SF6 gas status intelligent diagnosis and early warning system based on the Internet of Things

Abstract

The present application relates to the technical field of power equipment state monitoring and Internet of Things, and particularly discloses an SF6 gas state intelligent diagnosis and early warning system based on Internet of Things, which comprises a sensor array module, an edge computing module, an anti-interference communication module, a cloud platform intelligent analysis module, an early warning and linkage control module, an energy management module and a self-checking and fault-tolerant module; the sensor array module collects gas state and equipment environment data in real time, and the collected data are preprocessed by the edge computing module and then transmitted to the cloud platform through the anti-interference communication module; the cloud platform uses multi-modal data fusion and deep learning algorithm to realize gas leakage trend analysis, leakage source positioning and risk level evaluation; the early warning and linkage control module triggers graded early warning according to the diagnosis result and links related equipment for emergency response. The present application realizes all-round intelligent monitoring and management of SF6 gas equipment, and can effectively improve monitoring accuracy and system reliability and reduce equipment failure risk.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power equipment state monitoring and Internet of Things technology, in particular to an SF6 gas state intelligent diagnosis and early warning system based on Internet of Things. BACKGROUND

[0002] SF6 gas is widely used in power equipment such as circuit breakers and GIS combined electrical devices due to its excellent insulation and arc extinguishing performance. However, problems such as SF6 gas leakage and excessive micro-water content can cause a decrease in the insulation performance of the equipment, faults, and even explosions, and SF6 is a strong greenhouse gas, so leakage can cause environmental harm.

[0003] Currently, traditional SF6 gas monitoring mostly uses single-point sensors, which have the problems of limited monitoring range and single data, making it difficult to fully reflect the operating state of the equipment. Moreover, the communication of the monitoring system is easily affected by the strong electromagnetic interference in the substation, leading to unstable data transmission and high packet loss rate. In terms of data analysis, most rely on human experience for judgment, lack intelligent diagnosis and prediction capabilities, and cannot discover potential hazards in advance.

[0004] With the development of technologies such as Internet of Things and artificial intelligence, intelligent monitoring of power equipment has become a trend, but existing technologies still have deficiencies in multi-source data fusion analysis, reliable communication in complex environments, and system autonomous maintenance, making it difficult to meet the needs of the power system for efficient, safe, and intelligent monitoring of SF6 gas equipment. Therefore, the SF6 gas state intelligent diagnosis and early warning system based on Internet of Things is proposed to solve the above problems. SUMMARY

[0005] The purpose of the present application is to provide an SF6 gas state intelligent diagnosis and early warning system based on Internet of Things to solve the problems raised in the background.

[0006] To achieve the above purpose, the present application provides the following technical solutions:

[0007] The SF6 gas state intelligent diagnosis and early warning system based on Internet of Things comprises:

[0008] Sensor array module: deployed at key monitoring points of SF6 gas equipment, including SF6 gas concentration sensors, temperature and humidity sensors, pressure sensors, micro-water content sensors, and vibration sensors, for real-time collection of gas state parameters and equipment operating environment data;

[0009] Edge computing module: integrated in the local terminal, used for preprocessing of gas state parameters and equipment operating environment data, and generating preliminary diagnosis results based on a dynamic baseline model;

[0010] Anti-interference communication module: using LoRa wireless communication technology, integrating forward error correction coding FEC and orthogonal frequency division multiple access OFDMA modulation technology;

[0011] Cloud platform intelligent analysis module: receives data uploaded by the edge computing module, analyzes gas leakage trends, device aging states, and environmental coupling effects through multi-modal data fusion algorithms, and predicts leakage source locations and risk levels based on deep learning models;

[0012] Early warning and linkage control module: triggers graded early warning signals based on the leakage source location and risk level, and links the fan, air supplement device, and access control system to execute emergency response;

[0013] Energy management module: includes a solar power supply unit and an intelligent power distribution circuit to provide multi-mode power supply support for the system;

[0014] Self-checking and fault-tolerant module: periodically checks sensor accuracy, communication link integrity, and power supply status, and performs fault isolation and redundancy switching.

[0015] As a preferred solution, the dynamic baseline model is constructed by the following methods:

[0016] Based on historical data and device operating parameters, dynamic threshold intervals for SF6 gas concentration, pressure, and micro-water content are established;

[0017] Introduce environmental coupling factors, combine device vibration data and local weather information, and dynamically adjust the early warning sensitivity of the baseline model;

[0018] An improved hybrid time series prediction algorithm is used, and its modeling method is:

[0019] The prediction results of the ARIMA model are weighted and fused with the seasonal and trend components of the Prophet algorithm, the weight coefficient α is in the range of 0.6 to 0.8, and an environmental variable correction term is added, and the coupling coefficient β of the correction term is dynamically calculated by the ridge regression algorithm;

[0020] A double residual feedback mechanism is established:

[0021] Calculate the first-order residual between the real-time monitoring value and the model prediction value, decompose the second-order residual through wavelet transform, and inject the second-order residual into the model input layer in reverse to fine-tune the parameters.

[0022] As a preferred solution, the anti-interference communication module includes:

[0023] LoRa channel encoder based on convolutional code, the encoding process is: performing modulo operation on the information bit polynomial and the preset generator polynomial to generate encoded data with redundant check bits;

[0024] Adaptive frequency hopping mechanism, the frequency selection method is: according to the real-time signal strength RSSI of each frequency band, allocate the available frequency band resources in proportion, and select the communication frequency band with the smallest interference;

[0025] The encryption transmission protocol uses the AES-256 algorithm to perform end-to-end encryption on sensor data and control instructions.

[0026] As a preferred solution, the deep learning model comprises:

[0027] The leakage source positioning sub-model adopts an improved multi-scale attention convolutional neural network, dynamically calculates the attention weight matrix of each convolution kernel through a gated recurrent unit, and performs weighted fusion on convolution feature maps of different scales.

[0028] The risk prediction sub-model adopts a hybrid architecture of LSTM and a time convolution network, and the time convolution network captures multi-scale time sequence features through an exponentially increasing dilation factor.

[0029] The federal learning optimization mechanism performs weighted aggregation on local model parameters of each site, and the weight value is inversely proportional to the privacy protection strength of the site data distribution.

[0030] As a preferred solution, the early warning and linkage control module comprises:

[0031] The adaptive early warning strategy dynamically adjusts the early warning level according to the leakage rate, the equipment importance level, and the presence of personnel.

[0032] The air replenishment control logic adopts a PID-fuzzy compound controller, and the parameter tuning method is as follows:

[0033] The proportional coefficient Kp is dynamically adjusted according to the differential value of the pressure deviation, and the integral coefficient Ki is adaptively increased according to the absolute value of the pressure deviation.

[0034] The emergency ventilation mechanism is based on a dynamic weight path planning algorithm, and the path cost is composed of the actual moving cost, the estimated remaining cost, and the gas concentration penalty term, wherein the concentration penalty term weight exponentially decays over time.

[0035] As a preferred solution, the energy management module comprises:

[0036] The photovoltaic power supply unit adopts a maximum power point tracking algorithm to minimize the equivalent series resistance loss by optimizing the output voltage of the photovoltaic array.

[0037] The intelligent power distribution circuit proportionally allocates total power according to the priority weight of the equipment, and the equipment with high priority obtains power supply resources preferentially.

[0038] As a preferred solution, the self-checking and fault-tolerant module comprises:

[0039] The multi-index fusion diagnosis algorithm activates the difference between the sensor reading, the calibration coefficient, and the failure threshold through a rectified linear unit, and then outputs a fault probability distribution after dynamic weight weighting.

[0040] Fault recovery mechanism: when the communication is interrupted, automatically switch to local cache mode, the storage capacity does not exceed the maximum capacity of the edge device.

[0041] From the above technical solutions provided by the present application, the SF6 gas state intelligent diagnosis and early warning system based on the Internet of Things provided by the present application has the beneficial effects of:

[0042] Comprehensive and accurate monitoring to ensure equipment safety: the sensor array module deploys multiple high-precision sensors to collect real-time data on key parameters such as SF6 gas concentration, temperature and humidity, and pressure, providing comprehensive coverage of equipment operating status information; the edge computing module effectively filters interference through dynamic baseline modeling and data preprocessing, accurately determining the equipment operating condition; the cloud platform intelligent analysis module uses multi-modal data fusion and deep learning technology to analyze the gas leakage trend and equipment aging state, identify potential faults and hidden dangers in advance, and prevent accidents caused by equipment failure;

[0043] Efficient and reliable communication to ensure data transmission: the anti-interference communication module uses LoRa wireless communication technology, combined with forward error correction coding, orthogonal frequency division multiple access modulation and other technologies, to ensure the accuracy and stability of data transmission in a strong electromagnetic interference environment in a substation; adaptive frequency hopping mechanism and encryption transmission protocol further improve communication efficiency and data security, ensuring real-time and reliable transmission of monitoring data and control instructions, providing solid data support for accurate diagnosis and timely response of the system;

[0044] Intelligent early warning and linkage to improve emergency response capability: the early warning and linkage control module implements adaptive early warning strategies based on cloud platform analysis results, and transmits device risk information to relevant personnel in a timely manner through various means such as sound and light alarms and mobile terminal push; at the same time, it links the fan, gas supplement device and access control system, and automatically executes emergency response operations such as rapid dispersion of leaked gas, gas pressure supplement, personnel access control, etc. according to different risk levels, effectively reducing the damage of faults, improving the system's emergency handling capability, and reducing the loss of accidents;

[0045] Scientific energy management to enhance system endurance: the energy management module uses a solar power supply unit to convert solar energy into electrical energy, providing green and renewable energy for the system, reducing dependence on traditional electricity and saving energy costs; the intelligent power distribution circuit dynamically allocates power based on load priority to ensure priority power supply for critical modules, while realizing intelligent switching and management of multiple power sources, ensuring the continuous and stable operation of the system under various environmental conditions, enhancing the endurance of the system, and reducing the operation and maintenance cost;

[0046] Autonomous fault detection and tolerance, improve system robustness: self-checking and fault tolerance module periodically detects system sensor accuracy, communication link integrity and power state, through multi-index fusion diagnosis algorithm, timely discovery and positioning of faults; Once a fault is detected, quickly execute fault isolation and redundancy switching operation to ensure that the core functions of the system are not affected; Local storage-forwarding mode saves data when communication is interrupted to ensure data integrity, significantly improves the robustness and reliability of the system, reduces system downtime, and improves user experience. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 The figure is a schematic diagram of the overall structure of the SF6 gas state intelligent diagnosis and early warning system based on the Internet of Things. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical scheme and advantages of the present application clearer, the following will further describe the present application in combination with the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application and do not limit the present application.

[0049] In order to better understand the above technical scheme, the above technical scheme will be described in detail in combination with the drawings and specific embodiments of the specification.

[0050] As shown in Figure 1 The present application provides an SF6 gas state intelligent diagnosis and early warning system based on the Internet of Things, which comprises:

[0051] Sensor array module: deployed at key monitoring points of SF6 gas equipment, including SF6 gas concentration sensor, temperature and humidity sensor, pressure sensor, micro water content sensor and vibration sensor, real-time acquisition of gas state parameters and equipment operating environment data;

[0052] Edge computing module: integrated in the local terminal, used for preprocessing gas state parameters and equipment operating environment data, and generating preliminary diagnosis results based on a dynamic baseline model, the dynamic baseline model is constructed by the following methods:

[0053] Establishing dynamic threshold intervals of SF6 gas concentration, pressure and micro water content based on historical data and equipment operating parameters;

[0054] Introducing environmental coupling factors, combining with equipment vibration data and local weather information, dynamically adjusting the early warning sensitivity of the baseline model;

[0055] An improved hybrid time series prediction algorithm is used, and its modeling method is:

[0056] The prediction results of the ARIMA model are weighted and fused with the seasonal term and trend term components of the Prophet algorithm, the weight coefficient a is in the range of 0.6 to 0.8, and an environmental variable correction term is superimposed, and the coupling coefficient β of the correction term is dynamically calculated by the ridge regression algorithm;

[0057] A double residual feedback mechanism is established:

[0058] The first-order residual between the real-time monitoring value and the model prediction value is calculated, the second-order residual is decomposed by wavelet transform, and the second-order residual is injected into the model input layer in reverse to fine-tune the parameters;

[0059] Anti-interference communication module: using LoRa wireless communication technology, integrating forward error correction coding FEC and orthogonal frequency division multiple access OFDMA modulation technology;

[0060] Cloud platform intelligent analysis module: receiving the data uploaded by the edge computing module, analyzing the gas leakage trend, equipment aging state and environmental coupling influence through multi-modal data fusion algorithm, and predicting the leakage source position and risk level based on a deep learning model, the deep learning model includes:

[0061] Leakage source positioning sub-model: using an improved multi-scale attention convolutional neural network, dynamically calculating the attention weight matrix of each convolution kernel through a gated recurrent unit, and weighting and fusing convolution feature maps of different scales;

[0062] Risk prediction sub-model: using a hybrid architecture of LSTM and time convolution network, the time convolution network captures multi-scale time sequence features through an exponentially growing dilation factor;

[0063] Federal learning optimization mechanism: weighting and aggregating the local model parameters of each site, the weight value is inversely proportional to the privacy protection strength of the site data distribution;

[0064] Early warning and linkage control module: according to the leakage source position and risk level, triggering a hierarchical early warning signal, and linking the fan, air supplement device and access control system to execute emergency response;

[0065] Energy management module: including a solar power supply unit and an intelligent power distribution circuit, providing multi-mode power supply guarantee for the system;

[0066] Self-checking and fault-tolerant module: periodically detecting sensor accuracy, communication link integrity and power supply state, and performing fault isolation and redundancy switching.

[0067] In this embodiment, the sensor array module is the core component for precise data collection in the SF6 gas state intelligent diagnosis and early warning system based on the Internet of Things, like the system's "sensing tentacles", its performance directly affects the accuracy of the system's monitoring and diagnosis of the running state of SF6 gas equipment; in the complex electromagnetic environment of the substation, the running state of SF6 gas equipment is easily disturbed by many factors, the sensor array module works through scientific layout and multiple sensors to collect gas state and equipment running environment data in real time and accurately, providing a reliable basis for subsequent analysis and diagnosis, which is embodied in the following aspects:

[0068] I. Overall function overview:

[0069] This module deploys multiple types of sensors at key monitoring points of SF6 gas equipment to build a comprehensive data collection network; for key parameters such as SF6 gas concentration, temperature and humidity, pressure, micro-water content, and equipment vibration, advanced sensing technology is used for real-time monitoring, and the collected raw data is transmitted to the edge computing module through the data transmission link, providing basic data support for subsequent data processing, diagnosis analysis, and early warning control of the system;

[0070] II. Submodule composition and function:

[0071] (1) SF6 gas concentration sensing unit:

[0072] Sensor selection and deployment: high-precision infrared absorption or electrochemical principle SF6 gas concentration sensors are selected, and according to the structural characteristics of SF6 gas equipment, they are densely deployed at key locations such as circuit breaker gas chambers, GIS pipeline connection parts, and transformer gas chambers where gas leakage is prone to occur; the detection accuracy of this sensor can reach ppm level, and it can quickly respond to small changes in gas concentration;

[0073] Data collection and transmission: continuously collect SF6 gas concentration data, convert analog signals to digital signals through the built-in analog-to-digital conversion module, and then transmit them to the edge computing module through a dedicated data transmission line, providing core data for the system to determine gas leakage;

[0074] (2) Temperature and humidity sensing unit:

[0075] Sensor characteristics: digital temperature and humidity composite sensor with wide measurement range and high precision; temperature measurement range: -40℃-85℃, accuracy: ±0.3℃; humidity measurement range: 0%~100%RH, accuracy: ±2%RH; the sensor has an intelligent calibration algorithm built-in, which can automatically compensate for the influence of environmental factors on measurement results;

[0076] Data collection and role: Real-time monitoring of the temperature and humidity data of the equipment operating environment, and transmitting the packaged data to the edge computing module; the temperature and humidity data can be used to correct the measurement errors of other sensors caused by environmental factors, and provide environmental reference for analyzing the physical and chemical property changes of SF6 gas;

[0077] (Three) Pressure sensing unit:

[0078] Sensor principle and parameters: The pressure sensing unit adopts a piezoresistive pressure sensor with a range of 0-1.0MPa and an accuracy of 0.1%FS; based on the piezoresistive effect principle, when the internal pressure of the SF6 gas equipment changes, the resistance value of the pressure-sensitive resistor inside the sensor changes, and the resistance change is converted into a voltage signal through the measurement circuit, and then the pressure value is obtained;

[0079] Data collection and application: Real-time monitoring of the internal SF6 gas pressure of the equipment, and real-time transmission of the pressure data to the edge computing module; the pressure data is one of the key basis for judging whether the equipment has gas leakage, sealing failure and other faults;

[0080] (Four) Micro water content sensing unit:

[0081] Sensing technology and accuracy: The micro water content sensor based on the principle of capacitive or dew point method can accurately measure the trace water in SF6 gas as low as 10ppm; the capacitive sensor determines the micro water content by detecting the capacitance change caused by water in the gas; the dew point method sensor indirectly calculates the micro water content by measuring the dew point temperature of the gas;

[0082] Data collection and significance: Continuously collect SF6 gas micro water content data and transmit to the edge computing module; too high micro water content will cause the internal insulation performance of the equipment to decrease, and this data is of great significance for evaluating the insulation state of the equipment and preventing insulation faults;

[0083] (Five) Vibration sensing unit:

[0084] Sensor type and performance: The acceleration type vibration sensor can detect vibration acceleration in the range of 0.1m / s²-100m / s², and the frequency response range is 0.1Hz-10kHz; this sensor can sensitively capture the vibration signal changes caused by mechanical faults such as loose parts and bearing wear during equipment operation;

[0085] Data collection and analysis: Real-time collection of equipment vibration data, and transmission of the data to the edge computing module; through analysis and processing of the vibration data, the running state of the internal mechanical structure of the equipment can be judged, providing important clues for predicting equipment faults;

[0086] (Six) Data transmission and interface module:

[0087] Transmission line design: shielded cable or optical fiber is used as the data transmission line, which effectively resists the strong electromagnetic interference in the substation and ensures the stable transmission of sensor data collection; according to the data transmission needs of different types of sensors, the adaptive data interface protocol is designed to ensure the accuracy and consistency of data transmission;

[0088] Data aggregation and preprocessing: The data collected by each sensor is aggregated, and simple data format conversion and verification and other preprocessing operations are performed before transmission to the edge computing module, improving data transmission efficiency and reliability;

[0089] III. Key technical principles:

[0090] (1) SF6 gas concentration sensing principle:

[0091] The infrared absorption type SF6 gas concentration sensor is based on the absorption characteristics of SF6 gas to specific wavelength infrared light. When infrared light passes through the gas chamber containing SF6 gas, part of the infrared light is absorbed by SF6 gas. By detecting the attenuation degree of infrared light intensity, the gas concentration is calculated according to the Lambert-Beer law; the electrochemical sensor utilizes the relationship between the current generated by the electrochemical reaction of SF6 gas on the electrode and the gas concentration, realizing the measurement of gas concentration;

[0092] (2) Temperature and humidity sensing principle:

[0093] In the digital temperature and humidity composite sensor, the temperature sensing part usually uses a thermistor or a thermocouple. By detecting the resistance or voltage change caused by temperature change, the digital temperature signal is converted through the signal processing circuit; the humidity sensing part generally uses a high polymer polymer humidity-sensitive capacitor, whose capacitance value changes with the change of environmental humidity, and the humidity data is obtained by measuring the capacitance value change;

[0094] (3) Pressure sensing principle:

[0095] The piezoresistor inside the piezoresistive pressure sensor deforms under pressure, causing its resistance value to change. Through the Wheatstone bridge circuit, the resistance change is converted into a voltage signal, and then through the signal processing circuit such as amplification and filtering, the final electric signal output proportional to the pressure is obtained;

[0096] (4) Micro water content sensing principle:

[0097] In the capacitive micro water content sensor, the moisture-sensing film of the sensor adsorbs water in the gas, causing a change in its dielectric constant, which in turn changes the capacitance value. By measuring the capacitance value change, the gas micro water content is determined; the dew point method micro water content sensor cools the gas to reach the dew point temperature. When the water vapor in the gas begins to condense into dew, the dew point temperature is detected, and combined with the gas pressure and other parameters, the gas micro water content is calculated;

[0098] (V) Vibration sensing principle:

[0099] Acceleration type vibration sensor is based on piezoelectric effect or piezoresistive effect. When the sensor is subjected to vibration, the internal piezoelectric material or piezoresistive element generates electric charge or resistance change. The weak signal is amplified by charge amplifier or signal conditioning circuit, and then converted into voltage signal, and then the electric signal output related to vibration acceleration is obtained;

[0100] IV. Workflow of the module:

[0101] (I) Initialization phase:

[0102] After the system starts, the sensor array module completes the self-checking operation of each sensor, detects whether the sensor power supply is normal, the communication link is smooth, and the sensor itself performance meets the standard;

[0103] Initialize the working parameters of the sensor, such as sampling frequency, range, data transmission protocol, etc., and establish a data connection channel with the edge computing module;

[0104] (II) Data acquisition phase:

[0105] Each type of sensor continuously collects corresponding parameter data according to the set sampling frequency, such as SF6 gas concentration sensor collects gas concentration data every second, and temperature and humidity sensor collects temperature and humidity data every 5 seconds, etc.

[0106] The sensor performs preliminary processing on the collected raw data, such as filtering and denoising, range conversion, etc., to convert analog signals into digital signals;

[0107] (III) Data transmission phase:

[0108] The data transmission and interface module gathers the processed data of each sensor, and packages and encapsulates them according to a specific data format;

[0109] Through shielded cable or optical fiber transmission line, the encapsulated data is stably transmitted to the edge computing module. During the transmission process, a verification mechanism is used to ensure the accuracy and integrity of the data;

[0110] (IV) Data update and maintenance phase:

[0111] The sensor array module continuously collects data and updates the data transmitted to the edge computing module in real time;

[0112] Periodically calibrate and maintain the sensor, calibrate the accuracy of the sensor according to the usage time and environmental conditions, replace the aged or damaged sensor, and ensure the long-term stable operation of the module;

[0113] (V) End phase:

[0114] When the system stops running or receives a stop instruction, the sensor array module stops data collection and transmission operations, turns off the power of each sensor, saves sensor working parameters and state information, and waits for the next system startup.

[0115] In this embodiment, the edge computing module serves as the "local brain" of the SF6 gas state intelligent diagnosis and early warning system based on the Internet of Things, and undertakes the key task of data preprocessing and preliminary diagnosis. In the complex electromagnetic environment of a substation, the sensor array module will generate a large amount of raw data. The edge computing module processes the data on site to reduce the data transmission pressure, and quickly outputs the preliminary diagnosis results to provide strong support for the efficient operation of the system, which is specifically embodied in the following aspects:

[0116] I. Overall function overview:

[0117] This module is integrated into the local terminal and receives the SF6 gas state parameters and device operating environment data transmitted by the sensor array module. Through preprocessing methods such as noise filtering, data normalization, and outlier rejection, the quality of raw data is purified. At the same time, based on the dynamic baseline model, the preprocessed data is analyzed to generate preliminary diagnosis results, and key data and diagnosis information are uploaded to the cloud platform intelligent analysis module to realize hierarchical processing and efficient transmission of data.

[0118] II. Submodule composition and function:

[0119] (I) Data preprocessing unit:

[0120] Noise filtering: An adaptive filtering algorithm is used to filter out electromagnetic interference noise, environmental fluctuation noise, etc. in the sensor data. According to the time domain and frequency domain characteristics of the data, the filtering parameters are dynamically adjusted. For example, the Kalman filtering algorithm is used to combine the historical values and predicted values of the sensor data to effectively filter out random noise and retain the true signal characteristics.

[0121] Data normalization: The data collected by different types of sensors with different dimensions and numerical ranges are converted to a unified numerical interval, such as [0, 1] or a standard distribution with a mean of 0 and a standard deviation of 1, through the minimum-maximum normalization or Z-score normalization method, to facilitate subsequent data processing and analysis.

[0122] Outlier rejection: An outlier detection mechanism based on the combination of the 3σ principle and the Isolation Forest algorithm is established. Data points that are significantly deviated from the mean by 3 times the standard deviation are preliminarily determined as outliers. The Isolation Forest algorithm is then used to deeply mine the data to identify isolated points and outliers, which are then removed to ensure the reliability of the input data.

[0123] (II) Dynamic baseline model processing unit:

[0124] Model construction and update: Based on historical data and equipment operation parameters (device model, operation time, environmental temperature and humidity, etc.), the dynamic threshold interval of SF6 gas concentration, pressure and micro-water content is constructed; the environmental coupling factor is introduced, combined with equipment vibration data and local weather information (air pressure, wind speed), the warning sensitivity of the baseline model is dynamically adjusted; the improved hybrid time series prediction algorithm is adopted, such as formula

[0125] (wherein, is the gas state prediction value at time t; is the ARIMA model weight coefficient ; ARIMA is an autoregressive integrated moving average model, is the autoregressive order, is the difference order, is the moving average order; is the Prophet algorithm model, is the seasonal component of the Prophet algorithm, is the trend component of the Prophet algorithm, is the environmental variable correction term (including temperature, vibration and other parameters); is the environmental coupling coefficient (calculated dynamically by ridge regression)), and through a double residual feedback mechanism (wherein, is the first-order residual, is the true value of the gas state at time t, is the prediction value of the gas state at time t)、 (wherein, is the second-order residual, is the result of wavelet transform on the first-order residual ), the dynamic fine-tuning and optimization of model parameters are realized;

[0126] Preliminary diagnostic analysis: input the pretreated data into the dynamic baseline model, compare the real-time data with the dynamic threshold interval, analyze the change trend and deviation degree of the data; if the data exceeds the threshold range or shows abnormal change trend, combined with the prediction result of the model, generate the preliminary diagnostic conclusion, such as judging whether the equipment exists SF6 gas leakage, pressure abnormal fluctuation and other potential faults;

[0127] (III) Data transmission and interface unit:

[0128] ​Data filtering and packaging: According to the requirements of the cloud platform intelligent analysis module, the pre-processed data and preliminary diagnosis results are filtered to extract key data and diagnosis information. The filtered data is packaged according to a specific data format, with meta-information such as data identification and timestamp added for easy data analysis and processing by the cloud platform.

[0129] Communication protocol adaptation: LoRa wireless communication technology with anti-interference communication module is adopted to adapt its forward error correction coding (FEC) and orthogonal frequency division multiple access (OFDMA) modulation technology to ensure reliable data transmission in the strong electromagnetic interference environment of the substation. A stable data transmission link is established to monitor the data transmission status in real time. When transmission interruption or error occurs, automatic retransmission or error correction processing is performed.

[0130] (Four) Local storage and backup unit:

[0131] Data storage management: Set up local storage devices to store sensor raw data, pre-processed data and preliminary diagnosis results. Use a database management system to store and index data for quick query and call. According to the importance and frequency of use of data, set different storage periods, regularly clean up expired data and release storage resources.

[0132] Data backup mechanism: Establish a data backup strategy to regularly backup key data to external storage devices or cloud storage platforms. When the system fails or data is lost, the data can be quickly recovered from the backup to ensure the continuity of the system and the integrity of the data.

[0133] III. Key technical principles:

[0134] (I) Data preprocessing technology principle:

[0135] Adaptive filtering algorithm establishes a state space model of data, uses the state estimation at the previous time and the observation data at the current time to continuously update the filtering parameters, and realizes dynamic suppression of noise. The minimum-maximum normalization maps the data to the specified interval through linear transformation, and the formula is (Where, is the original data, and are the minimum and maximum values of the data, is the normalized data; and are the lower and upper limits of the target interval); Z-score normalization is based on the mean and standard deviation of the data for standardization, and the formula is (Where, is the mean of the data, For data standard deviation); 3σ principle is based on the assumption of normal distribution, considering that the probability of data falling outside the range of mean ± 3 times standard deviation is extremely small, which can be determined as an outlier; Isolation Forest algorithm builds a random binary tree to calculate the path length of data points, and the shorter the path, the more likely it is an outlier;

[0136] (II) Dynamic baseline model principle:

[0137] The dynamic baseline model combines traditional statistical models and machine learning algorithms to establish a basic threshold interval using historical data and device operating parameters. By introducing environmental coupling factors, it considers the impact of environmental factors on gas state parameters, allowing the model to adapt to changes under different working conditions. The improved hybrid time series prediction algorithm combines the ARIMA model's ability to capture time series trends and seasonality with the Prophet algorithm's advantage in handling complex trends and outliers, and further improves prediction accuracy through environmental variable correction terms. The dual residual feedback mechanism uses wavelet transform to analyze residuals at multiple scales, extract detailed information from residuals, and optimize model parameters in reverse to improve model fitting accuracy.

[0138] (III) Data transmission and storage principle:

[0139] In terms of data transmission, LoRa wireless communication technology uses spread spectrum modulation technology combined with forward error correction coding and orthogonal frequency division multiple access technology to enhance signal anti-interference capability and transmission efficiency. In terms of data storage, the database management system implements efficient storage, fast query and secure management of data through data file storage, index structure establishment and transaction processing mechanism. Backup technology ensures data security and recoverability through data replication, incremental backup and other methods.

[0140] Four, the working process of the module:

[0141] (I) Initialization phase:

[0142] After starting the edge computing module, it completes the self-checking of hardware devices, including the state detection of processors, memories, storage devices, communication interfaces, etc., to ensure normal operation of the device.

[0143] Load data preprocessing algorithms, initial parameters and configuration files of dynamic baseline model, establish communication connection with sensor array module and cloud platform intelligent analysis module, and initialize local storage database.

[0144] (II) Data receiving phase:

[0145] Real-time monitoring of data transmission channels of the sensor array module to receive raw data such as SF6 gas concentration, temperature and humidity, pressure, micro-water content and vibration;

[0146] The received data is checked for integrity and accuracy, and if errors or missing data are found, the sensor array module is requested to retransmit in a timely manner;

[0147] (Three) data preprocessing stage:

[0148] The original data is sequentially sent to the noise filtering, data normalization and outlier rejection unit, and processed according to the preset algorithm and parameters to purify the data quality;

[0149] The quality of the preprocessed data is evaluated, and if the data quality does not meet the requirements, it is returned for reprocessing or marked as abnormal data;

[0150] (Four) preliminary diagnosis stage:

[0151] The preprocessed data is input into the dynamic baseline model processing unit, and the model is used for data analysis and prediction to generate preliminary diagnosis results;

[0152] The preliminary diagnosis results are audited and confirmed, and diagnostic basis and related information are added;

[0153] (Five) data transmission stage:

[0154] According to the requirements of the cloud platform intelligent analysis module, the key data and diagnosis results are screened, packaged and encapsulated;

[0155] Through the anti-interference communication module, the encapsulated data is transmitted to the cloud platform intelligent analysis module, and the data transmission state is monitored in real time to ensure that the data is accurately and correctly delivered;

[0156] (Six) data storage and backup stage:

[0157] The original data, preprocessed data and preliminary diagnosis results are stored in the local database, and are classified and stored according to data type and time;

[0158] According to the set backup strategy, the key data is backed up regularly, and the backup time and state information are recorded;

[0159] (Seven) end stage:

[0160] When the system stops running or receives a stop instruction, the edge computing module stops data reception, processing and transmission operations, closes the communication connection with the sensor array module and the cloud platform intelligent analysis module, saves the current working state and parameters, and releases system resources.

[0161] In this embodiment, the anti-interference communication module includes:

[0162] The LoRa channel encoder based on convolution code, the encoding process is: the information bit polynomial is operated with the preset generator polynomial, and the encoding data with redundant check bits is generated;

[0163] Adaptive frequency hopping mechanism, the frequency selection method is: according to the real-time signal strength RSSI proportion of each frequency band, the available frequency band resource is allocated, the communication frequency band with the minimum interference is selected;

[0164] Encryption transmission protocol, using AES-256 algorithm to encrypt sensor data and control instructions end to end;

[0165] Further, the anti-interference communication module is the key hub to ensure reliable data transmission in the SF6 gas state intelligent diagnosis and early warning system based on the Internet of Things; In the complex environment of strong electromagnetic interference in the substation, the module ensures the stable and efficient transmission of sensor data and control instructions between the edge computing module and the cloud platform by integrating various advanced communication technologies, builds a data transmission defense line for the normal operation of the system, and specifically embodies as follows:

[0166] I. Overall function overview:

[0167] The module uses LoRa wireless communication technology as the foundation, integrates forward error correction coding (FEC), orthogonal frequency division multiple access (OFDMA) modulation technology, adaptive frequency hopping mechanism and encryption transmission protocol, and builds a complete anti-interference communication system; It is responsible for uploading the SF6 gas equipment monitoring data processed by the edge computing module to the cloud platform, and receiving the control instructions issued by the cloud platform and transmitting them to the early warning and linkage control module, ensuring the accuracy, integrity and security of data transmission in complex electromagnetic environment;

[0168] II. Submodule composition and function:

[0169] (1) LoRa channel coding unit:

[0170] Convolutional code encoder: the LoRa channel encoder based on convolutional code uses the formula

[0171] (wherein, is the encoded polynomial used for transmission in the channel; is the information bit, that is, the encoded representation of the original data to be transmitted: is the number of information bits; is the generator polynomial, which is the core of the pre-set encoding rule, determines the way of adding redundant information after encoding, through the formula calculation, the original information bit adds redundant check information according to a certain rule, generates error correction code, so that the receiving end can detect and correct errors according to the encoding rule) ; Before data transmission, the data output by the edge computing module is processed by convolutional coding to increase the redundancy of the data and improve the anti-interference ability and error correction ability of the data in the transmission process;

[0172] Coding parameter configuration: According to the actual communication environment and data transmission requirements, the coding rate and constraint length of convolutional code can be flexibly configured; higher coding rate can improve data transmission efficiency, but the error correction capability is relatively low; larger constraint length enhances the error correction capability, but increases the coding complexity and transmission delay; by reasonably adjusting the parameters, the balance between transmission efficiency and anti-interference performance is realized in different scenarios;

[0173] (II) Adaptive frequency hopping unit:

[0174] Frequency selection algorithm: formula (where, is the frequency point selected in the th frequency band, that is, the frequency finally used for data transmission; is the base frequency, which is the reference frequency for the entire frequency selection; is the signal strength of the th frequency band, reflecting the signal strength received at the moment; is the total number of available frequency bands, which clearly indicates the number of frequency ranges that can be selected; is the sum of the signal strengths of all available frequency bands, used to calculate the relative proportion of each frequency band signal strength; is the frequency interval, which determines the frequency difference between adjacent selectable frequency points; is the floor function, which ensures that the calculated frequency point is an integer); This algorithm monitors the signal strength (RSSI) of each frequency band in real time, dynamically selects the frequency band with the smallest interference and the best signal quality for data transmission according to the relative size of the signal strength; When a certain frequency band is affected by strong electromagnetic interference, causing the signal strength to drop, the system quickly switches to other high-quality frequency bands to avoid data transmission interruption or errors;

[0175] Frequency hopping period and strategy: Set a reasonable frequency hopping period to avoid continuous interference frequency bands in time, and not to increase communication overhead due to too frequent frequency hopping; At the same time, develop multiple frequency hopping strategies, such as random frequency hopping and sequence frequency hopping, to flexibly select according to different electromagnetic interference characteristics and communication needs, further enhancing the anti-interference ability;

[0176] (III) Modulation and demodulation unit:

[0177] OFDMA modulation technology: Use orthogonal frequency division multiple access (OFDMA) modulation technology to divide the total bandwidth into multiple orthogonal subcarriers, each of which can transmit data independently; At the sending end, the data to be transmitted is allocated to different subcarriers for parallel transmission, improving the frequency spectrum utilization and data transmission rate; At the same time, through the orthogonality between subcarriers, the multipath fading and inter-symbol interference are effectively suppressed, ensuring reliable data transmission in complex electromagnetic environments;

[0178] Demodulation and signal recovery: At the receiving end, the received signal is processed using corresponding demodulation algorithms to separate the original data from each subcarrier and restore complete and accurate information through signal processing techniques. During demodulation, the error correction capability of LoRa channel coding is combined to correct errors generated during transmission, ensuring data accuracy.

[0179] (Four) Encryption transmission unit:

[0180] AES-256 encryption algorithm: The sensor data and control instructions are encrypted end-to-end using the AES-256 encryption algorithm. AES-256 uses a 256-bit key to group encrypt data, and through multiple rounds of byte substitution, row shifting, column confusion, and round key addition operations, it converts the original data into ciphertext. Before data transmission, the data is encrypted using a pre-negotiated key. Only the receiving end with the correct key can decrypt and restore the data, ensuring data security during transmission and preventing data theft or tampering.

[0181] Key management: Establish a complete key management mechanism, including key generation, distribution, update, and storage. Use a secure key generation algorithm to generate high-strength keys and distribute them to both parties through a secure channel. Regularly update the keys to reduce the risk of key cracking. At the same time, encrypt the keys to ensure their security.

[0182] (Five) Communication link monitoring unit:

[0183] State monitoring: Real-time monitoring of communication link parameters such as signal strength, error rate, data transmission rate, etc. Analyze these parameters to determine the operating state of the communication link and promptly detect link failures or performance degradation.

[0184] Fault handling: When a communication link failure or anomaly is detected, automatically start the fault handling mechanism. For example, if the error rate is too high, trigger the retransmission mechanism to retransmit the error data. If the communication link is interrupted, try to re-establish the connection to ensure data transmission continuity. At the same time, report the fault information to the system management platform for timely troubleshooting and repair by maintenance personnel.

[0185] Three, key technology principles:

[0186] (One) Convolutional code error correction principle:

[0187] Convolutional codes are memory-based error-control codes that introduce redundant information into the encoding process by convolving the input information sequence with a generator polynomial. The receiver uses this redundant information and a specific decoding algorithm (such as the Viterbi decoding algorithm) to recover the original correct information sequence from the received sequence, which may contain errors, according to the maximum likelihood criterion. By increasing the constraint length of the code and properly designing the generator polynomial, convolutional codes can effectively improve the reliability of data transmission without significantly increasing the transmission bandwidth.

[0188] (2) Adaptive frequency hopping principle:

[0189] Adaptive frequency hopping technology is based on real-time perception of the channel environment. By continuously monitoring the signal strength and interference conditions in each frequency band, it dynamically adjusts the operating frequency according to a preset frequency selection algorithm. When a frequency band is interfered with, the system quickly switches to another idle or less interfered frequency band for communication, thus avoiding the interference source and achieving stable transmission of the communication link. Its core lies in fast and accurate channel status perception and efficient frequency switching mechanism, ensuring continuous communication in complex and changing electromagnetic environments.

[0190] (3) OFDMA modulation principle:

[0191] OFDMA technology divides high-speed data services into multiple low-speed sub-data streams, which are modulated onto multiple orthogonal subcarriers for parallel transmission. Because the subcarriers are orthogonal, orthogonal demodulation can be used at the receiver to accurately separate the data on each subcarrier, effectively overcoming the effects of multipath fading and inter-symbol interference. Furthermore, OFDMA can flexibly allocate subcarriers to different users or services, enabling multi-user sharing of spectrum resources and improving spectrum utilization and system capacity.

[0192] (IV) AES-256 encryption principle:

[0193] The AES-256 encryption algorithm is based on the principle of symmetric encryption and uses the same 256-bit key for encryption and decryption. The algorithm uses multiple rounds of complex transformations, including byte substitution (SubBytes), row shift (ShiftRows), column confusion (MixColumns), and round key addition (AddRoundKey), to scramble the original data and mix it with the key to generate ciphertext. Only by using the same key for reverse operations can the ciphertext be restored to the original data. The high-strength encryption characteristics and complex transformation process of AES-256 enable it to resist various cryptographic attacks and ensure the confidentiality and integrity of data.

[0194] In this embodiment, the cloud platform intelligent analysis module serves as the "smart hub" of the IoT-based SF6 gas status intelligent diagnosis and early warning system. It undertakes the core tasks of deep mining and intelligent decision-making for massive amounts of monitoring data. When faced with multi-source heterogeneous data uploaded by the edge computing module, this module uses advanced data fusion algorithms and deep learning models to accurately analyze SF6 gas leakage trends, equipment aging status, and environmental coupling effects, providing key support for the system's intelligent early warning and scientific decision-making. Specifically,

[0195] 1. Overview of overall functions:

[0196] This module receives pre-processed SF6 gas equipment monitoring data uploaded by the edge computing module, covering multi-dimensional information such as gas concentration, pressure, moisture content, temperature and humidity, and vibration. It integrates the correlation between different types of data through a multimodal data fusion algorithm and conducts in-depth data analysis combined with a deep learning model to achieve functions such as gas leakage trend prediction, leakage source location, equipment aging status assessment, and risk level classification. At the same time, it provides data query and analysis result display services for other modules of the system, and generates visual reports to assist operation and maintenance personnel in decision-making.

[0197] 2. Submodule composition and functions:

[0198] (1) Multimodal data fusion unit:

[0199] Data integration: Align and integrate different types of sensor data uploaded by the edge computing module according to time series to build a data set containing multi-dimensional information. Based on the characteristics of heterogeneous data, standardize the data and unify the data format and dimension, laying the foundation for subsequent fusion analysis.

[0200] Fusion algorithm: An algorithm based on the combination of feature-level fusion and decision-level fusion is adopted. In terms of feature-level fusion, dimensionality reduction algorithms such as principal component analysis (PCA) and independent component analysis (ICA) are used to extract key features of various types of data and fuse them. In terms of decision-level fusion, methods such as evidence theory and Bayesian networks are used to make comprehensive decisions based on the preliminary diagnostic results obtained by different sensors based on their respective data, thereby improving the accuracy and reliability of data fusion.

[0201] (2) Deep Learning Analysis Unit:

[0202] Leakage source localization sub-model: An improved multi-scale attention CNN model is used, and its feature map generation formula is:

[0203] (in, The final generated feature map is used for leak source location analysis; The number of convolution kernels determines the dimension of the features extracted by the model; The GRU dynamically calculates, The attention weight matrix corresponding to the first convolution kernel is used to highlight the key features related to the leakage source; Element-wise multiplication is used to fuse the attention weights and the convolution features; Conv2D is a two-dimensional convolution operation on the input feature map using the first convolution kernel The model processes the sensor data and combines the spatial layout information of the device to accurately locate the SF6 gas leakage source.

[0204] Risk prediction sub-model: using a hybrid architecture of LSTM-TCN, the dilated convolution formula of the temporal convolution network (TCN) is

[0205] (where, is the output at time t, used for risk prediction; is the length of the convolution kernel: is the weight of the first convolution kernel: is the input at time t; is an exponentially growing dilation factor, is the number of dilated convolution layers The dilated convolution increases the receptive field of the network and captures the long-term dependencies of the data); This sub-model combines the processing capability of LSTM for time series data and the dilated convolution characteristics of TCN to analyze the operation data of SF6 gas equipment, predict the future risk level of the equipment, and discover potential fault risks in advance; Model training and optimization: use a large amount of historical monitoring data and actual fault cases to train the deep learning model; adopt a federated learning optimization mechanism

[0206] (where,

[0207] is the global model parameter; is the number of clients participating in federated learning; is the model parameter of the first client; is the privacy protection intensity coefficient; is the data distribution of the first client is the average data distribution of all clients between and Divergence, used to balance the impact of client data differences on model training), while protecting data privacy, integrating multi-party data to optimize model parameters and improve the model's generalization ability and prediction accuracy;

[0208] (3) Data analysis and decision-making unit:

[0209] Trend analysis: Perform time series analysis on the fused data, plotting the changing trend of parameters such as SF6 gas concentration, pressure, and trace water content. Combining historical data with equipment operation patterns, analyze the changing trends of parameters and determine whether there are abnormal development trends in the equipment.

[0210] Risk assessment: Based on the prediction results of the deep learning model and the preset risk assessment indicators, the operating risks of SF6 gas equipment are quantitatively assessed and classified into risk levels (such as low risk, medium risk, and high risk), providing a decision-making basis for the early warning and linkage control modules;

[0211] Root cause analysis: Using the knowledge graph driven root cause analysis tool, the correlation calculation formula is

[0212] (in, For Entity and entities The correlation between them is used to analyze the cause of the failure is the number of feature dimensions: For Entity In the Dimensional feature encoding; For Entity In the By building a knowledge graph of equipment operating status, we can analyze the relationship between faults and various influencing factors, quickly locate the root cause of the fault, and provide guidance for equipment maintenance.

[0213] (IV) Data Management and Service Unit:

[0214] Data storage: Establish a distributed database to categorize and store data uploaded by edge computing modules, model training data, analysis results, etc.; use data compression and indexing technologies to improve data storage efficiency and query speed, and regularly back up data to ensure data security and integrity;

[0215] Data query and display: Provides a data query interface for system users, allowing users to query relevant data based on time, equipment type, parameter type, and other conditions; displays analysis results in the form of visual charts (such as line charts, bar charts, heat maps, etc.) and reports, allowing users to intuitively understand the equipment operating status and analysis results;

[0216] Compliance report generation: using the compliance report generator, automatically matching power industry standard clauses, conducting compliance checks on equipment operation data and analysis results, generating audit logs and compliance reports to meet power industry regulatory requirements;

[0217] III. Key technical principles:

[0218] (I) Multi-modal data fusion principle:

[0219] Multi-modal data fusion is a technology that combines multi-source heterogeneous data from different sensors to obtain more comprehensive and accurate information; feature-level fusion combines similar features by extracting features from each data to reduce data redundancy; decision-level fusion is based on independent decision results of each sensor to make a final decision through comprehensive judgment to improve the reliability and robustness of the decision; for example, by combining gas concentration sensor data and vibration sensor data, the fusion analysis can more accurately determine whether the equipment has a leak or mechanical failure;

[0220] (II) Deep learning model principle:

[0221] The improved multi-scale attention CNN model introduces an attention mechanism, enabling the model to automatically learn and focus on key features related to leak source positioning, improving positioning accuracy; in the LSTM-TCN hybrid architecture, LSTM can effectively handle long-term dependencies in time series data, and TCN's dilated convolution further expands the network's perception range of historical data, combining both to achieve accurate prediction of equipment risks; the federated learning optimization mechanism integrates multi-party data to update model parameters without sharing raw data, protecting data privacy while improving model performance;

[0222] (III) Knowledge graph root cause analysis principle:

[0223] The knowledge graph constructs a relationship network between entities and entities, representing various factors in the equipment operation process (such as device parameters, environmental conditions, fault phenomena, etc.) in a structured form; using the correlation degree calculation formula, the correlation between different entities is analyzed; when a device fails, the knowledge graph is used to quickly search for entities and relationships related to the failure, trace the root cause of the failure, and provide a basis for fault diagnosis and handling.

[0224] In this embodiment, the early warning and linkage control module is the core execution unit for risk response in the SF6 gas state intelligent diagnosis and early warning system based on the Internet of Things, like the "action center" of the system, which quickly responds and takes measures after the cloud platform intelligent analysis module outputs the diagnosis results; this module reduces the risk of SF6 gas equipment failure through hierarchical early warning and intelligent linkage, ensuring the safe and stable operation of the power system, which is specifically embodied in:

[0225] 1. Overview of overall functions:

[0226] This module receives the SF6 gas equipment operating status diagnosis results and risk level assessment information output by the cloud platform's intelligent analysis module. Based on preset rules and algorithms, it triggers warning signals of corresponding levels and notifies relevant personnel in a variety of ways. At the same time, based on the risk situation, it links equipment such as fans, gas supply devices, and access control systems to automatically execute emergency response operations, forming a closed-loop management from risk identification to disposal.

[0227] 2. Submodule composition and functions:

[0228] (1) Adaptive early warning strategy unit:

[0229] Warning level determination: Based on the gas leakage rate output by the cloud platform, the equipment importance level (such as the criticality of the equipment in the power grid, the scope of the fault impact, etc.), and the presence of personnel (obtained through the access control system or personnel positioning equipment), a multi-factor comprehensive evaluation algorithm is used to dynamically adjust the warning level. The warning level is divided into three levels: general warning, severe warning, and emergency warning, and different levels correspond to different risk levels.

[0230] Warning signal generation: For different warning levels, corresponding sound and light alarm signals, mobile terminal push information, and remote monitoring platform pop-up prompts are generated; general warnings use low-frequency sound and light alarms, and mobile terminals push brief risk warnings; severe warnings increase the frequency of sound and light alarms and push detailed risk analysis and response suggestions; emergency warnings trigger high-intensity sound and light alarms, and emergency warning information is pushed via SMS, phone calls, and other methods, and pop-up warnings are prominently displayed on the remote monitoring platform;

[0231] (2) Gas supply control logic unit:

[0232] Pressure monitoring and deviation calculation: Real-time acquisition of internal pressure sensor data of SF6 gas equipment, comparison with the normal operating pressure range of the equipment, and calculation of pressure deviation ; When the pressure is lower than the lower limit of the normal range, the air supply control process is started;

[0233] Improved PID-fuzzy composite control: using formula (in, for The proportional coefficient of the moment, is the initial proportional coefficient, is the adaptive coefficient, is the rate of change of pressure deviation) and for The integration coefficient at time, is the initial integration coefficient, is the adaptive coefficient, is the absolute value of the pressure deviation Dynamically adjust PID controller parameters; combined with fuzzy control rules, optimize the air supply rate and air supply volume of the air supply device in real time according to the pressure deviation and its rate of change, ensuring that the equipment pressure quickly returns to the normal range while avoiding excessive pressure fluctuations during the air supply process;

[0234] (3) Emergency ventilation mechanism unit:

[0235] Path planning: The fan ventilation path is planned based on the dynamic weight A* algorithm, and the cost function is

[0236] (in, For nodes The cost function value of From the starting node to the node The actual cost, is the time decay factor, is the attenuation coefficient, For time, as time goes by, path planning tends to reach the target quickly; For slave nodes Heuristic estimated cost to the target node, is the gas concentration influence coefficient, For nodes The gas concentration at is the maximum allowable gas concentration); comprehensively consider the ventilation path length, time factors and gas concentration distribution, calculate the optimal ventilation path, so that the fan can efficiently disperse the leaked SF6 gas and reduce the gas concentration in the environment;

[0237] Fan control: Controls the start and stop, speed, and wind direction of fans based on the planned path and gas concentration. When a high-concentration SF6 gas leak is detected, multiple fans are activated to work together to ventilate along the planned path, quickly reducing gas concentrations in hazardous areas and ensuring personnel safety and normal equipment operation.

[0238] (IV) Access control linkage control unit:

[0239] Personnel control: When the system issues a serious or emergency warning, it links to the access control system to restrict non-essential personnel from entering the area where the SF6 gas equipment is located, preventing people from being poisoned by leaking gas. At the same time, the access control system records personnel entry and exit information to facilitate subsequent safety management and accident tracing.

[0240] Emergency access management: While restricting access to personnel, ensure that emergency rescue access is unobstructed, providing assurance for professional maintenance personnel and rescue teams to quickly enter the scene; through the permission setting and automatic control of the access control system, the emergency access can be flexibly opened and closed;

[0241] (V) Information interaction and feedback unit:

[0242] Information sending: real-time sending of early warning information, device linkage status, etc. to the mobile terminal of the operation and maintenance personnel, the remote monitoring platform and the related management department system, to ensure that all parties can timely understand the device risk situation and the progress of emergency disposal;

[0243] Feedback receiving and processing: receiving the operation instructions, feedback information and on-site situation report sent by the operation and maintenance personnel through the mobile terminal or monitoring platform, analyzing and processing the instructions, and adjusting the early warning and linkage control strategy; for example, when the operation and maintenance personnel confirm the on-site situation, they can manually cancel the early warning or adjust the device linkage parameters.

[0244] In this embodiment, the energy management module is the "heart" of the SF6 gas state intelligent diagnosis and early warning system based on the Internet of Things, which undertakes the important mission of providing reliable energy guarantee for the stable operation of the system; in complex application scenarios such as substations, the module realizes multi-mode power supply through the cooperative work of the solar power supply unit and the intelligent power distribution circuit, effectively solves the problem of energy supply in the system operation process, and ensures the continuous and stable operation of the system under various environmental conditions, which is specifically embodied in:

[0245] I. Overall function overview:

[0246] The energy management module mainly consists of a solar power supply unit and an intelligent power distribution circuit; the solar power supply unit uses photovoltaic power generation technology to convert solar energy into electrical energy, providing green and renewable energy for the system; the intelligent power distribution circuit dynamically allocates electrical energy based on the load priority according to the power consumption demand of each module of the system, and at the same time has the functions of power state monitoring, switching and management, to ensure the power supply of the system under different working conditions and improve the energy utilization efficiency and reliability of the system;

[0247] II. Submodule composition and function:

[0248] (1) Solar power supply unit:

[0249] Photovoltaic module: high-efficiency monocrystalline silicon photovoltaic modules are selected, with a photoelectric conversion efficiency of more than 22%, and with characteristics such as high conversion efficiency, strong stability and long service life; according to the power consumption demand of the system and the installation environment, the installation number and layout of the photovoltaic module are reasonably planned to ensure that under good light conditions, solar energy can be maximally absorbed and converted into electrical energy;

[0250] Maximum power point tracking (MPPT) module: advanced MPPT algorithm is adopted, and its optimization formula is

[0251] (where, Vmpis the optimal working voltage of the photovoltaic cell, which is the voltage when the output power of the photovoltaic cell reaches the maximum value; Iscis the output current of the photovoltaic cell; Vocis the terminal voltage of the photovoltaic cell; Rs is the equivalent series resistance, representing the resistance loss inside the photovoltaic cell: η is the fill factor, which represents the maximum value of the expression in the parentheses; The MPPT module monitors the output voltage and current of the photovoltaic module in real time, dynamically adjusts the working point, and makes the photovoltaic module always work in the maximum power output state, thereby improving the utilization efficiency of solar energy. When the intensity of light, temperature and other environmental factors change, the MPPT module can quickly respond to ensure that the photovoltaic module stably outputs the maximum power;

[0252] The charge and discharge management module is responsible for charging and discharging control and management of the battery. When the output power of the photovoltaic module is sufficient, the excess power is stored in the battery, and intelligent charging strategies such as constant current charging, constant voltage charging, floating charging and other phased charging methods are adopted to ensure safe and efficient charging of the battery and prolong the service life of the battery. When the system power demand is greater than the output of the photovoltaic module, the charge and discharge management module controls the battery to supply power to the system, and at the same time monitors the battery power state in real time. When the power is lower than the set threshold, a low power warning is issued to avoid excessive discharge damage to the battery;

[0253] (II) Intelligent power distribution circuit:

[0254] Load priority division: according to the urgency and importance of power supply demand of each module in the system, the load is divided into different priorities; for example, the sensor array module, the edge computing module and other key functions that undertake data collection and preliminary processing are set as high-priority loads; while some auxiliary function modules, such as part of the display devices, are set as low-priority loads; through clear priority division, the normal operation of key modules is ensured in the case of insufficient power supply;

[0255] Dynamic allocation of electric energy: based on the load priority, the electric energy is dynamically allocated, and the formula is used (wherein, Pdis the electric energy allocated to the device ; wdis the priority weight of the device ; the higher the weight, the higher the priority of the device, and the greater the proportion obtained in the electric energy allocation: Pdis the electric energy demand of the device ; wdis the sum of the priority weights of all devices; ​​For total energy); intelligent power distribution circuit real-time monitoring system total energy and the energy demand of each load, according to the priority weight calculation and distribution of electric energy, realize the rational use of electric energy, avoid the overuse of electricity of some modules leading to other key modules power supply shortage;

[0256] Power switching and management: with multiple power input interfaces, can access solar power, mains power and other power; When solar power is insufficient or fails, the intelligent power distribution circuit automatically detects and switches to backup power (such as mains power), ensuring uninterrupted operation of the system; At the same time, the state of each power supply is monitored in real time, including voltage, current, frequency and other parameters. Once an abnormal power supply is found, appropriate measures such as alarm, automatic power switching, etc. are taken in time, and power supply state information is recorded for subsequent maintenance and troubleshooting.

[0257] In this embodiment, the self-checking and fault-tolerant module is the "safety guard" for the stable operation of the SF6 gas state intelligent diagnosis and early warning system based on the Internet of Things. It ensures the continuous and reliable work of the system in complex environments through real-time monitoring and fault handling of each key part of the system. From the whole to the details, the module will be described in detail as follows:

[0258] I. Overall function overview:

[0259] The self-checking and fault-tolerant module is mainly responsible for periodic detection of sensor accuracy, communication link integrity, and power supply state in the system, and timely detection of potential faults or abnormal conditions. Once a fault is detected, fault isolation and redundancy switching operations are performed quickly to ensure that the core functions of the system are not affected. At the same time, through data recording and analysis, it provides a basis for the maintenance and optimization of the system, and improves the overall robustness and reliability of the system;

[0260] II. Submodule composition and function:

[0261] (1) Multi-index fusion diagnosis unit:

[0262] Data acquisition and preprocessing: real-time acquisition of output data of sensor array module, signal state data of anti-interference communication module and power supply parameter data of energy management module, etc.; Filtering, noise reduction and other preprocessing operations are performed on the collected data to remove noise and interference in the data, ensuring the accuracy and effectiveness of the input data;

[0263] Multi-index fusion diagnosis algorithm: formula

[0264] (where, is the diagnosis result vector, each element corresponds to the probability of a fault type; is the number of monitoring indicators; is the The weight of an indicator reflects the importance of the indicator to fault diagnosis; It is a modified linear unit activation function used to introduce nonlinear relationships; For the The coefficient of the indicator; For the The actual measured value of each indicator; For the The algorithm comprehensively analyzes the correlation between multiple indicators to determine whether the system has a fault and the type of fault. For example, when sensor data is abnormal and the communication link signal strength decreases, the algorithm calculates the fault diagnosis result and determines that there may be a sensor fault or a communication link fault.

[0265] Weight update mechanism: In order to make the diagnosis algorithm adapt to the changes in the system operation process, the weight update formula is adopted

[0266] (in, For the The update amount of each indicator weight; is the learning rate, which controls the step size of weight update; is the loss function L about the Index weight The partial derivative of is used to measure the influence of the weight on the diagnosis result; is the entropy weight coefficient, which is used to balance the uncertainty of the diagnosis results; is the diagnosis result vector Entropy reflects the uncertainty of the diagnosis results); dynamically adjust the weight of each indicator according to the actual fault situation and diagnosis results to improve the accuracy and adaptability of the diagnosis algorithm;

[0267] (2) Fault recovery mechanism unit:

[0268] Fault detection and location: When the multi-indicator fusion diagnostic unit detects a system fault, it further analyzes and locates the fault in detail. Through in-depth investigation of sensor data, communication link status, power supply parameters, and other information, the specific location and cause of the fault are determined. For example, when abnormal sensor data is detected, the system compares data from multiple sensors with historical data to determine whether it is a single sensor fault or a sensor network fault.

[0269] Fault isolation: Once the fault location is determined, fault isolation is performed immediately. For sensor failures, the faulty sensor is disconnected from the system to prevent data from the faulty sensor from affecting the normal operation of the system. For communication link failures, the faulty link segment is isolated to prevent the fault from spreading to the entire communication network. For power supply failures, the faulty power module is isolated to prevent damage to other modules.

[0270] Redundancy switching: while fault isolation, start the redundancy switching mechanism; for key components such as sensors, communication modules, etc., the system is equipped with redundant devices; when the main device fails, automatically switch to the redundant device to ensure that the core functions of the system are not affected; for example, when the main sensor fails, the standby sensor automatically takes over the data acquisition task to ensure that the system can continuously obtain accurate monitoring data;

[0271] Local storage-forwarding mode: when communication interruption is detected, automatically switch to local storage-forwarding mode, using the formula

[0272] (where, is the amount of data stored in the local cache; is the upper limit of the edge storage capacity; is the time of communication interruption; is the current time; is the amount of data generated at the time; is the total amount of data generated from the time of communication interruption to the current time ); During the communication interruption, the data collected by the sensor is temporarily stored in the local cache, and when the communication is restored, the data in the cache is forwarded to the cloud platform intelligent analysis module according to the time sequence of data generation, ensuring the integrity and continuity of the data;

[0273] (Three) System state monitoring and recording unit:

[0274] Real-time state monitoring: continuously monitor the running state of each module of the system in real time, including the working state of the sensor, the connection state of the communication link, the output state of the power supply, etc.; Through real-time monitoring of these state information, abnormal conditions in the system running process are found in time, providing timely data support for fault diagnosis and processing;

[0275] Data recording and storage: record and store key data in the system running process, such as sensor acquisition data, fault diagnosis results, fault occurrence time and location, etc. in the local database; These data can not only be used for fault analysis and tracing, but also provide data basis for system optimization and improvement;

[0276] State report generation: generate system state report regularly, summarize the running situation of the system in a certain period of time, including fault occurrence times, fault types, fault handling conditions, etc. Send the state report to the system management personnel and maintenance personnel to facilitate them to understand the running condition of the system and formulate corresponding maintenance plan and optimization strategy.

[0277] While embodiments of the application have been shown and described, it is to be understood that the embodiments described are merely exemplary of the principles and application of the present application. Numerous modifications and adaptions can be effected without departing from the spirit and scope of the present application, which is not limited to the exact construction and arrangement described. It is intended, therefore, to cover all modifications and adaptions that fall within the scope of the claims and their equivalents.

Claims

1. The SF6 gas status intelligent diagnosis and early warning system based on the Internet of Things is characterized by: include: Sensor array module: Deployed at key monitoring points of SF6 gas equipment, including SF6 gas concentration sensors, temperature and humidity sensors, pressure sensors, trace water content sensors, and vibration sensors, to collect gas state parameters and equipment operating environment data in real time; Edge computing module: Integrated in the local terminal, it is used to pre-process gas state parameters and equipment operating environment data and generate preliminary diagnostic results based on a dynamic baseline model. The dynamic baseline model is constructed in the following way: Establish dynamic threshold ranges for SF6 gas concentration, pressure, and trace water content based on historical data and equipment operating parameters; Introducing environmental coupling factors, combining equipment vibration data with local meteorological information, and dynamically adjusting the early warning sensitivity of the baseline model; An improved hybrid time series prediction algorithm is used, and its modeling method is as follows: The prediction results of the ARIMA model are weighted and fused with the seasonal and trend components of the Prophet algorithm. The weight coefficient α ranges from 0.6 to 0.

8. The environmental variable correction term is superimposed, and the coupling coefficient β of the correction term is dynamically calculated using the ridge regression algorithm. Establish a dual residual feedback mechanism: Calculate the first-level residual between the real-time monitoring value and the model prediction value, decompose the second-level residual through wavelet transform, and inject the second-level residual back into the model input layer for parameter fine-tuning; Anti-interference communication module: Using LoRa wireless communication technology, integrating forward error correction coding FEC and orthogonal frequency division multiple access OFDMA modulation technology, the anti-interference communication module includes: The LoRa channel encoder based on convolutional code has the following encoding process: the information bit polynomial is modulo-operated with the preset generator polynomial to generate coded data with redundant check bits; Adaptive frequency hopping mechanism, which selects frequency points by allocating available frequency band resources based on the real-time signal strength RSSI ratio of each frequency band and selecting the communication band with the least interference; Encrypted transmission protocol, using AES-256 algorithm to perform end-to-end encryption of sensor data and control instructions; The cloud platform intelligent analysis module receives data uploaded by the edge computing module, analyzes gas leakage trends, equipment aging status, and environmental coupling effects through a multimodal data fusion algorithm, and predicts the location of the leakage source and risk level based on a deep learning model. The deep learning model includes: Leakage source localization sub-model: This model uses an improved multi-scale attention convolutional neural network to dynamically calculate the attention weight matrix of each convolution kernel through a gated recurrent unit, and performs weighted fusion of convolution feature maps of different scales. Risk prediction sub-model: This model uses a hybrid architecture of LSTM and temporal convolutional networks. The temporal convolutional network captures multi-scale temporal features through an exponentially growing dilation factor. Federated learning optimization mechanism: weighted aggregation of local model parameters at each site, with the weight value inversely proportional to the privacy protection strength of the site data distribution; Early warning and linkage control module: triggers graded early warning signals based on the location of the leak source and the risk level, and links the fan, air supply device and access control system to perform emergency response. The early warning and linkage control module includes: Adaptive warning strategy: Dynamically adjust the warning level based on the leakage rate, equipment importance level and personnel presence; Air supply control logic: PID-fuzzy composite controller is used, and its parameter setting method is as follows: The proportional coefficient Kp is dynamically adjusted according to the differential value of the pressure deviation, and the integral coefficient Ki is adaptively increased according to the absolute value of the pressure deviation; Emergency ventilation mechanism: Based on a dynamic weighted path planning algorithm, the path cost is composed of the actual movement cost, the estimated residual cost, and the gas concentration penalty term, where the weight of the concentration penalty term decays exponentially over time; Energy management module: includes solar power supply unit and intelligent power distribution circuit, providing multi-mode power supply guarantee for the system; Self-check and fault-tolerance module: Periodically checks sensor accuracy, communication link integrity, and power status, and performs fault isolation and redundancy switching.

2. The SF6 gas status intelligent diagnosis and early warning system based on the Internet of Things according to claim 1 is characterized by: The energy management module includes: Photovoltaic power supply unit: uses the maximum power point tracking algorithm to minimize the equivalent series resistance loss by optimizing the output voltage of the photovoltaic array; Intelligent power distribution circuit: Total power is distributed proportionally based on the device priority weight, and devices with higher priority are given priority in obtaining power resources.

3. The SF6 gas status intelligent diagnosis and early warning system based on the Internet of Things according to claim 1 is characterized by: The self-checking and fault-tolerant module includes: Multi-index fusion diagnosis algorithm: The difference between the sensor reading and the calibration coefficient and failure threshold is activated by a modified linear unit, and then dynamically weighted to output the fault probability distribution; Fault recovery mechanism: When communication is interrupted, it automatically switches to local cache mode, and the amount of stored data does not exceed the maximum capacity of the edge device.

Images