Leakage gas risk assessment method, device, equipment, medium and product

Abstract

The invention provides a gas leakage risk assessment method and device, equipment, a medium and a product. According to one example of the invention, the method can comprise the following steps: analyzing collected concentration data of leaked gas, and determining initial parameters of a concentration prediction model, the initial parameters comprising a difference order, an autoregression order and a moving average order; performing fitting optimization on the preliminary parameters by traversing different parameter combinations of the preliminary parameters; generating predicted concentration data in a future time period by using the optimized concentration prediction model; and matching the predicted concentration data and the gas type of the leaked gas with preset risk level data, and evaluating the predicted risk level of the leaked gas.

Description

Technical Field

[0001] This application relates to the technical field of risk prevention and control of hazardous chemicals, and particularly to a risk assessment method, device, equipment, medium and product for leaked gas. Background Art

[0002] In multiple high-risk fields such as industrial safety, emergency rescue, mine operations, petrochemical industry, etc., gas leakage accidents of harmful gases, combustible gases, etc. occur frequently, posing a serious threat to the lives of on-site personnel such as firefighters. Such accidents may not only lead to the rapid spread of fire, but the leaked gas is also highly toxic or asphyxiating to the human body, capable of causing serious health damage or even fatal injuries in a short period of time.

[0003] Currently, the assessment method of leaked gas mainly relies on on-site immediate measurement by gas detectors. Although this method can provide immediate gas concentration data, it cannot evaluate the potential personal safety risks caused by gas leakage. Summary of the Invention

[0004] To overcome the problems existing in the related art, this application provides a risk assessment method, device, equipment, medium and product for leaked gas.

[0005] According to the first aspect of any embodiment of this application, a risk assessment method for leaked gas is provided. The method includes:

[0006] Analyze the collected concentration data of the leaked gas to determine the preliminary parameters of the concentration prediction model, where the preliminary parameters include the difference order, autoregressive order, and moving average order;

[0007] By traversing different parameter combinations of the preliminary parameters, fit and optimize the preliminary parameters;

[0008] Use the optimized concentration prediction model to generate predicted concentration data for a future time period;

[0009] Match the predicted concentration data and the gas type of the leaked gas with the preset risk level data to evaluate the predicted risk level of the leaked gas.

[0010] According to the second aspect of any embodiment of this application, a risk assessment device for leaked gas is provided. The device includes:

[0011] An analysis module, configured to analyze the collected concentration data of the leaked gas to determine the preliminary parameters of the concentration prediction model, where the preliminary parameters include the difference order, autoregressive order, and moving average order;

[0012] A tuning module for fitting and tuning the preliminary parameters by traversing different parameter combinations of the preliminary parameters;

[0013] A prediction module for generating predicted concentration data within a future time period by using the tuned concentration prediction model;

[0014] An evaluation module for matching the predicted concentration data and the gas type of the leaked gas with preset risk level data to evaluate the predicted risk level of the leaked gas.

[0015] According to a third aspect of any embodiment of the present application, there is provided an electronic device, including:

[0016] A processor;

[0017] A memory for storing instructions executable by the processor;

[0018] Wherein, the processor realizes the method described in any embodiment of the present application by running the executable instructions.

[0019] According to a fourth aspect of any embodiment of the present application, there is provided a computer-readable storage medium, on which computer instructions are stored, and when the instructions are executed by a processor, the method described in any embodiment of the present application as above is realized.

[0020] According to a fifth aspect of any embodiment of the present application, there is provided a computer program product, on which a computer program / instruction is stored, and when the computer program / instruction is executed by a processor, the method described in any embodiment of the present application as above is realized.

[0021] The technical solution provided by the present application may include the following beneficial effects:

[0022] According to the above embodiments, by analyzing the collected concentration data of the leaked gas, the preliminary parameters of the concentration prediction model are determined, the preliminary parameters are fitted and tuned by traversing different parameter combinations of the preliminary parameters, the predicted concentration data within a future time period is generated by using the tuned concentration prediction model, the predicted concentration data and the gas type of the leaked gas are matched with the preset risk level data, and the predicted risk level of the leaked gas is evaluated, so that the potential gas leakage risk can be quantified and classified, enabling relevant personnel to have enough time to take preventive measures according to the predicted risk level, thereby effectively avoiding or reducing the personal safety risk caused by gas leakage in a targeted manner.

[0023] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings here are incorporated into the specification and form a part of this application, showing embodiments consistent with this application, and are used together with the specification to explain the principles of this application.

[0025] Figure 1 is a flowchart of a risk assessment method for leakage gas shown according to an exemplary embodiment of this application;

[0026] Figure 2 is a flowchart of a prediction method for predicted concentration data shown according to an exemplary embodiment of this application;

[0027] Figure 3 is a flowchart of another risk assessment method for leakage gas shown according to an exemplary embodiment of this application;

[0028] Figure 4 is a schematic structural diagram of an electronic device shown according to an exemplary embodiment of this application;

[0029] Figure 5 is a block diagram of a risk assessment device for leakage gas shown according to an exemplary embodiment of this application. Detailed Embodiments

[0030] Here, the exemplary embodiments will be described in detail, and the examples are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. On the contrary, they are merely examples of devices and methods consistent with some aspects of this application as detailed in the appended claims.

[0031] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms "a", "the", and "said" used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0032] It should be understood that although the terms first, second, third, etc. may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "when" or "while" or "in response to determining".

[0033] Currently, the evaluation method for monitoring leaked gas only measures the gas concentration instantaneously on-site and cannot predict the future change trend of the gas concentration. This not only poses challenges to the assessment of personal safety risks, especially in emergency rescue scenarios of gas leakage. For example, when firefighters are performing tasks, such limitations may pose a serious threat to their lives.

[0034] To solve the above problems, the present application proposes a risk assessment method for leaked gas. To further illustrate the present application, the following embodiments are provided:

[0035] Please refer to Figure 1 , Figure 1 FIG. is a flowchart of a risk assessment method for leaked gas shown according to an exemplary embodiment of the present application. This method can be applied to an evaluation system, and the evaluation system can be deployed on terminal devices such as personal computers, laptops, smartphones, tablets, Internet of Things devices, wearable devices, personal digital assistants, fire rescue equipment, etc., or can be deployed on a single server, a cluster server, a cloud server, etc. on the server side.

[0036] The method may include the following steps:

[0037] Step 101: Analyze the collected concentration data of the leaked gas to determine the preliminary parameters of the concentration prediction model. The preliminary parameters include the difference order, the autoregressive order, and the moving average order.

[0038] In this step, the evaluation system can collect the gas concentration data and gas type of the leaked gas in the current area through a collection device such as a sensor, and preprocess the gas concentration data. For example, perform data cleaning, noise reduction, etc. Among them, the collection device can be installed at a fixed position or carried by rescue personnel such as emergency responders for fire emergency and medical rescue.

[0039] The evaluation system arranges the gas concentration data of the leaked gas within a preset historical time period in chronological order to form the time-series collected concentration data. Use statistical methods, visualization tools, machine learning algorithms, etc. to analyze the time-series collected concentration data to determine the preliminary parameters of the concentration prediction model.

[0040] Among them, the leaked gas is the gas leaked in accident scenarios such as a fire scene. The leaked gas may include: toxic gases, and / or, combustible gases, etc. For example: hydrogen chloride (HCl), carbon monoxide (CO), formaldehyde (CH2O), sulfur dioxide (SO2), nitrogen dioxide (NO2), hydrogen fluoride (HF), hydrogen cyanide (HCN), and hydrogen sulfide (H2S), etc.

[0041] The gas type of the leaked gas is the type obtained by classifying according to the specific components of the gas, such as hydrogen chloride, carbon monoxide, formaldehyde, etc. The collected concentration data is the time-series gas concentration data of the leaked gas arranged in chronological order, and can be used to predict the future concentration data of the leaked gas.

[0042] The concentration prediction model is a time-series prediction model for predicting future gas concentration data, and can be an Autoregressive Integrated Moving Average Model (ARIMA) model, a Seasonal Autoregressive Integrated Moving Average (SARIMA) model, etc.

[0043] The initial parameters are the initial values of the model parameters set according to the data characteristics before starting the fitting and tuning, and are used to provide the initial range for the subsequent model parameter tuning. The initial parameters can include the differencing order (d), the autoregressive order (p), and the moving average order (q).

[0044] The differencing order is the number of times of differencing in the concentration prediction model, and is used to transform non-stationary time-series data into stationary data. The autoregressive order is the number of autoregressive terms in the concentration prediction model, that is, the number of correlations between the current data point and past data points. The moving average order is the number of moving average terms in the concentration prediction model, that is, the number of correlations between the current error term and past error terms.

[0045] Step 102: Fit and tune the initial parameters by traversing different parameter combinations of the initial parameters.

[0046] In this step, the evaluation system can adopt optimization methods such as grid search, random search, or genetic algorithm to traverse different parameter combinations of the initial parameters. The values of the initial parameters in different parameter combinations are different to find the optimal parameter combination. During the traversal process, techniques such as cross-validation are used to verify and evaluate the concentration prediction model to select the parameter combination with the best prediction performance, so as to obtain the tuned concentration prediction model.

[0047] Exemplarily, the evaluation system can perform fitting and tuning according to Formula 1 as follows:

[0048]

[0049] where y t represents the observed value of the collected concentration data at time t, μ represents the mean of the sequence, that is, the long-term average level of the collected concentration data, P represents the autoregressive order, q represents the moving average order, and i represents the differencing order. ε tDenote the error term at time t, i.e., the part in the model that cannot be explained by past observations, as θ t Denote the moving average coefficient, representing the past error term ε t-i On the current observation y t The degree of influence.

[0050] Step 103: Use the optimized concentration prediction model to generate predicted concentration data for a future time period.

[0051] In this step, the evaluation system can use the optimized concentration prediction model. Based on the pattern of the collected concentration data changing over time, the concentration prediction model generates predicted concentration data for a future time period. The evaluation system can also input parameters such as the gas type of the leaked gas and environmental information into the concentration prediction model to assist the concentration prediction model in making more accurate predictions.

[0052] Among them, the future time period is temporally after the historical time period and can be set based on the number of prediction periods specified by the user. The future time period can be several minutes or several hours in the future, etc. The predicted concentration data is the concentration data in time series generated by the concentration prediction model according to the actually collected historical concentration data, which can help rescue personnel judge the degree of danger of the environment they are in in real time and provide a reference for rescue personnel to formulate operation and evacuation plans.

[0053] Step 104: Match the predicted concentration data and the gas type of the leaked gas with the preset risk level data to evaluate the predicted risk level of the leaked gas.

[0054] In this step, the evaluation system can evaluate the predicted risk level of the leaked gas according to the predicted concentration data and the gas type of the leaked gas, in combination with the preset risk level data. The preset risk level data can be set according to factors such as gas type, concentration threshold, and diffusion speed.

[0055] By matching the predicted concentration data and the risk level data corresponding to the gas type, the predicted risk level of the leaked gas can be quickly predicted. The predicted risk level can reflect the urgency of the gas leakage incident and the level of response measures that need to be taken, providing a basis for emergency response and decision-making and improving the response speed to potential gas leakage hazards.

[0056] The evaluation system can also match the real-time collected gas concentration data and gas type of the leaked gas with the preset risk level data to determine the real-time risk level of the leaked gas and make real-time judgments and warnings about the degree of danger of the gas leakage scenario.

[0057] It can be understood that in an actual gas leakage scenario, there may be several leakage gases with different concentrations and different gas types simultaneously. The evaluation system can analyze, predict, and match the collected concentration data of the leakage gases of different gas types respectively to obtain the predicted risk level corresponding to each leakage gas. The evaluation system can also combine the multiple predicted risk levels corresponding to multiple leakage gases to obtain a comprehensive predicted risk level.

[0058] In one embodiment, the evaluation system can generate a warning message for the leakage gas according to the predicted risk level and send the warning message to the rescue personnel. Among them, the warning message includes at least one of the following: predicted risk level, gas type, emergency response plan, and maximum operation duration within the safe range.

[0059] The emergency response plan is a detailed action plan formulated for a specific leakage incident, aiming to minimize the harm of the leakage incident to personnel, the environment, and property. The emergency response plan can include information on evacuation routes, isolation measures, fire extinguishing methods, rescue operations, and other aspects. The maximum operation duration within the safe range is the longest time that rescue personnel can operate at the leakage site without being harmed by the leakage gas.

[0060] As described above, by generating a warning message for the leakage gas according to the determined predicted risk level, the predicted risk level can quickly convey the potential danger of the leakage gas, enabling relevant personnel to immediately identify and evaluate the current risk situation; the gas type information helps to guide emergency personnel to adopt the correct protective equipment and emergency measures; the emergency response plan can provide a direct basis for emergency decision-making, helping to reduce the decision-making time and improve the efficiency and effectiveness of emergency response; the maximum operation duration within the safe range information can help emergency personnel reasonably plan the operation time and avoid long-term exposure to a dangerous environment, thereby improving the reliability and comprehensiveness of risk assessment and helping rescue personnel quickly identify and respond to leakage incidents.

[0061] The risk assessment method for the leakage gas in this embodiment analyzes the collected concentration data of the leakage gas to determine the initial parameters of the concentration prediction model, fits and optimizes the initial parameters by traversing different parameter combinations of the initial parameters, uses the optimized concentration prediction model to generate predicted concentration data for a future time period, matches the predicted concentration data and the gas type of the leakage gas with the preset risk level data, and evaluates the predicted risk level of the leakage gas. It can quantitatively classify the potential gas leakage risk, enabling relevant personnel to have sufficient time to take preventive measures according to the predicted risk level, thereby effectively avoiding or reducing the personal safety risk caused by gas leakage in a targeted manner.

[0062] In the foregoing embodiments, the analysis of the collected concentration data was introduced, and the concentration prediction model was fitted and optimized to generate the predicted concentration data and the corresponding predicted risk levels within a future time period. In the following embodiments, the determination process of the preset risk level data will be described in more detail and can be applied to any of the above embodiments.

[0063] In one embodiment, for the leaked gas belonging to highly concentrated toxic gas, the evaluation system can obtain the gas concentration data at each time point within different preset exposure times of the leaked gas. Among them, the preset exposure time is the operation time of the rescue personnel in the gas leakage scenario, and each time point within the preset exposure time corresponds to a gas concentration data. The time interval between each time point can be set according to actual needs, such as every second, every minute or every hour, etc.

[0064] The Gaussian-Hermite integration method can be used to perform curve fitting and integration on the variation relationship of the gas concentration data with each time point to obtain the cumulative inhalation dose of the leaked gas by the rescue personnel within the preset exposure time.

[0065] Use the Hermite polynomial and the corresponding Gaussian nodes to perform curve fitting on the variation relationship of the gas concentration data with each time point. After the curve fitting is completed, use the Gauss-Hermite integration formula to integrate the fitted curve within the integration limits to obtain the cumulative inhalation dose.

[0066] Exemplarily, the evaluation system can calculate the cumulative inhalation dose according to the following formula 2:

[0067]

[0068] where X i represents the Gauss-Hermite node, W i represents the corresponding weight, and n represents the number of Gauss-Hermite nodes. The Hermite polynomial is a set of special orthogonal polynomials that are orthogonal to the weight function in the interval (-∞, ∞). orthogonal.

[0069] The Hermite polynomial can be defined by a recurrence relation to construct the Gauss-Hermite nodes and weights. The Gaussian-Hermite integration method can ensure that the fitted curve passes through each data point, and the derivative value of the fitted curve at each data point is the same as the original data, thereby improving the accuracy of the fitted curve and the stability of its change trend.

[0070] As described above, by obtaining the gas concentration data at each time point during the exposure time and using the Gauss-Hermite integration method to perform curve fitting and integration on the variation relationship of the gas concentration data with each time point, the cumulative inhalation dose is obtained. Since the Gauss-Hermite integration method has high precision in dealing with function integration and is particularly suitable for gas concentration data with exponential decay or similar forms, curve fitting and integration by this method can more accurately reflect the variation relationship of the gas concentration data with time, thereby improving the calculation accuracy of the cumulative inhalation dose; moreover, by avoiding complex mathematical operations and model assumptions, the risk level classification process can be simplified, and the efficiency and accuracy of risk level classification can be improved.

[0071] It can be understood that the above-mentioned Gauss-Hermite integration method is only a preferred example for calculating the cumulative inhalation dose, and other calculation methods can also be used. The embodiments of the present application do not limit this.

[0072] In one embodiment, the evaluation system can calculate the cumulative inhalation dose according to the gas concentration data of the leaked gas during the preset exposure time, combining the gas concentration data and the preset exposure time.

[0073] Using the Probit Model, based on the preset exposure time and the gas concentration data at each time point during the preset exposure time, the lethal factor corresponding to the leaked gas is determined. Among them, the Probit Model is a probability model reflecting human vulnerability and can consider the cumulative effect of working hours on the lethal factor.

[0074] Exemplarily, the evaluation system can determine the lethal factor according to the following formula 3:

[0075] Y = a + bln(c n t) Formula 3

[0076] Wherein, Y represents the lethal factor, c represents the gas concentration data at the time point during the preset exposure time, with the unit of ppm, t represents the preset exposure time, with the unit of min, and a, b, and n represent the constants of the harm of the leaked gas to the human body.

[0077] The evaluation system can use the probability analysis model to calculate the death probability of the leaked gas by combining the cumulative inhalation dose and the lethal factor, converting the quantified lethal factor into a more practical death risk index. Among them, the probability analysis model can evaluate the dose-effect relationship of the human body to the leaked gas.

[0078] Taking the calculation of the death probability based on the normal Gaussian probability distribution function as an example, the evaluation system can calculate the death probability according to the following formula 4:

[0079]

[0080] Among them, V represents the probability of death, Y represents the lethal factor, σ represents the standard deviation of the normal distribution, which represents the degree of dispersion of the probability of death, and u represents the integration variable.

[0081] The evaluation system can use the probability of death as an evaluation index to divide the danger level of the leaked gas into different predicted risk levels, and obtain the risk level data corresponding to the preset exposure time.

[0082] The evaluation system can also output the obtained risk level data to the user interface or related devices so that relevant personnel can take corresponding measures in a timely manner. The evaluation system can also be stored in a database or related files for subsequent evaluation and decision-making.

[0083] Exemplarily, taking the preset exposure time of 3 minutes as an example, according to the probability of death, the concentration thresholds of each predicted risk level are divided for high-concentration toxic gases, and the obtained risk level data can be shown in Table 1:

[0084] Table 1 Risk level data of high-concentration toxic gases when exposed for 3 minutes

[0085]

[0086] Taking the preset exposure time of 60 minutes as an example, according to the probability of death, the concentration thresholds of each predicted risk level are divided for high-concentration toxic gases, and the obtained risk level data can be shown in Table 2:

[0087] Table 2 Risk level data of high-concentration toxic gases when exposed for 60 minutes

[0088]

[0089] It can be understood that the above-mentioned preset exposure times of 3 minutes and 60 minutes are just examples, and risk level data can also be divided based on different preset exposure times. For subsequent prediction risk level evaluation, the risk level data corresponding to the actual exposure time can be determined according to the actual exposure time of the rescue personnel in the gas leakage scenario, and the predicted risk level of the leaked gas can be evaluated by matching the predicted concentration data and gas type with the risk level data.

[0090] As described above, by calculating the cumulative inhalation dose based on the gas concentration data of the leaked gas within the preset exposure time, determining the lethal factor corresponding to the leaked gas based on the preset exposure time and gas concentration data, calculating the probability of death of the leaked gas by combining the cumulative inhalation dose and the lethal factor, and dividing the risk level of the leaked gas into different risk levels according to the probability of death to obtain the risk level data, it is possible to comprehensively consider the symptoms of the human body in the toxic gas leakage environment and the time cumulative effect, quantify the risk levels of different leaked gases, and thus more accurately evaluate the potential threat of the leaked gas to human health, providing a scientific basis for risk assessment.

[0091] In one embodiment, for the leaked gas belonging to low-concentration toxic gas, the evaluation system can divide the predicted risk level of the toxic gas according to the preset national standards, occupational health guidelines and other industrial classification standards and the degree of human injury caused by the toxic gas to obtain the risk level data.

[0092] Exemplarily, the evaluation system can formulate the following classification rules according to the industrial classification standards and the degree of human injury caused by the toxic gas within the maximum eight-hour preset exposure time:

[0093] Level 1: The maximum allowable exposure concentration at which there is no harm to workers within eight hours.

[0094] Level 2: Staying in the environment for a period of time will cause certain damage to the human body (such as irritation to the skin or organs).

[0095] Level 3: Staying in the environment for a period of time will cause serious damage to the human body (such as damage to the respiratory tract or lungs).

[0096] Level 4: Death may occur after staying in the environment for a period of time (such as death may start to occur after more than 1 hour, that is, the lethal factor at 60 min ≥ 1).

[0097] Level 5: Staying in the environment for one to three minutes may cause death, and at this time, evacuation should be carried out at the fastest speed (such as the median lethal concentration (LC50) at 30 min).

[0098] Among them, the maximum allowable exposure concentration can be set and adjusted based on standard rules such as the Occupational Health and Safety Act of the United States.

[0099] Exemplarily, according to the above classification rules, the concentration thresholds for dividing each predicted risk level of the low-concentration toxic gas are obtained, and the risk level data can be shown in Table 3:

[0100] Table 3 Risk level data of low-concentration toxic gas

[0101]

[0102] As described above, by classifying the risk levels of toxic gases according to the preset industrial classification standards and the degree of human harm caused by toxic gases, predictive risk level data can be obtained, which can more scientifically classify the risk levels of toxic gases, more accurately reflect the actual harm degree of toxic gases, and the risk level data obtained by using this classification method is more accurate and reliable, which helps to improve the accuracy of risk assessment.

[0103] In one embodiment, for the leaked gas belonging to combustible gas, the evaluation system can classify the predictive risk level of the combustible gas according to the minimum explosion concentration of the combustible gas in the air and the preset safety threshold to obtain the risk level data.

[0104] Exemplarily, the reference safety threshold can be set in the way of the danger classification of gas leakage in a mine. When the methane concentration in the working environment reaches 2.0% or more and continues to increase, all personnel should immediately evacuate to a safe place. The alarm, power-off, and power-recovery concentrations of the methane sensor (portable instrument) are 1.0%, 1.5%, and <1.0% respectively. When gas accumulates, it must be discharged according to regulations. Only when the methane concentration in the return air current does not exceed 1.0% and the carbon dioxide concentration does not exceed 1.5% can work be carried out.

[0105] The reference safety threshold can be set to 1.0%, 1.5%, and 2%. Specifically, the gas concentration of methane cannot exceed 1.0%. If it exceeds 1.0%, work should be stopped. When using electrically charged equipment such as methane detectors, the gas concentration of methane cannot exceed 1.5%. When the methane concentration exceeds 2%, all rescue personnel should be organized to evacuate the rescue site.

[0106] For other combustible gases, taking hydrogen as an example, the minimum explosion concentration of hydrogen in the air is 4%, which is less than the minimum explosion concentration of methane in the air, which is 5%. According to the similarity of the properties of methane and hydrogen, the predictive risk level of hydrogen can be analogized based on the minimum explosion concentration of hydrogen in the air and the reference safety threshold of methane, and the predictive risk level of combustible gas can be classified to obtain the risk level data.

[0107] Exemplarily, the concentration thresholds for classifying each predictive risk level of combustible gas are obtained, and the risk level data can be as shown in Table 4:

[0108] Table 4 Risk Level Data of Combustible Gas

[0109]

[0110] As described above, by classifying the risk level of combustible gas based on the minimum explosion concentration of combustible gas in the air and a preset reference safety threshold, risk level data is obtained. Based on the explosion behavior of combustible gas under the minimum explosion concentration according to its physical and chemical properties, the risk level of combustible gas can be evaluated more scientifically, thereby improving the accuracy and reliability of risk level classification.

[0111] In the foregoing embodiments, the risk level data obtained by using different methods for combustible gas, highly concentrated toxic gas, and low-concentrated toxic gas is introduced. In the following embodiments, the prediction process of predicted concentration data will be described in more detail and can be applied to any of the above embodiments.

[0112] In one embodiment, for the gas leakage scenario indoors, as the leakage time increases, the concentration change of gas concentration data will gradually approach the saturation state, and a logarithmic model can be used to simulate and predict the concentration change trend of the leaked gas.

[0113] The evaluation system can evaluate the collected concentration data, perform a logarithmic transformation on the gas concentration data at each time point in the collected concentration data, and fit it to a logarithmic function. Using statistical methods such as R 2 value, mean square error MSE, and residual analysis, calculate the logarithmic goodness of fit between the transformed data points and the ideal logarithmic curve to obtain the logarithmic goodness of fit of the collected concentration data.

[0114] Compare the calculated logarithmic goodness of fit with the preset fitting conditions. The preset fitting conditions can evaluate whether the change trend of the collected concentration data conforms to the logarithmic model, such as a preset goodness of fit threshold or range, etc., which can be set based on the analysis of historical data, the experience of domain experts, or specific application requirements.

[0115] If the logarithmic goodness of fit does not meet the preset fitting conditions and the logarithmic goodness of fit is low, a tuned concentration prediction model can be used to predict the predicted concentration data of the leaked gas.

[0116] When it is determined that the logarithmic goodness of fit meets the preset fitting conditions and the logarithmic goodness of fit is high (such as the R 2 value is close to 1, the MSE is small, etc.), it can be determined that the change trend of the collected concentration data conforms to the logarithmic model. According to the characteristics of the collected concentration data and the evaluation results of the logarithmic goodness of fit, a suitable logarithmic model such as a natural logarithmic model or an exponential logarithmic model is selected.

[0117] The polyfit function in Python can be applied to logarithmically fit the collected concentration data using a logarithmic model to obtain the fitting line of the leaked gas. Based on the fitting line, the predicted concentration data for a future time period can be determined. For example, select the time point to be predicted and substitute it into the fitting line to calculate the corresponding predicted concentration data.

[0118] Among them, the logarithmic fitting degree is used to reflect the closeness between the collected concentration data and the logarithmic curve. The fitting line is used to identify the change trend of the leaked gas in the future time period.

[0119] As described above, by evaluating the collected concentration data and calculating the logarithmic fitting degree of the collected concentration data, the closeness between the collected concentration data and the logarithmic curve can be quantitatively evaluated, which helps to identify potential laws and trends in the data, thereby improving the scientific nature of data processing; when it is determined that the logarithmic fitting degree meets the preset fitting conditions, the logarithmic model is used to logarithmically fit the collected concentration data to obtain the fitting line of the leaked gas, which can visually display the dynamic changes of the gas concentration. Based on the fitting line, reliable predicted concentration data can be quickly determined.

[0120] In one embodiment, in an actual gas leakage scenario, due to the complexity and unpredictability of the operating environment, the collected concentration data of the leaked gas may be affected by various factors such as temperature, humidity, wind speed, etc., resulting in fluctuations.

[0121] To ensure the effectiveness of subsequent analysis, white noise detection can be performed on the collected concentration data, and the unit root test method can be applied for stationarity testing to determine the applicability of the data. Please refer to Figure 2 , Figure 2 which shows a flowchart of a prediction method for predicted concentration data. The method may include the following steps:

[0122] Step 201: Obtain the collected concentration data of the leaked gas.

[0123] In this step, the evaluation system can obtain the collected concentration data of the leaked gas in a historical time period.

[0124] Step 202: Determine whether the collected concentration data belongs to a white noise sequence.

[0125] In this step, the evaluation system can adopt the Ljung-Box test method. According to the autocorrelation coefficient of the collected concentration data, calculate the statistic of the autocorrelation coefficient, which can be used to determine whether the collected concentration data has autocorrelation. Compare the statistic with the preset autocorrelation condition to determine whether the collected concentration data is a white noise sequence.

[0126] When it is determined that the statistic satisfies the preset autocorrelation condition, it can be determined that the collected concentration data does not belong to the white noise sequence. When it is determined that the statistic does not satisfy the preset autocorrelation condition, it can be determined that the collected concentration data belongs to the white noise sequence.

[0127] Exemplarily, the preset autocorrelation condition can be that the statistic is greater than the critical value, determining that the time series has significant autocorrelation and does not belong to the white noise sequence; conversely, if the statistic is less than or equal to the critical value, it is determined that the time series is white noise, that is, there is no significant autocorrelation in the collected concentration data.

[0128] If the collected concentration data does not belong to the white noise sequence, step 203 can be continued;

[0129] If the collected concentration data belongs to the white noise sequence, step 201 can be executed.

[0130] Step 203: Use the unit root test method to perform a stationarity test on the collected concentration data.

[0131] In this step, the evaluation system can use the unit root test method to perform a stationarity test on the collected concentration data. For example, based on the Augmented Dickey-Fuller (ADF) detection method, a unit root test is performed on the collected concentration data, and the test threshold is set to 0.05 to detect whether there is a unit root in the collected concentration data.

[0132] Among them, the test threshold is the standard for judging whether the collected concentration data is stationary. For example: 0.01, 0.025, 0.05, 0.1, etc., which can be set based on the strictness of the unit root test.

[0133] Step 204: Determine whether there is a unit root in the collected concentration data.

[0134] In this step, the evaluation system can obtain the unit root test result of the collected concentration data. The unit root test result can include the ADF statistic. The ADF statistic is used to measure whether there is a unit root in the collected concentration data. The smaller the ADF statistic, the stronger the evidence to reject the null hypothesis (the collected concentration data has a unit root).

[0135] If the ADF statistic is lower than the test threshold, it can be determined that the collected concentration data does not have a unit root, and the collected concentration data is stationary data, meeting the stationarity condition; if the ADF statistic is higher than or equal to the test threshold, it can be determined that the collected concentration data has a unit root, and the collected concentration data is non-stationary data, not meeting the stationarity condition.

[0136] If the collected concentration data has a unit root, step 205 can be continued;

[0137] If the collected concentration data does not have a unit root, step 206 can be executed.

[0138] Step 205: Perform differencing on the collected concentration data.

[0139] In this step, when the collected concentration data has a unit root, perform differencing on the collected concentration data, and create the differenced collected concentration data by calculating the differences between adjacent data points in the collected concentration data, so as to eliminate the trend and periodic components of the collected concentration data.

[0140] During the differencing process, repeatedly perform differencing and stationarity tests on the collected concentration data until the stationarity test is passed to determine the differencing order that meets the stationarity condition.

[0141] Step 206: Traverse different parameter combinations of the preliminary parameters to obtain the optimal parameter combination.

[0142] In this step, when the collected concentration data does not have a unit root, it can be determined that the differencing order is 0. The evaluation system can construct an Autocorrelation Function (ACF) graph and a Partial Autocorrelation Function (PACF) graph based on the collected concentration data. Based on the ACF graph and PACF graph of the collected concentration data, determine the autoregressive order and moving average order in the preliminary parameters.

[0143] It can be understood that the above methods for determining the differencing order, autoregressive order, and moving average order are only examples, and other analysis methods can also be used. The embodiments of the present application do not limit this.

[0144] Exemplarily, the itertools module in python can be used to traverse different parameter combinations of the preliminary parameters, and the goodness or badness of the concentration prediction models fitted by different parameter combinations can be judged according to the Bayesian estimation values of the concentration prediction models until the optimal parameter combination is obtained.

[0145] Step 207: Use the optimal parameter combination to construct a tuned concentration prediction model.

[0146] In this step, the evaluation system can use the optimal parameter combination to construct a tuned concentration prediction model.

[0147] Step 208: Use the tuned concentration prediction model to generate predicted concentration data.

[0148] In this step, the evaluation system can generate predicted concentration data for a future time period according to the specified number of prediction periods by using the tuned concentration prediction model.

[0149] As described above, by calculating the statistic of the autocorrelation coefficient according to the autocorrelation coefficient of the collected concentration data, and using the unit root test method to perform a stationarity test on the collected concentration data when it is determined that the statistic meets the preset autocorrelation condition, data sequences that are random and without rules can be excluded, providing a more accurate data basis for subsequent analysis; when it is determined that the collected concentration data has a unit root, differencing the collected concentration data can effectively eliminate the random trend of the collected concentration data and determine the differencing order that meets the stationarity condition. This can not only serve as the basis for subsequent time series analysis, prediction, and modeling, but also ensure that the collected concentration data after differencing can accurately reflect the characteristics and trends of the original data.

[0150] To further introduce the process of predicting the concentration of leaked gas and classifying the prediction risk levels, Figure 3 The flowchart of another risk assessment method for leaked gas is shown. As Figure 3 shown, the method may include the following steps:

[0151] Step 301: Construct risk level data for high-concentration toxic gas based on the cumulative inhalation dose and the death probability corresponding to the lethality factor.

[0152] In this step, the evaluation system can obtain gas concentration data at each time point within different preset exposure times. Using the Gauss-Hermite integration method, curve fitting and integration are performed on the relationship between the gas concentration data and each time point to obtain the cumulative inhalation dose of the rescue personnel for the leaked gas.

[0153] The human vulnerability model can be used to determine the lethality factor corresponding to the leaked gas based on the preset exposure time and gas concentration data. Combining the cumulative inhalation dose and the lethality factor, the harm suffered by the human body due to the cumulative effect over time in the leakage environment of the toxic gas is quantified, and the death probability of the leaked gas is calculated. According to the death probability, the danger level of the leaked gas is divided into different prediction risk levels to obtain the risk level data of the high-concentration toxic gas at the preset exposure time.

[0154] For high-concentration toxic gas, risk level data corresponding to different preset exposure times can be constructed based on the cumulative inhalation dose and the lethality factor at different preset exposure times, and the risk level data of the high-concentration toxic gas can be stored in the database.

[0155] Step 302: Construct risk level data for low-concentration toxic gas based on the preset industrial classification standard and the degree of human injury.

[0156] In this step, the evaluation system can classify the predicted risk levels of low-concentration toxic gases according to the preset industrial classification standards and the degree of human harm caused by toxic gases, construct risk level data for low-concentration toxic gases, and store the risk level data for low-concentration toxic gases in the database.

[0157] Step 303: Construct risk level data for combustible gases based on the minimum explosion concentration of combustible gases in the air and the reference safety threshold.

[0158] In this step, the evaluation system can classify the predicted risk levels of combustible gases according to the minimum explosion concentration of combustible gases in the air and the reference safety threshold, construct risk level data for combustible gases, and store the risk level data for combustible gases in the database.

[0159] Step 304: Collect the gas concentration data and gas type of the leaked gas.

[0160] In this step, the evaluation system can use the collection device to collect the gas concentration data and gas type of the leaked gas in real time. The gas concentration data and gas type can be compared with the risk level data in the database to evaluate the real-time risk level of the leaked gas.

[0161] Exemplarily, based on the gas concentration data and gas type collected in real time, it can be determined whether the leaked gas is a combustible gas, a high-concentration toxic gas, or a low-concentration toxic gas. If the leaked gas is a combustible gas or a low-concentration toxic gas, the gas concentration data and gas type can be matched with the corresponding risk level data to determine the real-time risk level of the leaked gas.

[0162] If the leaked gas is a high-concentration toxic gas, the actual exposure time of the rescue personnel working in the gas leakage scenario until the gas concentration data is collected can be calculated, and the risk level data corresponding to the actual exposure time can be searched in the database. The gas concentration data and gas type can be matched with the risk level data corresponding to the actual exposure time to determine the real-time risk level of the leaked gas.

[0163] The evaluation system can generate real-time warning information for the leaked gas according to the real-time risk level of the leaked gas in the gas leakage scenario. Exemplarily, the real-time warning information can be "The gas around you is currently a toxic gas, and the real-time risk level is high risk. Please leave this area within a short time"; there may be multiple leaked gases in the gas leakage scenario, and the real-time warning information can also be "The comprehensive risk level for working here is: the current real-time risk level of hydrogen is low risk, and normal work can be carried out; the current real-time risk level of methane is low risk, and normal work can be carried out".

[0164] Real-time warning information can also be "The cumulative inhalation dose of the leaked gas in the past time is 271.58, the death probability is 0, and the real-time risk level is low risk; the cumulative inhalation dose of the leaked gas in the past time is 289.58, the death probability is 0, and the real-time risk level is low risk; the cumulative inhalation dose of the leaked gas in the past time is 424.58, the death probability is 0, and the real-time risk level is low risk; the cumulative inhalation dose of the leaked gas in the past time is 1173.16, the death probability is 0, and the real-time risk level is low risk".

[0165] Step 305: Determine whether the collected concentration data meets the preset fitting conditions.

[0166] In this step, the evaluation system can collect the gas concentration data with a concentration exceeding the set threshold within the historical time period to form the collected concentration data. Evaluate the collected concentration data, calculate the logarithmic fitting degree of the collected concentration data, and determine whether the logarithmic fitting degree of the collected concentration data meets the preset fitting conditions.

[0167] If the logarithmic fitting degree meets the preset fitting conditions, step 306 can be continued;

[0168] If the logarithmic fitting degree does not meet the preset fitting conditions, step 308 can be executed.

[0169] Step 306: Perform logarithmic fitting on the collected concentration data using the logarithmic model to obtain the fitting line of the leaked gas.

[0170] In this step, when the logarithmic fitting degree meets the preset fitting conditions, the change trend of the collected concentration data conforms to the logarithmic model, and the evaluation system can perform logarithmic fitting on the collected concentration data using the logarithmic model to obtain the fitting line of the leaked gas.

[0171] Step 307: Determine the predicted concentration data in the future time period according to the fitting line.

[0172] In this step, the evaluation system can determine the predicted concentration data in the future time period according to the fitting line obtained by logarithmic fitting.

[0173] Step 308: Perform white noise detection and stationarity test on the collected concentration data of the leaked gas.

[0174] In this step, when the logarithmic fitting degree does not meet the preset fitting conditions, the evaluation system can use the concentration prediction model for concentration prediction. White noise detection can be performed on the collected concentration data of the leaked gas, and according to the autocorrelation coefficient of the collected concentration data, the statistic of the autocorrelation coefficient is calculated to determine whether the statistic meets the preset autocorrelation conditions.

[0175] If the statistic satisfies the preset autocorrelation condition, it can be determined that the collected concentration data does not belong to a purely random white noise sequence. The unit root test method is used to perform a stationarity test on the collected concentration data.

[0176] When it is determined that the collected concentration data has a unit root, the collected concentration data is differenced to determine the differencing order that satisfies the stationarity condition. Using the autocorrelation function graph and partial autocorrelation function graph of the collected concentration data, the autoregressive order and moving average order in the preliminary parameters are determined.

[0177] Step 309: By traversing different parameter combinations of the preliminary parameters, the preliminary parameters are fitted and optimized.

[0178] In this step, the evaluation system can traverse different parameter combinations of the preliminary parameters to fit and optimize the preliminary parameters and determine the optimal parameter combination.

[0179] Step 310: Using the optimized concentration prediction model, predicted concentration data for a future time period is generated.

[0180] In this step, the evaluation system can use the optimal parameter combination to construct an optimized concentration prediction model. Using the optimized concentration prediction model, according to the required number of prediction periods, predicted concentration data for a future time period is generated.

[0181] Step 311: The predicted concentration data and the gas type are matched with the risk level data to evaluate the predicted risk level of the leaked gas.

[0182] In this step, the evaluation system can match the predicted concentration data and the gas type with the risk level data in the database to evaluate the predicted risk level of the leaked gas.

[0183] Exemplarily, the specified number of prediction periods by the user is 3. The optimized concentration prediction model can generate predicted concentration data for the next 3 periods. The predicted concentration data for the next 3 periods may include the gas concentration data for the 0th future period, the gas concentration data for the 1st future period, and the gas concentration data for the 2nd future period.

[0184] The evaluation system respectively matches the gas concentration data for the 0th future period, the gas concentration data for the 1st future period, and the gas concentration data for the 2nd future period with the risk level data corresponding to the gas type to evaluate the predicted risk level for the next 3 periods.

[0185] Specifically, according to the gas type of the leaked gas, it can be determined whether the leaked gas is a combustible gas, a highly concentrated toxic gas, or a low-concentration toxic gas. If the leaked gas is a combustible gas or a low-concentration toxic gas, the predicted concentration data and the gas type can be matched with the corresponding risk level data to determine the predicted risk level for the next 3 periods.

[0186] If the leaked gas is a highly concentrated toxic gas, the actual exposure time of the rescue personnel during operation in the gas leakage scenario to a future time period can be calculated, and the risk level data corresponding to the actual exposure time can be searched in the database. Match the predicted concentration data and gas type with the risk level data corresponding to the actual exposure time to determine the predicted risk levels for the next three periods.

[0187] Step 312: Generate early warning information for the leaked gas according to the predicted risk level.

[0188] In this step, the evaluation system can generate early warning information such as the predicted risk level of the leaked gas, gas type, cumulative inhalation dose, emergency response plan, and maximum operation duration within the safe range according to the predicted risk level.

[0189] Figure 4 This is a schematic structural diagram of an electronic device shown according to an exemplary embodiment of the present application. The electronic device can be, for example, a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a personal digital assistant, a server, a smart home appliance, a vehicle-mounted computer, a fire rescue device, etc. Refer to Figure 4 , at the hardware level, the electronic device includes a processor 401, an internal bus 402, a network interface 403, a memory 404, and a non-volatile memory 405. Of course, it may also include other hardware required for other services. The processor 401 reads the corresponding computer program from the non-volatile memory 405 into the memory 404 and then runs it, forming a risk assessment device for leaked gas at the logical level. Of course, in addition to the software implementation method, the present application does not exclude other implementation methods, such as logical devices or a combination of software and hardware. That is to say, the execution subject of the following processing flow is not limited to each logical unit, and can also be hardware or a logical device.

[0190] Figure 5 This is a block diagram of a risk assessment device for leaked gas shown according to an exemplary embodiment of the present application. Refer to Figure 5 , the device may include: an analysis module 501, a tuning module 502, a prediction module 503, and an evaluation module 504, where:

[0191] The analysis module 501 is used to analyze the collected concentration data of the leaked gas to determine the preliminary parameters of the concentration prediction model, and the preliminary parameters include the difference order, the autoregressive order, and the moving average order;

[0192] The tuning module 502 is used to perform fitting tuning on the preliminary parameters by traversing different parameter combinations of the preliminary parameters;

[0193] The prediction module 503 is configured to generate predicted concentration data for a future time period by using the optimized concentration prediction model;

[0194] The evaluation module 504 is configured to match the predicted concentration data and the gas type of the leaked gas with preset risk level data, and evaluate the predicted risk level of the leaked gas.

[0195] In one example, before the evaluation module 504 is configured to match the predicted concentration data and the gas type of the leaked gas with preset risk level data and evaluate the predicted risk level of the leaked gas, the evaluation module 504 further includes: calculating an accumulated inhalation dose according to the gas concentration data of the leaked gas within a preset exposure time; determining a lethality factor corresponding to the leaked gas based on the preset exposure time and the gas concentration data; calculating a fatality probability of the leaked gas by combining the accumulated inhalation dose and the lethality factor; and dividing the risk degree of the leaked gas into different risk levels according to the fatality probability to obtain the risk level data.

[0196] In one example, when the evaluation module 504 is configured to calculate the accumulated inhalation dose according to the gas concentration data of the leaked gas within a preset exposure time, the evaluation module 504 includes: obtaining the gas concentration data at each time point within the preset exposure time; and using the Gauss-Hermite integration method to perform curve fitting and integration on the variation relationship of the gas concentration data with respect to each time point to obtain the accumulated inhalation dose.

[0197] In one example, before the evaluation module 504 is configured to match the predicted concentration data and the gas type of the leaked gas with preset risk level data and evaluate the predicted risk level of the leaked gas, the evaluation module 504 further includes: evaluating the collected concentration data and calculating a logarithmic goodness of fit of the collected concentration data, where the logarithmic goodness of fit is used to reflect the degree of closeness between the collected concentration data and a logarithmic curve; in the case where it is determined that the logarithmic goodness of fit meets a preset fitting condition, performing logarithmic fitting on the collected concentration data by using a logarithmic model to obtain a fitting line of the leaked gas, where the fitting line is used to identify the variation trend of the leaked gas in the future time period; and determining the predicted concentration data according to the fitting line.

[0198] In one example, when the analysis module 501 is used to analyze the collected concentration data of the leaked gas to determine the preliminary parameters of the concentration prediction model, it includes: calculating the statistic of the autocorrelation coefficient according to the autocorrelation coefficient of the collected concentration data; when determining that the statistic meets the preset autocorrelation condition, using the unit root test method to perform a stationarity test on the collected concentration data; when determining that the collected concentration data has a unit root, performing a differencing process on the collected concentration data to determine the differencing order that meets the stationarity condition.

[0199] In one example, the leaked gas includes a toxic gas; before the evaluation module 504 is used to match the predicted concentration data and the gas type of the leaked gas with the preset risk level data to evaluate the predicted risk level of the leaked gas, it further includes: dividing the risk level of the toxic gas according to the preset industrial classification standard and the degree of human injury of the toxic gas to obtain the risk level data.

[0200] In one example, the leaked gas includes a combustible gas; before the evaluation module 504 is used to match the predicted concentration data and the gas type of the leaked gas with the preset risk level data to evaluate the predicted risk level of the leaked gas, it further includes: dividing the risk level of the combustible gas according to the minimum explosion concentration of the combustible gas in the air and the preset reference safety threshold to obtain the risk level data.

[0201] In one example, the evaluation module 504 is further used to generate a warning message for the leaked gas according to the predicted risk level, and the warning message includes at least one of the following: the predicted risk level, the gas type, the emergency response plan, and the maximum operation duration within the safe range.

[0202] The implementation processes of the functions and roles of each unit in the above device are specifically described in detail in the implementation processes of the corresponding steps in the above method, and will not be elaborated here.

[0203] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiments described above are only illustrative. The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of the present application. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0204] In an exemplary embodiment, there is also provided a non-transitory computer-readable storage medium including instructions, such as a memory including instructions, which can be executed by a processor of a leakage gas risk assessment device to implement the method described in any one of the above embodiments.

[0205] Among them, the non-transitory computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc., and the present application does not limit this.

[0206] In an exemplary embodiment, there is also provided a computer program product including a computer program / instructions, which can be executed by a processor of a leakage gas risk assessment device to implement the method described in any one of the above embodiments.

[0207] The above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired result. In certain implementations, multitasking and parallel processing are also possible or may be advantageous.

[0208] Those skilled in the art will readily conceive of other embodiments of the present application after considering the specification and practicing the invention herein. The present application is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present application is only limited by the appended claims.

[0209] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the scope of protection of the present application.

Claims

1. A risk assessment method for leaking gas, characterized in that, The method includes: Analyze the collected concentration data of the leaked gas to determine the preliminary parameters of the concentration prediction model, where the preliminary parameters include the difference order, autoregressive order, and moving average order; Fit and optimize the preliminary parameters by traversing different parameter combinations of the preliminary parameters; Use the optimized concentration prediction model to generate predicted concentration data for a future time period; Match the predicted concentration data and the gas type of the leaked gas with preset risk level data to evaluate the predicted risk level of the leaked gas.

2. The method according to claim 1, wherein Before the step of matching the predicted concentration data and the gas type of the leaked gas with preset risk level data to evaluate the predicted risk level of the leaked gas, the method further includes: Calculate the cumulative inhalation dose according to the gas concentration data of the leaked gas within a preset exposure time; Determine the lethality factor corresponding to the leaked gas based on the preset exposure time and the gas concentration data; Calculate the fatality probability of the leaked gas by combining the cumulative inhalation dose and the lethality factor; Divide the risk level of the leaked gas into different risk levels according to the fatality probability to obtain the risk level data.

3. The method according to claim 2, characterized in that, The step of calculating the cumulative inhalation dose according to the gas concentration data of the leaked gas within a preset exposure time includes: Obtain the gas concentration data at each time point within the preset exposure time; Use the Gauss-Hermite quadrature method to perform curve fitting and integration on the variation relationship of the gas concentration data with respect to each time point to obtain the cumulative inhalation dose.

4. The method according to claim 1, characterized in that, Before the step of matching the predicted concentration data and the gas type of the leaked gas with preset risk level data to evaluate the predicted risk level of the leaked gas, the method further includes: Evaluate the collected concentration data and calculate the logarithmic goodness of fit of the collected concentration data, where the logarithmic goodness of fit is used to reflect the closeness between the collected concentration data and the logarithmic curve; When it is determined that the logarithmic goodness of fit meets the preset fitting condition, use a logarithmic model to perform logarithmic fitting on the collected concentration data to obtain the fitting line of the leaked gas, and the fitting line is used to identify the change trend of the leaked gas in the future time period; Determine the predicted concentration data according to the fitting line.

5. The method according to claim 1, wherein The step of analyzing the collected concentration data of the leaked gas to determine the preliminary parameters of the concentration prediction model includes: Calculate the statistic of the autocorrelation coefficient according to the autocorrelation coefficient of the collected concentration data; When it is determined that the statistic meets the preset autocorrelation condition, use the unit root test method to perform a stationarity test on the collected concentration data; When it is determined that the collected concentration data has a unit root, perform differencing on the collected concentration data to determine the difference order that meets the stationarity condition.

6. The method according to claim 1, wherein The leaked gas includes toxic gas; Before the step of matching the predicted concentration data and the gas type of the leaked gas with preset risk level data to evaluate the predicted risk level of the leaked gas, the method further includes: Divide the risk level of the toxic gas according to the preset industrial classification standard and the degree of human injury caused by the toxic gas to obtain the risk level data.

7. The method according to claim 1, characterized in that The leaked gas includes combustible gas; Before matching the predicted concentration data and the gas type of the leaked gas with the preset risk level data to evaluate the predicted risk level of the leaked gas, the method further includes: Divide the risk level of the combustible gas according to the minimum explosion concentration of the combustible gas in the air and the preset reference safety threshold to obtain the risk level data.

8. The method according to claim 1, wherein The method further includes: Generate a warning message for the leaked gas according to the predicted risk level, where the warning message includes at least one of the following: the predicted risk level, the gas type, the emergency response plan, and the maximum operation duration within the safe range.

9. A risk assessment device for leaked gas, characterized in that, The device includes: An analysis module for analyzing the collected concentration data of the leaked gas to determine the preliminary parameters of the concentration prediction model, where the preliminary parameters include the difference order, the autoregressive order, and the moving average order; An optimization module for fitting and optimizing the preliminary parameters by traversing different parameter combinations of the preliminary parameters; A prediction module for generating predicted concentration data for a future time period by using the optimized concentration prediction model; An evaluation module for matching the predicted concentration data and the gas type of the leaked gas with the preset risk level data to evaluate the predicted risk level of the leaked gas.

10. An electronic device, characterized in that, Includes: A processor; A memory for storing instructions executable by the processor; Wherein, the processor realizes the method according to any one of claims 1-8 by running the executable instructions.

11. A computer-readable storage medium having computer instructions stored thereon, characterized in that, When the instruction is executed by the processor, the method according to any one of claims 1-8 is realized.

12. A computer program product, on which a computer program / instructions are stored, characterized in that, When the computer program / instruction is executed by the processor, the method according to any one of claims 1-8 is realized.

Images