conditions, effectively avoiding the problems of excessive alarms or response delays in traditional systems; and through the Internet of Things technology, the early warning signal can be transmitted to the user terminal and the emergency response center in real time, which can achieve a fast and accurate response, and significantly improve the safety and management efficiency of fire prevention. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A schematic diagram of a real-time early warning system for electrical fires according to an embodiment of the present invention; Figure 2 Schematic diagram of a fire risk assessment module according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0020] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0021] like Figure 1-Figure 2 As shown, the real-time early warning system for electrical fires based on the Internet of Things includes an environmental monitoring module, an equipment status monitoring module, a fire risk assessment module, a fire source location module, and an early warning module; among which: Environmental monitoring module: used to collect environmental data around electrical equipment in real time, including temperature, humidity and oxygen content, and transmit the collected environmental data to the fire risk assessment module through the Internet of Things; Equipment status monitoring module: used to monitor the operating data of electrical equipment in real time, including current, voltage and surface temperature, and transmit the monitored operating data to the fire risk assessment module; Fire risk assessment module: used to receive environmental data from the environmental monitoring module and the environmental data from the equipment status monitoring module, use the deep belief network algorithm to assess the fire risk, generate the fire risk level, and transmit the assessment results to the fire source location module and the early warning module; Fire source location module: Based on the assessment results of the fire risk assessment module and combined with the surface temperature and current data provided by the equipment status monitoring module, the specific location of the potential fire source is located and the location information is transmitted to the early warning module; Early warning module: used to generate early warning signals according to the fire risk level and fire source location information, and send the early warning information to the user terminal and emergency response center through the Internet of Things.

[0022] The environment monitoring module includes a sensor unit, a data processing unit and a communication unit; wherein: Sensor unit: including a temperature sensor, a humidity sensor and an oxygen concentration sensor, which are used to collect the temperature, humidity and oxygen content around the electrical equipment respectively, and convert the collected data into electrical signals; Data processing unit: used to filter, correct and convert the electrical signals collected by the sensor unit to form standardized digital environmental data; Communication unit: including wireless communication submodule or wired communication submodule; the wireless communication submodule adopts wireless communication technology, including Wi-Fi, Bluetooth and Zigbee, to transmit digital environmental data to the fire risk assessment module through the wireless network, so as to realize long-distance and seamless transmission; the wired communication submodule adopts wired communication technology, including Ethernet and industrial bus protocol, to transmit data to the fire risk assessment module through the local area network, so as to ensure stable and fast data transmission in the local environment; the above-mentioned units are connected by wired or wireless means and work together to realize real-time collection, processing and transmission of environmental data around electrical equipment; the environmental data collected by the environmental monitoring module is transmitted to the fire risk assessment module through the communication unit after processing, so as to provide environmental data support for subsequent fire risk assessment.

[0023] The equipment status monitoring module includes a current monitoring unit, a voltage monitoring unit and a temperature monitoring unit, wherein: Current monitoring unit: including a current sensor, which is used to monitor the current data of the electrical equipment in real time and convert the current signal into an electrical signal; the current sensor is a Hall effect sensor or a current transformer, which can collect the current data of the electrical equipment with high precision; Voltage monitoring unit: including voltage sensor, used to monitor the voltage data of electrical equipment in real time, detect the input voltage or output voltage of electrical equipment through the voltage sensor, and convert the voltage signal into an electrical signal; the voltage sensor is a voltage transmitter or a voltage drop sensor, which can collect voltage changes in the operation of the equipment in real time; Temperature monitoring unit: used to collect the surface temperature of the equipment through the temperature sensor and convert the temperature signal into an electrical signal. The temperature sensor is a thermocouple or thermistor, which can accurately reflect the working status of the equipment; the above units are connected through wired or wireless communication, and the collected current, voltage and surface temperature data are transmitted to the fire risk assessment module in real time, so as to comprehensively monitor and evaluate the operating status of the equipment.

[0024] The fire risk assessment module includes a data receiving unit, a data processing unit, a risk assessment unit and a risk level generating unit, wherein: Data receiving unit: used to receive environmental data from the environmental monitoring module and operating data from the equipment status monitoring module, including temperature, humidity, oxygen content, current, voltage and surface temperature, and transmit the received data to the data processing unit; Data processing unit: used to pre-process the environmental data and operation data obtained by the data receiving unit, including data cleaning, denoising and standardization, to form a standardized data set, providing a reliable data basis for subsequent analysis and modeling; Risk Assessment Unit: The Deep Belief Network (DBN) algorithm is used to analyze and model standardized data sets, and to predict electrical fire risks based on historical and real-time data. The DBN algorithm integrates multi-source heterogeneous data and adds an adaptive weight adjustment mechanism, which can more effectively capture the nonlinear relationship between data and improve the accuracy and real-time performance of fire risk prediction. Risk level generation unit: Based on the fire risk prediction results output by the risk assessment unit, the fire risk level is generated according to the preset risk assessment standards, including low risk, medium risk and high risk, and the assessment results are transmitted to the fire source location module and the early warning module; through the above scheme, the fire risk assessment module can efficiently process data from different sources and perform analysis and modeling through the deep belief network algorithm. The algorithm can better adapt to the characteristics of multi-source data, capture complex nonlinear relationships, improve the accuracy of fire risk assessment, and thus enhance the early warning capability of electrical fires.

[0025] The data processing unit includes: Data cleaning: used to remove missing values, outliers, and duplicate data. Specifically, missing values ​​are filled by the mean interpolation method, and outliers are identified and removed using the Z-score method. For duplicate data, a hash value-based deduplication algorithm is used to remove redundant data. For example, the mean interpolation method presets the missing data value to , then the missing values ​​are filled by the mean value of the field, the formula is: ,in, is the value after filling, is the existing data of this field. is the number of samples; Denoising: The mean filter algorithm is used to remove noise from the signal and enhance the stability of the data. Specifically, a sliding window is used to process the data, and the filtered signal is the smoothed value of the original signal. The mean filter formula is: ,in, is the window size, is the original signal, is the signal after filtering; Standardization: The data is processed by the Z-score standardization method, and the mean of all features is adjusted to 0 and the standard deviation is adjusted to 1 to eliminate the influence of different feature scales and ensure that the data is analyzed within a unified standard range; the standardization formula is: ,in, is the original data, is the mean value of this feature, is the standard deviation of the feature, The data is standardized; through the above steps, the data processing unit can effectively clean, denoise and standardize the raw data from the environmental monitoring module and the equipment status monitoring module to form a standardized data set, providing accurate and stable input data for subsequent risk assessment.

[0026] The risk assessment unit includes: Data input layer construction: First, the standardized data set is divided into training data and test data; the training data includes historical data from the environmental monitoring module and the equipment status monitoring module, and the test data is the real-time collected environmental data and operation data; and each piece of data is input into the input layer of the deep belief network ,in is the data feature vector after normalization, is the dimension of data features; Pre-training of hidden and visible layers: By performing unsupervised learning on the training data, the hidden layers of the deep belief network are pre-trained using a greedy layer-by-layer training strategy. First, the weight matrix of each layer is and the bias vector Initialize and train using the contrast divergence algorithm; the activation output of each layer Calculated by the following formula: ,in, For the The activation output of the layer, for Activation function, is the output of the previous layer, For the The weight matrix of the layer, For the The bias vector of the layer; Supervised learning and error back propagation: After pre-training the hidden layer, supervised learning is used to fine-tune the deep belief network, and the error back propagation algorithm is used to adjust the weights and biases; specifically, by calculating the error of the output layer , and adjust the parameters of each layer through back propagation according to the error. The formula is: ,in, is the actual output value, To predict the output value, is the error term; the formula for updating the weight matrix and bias vector through the back propagation algorithm is: and ,in, is the learning rate, and Represent the gradients of weights and biases respectively; Fire risk prediction: The trained deep belief network will analyze the test data through forward propagation and calculate the predicted fire risk value; specifically, the prediction result of electrical fire is generated by calculating the activation output of each layer and passing it to the final output layer; the predicted output value represents the probability or risk value of fire in electrical equipment, and the calculation formula of the output value is: ,in, is the total number of layers in the deep belief network, is the output of the last hidden layer, and are the weight matrix and bias vector of the output layer respectively, is the predicted fire risk value; in the above steps, by adopting the deep belief network algorithm, the risk assessment unit can efficiently extract potential fire risk factors from the multidimensional information in the standardized data set and accurately predict the fire risk of electrical equipment; this method improves the accuracy and real-time performance of the prediction by optimizing the network parameters through layer-by-layer training and back propagation, and significantly enhances the system's fire prediction ability in different environments and equipment states.

[0027] The risk level generation unit includes: Receive prediction results: Receive the fire risk prediction results from the risk assessment unit and set them as , the result It is a numerical value ranging from 0 to 1, which is used to indicate the intensity of the current electrical fire risk. The higher the value, the greater the fire risk. Fire risk level calculation: by comparing the predicted results Classify the fire risk level according to the preset risk assessment standards; When , it means the fire risk is low and the level is low risk; when , indicating that the fire risk is at a medium level, the level is medium risk; when , indicating that the fire risk is high and the level is high risk; through the above scheme, the risk level generation unit can accurately classify the fire risk level according to the fire risk prediction results and the preset evaluation criteria, provide clear fire risk information, and ensure the timely response and effective management of the electrical fire warning system.

[0028] The fire source positioning module includes a data fusion unit, a positioning algorithm unit and a position output unit, wherein: Data fusion unit: used to receive the assessment results from the fire risk assessment module and the surface temperature and current data provided by the equipment status monitoring module, fuse these data to form fused data, and provide basic information for fire source positioning; Positioning algorithm unit: Based on the fusion data provided by the data fusion unit, the unit uses multi-sensor data fusion and positioning algorithms to determine the specific location of the potential fire source. By analyzing the abnormality degree and spatial location information of each monitoring point, the unit calculates the location coordinates of the fire source and realizes accurate positioning of the fire source. Position output unit: formats the fire source location information calculated by the positioning algorithm unit, and transmits the location information to the early warning module through the Internet of Things, so that the early warning module can generate corresponding early warning signals; the above units are connected through an internal data bus or wireless communication, and work together to achieve accurate positioning of potential fire sources; the data fusion unit ensures the effective integration of multi-source data, the positioning algorithm unit calculates the location of the fire source through comprehensive analysis, and the position output unit ensures the timely transmission of positioning information, providing reliable position information support for the early warning module.

[0029] The positioning algorithm unit specifically includes: Abnormal signal calculation: According to the surface temperature, current data and fire risk prediction results in the fused data, the abnormal signal strength of each monitoring point is calculated. The calculation is performed by weighted average. The formula is: ,in, Indicates abnormal signal strength; and are weight coefficients, which respectively represent the contribution of surface temperature, current and fire risk prediction results to the abnormal signal, and ; is the surface temperature; is the current; The fire risk prediction results; Sensor position calibration: Receive the coordinates of each sensor position and the corresponding abnormal signal strength , calculate the contribution of each sensor based on its spatial position and signal strength , the formula is: ,in, For sensor Contribution to the fire source location, is the abnormal signal strength of the sensor, is the distance from the sensor to the fire source; Multi-sensor data fusion: The contribution from different sensors is fused through weighted averaging to ultimately determine the coordinates of potential fire sources , the formula is: and ,in, is the final location coordinate of the fire source, For the contribution of each sensor, and For the The position coordinates of the sensors, is the total number of sensors; through the above scheme, the positioning algorithm unit can accurately locate potential fire sources based on multi-sensor data fusion technology and combined with the data provided by the equipment status monitoring module and the fire risk assessment module, thereby providing more accurate support for fire warning and response.

[0030] The early warning module includes an early warning signal generation unit and an early warning information transmission unit, wherein: Warning signal generating unit: used to generate a warning signal according to the fire risk level provided by the fire risk assessment module and the fire source location information provided by the fire source location module in accordance with the preset fire warning rules; Rule 1: When the fire risk level is high and the fire source is less than 5 meters away from the electrical equipment, an emergency warning signal is generated to prompt immediate emergency measures; Rule 2: When the fire risk level is medium risk and the distance between the fire source and the electrical equipment is between 5 meters and 20 meters, a medium priority warning signal is generated and enhanced monitoring is recommended; Rule 3: When the fire risk level is low and the fire source is more than 20 meters away from the electrical equipment, a general warning signal is generated to alert users; Early warning information transmission unit: used to transmit the generated early warning signal to the user terminal and emergency response center through the Internet of Things. The early warning information transmission unit adopts wireless communication protocol or wired communication protocol to ensure the rapid transmission of early warning information. Through the above scheme, the early warning module can accurately generate early warning signals of different levels based on the dynamic changes of fire risk level and fire source location, and transmit them to relevant personnel in time, so as to improve the response efficiency and accuracy of the system and effectively prevent the occurrence of fire accidents.

[0031] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present invention is limited to these examples. Under the concept of the present invention, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the present invention as described above, which are not provided in detail for the sake of simplicity.

[0032] The present invention is intended to cover all such substitutions, modifications and variations that fall within the broad scope of the claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. The real-time early warning system for electrical fires based on the Internet of Things is characterized by: It includes environment monitoring module, equipment status monitoring module, fire risk assessment module, fire source location module and early warning module; among which: Environmental monitoring module: used to collect environmental data around electrical equipment in real time, including temperature, humidity and oxygen content, and transmit the collected environmental data to the fire risk assessment module through the Internet of Things; Equipment status monitoring module: used to monitor the operating data of electrical equipment in real time, including current, voltage and surface temperature, and transmit the monitored operating data to the fire risk assessment module; Fire risk assessment module: used to receive environmental data from the environmental monitoring module and the environmental data from the equipment status monitoring module, use the deep belief network algorithm to assess the fire risk, generate the fire risk level, and transmit the assessment results to the fire source location module and the early warning module; Fire source location module: Based on the assessment results of the fire risk assessment module and combined with the surface temperature and current data provided by the equipment status monitoring module, the specific location of the potential fire source is located and the location information is transmitted to the early warning module; Early warning module: used to generate early warning signals according to the fire risk level and fire source location information, and send the early warning information to the user terminal and emergency response center through the Internet of Things.

2. The real-time electrical fire warning system based on the Internet of Things according to claim 1 is characterized in that: The environment monitoring module includes a sensor unit, a data processing unit and a communication unit; wherein: Sensor unit: including a temperature sensor, a humidity sensor and an oxygen concentration sensor, which are used to collect the temperature, humidity and oxygen content around the electrical equipment respectively, and convert the collected data into electrical signals; Data processing unit: used to filter, correct and convert the electrical signals collected by the sensor unit to form standardized digital environmental data; Communication unit: includes a wireless communication submodule or a wired communication submodule; the wireless communication submodule adopts wireless communication technology, including Wi-Fi, Bluetooth and Zigbee, to transmit digital environmental data to the fire risk assessment module through a wireless network; the wired communication submodule adopts wired communication technology, including Ethernet and industrial bus protocol, to transmit data to the fire risk assessment module through a local area network.

3. The real-time electrical fire warning system based on the Internet of Things according to claim 1 is characterized in that: The equipment status monitoring module includes a current monitoring unit, a voltage monitoring unit and a temperature monitoring unit, wherein: Current monitoring unit: including a current sensor, used to monitor the current data of the electrical equipment in real time and convert the current signal into an electrical signal; the current sensor is a Hall effect sensor or a current transformer; Voltage monitoring unit: including a voltage sensor, used to monitor the voltage data of the electrical equipment in real time, detect the input voltage or output voltage of the electrical equipment through the voltage sensor, and convert the voltage signal into an electrical signal; the voltage sensor is a voltage transmitter or a voltage drop sensor; Temperature monitoring unit: used to collect the surface temperature of the equipment through a temperature sensor and convert the temperature signal into an electrical signal. The temperature sensor is a thermocouple or a thermistor.

4. The real-time electrical fire warning system based on the Internet of Things according to claim 1 is characterized in that: The fire risk assessment module includes a data receiving unit, a data processing unit, a risk assessment unit and a risk level generating unit, wherein: Data receiving unit: used to receive environmental data from the environmental monitoring module and operating data from the equipment status monitoring module, and transmit the received data to the data processing unit; Data processing unit: used to pre-process the environmental data and operation data obtained by the data receiving unit, including data cleaning, denoising and standardization, to form a standardized data set; Risk Assessment Unit: Uses deep belief network algorithms to analyze and model standardized data sets, and predicts electrical fire risks based on historical and real-time data; Risk level generation unit: Based on the fire risk prediction results output by the risk assessment unit, the fire risk level is generated according to the preset risk assessment standards, including low risk, medium risk and high risk, and the assessment results are transmitted to the fire source location module and the early warning module.

5. The real-time electrical fire warning system based on the Internet of Things according to claim 4 is characterized in that: The data processing unit comprises: Data cleaning: used to remove missing values, outliers and duplicate data. Specifically, missing values ​​are filled by the mean interpolation method, and outliers are identified and removed using the Z-score method. For duplicate data, a hash value-based deduplication algorithm is used to remove redundant data. Denoising: remove noise from the signal through the mean filtering algorithm to enhance the stability of the data; Standardization: The data was processed using the Z-score standardization method, with the mean of all features adjusted to 0 and the standard deviation adjusted to 1 to eliminate the impact of different feature scales and ensure that the data was analyzed within a unified standard range.

6. The real-time electrical fire warning system based on the Internet of Things according to claim 5 is characterized in that: The risk assessment unit comprises: Data input layer construction: First, the standardized data set is divided into training data and test data; the training data includes historical data from the environmental monitoring module and the equipment status monitoring module, and the test data is the environmental data and operation data collected in real time; Pre-training of hidden layers and visible layers: by performing unsupervised learning on the training data, specifically using a greedy layer-by-layer training strategy to pre-train each hidden layer of the deep belief network; Supervised learning and error back propagation: After pre-training the hidden layers, supervised learning is used to fine-tune the deep belief network, and the error back propagation algorithm is used to adjust the weights and biases; Fire risk prediction: The trained deep belief network will analyze the test data through forward propagation to calculate the predicted fire risk value.

7. The real-time electrical fire warning system based on the Internet of Things according to claim 6 is characterized in that: The risk level generating unit comprises: Receive prediction results: Receive the fire risk prediction results from the risk assessment unit and set them as , the result is a numerical value in the range of 0 to 1; Fire risk level calculation: by comparing the predicted results Classify the fire risk level according to the preset risk assessment standards; When , it means the fire risk is low and the level is low risk; when , indicating that the fire risk is at a medium level, the level is medium risk; when , indicating that the fire risk is high and the level is high risk.

8. The real-time electrical fire warning system based on the Internet of Things according to claim 1 is characterized in that: The fire source positioning module includes a data fusion unit, a positioning algorithm unit and a position output unit, wherein: Data fusion unit: used to receive the assessment results from the fire risk assessment module and the surface temperature and current data provided by the equipment status monitoring module, and fuse these data to form fused data; Positioning algorithm unit: Based on the fusion data provided by the data fusion unit, it uses multi-sensor data fusion and positioning algorithm to determine the specific location of the potential fire source; Position output unit: formats the fire source location information calculated by the positioning algorithm unit, and transmits the location information to the early warning module through the Internet of Things.

9. The real-time electrical fire warning system based on the Internet of Things according to claim 8 is characterized in that: The positioning algorithm unit specifically includes: Abnormal signal calculation: According to the surface temperature, current data and fire risk prediction results in the fused data, the abnormal signal strength of each monitoring point is calculated. The calculation is performed by weighted average. The formula is: ,in, Indicates abnormal signal strength; and are weight coefficients, which respectively represent the contribution of surface temperature, current and fire risk prediction results to the abnormal signal, and ; is the surface temperature; is the current; The fire risk prediction results; Sensor position calibration: Receive the coordinates of each sensor position and the corresponding abnormal signal strength , calculate the contribution of each sensor based on its spatial position and signal strength ; Multi-sensor data fusion: The contribution from different sensors is fused through weighted averaging to ultimately determine the coordinates of potential fire sources , the formula is: and ,in, is the final location coordinate of the fire source, For the contribution of each sensor, and For the The position coordinates of the sensors, is the total number of sensors.

10. The real-time electrical fire warning system based on the Internet of Things according to claim 1 is characterized in that: The warning module includes a warning signal generating unit and a warning information transmitting unit, wherein: Warning signal generating unit: used to generate a warning signal according to the fire risk level provided by the fire risk assessment module and the fire source location information provided by the fire source location module in accordance with the preset fire warning rules; Rule 1: When the fire risk level is high and the fire source is less than 5 meters away from the electrical equipment, an emergency warning signal is generated to prompt immediate emergency measures; Rule 2: When the fire risk level is medium risk and the distance between the fire source and the electrical equipment is between 5 meters and 20 meters, a medium priority warning signal is generated and enhanced monitoring is recommended; Rule 3: When the fire risk level is low and the fire source is more than 20 meters away from the electrical equipment, a general warning signal is generated to alert users; Warning information transmission unit: used to transmit the generated warning signal to the user terminal and emergency response center through the Internet of Things.

Images