A safe intelligent early warning method, system, device and medium for a very high sulfur-containing gas field

Abstract

This invention discloses a method, system, equipment, and medium for intelligent early warning of safety in ultra-high sulfur-containing gas fields, relating to the field of information and communication technology. The method includes: collecting and merging multi-source datasets from the gas field for preprocessing; performing anomaly detection based on a support vector machine algorithm incorporating radial basis function kernels and optimization, identifying potential hazardous areas and hazard types according to anomaly indices; quantifying and calculating the instantaneous diffusion impact, delayed diffusion impact, and emergency response difficulty coefficient when a hazard occurs; inputting the above three key indicators into a trained deep convolutional neural network model, dynamically determining the weights of each indicator through network feature extraction and nonlinear mapping of fully connected layers, outputting a comprehensive risk index, and generating graded early warning information; finally, pushing the early warning information to terminal devices in real time. This invention effectively overcomes the shortcomings of traditional threshold methods (high false alarm rate) and static models (inability to adapt to complex dynamic conditions), significantly improving early warning accuracy and emergency response timeliness.

Description

Technical Field

[0001] This invention relates to the field of information and communication technology, specifically to a method, system, equipment, and medium for intelligent early warning of safety in ultra-high sulfur gas fields. Background Technology

[0002] The statements in this section are provided only as background information in relation to this disclosure and may not constitute prior art.

[0003] During the development and production of ultra-high sulfur gas fields, the presence of large concentrations of toxic and harmful gases such as hydrogen sulfide poses a serious and irreversible threat to the lives of on-site personnel, equipment, and the surrounding ecological environment should leaks or abnormalities occur. Therefore, establishing an efficient and accurate safety early warning mechanism is crucial for ensuring the safe operation of gas fields.

[0004] However, existing safety monitoring technologies for ultra-high sulfur gas fields generally suffer from the following serious defects: 1. Monitoring methods are limited and have poor ability to identify hidden anomalies. Existing technologies mostly use single-dimensional concentration threshold alarm methods, which cannot capture early, subtle coupling distortions of multi-dimensional parameters (such as pressure, temperature, and flow rate) under complex operating conditions. This crude monitoring method leads to frequent false alarms and missed alarms; and missed alarms may lead to the spread of high-concentration hydrogen sulfide pollution.

[0005] 2. Rigid risk assessment models are detached from actual on-site rescue obstacles. Traditional risk assessments often only consider the diffusion range of harmful gases and use manually set static weight formulas. They fail to incorporate dynamic considerations such as constantly changing wind speeds, equipment status, and real-world rescue bottlenecks such as "the professional skill level of emergency personnel" and "the distance to emergency stations," resulting in a serious disconnect between the warning levels given and the actual risks.

[0006] 3. Severely delayed response. When abnormal situations occur, the existing monitoring system lacks in-depth intelligent analysis and end-to-end automated information flow, relying heavily on manual judgment. This results in an excessively long time span from the discovery of an anomaly to the issuance of a tiered warning, missing the best "golden window" for controlling potential risks.

[0007] In summary, there is an urgent need for an intelligent early warning scheme for ultra-high sulfur gas fields that can achieve multi-dimensional data fusion, dynamic weight evaluation, and accurate and low-latency performance. Summary of the Invention

[0008] The purpose of this invention is to provide a method, system, device, and medium for intelligent early warning of safety in ultra-high sulfur gas fields, aiming to solve the technical problems in the existing technology of high false alarm rate of threshold monitoring, inaccurate early warning due to static evaluation model deviating from actual working conditions, and serious response lag when issuing early warnings for abnormal situations in high sulfur gas fields.

[0009] The technical solution of the present invention is as follows: A smart early warning method for ultra-high sulfur gas fields, executed by a processor, includes: Collect multi-source data sets during the operation of the gas field; Preprocessing is performed on the multi-source dataset to obtain standardized data; Anomaly detection is performed on the standardized data based on the support vector machine algorithm to identify potential hazardous areas and their corresponding hazard types. Calculate the instantaneous diffusion impact, delayed diffusion impact, and emergency response difficulty coefficient when a hazard occurs in the potentially hazardous area; The instantaneous diffusion impact, the delayed diffusion impact, and the emergency response difficulty coefficient are input into a deep convolutional neural network model to output a comprehensive risk index for the potential dangerous area, and a graded early warning information is generated based on the comprehensive risk index. The tiered early warning information will be pushed to the terminal devices of relevant management personnel and emergency response departments in real time.

[0010] Furthermore, the multi-source data set includes hydrogen sulfide concentration, ambient temperature, pressure value, gas flow rate, equipment operating parameters, operation log information, and external environmental data; wherein, the external environmental data includes at least wind speed, wind direction, and humidity.

[0011] Furthermore, the preprocessing of the multi-source dataset to obtain standardized data specifically includes: Invalid or erroneous data in the multi-source dataset is removed using logical judgment rules. Missing values ​​in the missing data can be imputed using the mean, median, or interpolation method. Use filtering algorithms or wavelet transform methods to remove noise interference from the data; The processed data is mapped to the [0,1] interval using the min-max normalization method to obtain the standardized data.

[0012] Furthermore, the step of performing anomaly detection on the standardized data based on the support vector machine algorithm to identify potential hazardous areas and their corresponding hazard types includes: Construct a support vector machine algorithm model that incorporates a radial basis function kernel; A grid search combined with cross-validation method is used to perform a combined search within a preset hyperparameter range in order to optimize the hyperparameters of the support vector machine algorithm model. and value; The standardized data is input into the optimized support vector machine algorithm model to calculate the degree of deviation between the current data distribution and the historical normal data distribution, thereby obtaining the anomaly index. Areas with an anomaly index greater than a preset threshold are identified as potentially dangerous areas, and are classified into corresponding hazard types based on their anomaly characteristics.

[0013] Furthermore, the specific formula for calculating the anomaly index is as follows:

[0014] in: Indicates an abnormality index; The number of samples; For feature dimensions; For the first The first sample 3D eigenvalues; The first in historical normal data The mean of the dimensional features.

[0015] Furthermore, the calculation of the instantaneous diffusion impact, delayed diffusion impact, and emergency response difficulty coefficient when a hazard occurs in the potentially hazardous area specifically includes: Based on the initial diffusion radius data and concentration gradient change data of harmful gases in the gas field, the instantaneous diffusion impact is calculated using the following formula:

[0016] in: The instantaneous diffusion effect; The concentration of hazardous substances; The rate at which hazardous substances diffuse; Wind speed; , , These are the weight coefficients of each factor, and they satisfy... ; Based on numerical simulation methods to predict the coverage area and concentration distribution data of hazardous substances spreading over a wider area over time, the delayed diffusion impact is calculated using the following formula:

[0017] in: The degree of delayed diffusion effect; Indicates the start time; Indicates the end time; Indicates the concentration as it changes over time; Indicates the diffusion rate as it changes over time; Indicates wind speed as it changes over time; , , Let be the weight coefficient, and satisfy... ; Based on the weighted scoring method, the emergency response difficulty coefficient is calculated using the following formula:

[0018] in: The emergency response difficulty coefficient is mentioned above. The distance between the potentially hazardous area and the nearest emergency station; Time required for emergency resource allocation; The professional skill level of emergency personnel; , , Let be the weight coefficient, and satisfy... .

[0019] Furthermore, the instantaneous diffusion impact, the delayed diffusion impact, and the emergency response difficulty coefficient are input into a deep convolutional neural network model to output a comprehensive risk index for the potentially hazardous area, and a graded early warning information is generated based on the comprehensive risk index, including: A training set is constructed by acquiring historical accident data, simulation data, and laboratory test data, and the deep convolutional neural network model is trained using the training set. The instantaneous diffusion impact, the delayed diffusion impact, and the emergency response difficulty coefficient are input into the trained deep convolutional neural network model, and features are extracted through multiple convolutional layers to form a feature set; The feature set is input into a fully connected layer to perform a nonlinear mapping, and the weight coefficients are dynamically determined based on the mapping result. , , The comprehensive risk index is output, and the calculation relationship of the comprehensive risk index satisfies:

[0020] in: This represents the overall risk index; This indicates the degree of instantaneous diffusion influence; This indicates the degree of delayed diffusion influence; This indicates the difficulty level of the emergency response; , , Let be the weight coefficient, and satisfy... ; When the comprehensive risk index is less than the first threshold, a level 1 warning message is generated; when the comprehensive risk index is between the first threshold and the second threshold, a level 2 warning message is generated; when the comprehensive risk index is greater than the second threshold, a level 3 warning message is generated.

[0021] This invention also proposes a smart early warning system for ultra-high sulfur gas fields, capable of implementing the methods described above, including: The data acquisition module is used to collect multi-source data sets during the operation of the gas field and preprocess them to obtain standardized data; Anomaly detection module is used to perform anomaly detection on the standardized data based on support vector machine algorithm, and identify potential dangerous areas and their corresponding hazard types; The indicator calculation module is used to calculate the instantaneous diffusion impact, delayed diffusion impact, and emergency response difficulty coefficient when a hazard occurs in the potential hazardous area. The risk assessment module is used to input the instantaneous diffusion impact, the delayed diffusion impact, and the emergency response difficulty coefficient into a deep convolutional neural network model, output a comprehensive risk index of the potential dangerous area, and generate graded early warning information based on the comprehensive risk index; The early warning push module is used to push the tiered early warning information to the terminal devices of relevant management personnel and emergency response departments in real time.

[0022] The present invention also proposes an electronic device, comprising: At least one processor; and a memory communicatively connected to said at least one processor; The memory stores instructions that can be executed by the at least one processor, and the at least one processor executes the instructions stored in the memory to perform the method described above.

[0023] The present invention also proposes a computer-readable storage medium for storing instructions that, when executed, cause the method described above to be implemented.

[0024] Compared with existing technologies, the advantages of this invention are: 1. Accurately identifies hidden risks, significantly reducing false alarms and false negatives. This invention overcomes the limitations of traditional single-point threshold methods by deeply cleaning and normalizing multi-source data and introducing radial basis function (RBF) and hyperparameter grid-optimized support vector machine algorithms. It can not only calculate anomaly indices that accurately reflect deviations from the normal baseline and precisely identify abnormal fluctuations in hydrogen sulfide at the 0.1 ppm level, significantly reducing the false alarm rate by 62% compared to traditional methods; it can also accurately determine specific hazard types (such as leaks, corrosion, and pressure anomalies) by tracing the characteristic dimensions of the dominant anomalies, providing precise guidance for subsequent emergency response.

[0025] 2. Multi-dimensional indicator quantification and dynamic weight mapping ensure that risk assessment is highly relevant to reality. This invention quantifies the instantaneous diffusion impact, delayed diffusion impact, and emergency response difficulty coefficient from three dimensions: "instantaneous impact," "continuous impact," and "real-world rescue resistance." In particular, the emergency response difficulty coefficient introduces an inversely proportional, step-like quantification mechanism for the professional skill level of emergency personnel; more importantly, utilizing the multi-layer convolutional and fully connected nonlinear mapping capabilities of deep convolutional neural networks, the system can dynamically determine the weight coefficients of each indicator based on the complex and ever-changing context of the situation. , , This breaks through the rigid and inflexible technical bottleneck of traditional static weight formulas, greatly improving the scientific nature and accuracy of hierarchical early warning.

[0026] 3. Rapid Response and Seamless End-to-End Integration: Building a Safe Production Barrier. This invention constructs an end-to-end architecture encompassing "data acquisition - anomaly detection - intelligent fusion - tiered push," possessing powerful self-learning and adaptive capabilities. In practical gas field applications, it successfully improved the accuracy of accident early warning to 98.7% (a 23 percentage point improvement over the original system), and reduced the early warning response time to less than 8 seconds. Information directly reaches the terminal equipment of management personnel and emergency departments, securing valuable golden rescue time for preventing severe explosions, poisoning accidents, and ecological disasters, effectively avoiding huge economic losses and environmental accountability. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments recorded in the embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0028] Figure 1 A flowchart of a safety intelligent early warning method for ultra-high sulfur gas fields; Figure 2 This is a block diagram of a safety intelligent early warning system for ultra-high sulfur gas fields; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0029] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0030] The features and performance of the present invention will be further described in detail below with reference to embodiments.

[0031] Example 1 Please refer to Figure 1 To address the aforementioned pain points, this embodiment provides an intelligent early warning method that combines high timeliness and high accuracy. This method is executed by a processor in an electronic device. To achieve a four-dimensional quantitative assessment integrating over 300 indicators such as device health, environmental parameters, and emergency resources, and to accurately determine the risk level, the method specifically includes the following steps: Step S1: Collect a multi-source data set during the operation of the gas field.

[0032] Specifically, due to the complex environment of ultra-high sulfur gas fields, single-dimensional monitoring often fails to reflect the true state of the system. Therefore, the multi-source data set includes hydrogen sulfide concentration, ambient temperature, pressure, gas flow rate, equipment operating parameters, operation log information, and external environmental data; wherein the external environmental data includes at least wind speed, wind direction, and humidity.

[0033] In practical implementation, the aforementioned data can be acquired in real time (within seconds) through sensor networks and industrial IoT technology deployed at the gas field. This enables the acquisition of more than 20 key parameters, such as wellhead pressure, hydrogen sulfide concentration, and temperature, within seconds, breaking through the time delay bottleneck of traditional monitoring and ensuring the comprehensiveness, multidimensionality, and real-time nature of the data.

[0034] Step S2: Perform preprocessing on the multi-source data set to obtain standardized data.

[0035] Specifically, since raw data collected by field sensors often contains issues such as noise, missing data, or inconsistent formats, directly using it for model analysis can lead to decreased accuracy. Therefore, preprocessing is crucial for improving data quality and analytical accuracy. The preprocessing specifically includes: Invalid or erroneous data in the multi-source dataset is removed using logical judgment rules. For example, abnormally high values ​​of hydrogen sulfide concentration exceeding physically reasonable ranges or garbled characters in equipment operating parameters are removed. Missing data can be imputed using the mean, median, or interpolation method. For example, for missing wind speed data within a certain time period, the average wind speed of the preceding and following time periods can be used to impute the missing data. Filtering algorithms or wavelet transform methods are used to remove noise interference from the data. For example, a low-pass filter is used to remove high-frequency environmental noise interference from gas flow rate data; The processed data is mapped to the [0,1] interval using the Min-Max Normalization method to obtain the standardized data. Normalization can eliminate the differences between data of different dimensions (such as pressure values ​​in megapascals and concentrations in ppm), effectively accelerating the convergence speed of subsequent machine learning models.

[0036] Step S3: Perform anomaly detection on the standardized data based on the support vector machine algorithm to identify potential hazardous areas and their corresponding hazard types.

[0037] Specifically, traditional thresholding methods struggle to identify hidden anomalies under complex operating conditions. This embodiment constructs an anomaly detection model based on cross-validation of deep learning and machine learning, specifically including: A support vector machine algorithm model incorporating a radial basis function (RBF) kernel is constructed. The RBF kernel function can effectively classify nonlinearly inseparable field data into a high-dimensional space. A grid search combined with cross-validation method is used to perform a combinatorial search within a preset range of hyperparameters (i.e., searching for the optimal combination point by point and evaluating the classification performance of each hyperparameter group through cross-validation) to optimize the hyperparameters of the support vector machine algorithm model. and Values ​​to prevent model overfitting; The standardized data is input into the optimized support vector machine algorithm model to calculate the deviation between the current data distribution and the historical normal data distribution, thus obtaining the anomaly index. In one specific embodiment, the formula for calculating the anomaly index is as follows:

[0038] in: Indicates an abnormality index; The number of samples; For feature dimensions; For the first The first sample 3D eigenvalues; The first in historical normal data The mean of the dimensional features; This formula quantifies the degree to which the current operating condition deviates from the normal baseline by calculating the mean Euclidean distance in the feature space in an intuitive and accurate manner.

[0039] In this embodiment, it should be noted that the classification performance is evaluated based on an anomaly index, which is the degree of deviation between the current data distribution and the historical normal data distribution. The greater the deviation, the more abnormal the data, and the higher the probability of potential dangerous areas. Therefore, the quality of classification can be indirectly reflected by this degree of deviation; models that can more accurately distinguish between normal and abnormal data have better classification performance.

[0040] Furthermore, areas where the anomaly index exceeds a preset threshold are identified as potential hazardous areas, and these areas are classified into corresponding hazard types (e.g., leakage, corrosion, or abnormal pressure) based on their abnormal characteristics. Through this improved algorithm, the present invention can accurately identify abnormal fluctuations in hydrogen sulfide at the 0.1 ppm level, significantly reducing the false alarm rate by 62% compared to the traditional single-point threshold method. In this embodiment, it should be noted that the classification of hazard types can be specifically determined by analyzing the characteristics of abnormal data, such as hydrogen sulfide concentration, pressure value, and gas flow rate. For example, if the hydrogen sulfide concentration is abnormally high, it may correspond to a leakage hazard; if the pressure value is abnormally high, it may correspond to a pressure anomaly hazard. By comparing current abnormal data with historical data, specific types or characteristics of danger can be identified.

[0041] Step S4: Calculate the instantaneous diffusion impact, delayed diffusion impact, and emergency response difficulty coefficient when a hazard occurs in the potential hazardous area.

[0042] After identifying the hazardous area and its corresponding hazard type, the next step is to quantify the multi-dimensional risk indicators. Considering the physical characteristics of high-sulfur gas diffusion and the actual conditions at the rescue site, this embodiment performs calculations from three dimensions: "instantaneous impact," "lasting impact," and "rescue resistance." First, based on the initial diffusion radius data and concentration gradient change data of harmful gases in the gas field, the instantaneous diffusion impact is calculated using the following formula:

[0043] in: The instantaneous diffusion effect degree; The concentration of hazardous substances; The rate at which hazardous substances diffuse; Wind speed; , , These are the weight coefficients of each factor, and they satisfy... .

[0044] This step, by simulating the initial diffusion range and combining it with concentration distribution characteristics, can accurately reflect the scope and intensity of the impact of a dangerous event on the surrounding area at the moment of its occurrence.

[0045] Secondly, based on numerical simulation methods to predict the coverage area and concentration distribution data of hazardous substances spreading over a wider area over time, the delayed diffusion impact is calculated using the following formula:

[0046] in: The degree of delayed diffusion influence; Indicates the start time; Indicates the end time; Indicates the concentration as it changes over time; Indicates the diffusion rate as it changes over time; Indicates wind speed as it changes over time; , , Let be the weight coefficient, and satisfy... .

[0047] This formula utilizes integral calculations to fully consider the cumulative coverage effect of harmful gases over a period of time after a hazardous event occurs.

[0048] Finally, based on the weighted scoring method, the emergency response difficulty coefficient is calculated using the following formula:

[0049] in: The emergency response difficulty coefficient is mentioned above. The distance between the potentially hazardous area and the nearest emergency station; Time required for emergency resource allocation; This refers to the professional skill level of emergency response personnel. In this embodiment, it should be specifically noted that, to avoid data ambiguity and incalculable problems caused by subjective factors, the professional skill level of the emergency response personnel in this embodiment... Structured values ​​are obtained through a pre-built skill quantification mapping table. Because... This represents the difficulty of the response; therefore, the higher the professional skill level of emergency personnel, the smaller their contribution to the difficulty of the rescue should be. The lower the value, the better. In a specific application example, the system maps the skill level of the available emergency personnel to a value in the range [0,1]: (1) If the currently deployed personnel is a main professional rescue team (with professional rescue qualifications for high-sulfur gas fields and rich practical / drill experience), the corresponding professional skill level is extremely high, then The value is 0.2; (2) If the dispatched personnel are regular auxiliary rescue personnel (who have passed standardized safety training, have basic gas mask usage skills but lack experience in handling major accidents), then The value is 0.5; (3) If only temporary security or evacuation guidance personnel can be deployed, and professional rescue skills are weak, then The value is set to 0.8. Through this inversely proportional, step-like quantification mechanism, this invention successfully transforms the unstructured "personnel skill level" into a numerical characteristic that can be precisely calculated by a computer, ensuring the accuracy of the emergency response difficulty coefficient. The objectivity, clarity, and feasibility of the calculation; , , Let be the weight coefficient, and satisfy... .

[0050] The difficulty level takes into account practical rescue bottlenecks such as geographical environment, transportation conditions, and personnel allocation.

[0051] Step S5: Input the instantaneous diffusion impact, the delayed diffusion impact, and the emergency response difficulty coefficient into the deep convolutional neural network model, and output the comprehensive risk index of the potential dangerous area.

[0052] Specifically, traditional risk assessments often employ static weighting formulas, which are ill-suited to the complex and dynamic spatiotemporal coupling characteristics of gas field environments. This embodiment cleverly combines a deep learning framework with an explicit physical evaluation model, specifically including: A training set is constructed by acquiring historical accident data, simulation data, and laboratory test data, and the deep convolutional neural network model is iteratively trained and optimized using the training set based on the backpropagation algorithm. The instantaneous diffusion impact, the delayed diffusion impact, and the emergency response difficulty coefficient obtained in step S4 are input into the trained deep convolutional neural network model. Deep, non-linear coupled correlation features in the training set data are extracted through multiple convolutional layers to form a feature set. The feature set is input into a fully connected layer to perform a nonlinear mapping, and the weight coefficients are dynamically determined based on the mapping result. , , The system outputs the comprehensive risk index. Its core logic lies in the fact that the neural network intelligently allocates the weight ratios of the three major influencing indicators based on the contextual characteristics of the current operating conditions (for example, in scenarios with high wind speeds and nearby residential areas, the network automatically increases the weight of the instantaneous diffusion impact). To emphasize the importance of immediate response, the calculation relationship of the comprehensive risk index satisfies:

[0053] in: This represents the overall risk index; This indicates the degree of instantaneous diffusion influence; This indicates the degree of delayed diffusion influence; This indicates the difficulty level of the emergency response; , , Let be the weight coefficient, and satisfy... .

[0054] Step S6: Generate graded early warning information based on the comprehensive risk index, and push the graded early warning information to the terminal devices of relevant management personnel and emergency response departments in real time.

[0055] Specifically, this embodiment establishes refined early warning classification rules: When the comprehensive risk index When the value is less than the first threshold (preferably 0.3), it is determined to be low risk and a level 1 warning message is generated; When the comprehensive risk index When the risk level falls between the first threshold and the second threshold (preferably 0.6), it is determined to be medium risk and a secondary warning message is generated. When the comprehensive risk index If the value is greater than the second threshold (0.6), it is judged as high risk and a level 3 warning message is generated.

[0056] During the information push and processing phase, the tiered early warning information is not only pushed to the smartphones and other terminal devices of relevant management personnel via SMS, email, or dedicated applications, but also displayed on a large screen in the monitoring center. Preferably, combined with a digital twin platform, the system can achieve three-dimensional visualization and positioning of high-risk areas. With the help of the intelligent analysis architecture of this invention, the entire early warning response time is significantly shortened to less than 8 seconds.

[0057] In addition, for high-risk areas (Level 3 warning), the system is equipped with linkage control function, which will automatically trigger the corresponding emergency plan. Actions include, but are not limited to: closing relevant pipeline valves, starting on-site ventilation equipment, and dispatching emergency teams to the scene through command and dispatch.

[0058] The method provided in this invention has strong self-learning and adaptive capabilities. Through continuous iteration of the early warning model, in the actual application test of the applicant in the Northeast Sichuan Gas Field, the accuracy of accident early warning has been successfully improved to 98.7%, which is 23 percentage points higher than the original system. This provides extremely reliable technical support for the safe and efficient production of gas fields in ultra-high sulfur environments.

[0059] Example 2 Based on the same inventive concept as in the foregoing embodiments, please refer to Figure 2 Embodiment 2 of the present invention provides a smart early warning system for ultra-high sulfur gas fields. Each functional module within this system corresponds one-to-one with each operational step of the smart early warning method for ultra-high sulfur gas fields described in Embodiment 1. Specifically, the system includes: The data acquisition module is used to collect multi-source data sets during the operation of the gas field and preprocess them to obtain standardized data; Anomaly detection module is used to perform anomaly detection on the standardized data based on support vector machine algorithm, and identify potential dangerous areas and their corresponding hazard types; The indicator calculation module is used to calculate the instantaneous diffusion impact, delayed diffusion impact, and emergency response difficulty coefficient when a hazard occurs in the potential hazardous area. The risk assessment module is used to input the instantaneous diffusion impact, the delayed diffusion impact, and the emergency response difficulty coefficient into a deep convolutional neural network model, output a comprehensive risk index of the potential dangerous area, and generate graded early warning information based on the comprehensive risk index; The early warning push module is used to push the tiered early warning information to the terminal devices of relevant management personnel and emergency response departments in real time.

[0060] It should be noted that the modules in the intelligent early warning system for ultra-high sulfur gas fields provided in this embodiment are merely virtual device modules divided according to logical functions. In actual physical equipment, these modules can be executed by the same processor, or they can be distributed across multiple processors or cloud servers to work together.

[0061] Furthermore, the specific internal operating rules, formula derivation principles, and technical effects such as preventing missed alarms and false alarms achieved by each module in this embodiment are completely one-to-one corresponding to and identical to the method steps (steps S1-S6) in Embodiment 1. For example, the source tracing mechanism for determining hazard types (leakage, corrosion, pressure anomaly) based on anomaly indices in the anomaly detection module, and the professional skill level of emergency personnel in the index calculation module... The inverse proportional step-wise quantification assignment mechanism, and the dynamic determination of weight coefficients in the risk assessment module. , , For details regarding the principles of nonlinear network mapping, please refer directly to the extremely detailed description in the aforementioned Example 1. To avoid repetition and maintain the conciseness of the text, these details will not be repeated here.

[0062] Example 3 Based on the same inventive concept as in the foregoing embodiments, this embodiment provides an electronic device that can implement the intelligent early warning method for ultra-high sulfur gas fields provided in the above embodiments of the present invention. In one embodiment, the electronic device can be a server, a terminal device, or other electronic equipment. Figure 3 As shown, the electronic device may include: At least one processor and a memory connected to the at least one processor. In this embodiment of the invention, the specific connection medium between the processor and the memory is not limited. Figure 3 The example used is the connection between the processor and memory via a bus. The bus... Figure 3 The connections between other components are indicated by thick lines and are for illustrative purposes only, not as limiting information. Buses can be divided into address buses, data buses, control buses, etc., but for ease of representation, [the specific bus type is not shown here]. Figure 3The processor is represented by a single thick line, but this does not imply that there is only one bus or one type of bus. Alternatively, a processor can also be called a controller; there are no restrictions on the name.

[0063] In this embodiment of the invention, the memory stores instructions executable by at least one processor. By executing the instructions stored in the memory, the at least one processor can perform the aforementioned intelligent early warning method for ultra-high sulfur gas fields. The processor can implement... Figure 3 The functions of each module in the device shown.

[0064] The processor is the control center of the device. It can connect to various parts of the control device through various interfaces and lines. By running or executing instructions stored in memory and calling data stored in memory, it can monitor the device's various functions and process data, thereby enabling overall monitoring of the device.

[0065] In an alternative design, the processor may include one or more processing units. The processor may integrate an application processor and a modem processor, wherein the application processor primarily handles the operating system, user interface, and applications, while the modem processor primarily handles wireless communication. It is understood that the modem processor may also not be integrated into the processor. In some embodiments, the processor and memory may be implemented on the same chip; in some embodiments, they may also be implemented separately on separate chips.

[0066] The processor can be a general-purpose processor, such as a CPU, digital signal processor, application-specific integrated circuit, field-programmable gate array or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the intelligent early warning method for ultra-high sulfur gas fields disclosed in the embodiments of this invention can be directly manifested as being executed by a hardware processor, or executed by a combination of hardware and software modules within the processor.

[0067] Memory, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. Memory can include at least one type of storage medium, such as flash memory, hard disk, multimedia cards, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read-only memory (PROM), read-only memory (ROM), and electrically erasable programmable read-only memory (EPROM). Only memory (EEPROM), magnetic storage, magnetic disks, optical disks, etc. A memory is any other medium capable of carrying or storing desired program code in the form of instructions or data structures, and accessible by a computer, but is not limited thereto. The memory in embodiments of this invention can also be a circuit or any other device capable of performing storage functions for storing program instructions and / or data.

[0068] By designing and programming the processor, the code corresponding to the intelligent early warning method for ultra-high sulfur gas fields described in the foregoing embodiments can be embedded into the chip, enabling the chip to execute the steps of the method described in the foregoing embodiments during operation. How to design and program the processor is a technique well-known to those skilled in the art and will not be elaborated upon here.

[0069] Example 4 Based on the same inventive concept, embodiments of the present invention also provide a storage medium storing computer instructions, which, when executed on a computer, cause the computer to perform the aforementioned intelligent early warning method for ultra-high sulfur gas field safety.

[0070] In some alternative embodiments, the present invention also provides a method for intelligent early warning of safety in ultra-high sulfur gas fields, which can also be implemented in the form of a program product, including program code. When the program product is run on a device, the program code is used to cause the control device to perform the steps in the method for intelligent early warning of safety in ultra-high sulfur gas fields according to various exemplary embodiments of the present invention as described above.

[0071] It should be noted that although several units or sub-units of the apparatus have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of the invention, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided and embodied by multiple units. Furthermore, although the operation of the method of the invention is described in a specific order in the drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0072] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can be implemented in one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs) containing computer-usable program code. The form of a computer program product implemented on ROM, optical memory, etc.

[0073] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a server, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0074] Program code for performing the operations of this invention can be written using any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0075] In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0076] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0077] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0078] The embodiments described above merely illustrate specific implementation methods of this application, and while the descriptions are detailed and specific, they should not be construed as limiting the scope of protection of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the technical solution of this application, and these modifications and improvements all fall within the scope of protection of this application.

[0079] This background section is provided to generally present the context of the invention. The work of the currently named inventors, the work to the extent described in this background section, and aspects of this section that did not constitute prior art at the time of application are neither expressly nor impliedly acknowledged as prior art to the invention.

Claims

1. A method for intelligent early warning of safety in ultra-high sulfur gas fields, characterized in that, This method is executed by the processor and includes: Collect multi-source data sets during the operation of the gas field; Preprocessing is performed on the multi-source dataset to obtain standardized data; Anomaly detection is performed on the standardized data based on the support vector machine algorithm to identify potential hazardous areas and their corresponding hazard types. Calculate the instantaneous diffusion impact, delayed diffusion impact, and emergency response difficulty coefficient when a hazard occurs in the potentially hazardous area; The instantaneous diffusion impact, the delayed diffusion impact, and the emergency response difficulty coefficient are input into a deep convolutional neural network model to output a comprehensive risk index for the potential dangerous area, and a graded early warning information is generated based on the comprehensive risk index. The tiered early warning information will be pushed to the terminal devices of relevant management personnel and emergency response departments in real time.

2. The intelligent early warning method for ultra-high sulfur gas field safety according to claim 1, characterized in that, The multi-source data set includes hydrogen sulfide concentration, ambient temperature, pressure, gas flow rate, equipment operating parameters, operation log information, and external environmental data; wherein, the external environmental data includes at least wind speed, wind direction, and humidity.

3. The intelligent early warning method for ultra-high sulfur gas fields according to claim 1, characterized in that, The preprocessing of the multi-source dataset to obtain standardized data specifically includes: Invalid or erroneous data in the multi-source dataset is removed using logical judgment rules. Missing values ​​in the missing data can be imputed using the mean, median, or interpolation method. Use filtering algorithms or wavelet transform methods to remove noise interference from the data; The processed data is mapped to the [0,1] interval using the min-max normalization method to obtain the standardized data.

4. The intelligent early warning method for ultra-high sulfur gas field safety according to claim 1, characterized in that, The step of performing anomaly detection on the standardized data based on the support vector machine algorithm to identify potential hazardous areas and their corresponding hazard types includes: Construct a support vector machine algorithm model that incorporates a radial basis function kernel; A grid search combined with cross-validation method is used to perform a combined search within a preset hyperparameter range in order to optimize the hyperparameters of the support vector machine algorithm model. and value; The standardized data is input into the optimized support vector machine algorithm model to calculate the degree of deviation between the current data distribution and the historical normal data distribution, thereby obtaining the anomaly index. Areas with an anomaly index greater than a preset threshold are identified as potentially dangerous areas, and are classified into corresponding hazard types based on their anomaly characteristics.

5. The intelligent early warning method for ultra-high sulfur gas field safety according to claim 4, characterized in that, The specific formula for calculating the anomaly index is as follows: in: Indicates an abnormality index; The number of samples; For feature dimensions; For the first The first sample 3D eigenvalues; The first in historical normal data The mean of the dimensional features.

6. The intelligent early warning method for ultra-high sulfur gas field safety according to claim 1, characterized in that, The calculation of the instantaneous diffusion impact, delayed diffusion impact, and emergency response difficulty coefficient when a hazard occurs in the potentially hazardous area specifically includes: Based on the initial diffusion radius data and concentration gradient change data of harmful gases in the gas field, the instantaneous diffusion impact is calculated using the following formula: in: The instantaneous diffusion effect; The concentration of hazardous substances; The rate at which hazardous substances diffuse; Wind speed; , , These are the weight coefficients of each factor, and they satisfy... ; Based on numerical simulation methods to predict the coverage area and concentration distribution data of hazardous substances spreading over a wider area over time, the delayed diffusion impact is calculated using the following formula: in: The degree of delayed diffusion effect; Indicates the start time; Indicates the end time; Indicates the concentration as it changes over time; Indicates the diffusion rate as it changes over time; Indicates wind speed as it changes over time; , , Let be the weight coefficient, and satisfy... ; Based on the weighted scoring method, the emergency response difficulty coefficient is calculated using the following formula: in: The emergency response difficulty coefficient is mentioned above. The distance between the potentially hazardous area and the nearest emergency station; Time required for emergency resource allocation; The professional skill level of emergency personnel; , , Let be the weight coefficient, and satisfy... .

7. The intelligent early warning method for ultra-high sulfur gas field safety according to claim 1, characterized in that, The process involves inputting the instantaneous diffusion impact, the delayed diffusion impact, and the emergency response difficulty coefficient into a deep convolutional neural network model to output a comprehensive risk index for the potentially hazardous area. Based on this comprehensive risk index, tiered early warning information is generated, including: A training set is constructed by acquiring historical accident data, simulation data, and laboratory test data, and the deep convolutional neural network model is trained using the training set. The instantaneous diffusion impact, the delayed diffusion impact, and the emergency response difficulty coefficient are input into the trained deep convolutional neural network model, and features are extracted through multiple convolutional layers to form a feature set; The feature set is input into a fully connected layer to perform a nonlinear mapping, and the weight coefficients are dynamically determined based on the mapping result. , , The comprehensive risk index is output, and the calculation relationship of the comprehensive risk index satisfies: in: This represents the overall risk index; This indicates the degree of instantaneous diffusion influence; This indicates the degree of delayed diffusion effect; This indicates the difficulty level of the emergency response; , , Let be the weight coefficient, and satisfy... ; When the comprehensive risk index is less than the first threshold, a level 1 warning message is generated; when the comprehensive risk index is between the first threshold and the second threshold, a level 2 warning message is generated; when the comprehensive risk index is greater than the second threshold, a level 3 warning message is generated.

8. A smart early warning system for the safety of ultra-high sulfur gas fields, characterized in that, A method capable of implementing any one of claims 1-7 includes: The data acquisition module is used to collect multi-source data sets during the operation of the gas field and preprocess them to obtain standardized data; Anomaly detection module is used to perform anomaly detection on the standardized data based on support vector machine algorithm, and identify potential dangerous areas and their corresponding hazard types; The indicator calculation module is used to calculate the instantaneous diffusion impact, delayed diffusion impact, and emergency response difficulty coefficient when a hazard occurs in the potential hazardous area. The risk assessment module is used to input the instantaneous diffusion impact, the delayed diffusion impact, and the emergency response difficulty coefficient into a deep convolutional neural network model, output a comprehensive risk index of the potential dangerous area, and generate graded early warning information based on the comprehensive risk index; The early warning push module is used to push the tiered early warning information to the terminal devices of relevant management personnel and emergency response departments in real time.

9. An electronic device, characterized in that, include: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores instructions executable by the at least one processor, which executes the instructions stored in the memory to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store instructions that, when executed, cause the method as described in any one of claims 1-7 to be implemented.

Images