Intelligent toxic gas early warning platform based on sensor monitoring

Abstract

The invention provides a toxic gas intelligent early warning platform based on sensor monitoring, and relates to the technical field of gas monitoring. An initial two-dimensional material is set through a two-dimensional material setting unit; the doping module acquisition unit acquires a doping module; the doping processing unit outputs a doped two-dimensional material sample according to the doping module, sensor simulation is carried out through the sensor structure simulation platform, and a plurality of simulation sensors are output; the toxic gas sample collection unit collects toxic gas samples of a target monitoring environment; the adsorption capacity evaluation unit is used for evaluating the toxic gas adsorption capacity of the plurality of simulation sensors to obtain a preferable sensor; and the toxic gas early warning unit performs toxic gas early warning on the target monitoring environment according to the preferred sensor. The technical problem that the accuracy of the monitoring result is limited due to the fact that the configuration of the sensor cannot be adjusted according to the monitoring environment adaptability in existing toxic gas monitoring and early warning is solved, and the accuracy and the environment adaptability of toxic gas early warning are improved.

Description

Technical Field

[0001] The present application relates to the field of gas monitoring technology, and in particular to an intelligent early warning platform for toxic gases based on sensor monitoring. Background Art

[0002] With the acceleration of industrialization and urbanization, toxic gas leakage accidents occur frequently, posing a serious threat to human health and the ecological environment. Therefore, the development of an efficient and accurate toxic gas monitoring and early warning system has become an urgent need.

[0003] Existing toxic gas monitoring and early warning methods are mostly based on traditional sensor technologies such as metal oxide sensors, semiconductor sensors, and electrochemical sensors. These methods usually use conventional materials to make sensors, and the sensors are directly installed in the monitoring environment for gas detection. In terms of data processing, they rely on simple threshold judgments and other methods for early warning. However, traditional sensors have poor sensitivity and selectivity, are not sensitive to certain low-concentration toxic gases, and have difficulty distinguishing gases of similar properties. When multiple gases coexist, cross-interference is prone to occur, which can easily lead to misjudgment. Secondly, existing sensors have poor adaptability to environmental conditions. Under complex and changeable environmental conditions, such as high temperature, high humidity, or highly corrosive environments, the sensor performance is unstable, affecting the accuracy of the monitoring results. Summary of the invention

[0004] The present application provides a toxic gas intelligent early warning platform based on sensor monitoring, which solves the technical problem that the existing toxic gas monitoring and early warning are mostly based on conventional detection by traditional sensors, and the sensor configuration cannot be adjusted according to the adaptability of the monitoring environment, resulting in limited accuracy of the monitoring results. The technical effect of improving the accuracy and environmental adaptability of toxic gas early warning is achieved.

[0005] In view of the above problems, the present application provides a toxic gas intelligent early warning platform based on sensor monitoring, the platform comprising: a two-dimensional material setting unit, the two-dimensional material setting unit is used to construct a sensor structure simulation platform, and the initial two-dimensional material is set according to the sensor structure simulation platform; a doping module acquisition unit, the doping module acquisition unit is used to acquire a doping module, and the doping module is connected to the sensor structure simulation platform; a doping processing unit, the doping processing unit is used to perform doping processing on the initial two-dimensional material according to the doping module, and output a doped two-dimensional material sample, the sensor structure simulation platform performs sensor simulation based on the doped two-dimensional material sample, and outputs multiple simulated sensors; a toxic gas sample collection unit, the toxic gas sample collection unit is used to determine the target monitoring environment and collect toxic gas samples of the target monitoring environment; an adsorption capacity evaluation unit, the adsorption capacity evaluation unit is used to evaluate the toxic gas adsorption capacity of the multiple simulated sensors according to the toxic gas samples, and obtain a preferred sensor, wherein the preferred sensor is a sensor with the largest toxic gas adsorption index; a toxic gas early warning unit, the toxic gas early warning unit is used to perform a toxic gas early warning on the target monitoring environment according to the preferred sensor.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The two-dimensional material setting unit constructs a sensor structure simulation platform, sets the initial two-dimensional material according to the sensor structure simulation platform, and provides an initial model and material basis for the subsequent sensor design. The doping module acquisition unit acquires the doping module, and the doping module is connected to the sensor structure simulation platform. The doping processing unit performs doping treatment on the initial two-dimensional material according to the doping module, changes the electronic structure, chemical activity and other properties of the material, and outputs a doped two-dimensional material sample. The sensor structure simulation platform performs sensor simulation based on the doped two-dimensional material sample, outputs multiple simulated sensors, improves sensor performance, and provides a diversified selection basis. The toxic gas sample collection unit determines the target monitoring environment and collects toxic gas samples of the target monitoring environment, so as to facilitate the subsequent screening of suitable sensors according to the types and characteristics of toxic gases in a specific environment, and improve the pertinence, adaptability and accuracy of the early warning in the actual environment. The adsorption capacity evaluation unit evaluates the toxic gas adsorption capacity of the multiple simulated sensors according to the toxic gas samples, obtains the preferred sensor with the largest toxic gas adsorption index, improves the accuracy of sensor selection and environmental pertinence, and thus improves the accuracy of the early warning. The toxic gas early warning unit performs a toxic gas early warning on the target monitoring environment according to the preferred sensor.

[0008] In summary, this application significantly improves the accuracy and flexibility of the toxic gas monitoring system by introducing two-dimensional materials and doping technology, combined with sensor structure simulation and intelligent early warning mechanism. This solution optimizes sensor performance through two-dimensional materials and doping treatment, and enhances the detection ability of various toxic gases. At the same time, combined with the target environmental samples and simulation evaluation mechanism for precise screening, the sensor can be optimized according to the specific monitoring environment, and the optimal sensor can be automatically selected, which greatly improves the environmental adaptability of toxic gas monitoring, so that accurate toxic gas detection and early warning can be achieved under different gas concentrations and environmental conditions, ensuring environmental safety and personnel health.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A schematic diagram of the structure of a toxic gas intelligent early warning platform based on sensor monitoring provided in an embodiment of the present application.

[0011] Figure 2 A schematic diagram of the process of providing toxic gas warning for a target monitoring environment according to a preferred sensor in a sensor-based toxic gas intelligent warning platform provided in an embodiment of the present application.

[0012] Figure 3 A schematic diagram of the process of evaluating the toxic gas adsorption capacity in the toxic gas intelligent early warning platform based on sensor monitoring provided in an embodiment of the present application.

[0013] Explanation of the reference numerals: two-dimensional material setting unit 10 , doping module acquisition unit 20 , doping processing unit 30 , toxic gas sample collection unit 40 , adsorption capacity evaluation unit 50 , toxic gas early warning unit 60 . DETAILED DESCRIPTION

[0014] The embodiment of the present application provides a toxic gas intelligent early warning platform based on sensor monitoring. By constructing a sensor structure simulation platform, utilizing the characteristics of two-dimensional materials, and obtaining a variety of simulated sensors through doping treatment, the simulated sensors are then evaluated for adsorption capacity in combination with toxic gas samples in the target monitoring environment, and finally the preferred sensor with the largest adsorption index is selected to provide toxic gas early warning for the target environment. The embodiment of the present application solves the technical problem that the existing toxic gas monitoring and early warning are mostly based on conventional detection by traditional sensors, and the sensor configuration cannot be adjusted according to the adaptability of the monitoring environment, resulting in limited accuracy of the monitoring results, and achieves the technical effect of improving the accuracy of toxic gas early warning and environmental adaptability.

[0015] like Figure 1 As shown, the embodiment of the present application provides a toxic gas intelligent early warning platform based on sensor monitoring, and the platform includes:

[0016] The two-dimensional material setting unit 10 is used to construct a sensor structure simulation platform and set initial two-dimensional materials according to the sensor structure simulation platform.

[0017] Specifically, two-dimensional materials refer to materials with atomic or molecular thickness, usually extending on a 2D plane. Common two-dimensional materials include graphene, germanium, boron nitride, molybdenum disulfide, tungsten disulfide, etc. These materials have ultra-high specific surface area, excellent electrical conductivity and strong chemical reactivity.

[0018] First, a sensor structure simulation platform is constructed using material simulation software (such as Materials Studio, etc.) and computer-aided design (CAD) software. This platform is established through computer simulation technology, which simulates and predicts the structural characteristics and performance of different sensors based on physical and chemical laws, and helps design and optimize sensor components. According to the design goals of the sensor, such as the type of gas to be detected, sensitivity requirements, etc., a suitable two-dimensional material is selected from the known two-dimensional material library as the initial two-dimensional material, which is the basic material for subsequent operations such as sensor performance optimization. For example, to design a sensor with high sensitivity to nitrogen dioxide detection, graphene with strong electronic conductivity and high surface area may be selected as the initial two-dimensional material.

[0019] The doping module acquisition unit 20 is used to acquire a doping module, and the doping module is connected to the sensor structure simulation platform.

[0020] Specifically, the doping module acquisition unit 20 selects and acquires a doping module from a pre-designed module library, which can be stored locally or connected to a remote database via a network. During the acquisition process, a suitable doping module is selected according to the requirements of the sensor structure simulation platform and the characteristics of the initial two-dimensional material. Among them, the doping module is a tool containing different doping elements, which is connected to the sensor structure simulation platform. By selecting doping elements, the chemical properties of the two-dimensional material can be changed, thereby optimizing the performance of the material. For example, if the initial two-dimensional material is graphene, and it is hoped to improve its adsorption performance for carbon monoxide gas, then a doping module capable of doping graphene with specific atoms (such as nitrogen atoms) will be obtained from the module library, thereby improving the sensitivity to carbon monoxide.

[0021] The doping processing unit 30 is used to perform doping processing on the initial two-dimensional material according to the doping module and output a doped two-dimensional material sample. The sensor structure simulation platform performs sensor simulation based on the doped two-dimensional material sample and outputs multiple simulated sensors.

[0022] Specifically, in the doping processing unit 30, the doping module provides a specific doping scheme, and the unit is responsible for introducing dopants into the initial two-dimensional material according to these schemes and outputting a doped two-dimensional material sample. The surface properties and electronic structure of the doped two-dimensional material will change, thereby improving the material's responsiveness to toxic gases. For example, graphene is used as the initial two-dimensional material, and in order to improve its detection sensitivity to ammonia, nitrogen is selected to be doped. After nitrogen doping, the electronic structure of graphene changes, increasing the active sites of nitrogen atoms, thereby enhancing the interaction between ammonia molecules and the surface of the material. This doping treatment can significantly improve the sensor's adsorption and detection sensitivity to ammonia.

[0023] After the doping process is completed, the doped two-dimensional material samples are sent to the sensor structure simulation platform for further performance simulation. The simulation platform analyzes the performance of the doped material samples in different gas environments to verify whether the doped materials have achieved the expected effect, such as improving the sensitivity or selectivity to certain toxic gases. Through the sensor structure simulation platform, multiple virtual sensor models, that is, multiple simulated sensors, are generated. Each simulated sensor is based on different doped material samples and design schemes, and its performance characteristics are different, providing a variety of selectable samples for subsequent screening.

[0024] The toxic gas sample collection unit 40 is used to determine a target monitoring environment and collect toxic gas samples of the target monitoring environment.

[0025] Specifically, the target monitoring environment refers to the specific area or scene that needs to be monitored, such as industrial plants, chemical laboratories, underground mines, urban air, etc. By collecting toxic gas samples in these environments, real test data can be provided for sensor performance evaluation.

[0026] The toxic gas sample collection unit 40 first needs to determine the target monitoring environment, that is, to clarify in which specific environments gas detection is performed. The selection of the target environment can be based on the risk assessment of gas leakage, or can be determined according to the type of gas to be detected. For example, if the goal is to monitor chlorine leakage in a factory, the target monitoring environment is the working area of ​​the factory. Then, collect toxic gas samples in the target environment. The collection of gas samples can be carried out through special gas sampling equipment, such as gas sampling bags, gas detection tubes and other equipment, to obtain samples of toxic gases in the environment. These samples will provide basic data for subsequent performance evaluation and sensor optimization.

[0027] The adsorption capacity evaluation unit 50 is used to evaluate the toxic gas adsorption capacity of the multiple simulation sensors according to the toxic gas sample to obtain a preferred sensor, wherein the preferred sensor is a sensor with the largest toxic gas adsorption index.

[0028] Specifically, in gas sensors, the adsorption of gas molecules on the surface of the material determines the sensitivity of the sensor to the gas. Materials with strong adsorption capacity can detect the presence of gas at lower concentrations. The toxic gas adsorption index is a quantitative indicator used to measure the adsorption capacity of analog sensors for toxic gases. For example, adsorption amount, adsorption rate, etc. can be used as components of the adsorption index.

[0029] The collected toxic gas samples are converted into simulated data, and multiple simulated sensors are simulated and tested on the sensor structure simulation platform. The same environmental conditions (such as the same temperature, pressure and humidity) are set to detect the adsorption capacity of each simulated sensor for toxic gas samples, and the corresponding toxic gas adsorption index is calculated. Then these indexes are compared to find the largest toxic gas adsorption index, and the simulated sensor corresponding to the largest index is determined as the preferred sensor. This sensor theoretically has the best adsorption capacity for toxic gases in the target monitoring environment, and is an ideal choice for subsequent actual toxic gas warnings.

[0030] A toxic gas warning unit 60 is used to provide a toxic gas warning for the target monitoring environment according to the preferred sensor.

[0031] Specifically, the configuration data of the determined preferred sensor is obtained from the sensor structure simulation platform, the sensor is actually deployed in the target monitoring environment according to the configuration data, and the real-time gas concentration detection is performed in the target monitoring environment. The toxic gas early warning unit 60 is connected to the deployed sensor, obtains the real-time gas concentration data collected by the sensor in real time, and then compares the real-time received data with the preset safety threshold. If the gas concentration exceeds the set safety limit, the toxic gas early warning unit 60 will immediately send out an alarm signal, such as an audible and visual alarm, a text message or email notification, a system display warning, etc., to remind relevant personnel to take appropriate countermeasures (such as ventilation, personnel evacuation, etc.) to ensure the safety of the staff and reduce the occurrence of accidents.

[0032] Further, such as Figure 2 As shown, the poisonous gas warning unit 60 in the embodiment of the present application is also used to perform the following steps:

[0033] Step P61: deploy multiple preferred sensors in the target monitoring environment.

[0034] Step P62: Construct an intelligent monitoring network based on the multiple preferred sensors.

[0035] Step P63: Use the intelligent monitoring network to monitor the target monitoring environment in real time and output a real-time sensing data set of toxic gases.

[0036] Step P64: Input the toxic gas real-time sensing data set into the toxic gas risk identification model, and output a first risk indicator according to the toxic gas risk identification model.

[0037] Step P65: When the first risk indicator is greater than the preset risk indicator, a first warning signal is generated.

[0038] Specifically, the toxic gas risk identification model is a model based on big data analysis and machine learning. It can analyze the input gas concentration data, comprehensively consider information such as gas concentration, time, and environmental factors, identify the risk of toxic gas leakage in the environment, and output a quantitative risk indicator. The first risk indicator is a quantitative indicator output by the toxic gas risk identification model based on the real-time sensing data set of toxic gases, which is used to indicate the risk level of toxic gases in the current target monitoring environment monitored by the preferred sensor. The preset risk indicator is a pre-set critical value of the risk indicator. When the first risk indicator exceeds this value, it means that the current toxic gas risk has reached the level that requires early warning. The first early warning signal is an alarm signal triggered when the first risk indicator is greater than the preset risk indicator, which is used to prompt that the risk of toxic gases in the target monitoring environment has exceeded the safety range and corresponding measures need to be taken.

[0039] According to the layout, area, possible distribution of toxic gas leakage sources and other factors of the target monitoring environment, multiple preferred sensors are arranged in the target monitoring environment. These sensors cover different parts of the monitoring area in a distributed manner. The multiple preferred sensors arranged are connected to each other through wired or wireless communication, and connected to the central processing system to build an intelligent monitoring network, and the real-time monitoring data is uploaded to the toxic gas early warning unit 60. During the construction process, each sensor needs to be numbered and the network parameters configured so that the data transmission of each sensor can be accurately identified and managed.

[0040] The intelligent monitoring network continuously monitors the target monitoring environment in real time and outputs a real-time sensor data set about the toxic gases in the environment, including the concentration, time, location and other information of the toxic gases detected by each sensor, to ensure comprehensive monitoring of the entire target environment.

[0041] The toxic gas data set collected in real time is input into the toxic gas risk identification model for analysis. Based on the data provided by the sensor and environmental factors such as temperature, humidity, and gas diffusion, the model outputs a quantitative first risk indicator to indicate the risk level of toxic gas leakage in the environment.

[0042] The first risk indicator is compared with the preset risk indicator. If the first risk indicator is greater than the preset risk indicator, a first warning signal is generated according to a predetermined rule. This warning signal can be in the form of sound and light alarm, email notification, SMS alarm, etc., to remind relevant personnel of the risk of toxic gas leakage and the need to take emergency measures as soon as possible.

[0043] Further, such as Figure 3 As shown, the adsorption capacity evaluation unit 50 in the embodiment of the present application is also used to perform the following steps:

[0044] Step P51: training an adsorption capacity evaluation model based on the toxic gas sample.

[0045] Step P52: Download the trained adsorption capacity evaluation model to the sensor structure simulation platform.

[0046] Step P53: According to the adsorption capacity evaluation model, a plurality of key adsorption capacity indicators of each simulated sensor based on the toxic gas sample are obtained, including at least electronic structure change characteristics and adsorption energy characteristics.

[0047] Step P54: fitting is performed according to the multiple adsorption capacity key indicators to obtain multiple toxic gas adsorption indicators corresponding to the multiple simulated sensors, and the preferred sensor with the largest toxic gas adsorption indicator is selected.

[0048] Specifically, first collect a large amount of toxic gas sample data in the target monitoring environment. These data should contain information such as the type, concentration, molecular structure of the toxic gas, and various parameters related to adsorption under different conditions. Then select a suitable model structure, and build an adsorption capacity evaluation model based on machine learning algorithms such as neural networks (such as multi-layer perceptrons) or support vector machines. The collected toxic gas sample data is divided into a training set and a validation set. The model is trained using the training set. By adjusting the parameters of the model (such as weights and biases in the neural network), the model can accurately predict the results related to the adsorption capacity based on the input toxic gas sample characteristics. During the training process, the validation set is used to evaluate the performance of the model to prevent overfitting. When the error of the model on the validation set tends to be stable, the training is stopped and the trained adsorption capacity evaluation model is output.

[0049] The trained adsorption capacity evaluation model is downloaded to the sensor structure simulation platform. The downloaded model will be used in conjunction with the sensor structure modeling tools and simulation tools on the platform for subsequent simulation and evaluation work.

[0050] On the sensor structure simulation platform, input the toxic gas sample into the downloaded adsorption capacity evaluation model. The model analyzes each simulated sensor based on the input toxic gas sample and obtains multiple key adsorption capacity indicators based on the sample, such as obtaining the electronic structure change characteristics by calculating the quantum mechanical parameters when the simulated sensor interacts with the toxic gas molecules, and obtaining the adsorption energy characteristics by energy calculation methods (such as density functional theory calculations). Among them, the electronic structure change characteristics refer to the changes in the electronic structure of the material when the gas molecules interact with the sensor material. The change in electronic structure is often directly related to whether the gas molecules are easily adsorbed and the strength of the adsorption. The adsorption energy characteristics refer to the energy released or absorbed when the sensor material adsorbs the gas. The higher the adsorption energy, the stronger the material's adsorption capacity for the gas.

[0051] For each simulated sensor, a fitting operation is performed according to its corresponding multiple key adsorption capacity indicators. The fitting method can be based on mathematical methods such as multivariate linear regression and polynomial fitting to construct a functional relationship, and these key indicators are combined to obtain a toxic gas adsorption index. The toxic gas adsorption indicators of all simulated sensors are compared, and the simulated sensor with the largest adsorption index value is selected. According to the simulated sensor, the actual preferred sensor is configured for subsequent performance evaluation and application.

[0052] By training the adsorption capacity evaluation model, the adsorption capacity of multiple simulated sensors is evaluated and screened, and the optimal sensor with the best performance in the target monitoring environment is determined to achieve accurate monitoring of toxic gases in the target monitoring environment, thereby improving the accuracy of early warning.

[0053] Furthermore, the adsorption capacity evaluation model further includes an external condition adjustment network. The adsorption capacity evaluation unit 50 of the embodiment of the present application further includes the following steps of training the adsorption capacity evaluation model:

[0054] Step P511: configuring multiple external condition factors, wherein the multiple external condition factors include stress magnitude, external electric field strength and operating temperature.

[0055] Step P512: Establish a conversion relationship between the multiple external condition factors and the toxic gas adsorption index, perform network model training based on the conversion relationship, output the external condition adjustment network, and use the external condition adjustment network to feedback optimize the adsorption capacity evaluation model.

[0056] Specifically, the adsorption capacity evaluation model also includes an external condition adjustment network, which is used to predict the impact of external condition factors on the toxic gas adsorption index. Among them, the external condition factors are external factors that affect the ability of the simulated sensor to adsorb toxic gases, including stress magnitude, external electric field strength, and operating temperature. The stress magnitude will change the microstructure of the sensor, thereby affecting the adsorption site; the external electric field strength will affect the polarization degree of the gas molecules and the electrical properties of the sensor, thereby affecting the adsorption process; the operating temperature will change the movement speed of the gas molecules and affect the interaction between the gas and the material surface.

[0057] According to the target monitoring environment and the characteristics of the analog sensor, determine the range of external condition factors such as stress magnitude, external electric field strength and operating temperature. For example, if the analog sensor is used for monitoring a chemical plant in a high-temperature environment, the range of the operating temperature may be (300K, 800K); if the sensor works in an environment with electric field interference, the range of the external electric field strength may be set to (0V / m, 1000V / m) according to the actual possible electric field strength.

[0058] Collect data related to external conditions (stress, external electric field strength and operating temperature) and the adsorption capacity of the sensor. These data usually come from experimental tests, simulations or historical data. Analyze the collected data through data visualization, correlation analysis and other methods to find out the possible correlation patterns between external condition factors and toxic gas adsorption indicators, and establish the conversion relationship between external condition factors and toxic gas adsorption indicators. If it is a simple linear relationship, it can be expressed in the form of y=ax+b, where y represents the toxic gas adsorption index, x represents the external condition factor (such as temperature), and a and b are parameters that need to be determined through training. If the relationship is more complex, it may be necessary to use polynomial functions, exponential functions or other nonlinear function forms. For example, for some cases where the adsorption process is synergistically affected by multiple factors, y=a may be used.1 x 1 2 +a 2 x 2 +a 3 e x3 +b form, where x 1 、x 2 、x 3 They represent different external condition factors respectively.

[0059] The external condition adjustment network is a feedback network, which usually adopts a neural network model, such as a multi-layer perceptron (MLP) or a convolutional neural network (CNN). The external condition factor data is used as the input of the network, and the corresponding toxic gas adsorption index data is used as the output. During the training process, the loss function is defined according to the mathematical form of the determined conversion relationship. The weights and biases of the network are adjusted by the back propagation algorithm to minimize the loss function. In each iteration, the output of the network is calculated according to the current weights and biases, and then the loss function is calculated. The weights and biases are updated according to the loss function, and the update step size is controlled by the learning rate. This process is repeated until the preset training stop condition is reached, such as reaching the maximum number of training rounds or the loss function converges to a sufficiently small value, and the trained external condition adjustment network is output. The network can output the adjustment result of the adsorption capacity evaluation model according to the input external condition factors, which is usually an influence coefficient on the toxic gas adsorption index.

[0060] The external condition adjustment network is combined with the adsorption capacity evaluation model for joint training. The external condition adjustment network will adjust the output of the adsorption capacity evaluation model according to different external conditions, so that the toxic gas adsorption index output by the model is more in line with the actual situation, thereby realizing feedback optimization of the adsorption capacity evaluation model. Through the external condition adjustment network, the adsorption capacity evaluation model can dynamically adapt to changes in the external environment and can more accurately evaluate the adsorption capacity of simulated sensors for toxic gases under different external conditions.

[0061] Furthermore, the adsorption capacity evaluation unit 50 of the embodiment of the present application is also used to perform the following steps:

[0062] Step P531: Collect multiple environmental parameters corresponding to the multiple external condition factors in the target monitoring environment.

[0063] Step P532: The network is adjusted according to the external conditions to analyze the multiple environmental parameters and output an adsorption influence coefficient, where the adsorption influence coefficient is a coefficient that affects the size of the toxic gas adsorption index.

[0064] Step P533: adjusting the plurality of toxic gas adsorption indices according to the adsorption influence coefficient, and outputting the adjusted plurality of toxic gas adsorption indices.

[0065] Specifically, environmental parameters are actual measured values ​​corresponding to external condition factors in the target monitoring environment. The adsorption influence coefficient is a coefficient obtained by the external condition adjustment network based on the environmental parameter analysis, which indicates the degree of influence of external conditions on the size of the toxic gas adsorption index. For example, if the adsorption influence coefficient is 1.2, it means that under the current environmental conditions, the toxic gas adsorption index may increase to 1.2 times the original value.

[0066] For each external condition factor, use the corresponding sensor or measuring equipment to collect environmental parameters in the target monitoring environment. For stress magnitude, a stress sensor can be used; for external electric field strength, an electric field strength measuring instrument can be used; for operating temperature, a temperature sensor can be used, etc. The collected environmental parameters are sorted and marked to ensure that each parameter is associated with the corresponding external condition factor for subsequent analysis.

[0067] The collected multiple environmental parameters are input into the trained external condition adjustment network. The external condition adjustment network analyzes and processes the input environmental parameters according to its internal network structure and the parameters obtained through training. For example, the neurons in the network will perform weighted calculations, activation function processing and other operations according to the input environmental parameter values, and finally output the adsorption influence coefficient.

[0068] For the toxic gas adsorption index corresponding to each simulated sensor, multiply it by the adsorption influence coefficient to obtain the adjusted toxic gas adsorption index. For example, if the original toxic gas adsorption index of a simulated sensor is 0.6 and the adsorption influence coefficient is 1.2, then the adjusted toxic gas adsorption index is 0.6×1.2=0.72.

[0069] Furthermore, the poisonous gas warning unit 60 of the embodiment of the present application is also used to perform the following steps:

[0070] Step P66: Collect toxic gas samples from various areas in the target monitoring environment.

[0071] Step P67: Detect the similarity of toxic gas samples in each area. If the similarity of toxic gas samples in each area is less than a preset threshold, obtain multiple types of toxic gas samples.

[0072] Step P68: Evaluate the toxic gas adsorption capacity of the multiple simulated sensors according to the multiple types of toxic gas samples to obtain multiple matching preferred sensors, wherein each matching preferred sensor corresponds to one area.

[0073] Step P69: issuing poisonous gas warnings to each area based on the multiple matching preferred sensors.

[0074] Specifically, the toxic gas sample similarity is an indicator used to measure the similarity between toxic gas samples in different areas of the target monitoring environment. Similarity can be evaluated based on the type of toxic gas, concentration ratio, etc. For example, if the toxic gas in two areas is carbon monoxide and the concentration is similar, then their toxic gas sample similarity is high; conversely, if one area is carbon monoxide and the other area is hydrogen sulfide, then the similarity is low.

[0075] The preset threshold is a pre-set value used to determine whether the toxic gas samples in each area are similar. When the similarity of the toxic gas samples is less than this preset threshold, it is considered that there are large differences in the toxic gas samples in each area and classification processing is required. The matching preferred sensor is the preferred sensor corresponding to each area obtained by evaluating the toxic gas adsorption capacity of multiple simulated sensors according to multiple types of toxic gas samples. Each matching preferred sensor has a good adsorption capacity for the toxic gas in the corresponding area and can provide more effective toxic gas warnings.

[0076] According to the layout of the target monitoring environment, regional division and possible distribution of toxic gases, the location and number of sampling points in each area are determined, and toxic gas samples are collected at each sampling point. The samples in each area contain information such as the type, concentration, and distribution of the gas.

[0077] The collected gas samples are analyzed to extract relevant characteristic data, including the type, concentration, chemical composition ratio, etc. of the gas. For example, chemical analysis methods can be used to determine the composition of toxic gases; or spectral analysis methods (such as infrared spectroscopy, mass spectrometry, etc.) can be used to analyze the spectral characteristics of the samples. The characteristic data of the toxic gas samples in each region are quantified into digital vectors, and the similarity of each toxic gas sample is determined by a similarity algorithm. For example, the Euclidean distance can be used to calculate the distance between two samples in the feature space. The smaller the distance, the higher the similarity; or the cosine similarity can be used to measure the similarity of the two samples in the direction of the feature vector. If the directions of the two samples are consistent, the similarity is high; the Jaccard similarity can also be used to evaluate the proportion of common components in the two samples to determine the similarity between the samples.

[0078] The calculated similarity is compared with the preset threshold. If the similarity is less than the preset threshold, the toxic gas samples are divided into multiple categories of toxic gas samples according to the difference characteristics of the samples. For each category of toxic gas samples, multiple simulated sensors are analyzed through the existing adsorption capacity evaluation model to evaluate the adsorption capacity of each simulated sensor for this category of toxic gas. According to the evaluation results of the adsorption capacity, the simulated sensor with the strongest adsorption capacity for each category of toxic gas is selected, and the actual matching preferred sensor is deployed in the corresponding area according to the configuration of the simulated sensor.

[0079] Corresponding matching preferred sensors are deployed in each area, and toxic gas warnings are issued to each area according to steps P62 to P65 to ensure that gas monitoring in each area of ​​the target monitoring environment is more accurate and targeted, thereby improving the accuracy of toxic gas monitoring and warning.

[0080] Furthermore, the doping processing unit 30 in the embodiment of the present application is also used to perform the following steps:

[0081] Step P31: Processing the surface microstructure of the initial two-dimensional material to obtain the processed initial two-dimensional material.

[0082] Step P32: The doping module performs element doping treatment on the processed initial two-dimensional material according to multiple sets of doping treatment conditions, and outputs a doped two-dimensional material sample, wherein the doping treatment conditions include the doping element type or doping ion type, doping concentration and doping distribution.

[0083] Specifically, the doping treatment unit 30 first performs surface microstructure treatment on the initial two-dimensional material to optimize the properties of the material surface and improve its interaction with toxic gases in the environment. Common surface treatment methods include physical or chemical etching, redox treatment, heat treatment, etc. These methods can change the surface roughness and pore structure of the two-dimensional material or introduce functional groups to improve its adsorption or reactivity.

[0084] Next, doping treatment is performed on the treated initial two-dimensional material according to multiple sets of doping treatment conditions. By introducing different types and concentrations of doping elements or ions into the two-dimensional material, the electronic structure, surface characteristics and catalytic properties of the material are changed to improve the specific properties of the material, such as enhancing its adsorption capacity for specific gases or changing its electrical, magnetic and other properties. During the doping process, the distribution of doping elements or ions on the surface or structure of the material can also be adjusted, and the uniformity of the distribution will affect the performance of the material. For example, it can be evenly distributed on the surface or inside of the two-dimensional material, or form a local high-concentration area.

[0085] Through doping treatment, multiple doped two-dimensional material samples under different doping treatment conditions are output to provide a basis for selecting a variety of sensor materials to meet the needs of different monitoring environments.

[0086] Furthermore, the sensor structure simulation platform described in the embodiment of the present application also includes an electrode model, and the electrode model is used to quantify the gas concentration of the sensor signal output by each sensor.

[0087] Specifically, the sensor structure simulation platform also includes an electrode model, which is constructed based on specific physical or chemical principles and whose main function is to convert the sensing signal output by the sensor into a quantifiable gas concentration value.

[0088] When detecting toxic gases, sensors usually generate signals through electrochemical reactions or physical changes (such as changes in resistance, current or voltage). Since different gas concentrations have different effects on sensors, the electrode model simulates the changes in the electrical properties of the material (such as changes in resistance, current or voltage) when the sensor adsorbs toxic gas molecules during the actual monitoring process to infer the specific gas concentration. Through precise signal quantification, the electrode model can not only improve the sensitivity and accuracy of the sensor, but also ensure the reliability of gas concentration data, thereby providing a scientific basis for subsequent monitoring and early warning.

[0089] After receiving the sensor signal of the actually deployed sensor, the toxic gas warning unit 60 calls the electrode model in the sensor structure simulation platform, inputs the sensor signal into the electrode model, obtains the corresponding toxic gas concentration value, and then completes the toxic gas warning.

[0090] In summary, the toxic gas intelligent early warning platform based on sensor monitoring provided by the embodiment of the present application has the following technical effects:

[0091] The two-dimensional material setting unit 10 constructs a sensor structure simulation platform, sets the initial two-dimensional material according to the sensor structure simulation platform, and provides an initial model and material basis for the subsequent sensor design. The doping module acquisition unit 20 acquires the doping module, and the doping module is connected to the sensor structure simulation platform. The doping processing unit 30 performs doping treatment on the initial two-dimensional material according to the doping module, changes the electronic structure, chemical activity and other properties of the material, and outputs a doped two-dimensional material sample. The sensor structure simulation platform performs sensor simulation based on the doped two-dimensional material sample, outputs multiple simulated sensors, improves sensor performance, and provides a diversified selection basis. The toxic gas sample collection unit 40 determines the target monitoring environment and collects toxic gas samples of the target monitoring environment, so as to facilitate the subsequent screening of suitable sensors according to the types and characteristics of toxic gases in a specific environment, and improve the pertinence, adaptability and accuracy of the early warning in the actual environment. The adsorption capacity evaluation unit 50 evaluates the toxic gas adsorption capacity of the multiple simulated sensors according to the toxic gas samples, obtains the preferred sensor with the largest toxic gas adsorption index, improves the accuracy of sensor selection and environmental pertinence, thereby improving the accuracy of early warning. The toxic gas warning unit 60 performs toxic gas warning on the target monitoring environment according to the preferred sensor.

[0092] Overall, the embodiments of the present application significantly improve the accuracy and flexibility of the toxic gas monitoring system and reduce the waste of resources in actual testing by introducing two-dimensional materials and doping technology, combined with sensor structure simulation and intelligent early warning mechanism. This solution optimizes sensor performance through two-dimensional materials and doping treatment, and enhances the detection capability of a variety of toxic gases. At the same time, precise screening is carried out in combination with target environmental samples and simulation evaluation mechanisms, so that the sensor can be optimized according to the specific monitoring environment, and the optimal sensor can be automatically selected, which greatly improves the environmental adaptability of toxic gas monitoring, thereby achieving accurate toxic gas detection and early warning under different gas concentrations and environmental conditions, ensuring environmental safety and personnel health.

[0093] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. Intelligent early warning platform for toxic gases based on sensor monitoring, characterized in that: The platform includes: A two-dimensional material setting unit, the two-dimensional material setting unit is used to construct a sensor structure simulation platform and set an initial two-dimensional material according to the sensor structure simulation platform; A doping module acquisition unit, the doping module acquisition unit is used to acquire a doping module, the doping module is connected to the sensor structure simulation platform; A doping processing unit, wherein the doping processing unit is used to perform doping processing on the initial two-dimensional material according to the doping module, and output a doped two-dimensional material sample, and the sensor structure simulation platform performs sensor simulation based on the doped two-dimensional material sample, and outputs a plurality of simulated sensors; A toxic gas sample collection unit, the toxic gas sample collection unit is used to determine a target monitoring environment and collect toxic gas samples of the target monitoring environment; an adsorption capacity evaluation unit, the adsorption capacity evaluation unit being used to evaluate the toxic gas adsorption capacity of the plurality of simulated sensors according to the toxic gas sample, and obtain a preferred sensor, wherein the preferred sensor is a sensor with the largest toxic gas adsorption index; A toxic gas warning unit is used to provide a toxic gas warning for the target monitoring environment according to the preferred sensor.

2. The platform according to claim 1, characterized in that The toxic gas early warning unit performs a toxic gas early warning on the target monitoring environment according to the preferred sensor, and the execution steps include: Arrange a plurality of preferred sensors in the target monitoring environment; Building an intelligent monitoring network based on the multiple preferred sensors; Using the intelligent monitoring network to monitor the target monitoring environment in real time, and outputting a real-time sensing data set of toxic gases; Inputting the toxic gas real-time sensing data set into a toxic gas risk identification model, and outputting a first risk indicator according to the toxic gas risk identification model; When the first risk indicator is greater than a preset risk indicator, a first warning signal is generated.

3. The platform according to claim 1, characterized in that The adsorption capacity evaluation unit evaluates the toxic gas adsorption capacity of the plurality of simulation sensors according to the toxic gas sample, and the execution steps include: Training an adsorption capacity evaluation model based on the toxic gas sample; Downloading the trained adsorption capacity evaluation model to the sensor structure simulation platform; According to the adsorption capacity evaluation model, a plurality of key adsorption capacity indicators of each simulated sensor based on the toxic gas sample are obtained, including at least electronic structure change characteristics and adsorption energy characteristics; Fitting is performed according to the multiple adsorption capacity key indicators to obtain multiple toxic gas adsorption indicators corresponding to the multiple simulated sensors, and the preferred sensor with the largest toxic gas adsorption indicator is screened.

4. The platform according to claim 3, characterized in that The adsorption capacity evaluation model further includes an external condition adjustment network. The adsorption capacity evaluation unit trains the adsorption capacity evaluation model. The execution steps include: Configuring multiple external condition factors, wherein the multiple external condition factors include stress magnitude, external electric field strength and operating temperature; A conversion relationship between the multiple external condition factors and the toxic gas adsorption index is established, a network model is trained according to the conversion relationship, an external condition adjustment network is output, and the adsorption capacity evaluation model is feedback optimized by the external condition adjustment network.

5. The platform according to claim 4, characterized in that The adsorption capacity evaluation unit obtains a plurality of key adsorption capacity indicators of each simulated sensor based on the toxic gas sample, and the execution step further includes: Collecting multiple environmental parameters corresponding to the multiple external condition factors in the target monitoring environment; Analyzing the multiple environmental parameters according to the external condition adjustment network, and outputting an adsorption influence coefficient, wherein the adsorption influence coefficient is a coefficient that affects the size of the toxic gas adsorption index; The multiple toxic gas adsorption indices are adjusted according to the adsorption influence coefficient, and the adjusted multiple toxic gas adsorption indices are output.

6. The platform according to claim 3, characterized in that The toxic gas early warning unit performs a toxic gas early warning on the target monitoring environment, and the execution step further includes: Collect toxic gas samples from various areas of the target monitoring environment; Detect the similarity of toxic gas samples in each area. If the similarity of toxic gas samples in each area is less than a preset threshold, obtain multiple types of toxic gas samples; According to the multiple types of toxic gas samples, the multiple simulated sensors are evaluated for toxic gas adsorption capacity to obtain multiple matching preferred sensors, wherein each matching preferred sensor corresponds to one area; According to the multiple matching preferred sensors, poisonous gas warnings are respectively issued to each area.

7. The platform according to claim 1, characterized in that The doping processing unit performs doping processing on the initial two-dimensional material according to the doping module, and the execution steps include: Processing the surface microstructure of the initial two-dimensional material to obtain a processed initial two-dimensional material; The doping module performs element doping on the processed initial two-dimensional material according to multiple sets of doping conditions, and outputs a doped two-dimensional material sample, wherein the doping conditions include the doping element type or doping ion type, doping concentration and doping distribution.

8. The platform according to claim 1, characterized in that The sensor structure simulation platform also includes an electrode model, which is used to quantify the gas concentration of the sensing signal output by each sensor.

Images