Safety risk assessment method and system for electronic atomization apparatus, and readable storage medium
By collecting and fuzzing data from electronic atomizing devices and using Bayesian networks for risk calculation, the problem of difficulty in assessing the safety of electronic atomizing devices in existing technologies is solved, enabling dynamic, real-time risk assessment and scientific risk warning during their use.
Patent Information
- Application Number
- PCT/CN2025/135482
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-09
- Filing Date
- 2025-11-17
- Publication Date
- 2026-11-12
AI Technical Summary
Existing technologies are insufficient to effectively assess the potential safety risks of using electronic atomization devices, thus affecting their safety and health risks.
By collecting battery data, atomization data, chemical composition data, and usage behavior data of electronic atomization devices, and after fuzzification processing, the data is input into a Bayesian network for risk calculation to obtain safety risk assessment results, including battery safety risk level, atomization safety risk level, chemical composition risk level, and usage behavior risk level. The results are then weighted and summed to provide scientific risk warnings and prevention and control suggestions.
It enables dynamic, real-time safety risk assessment during the use of electronic atomization devices, provides scientific risk warnings and prevention and control suggestions, and improves users' safety awareness and risk management capabilities.
Smart Images

Figure CN2025135482_12112026_PF_FP_ABST
Abstract
Description
Method and system for safety risk assessment of electronic atomization device, readable storage medium
[0001]
[0002] This application claims priority to Chinese Patent Application No. 2025105996197, filed on May 09, 2025, the contents of which are incorporated herein by reference in its entirety.
[0003] The present application relates to the technical field of electronic atomization devices, and in particular to a method and system for safety risk assessment of electronic atomization devices, and a readable storage medium.
[0004] With the rapid development of electronic atomization devices, electronic atomization devices, as a product that can heat a target liquid to atomization to form an aerosol for users to smoke, have attracted widespread attention because they may contain nicotine, flavorings, and other chemical substances. However, there are still many uncertainties in the safety of electronic atomization devices. For example, the use environment of electronic atomization devices, user habits, the quality of the device, and different liquid components can all affect the safety and health risks of electronic atomization devices.
[0005] Therefore, how to effectively evaluate the potential risks in the use of electronic atomization devices has become a technical problem to be solved.
[0006] The present application provides a method and system for safety risk assessment of electronic atomization devices, and a readable storage medium, which can dynamically and in real time evaluate the safety risks in the use of electronic atomization devices.
[0007] To solve the above technical problems, one technical solution adopted by the present application is to provide a method for safety risk assessment of electronic atomization devices, which includes: collecting a plurality of data corresponding to the electronic atomization device; wherein the plurality of data includes at least two of battery data, atomization data, chemical component data, and use behavior data; performing fuzzy processing on the plurality of data to obtain a fuzzy value corresponding to each data; inputting the plurality of fuzzy values into a Bayesian network for risk calculation to obtain a safety risk assessment result corresponding to the plurality of data.
[0008] The process involves inputting several fuzzy values into a Bayesian network for risk calculation to obtain a safety risk assessment result. This includes: using a Bayesian network to infer the fuzzy values to obtain a preliminary risk assessment result, which includes at least two of the following: battery safety risk level, atomization safety risk level, chemical composition risk level, and usage behavior risk level; and then weighting and summing the various risk levels in the preliminary risk assessment result according to their respective risk level weights to obtain the final safety risk assessment result.
[0009] The process involves using a Bayesian network to reason about several fuzzy values to obtain preliminary risk assessment results. This includes using a conditional probability table in a Bayesian network to reason about several fuzzy values to obtain preliminary risk assessment results. The conditional probability table includes the probability of each fuzzy value corresponding to various risk levels.
[0010] Among them, the conditional probability table in Bayesian network is used to reason about several fuzzy values to obtain preliminary risk assessment results, including: selecting the risk level with the highest probability from the probabilities of various risk levels corresponding to each fuzzy value as the preliminary risk assessment result; when there are two identical and largest probabilities among the probabilities of various risk levels corresponding to each fuzzy value, the risk level with the higher probability is taken as the preliminary risk assessment result.
[0011] And / or, when the probability of the highest risk level corresponding to each fuzzy value exceeds the threshold, an early warning will be issued.
[0012] The safety risk assessment result is obtained by weighting and summing the various risk levels in the preliminary risk assessment result according to the risk level weights. This includes: quantifying at least two risk levels in the preliminary risk assessment result to obtain at least two risk level quantification values; weighting and summing the quantification values of each risk level according to the risk level weights to obtain a risk score; and weighting and summing the risk scores corresponding to each risk level to obtain the overall risk score and overall risk level, which serve as the safety risk assessment result.
[0013] The process involves fuzzifying several data points to obtain a fuzzy value for each data point, including: obtaining a fuzzy set of the corresponding type according to the type of each data point, wherein the fuzzy set includes different fuzzy values; and determining the corresponding fuzzy value from the fuzzy set based on the value of each data point.
[0014] The process involves inputting several fuzzy values into a Bayesian network to calculate the risk and obtain a security risk assessment result. This includes: displaying the security risk assessment result; and / or providing risk warnings corresponding to the security risk assessment result.
[0015] The battery data includes at least one of voltage fluctuation coefficient and surface temperature rise rate; the atomization data includes at least one of atomization temperature deviation, atomization particle size distribution and aerosol generation efficiency; the chemical composition data includes at least one of nicotine concentration fluctuation, aldehyde and ketone content and heavy metal migration; and the usage behavior data includes at least one of single inhalation duration, average daily usage frequency and abnormal usage pattern identification.
[0016] To solve the above-mentioned technical problems, another technical solution adopted in this application is: to provide a safety risk assessment system for an electronic atomizing device, the safety risk assessment system including a memory and a processor coupled to the memory, the memory being used to store a computer program, and the computer program being executed by the processor to implement the safety risk assessment method provided by the above technical solution.
[0017] To address the aforementioned technical problems, another technical solution adopted in this application is to provide a computer-readable storage medium for storing a computer program, which, when executed by a processor, is used to implement the security risk assessment method provided by the above technical solution.
[0018] The safety risk assessment method, system, and readable storage medium for electronic atomization devices provided in this application dynamically and in real time assess the safety risks of using electronic atomization devices by collecting several data points corresponding to the electronic atomization device, including at least two of battery data, atomization data, chemical composition data, and usage behavior data; fuzzifying the data points to obtain fuzzy values for each data point; and inputting the fuzzy values into a Bayesian network for risk calculation to obtain the safety risk assessment results corresponding to the data points. This approach can provide users with scientific risk warnings and prevention and control suggestions. [Attached Image Description]
[0019] Figure 1 is a flowchart illustrating an embodiment of the safety risk assessment method for the electronic atomizing device provided in this application;
[0020] Figure 2 is a flowchart of an embodiment of step 12 in Figure 1;
[0021] Figure 3 is a flowchart illustrating another embodiment of the safety risk assessment method for the electronic atomizing device provided in this application;
[0022] Figure 4 is a flowchart of an embodiment of step 34 in Figure 3;
[0023] Figure 5 is a structural schematic diagram of an embodiment of the safety risk assessment system for the electronic atomizing device provided in this application;
[0024] Figure 6 is a structural schematic diagram of an embodiment of the safety risk assessment device provided in this application;
[0025] Figure 7 is a schematic diagram of an embodiment of the computer-readable storage medium provided in this application.
Detailed Implementation Methods
[0026] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. It is understood that the specific embodiments described herein are only for explaining this application and not for limiting it. Furthermore, it should be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all structures. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.
[0027] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0028] With the rapid development of electronic atomization devices, these devices, which heat a target liquid to atomize it into an aerosol for users to inhale, have attracted widespread attention due to their potential presence of nicotine, flavorings, and other chemicals. However, the safety of electronic atomization devices remains highly uncertain. For example, factors such as the usage environment, user habits, device quality, and the composition of the liquid can all affect the safety and health risks associated with electronic atomization devices.
[0029] Therefore, how to effectively assess the potential risks in the use of electronic atomization devices has become an urgent technical challenge.
[0030] Based on this, this application proposes a method to dynamically and in real-time assess the safety risks of using electronic atomizing devices by collecting several data points corresponding to the device, including at least two of battery data, atomization data, chemical composition data, and usage behavior data; fuzzifying the data points to obtain a fuzzy value for each data point; and inputting these fuzzy values into a Bayesian network for risk calculation to obtain a safety risk assessment result for the data points. This method can provide users with scientific risk warnings and prevention suggestions. See any of the following embodiments for details.
[0031] Referring to Figure 1, Figure 1 is a flowchart illustrating an embodiment of the safety risk assessment method for an electronic atomizing device provided in this application. The safety risk assessment method includes:
[0032] Step 11: Collect several data points corresponding to the electronic atomization device; among which, the several data points include at least two of the following: battery data, atomization data, chemical composition data, and usage behavior data.
[0033] In some embodiments, battery data includes at least one of voltage fluctuation coefficient and surface temperature rise rate; atomization data includes at least one of atomization temperature deviation value, atomization particle size distribution and aerosol generation efficiency; chemical composition data includes at least one of nicotine concentration fluctuation, aldehyde and ketone content and heavy metal migration; and usage behavior data includes at least one of single suction duration, average daily usage frequency and abnormal usage pattern recognition.
[0034] In some embodiments, the surface temperature rise rate of the battery can be measured using IR (Infrared) thermal imaging technology. The voltage fluctuation coefficient can be calculated using the number of battery charge-discharge cycles. Atomization temperature deviation can be detected using a temperature sensor and a PID (Proportional Integral Derivative) control system. The particle size distribution of atomized particles can be monitored using a laser particle size analyzer. Aerosol generation efficiency can be measured using gas chromatography. Nicotine concentration fluctuations can be detected using HPLC-MS (High Performance Liquid Chromatography-Mass Spectrometry). Aldehyde and ketone content can be analyzed using DNPH (2,4-dinitrophenylhydrazine) derivatization-HPLC. Heavy metal migration can be detected using ICP-MS (Inductively Coupled Plasma-Mass Spectrometry). The duration of a single aspiration session can be recorded using a pressure sensor. Daily usage frequency can be analyzed synchronously via an application. Abnormal usage patterns can be identified using a machine learning behavior analysis system.
[0035] In some embodiments, a corresponding sensor can be installed on the electronic atomizing device to collect any of the above-mentioned data, or to interact with the corresponding device to obtain the corresponding data.
[0036] In some embodiments, the data may include battery data and atomization data.
[0037] In some embodiments, the data may include battery data and chemical composition data.
[0038] In some embodiments, the data may include battery data and usage behavior data.
[0039] In some embodiments, the data may include atomization data and chemical composition data.
[0040] In some embodiments, the data may include atomization data and usage behavior data.
[0041] In some embodiments, the data may include chemical composition data and usage behavior data.
[0042] In some embodiments, the data may include battery data, atomization data, and chemical composition data.
[0043] In some embodiments, the data may include battery data, atomization data, and usage behavior data.
[0044] In some embodiments, the data may include battery data, chemical composition data, and usage behavior data.
[0045] In some embodiments, the data may include atomization data, chemical composition data, and usage behavior data.
[0046] In some embodiments, the data may include battery data, atomization data, chemical composition data, and usage behavior data.
[0047] Step 12: Perform fuzzification on several data points to obtain the fuzzy value corresponding to each data point.
[0048] In some embodiments, referring to FIG2, step 12 may be the following process:
[0049] Step 121: Obtain the corresponding fuzzy set according to the type of each data; wherein, the fuzzy set includes different fuzzy values.
[0050] In some embodiments, the fuzzy sets corresponding to the battery data, atomization data, chemical composition data, and usage behavior data described above can be shown in Table 1 below:
[0051] Table 1
[0052] Step 122: Determine the corresponding fuzzy value from the corresponding fuzzy set based on the value of each data point.
[0053] Based on Table 1 above, the corresponding fuzzy value can be determined from the corresponding fuzzy set according to the value of each data point. That is, the fuzzy values in the fuzzy set all correspond to a certain numerical range, so that a one-to-one correspondence can be made.
[0054] Step 13: Input several fuzzy values into a Bayesian network to calculate the risk and obtain the security risk assessment results corresponding to several data.
[0055] In some embodiments, a Bayesian network is used to synthesize several fuzzy values to establish a Conditional Probability Table (CPT) for each risk factor (data). Each node represents an evaluation index, and directed edges represent the conditional dependencies between nodes. The CPT describes the probability distribution of the current node given the state of its parent node. For example, through historical data, experimental studies, or expert experience, the probability of battery safety risks, atomization safety risks, chemical composition safety risks, and usage behavior safety risks under different parent node combinations can be determined.
[0056] In some embodiments, the following example illustrates the fuzzy values corresponding to the input Bayesian network, which are voltage fluctuation coefficient, surface temperature rise rate, atomization temperature deviation, atomization particle size distribution, aerosol generation efficiency, nicotine concentration fluctuation, aldehyde and ketone content, heavy metal migration, single suction duration, average daily usage frequency, and abnormal usage pattern recognition:
[0057] A Bayesian network can include input nodes, intermediate nodes, and output nodes.
[0058] The input nodes include the voltage fluctuation coefficient, surface temperature rise rate, atomization temperature deviation, atomization particle size distribution, aerosol generation efficiency, nicotine concentration fluctuation, aldehyde and ketone content, heavy metal migration, single suction duration, average daily usage frequency, and nodes corresponding to abnormal usage patterns.
[0059] Intermediate nodes are used to represent preliminary risk assessment results. For example, fuzzy values of voltage fluctuation coefficient and surface temperature rise rate can be used to infer the battery safety risk level; fuzzy values corresponding to atomization temperature deviation, atomization particle size distribution, and aerosol generation efficiency can be used to infer the atomization safety risk level; fuzzy values corresponding to nicotine concentration fluctuation, aldehyde and ketone content, and heavy metal migration can be used to infer the chemical composition risk level; and fuzzy values corresponding to single suction duration, average daily usage frequency, and abnormal usage pattern recognition can be used to infer the usage behavior risk level.
[0060] The output node outputs the overall risk level (safety risk assessment result) based on the preliminary risk assessment results.
[0061] In some embodiments, the risk level may include a first risk level, a second risk level, and a third risk level. The risk level of the first risk level is less than that of the second risk level, and the risk level of the second risk level is less than that of the third risk level. In some embodiments, the first risk level may be defined as a low risk level, the second risk level as a medium risk level, and the third risk level as a high risk level.
[0062] In this embodiment, by collecting several data points corresponding to the electronic atomizing device, including at least two of battery data, atomization data, chemical composition data, and usage behavior data; fuzzifying the data points to obtain a fuzzy value for each data point; and inputting the fuzzy values into a Bayesian network to obtain the safety risk assessment results corresponding to the data points, the safety risks in the use of the electronic atomizing device can be dynamically and in real time assessed, thereby providing users with scientific risk warnings and prevention and control suggestions.
[0063] Referring to Figure 3, Figure 3 is a flowchart illustrating another embodiment of the safety risk assessment method for the electronic atomizing device provided in this application. This safety risk assessment method includes:
[0064] Step 31: Collect several data points corresponding to the electronic atomization device; among which, the several data points include at least two of the following: battery data, atomization data, chemical composition data, and usage behavior data.
[0065] Step 32: Perform fuzzification on several data points to obtain the fuzzy value corresponding to each data point.
[0066] In some embodiments, steps 31 to 32 have the same or similar technical solutions as any embodiment of this application, and will not be described in detail here.
[0067] Step 33: Use a Bayesian network to reason about several fuzzy values to obtain preliminary risk assessment results. The preliminary risk assessment results include at least two of the following: battery safety risk level, atomization safety risk level, chemical composition risk level, and usage behavior risk level.
[0068] In some embodiments, the data may include battery data and atomization data, in which case the preliminary risk assessment results include the battery safety risk level and the atomization safety risk level.
[0069] In some embodiments, the data may include battery data and chemical composition data, in which case the preliminary risk assessment results include the battery safety risk level and the chemical composition risk level.
[0070] In some embodiments, the data may include battery data and usage behavior data, in which case the preliminary risk assessment results include battery safety risk level and usage behavior risk level.
[0071] In some embodiments, the data may include atomization data and chemical composition data, and the preliminary risk assessment results include atomization safety risk level and chemical composition risk level.
[0072] In some embodiments, the data may include atomization data and usage behavior data, and the preliminary risk assessment results include atomization safety risk level and usage behavior risk level.
[0073] In some embodiments, the data may include chemical composition data and usage behavior data, in which case the preliminary risk assessment results include chemical composition risk level and usage behavior risk level.
[0074] In some embodiments, the data may include battery data, atomization data, and chemical composition data, and the preliminary risk assessment results may include battery safety risk level, atomization safety risk level, and chemical composition risk level.
[0075] In some embodiments, the data may include battery data, atomization data, and usage behavior data, and the preliminary risk assessment results may include battery safety risk level, atomization safety risk level, and usage behavior risk level.
[0076] In some embodiments, the data may include battery data, chemical composition data, and usage behavior data, and the preliminary risk assessment results may include battery safety risk level, chemical composition risk level, and usage behavior risk level.
[0077] In some embodiments, the data may include atomization data, chemical composition data, and usage behavior data, and the preliminary risk assessment results may include atomization safety risk level, chemical composition risk level, and usage behavior risk level.
[0078] In some embodiments, the data may include battery data, atomization data, chemical composition data, and usage behavior data, and the preliminary risk assessment results may include battery safety risk level, atomization safety risk level, chemical composition risk level, and usage behavior risk level.
[0079] In some embodiments, a conditional probability table in a Bayesian network can be used to infer several fuzzy values to obtain a preliminary risk assessment result; wherein, the conditional probability table includes the probability of different risk levels corresponding to each fuzzy value.
[0080] In some embodiments, the conditional probability table corresponding to the battery data is shown in Table 2 below:
[0081] Table 2
[0082] In some embodiments, the following process can be used to infer preliminary risk assessment results from several fuzzy values using conditional probability tables in a Bayesian network:
[0083] From the probabilities of various risk levels corresponding to each fuzzy value, the risk level with the highest probability is selected as the preliminary risk assessment result. Referring to Table 2: Voltage fluctuation coefficient and surface temperature rise rate have corresponding fuzzy values. Therefore, the probabilities of low, medium, and high battery safety risk levels corresponding to each fuzzy value can be found in the conditional probability table. The risk level with the highest probability is selected as the preliminary risk assessment result. For example, when the fuzzy values corresponding to voltage fluctuation coefficient and surface temperature rise rate are normal, the probability of 0.8 for the low risk level is the highest probability; therefore, the low risk level is selected as the preliminary risk assessment result corresponding to the battery data.
[0084] When each fuzzy value corresponds to two identical and maximum probabilities among the various risk levels, the risk level with the higher probability is taken as the preliminary risk assessment result. Referring to Table 2: Voltage fluctuation coefficient and surface temperature rise rate have corresponding fuzzy values. Therefore, the probabilities of low, medium, and high battery safety risk levels corresponding to these fuzzy values can be found in the conditional probability table. The risk level with the highest probability is selected as the preliminary risk assessment result. For example, if the fuzzy values corresponding to voltage fluctuation coefficient and surface temperature rise rate are relatively large, the probability corresponding to the low risk level is 0.4, and the probability corresponding to the medium risk level is also 0.4, both being the maximum probabilities. In this case, the medium risk level is taken as the preliminary risk assessment result corresponding to the battery data.
[0085] Furthermore, threshold settings can be implemented during the risk assessment process. When the probability of the highest risk level corresponding to each fuzzy value exceeds the threshold, an early warning is issued. For example, a threshold can be set for high-risk levels; if the probability of a high-risk level corresponding to a fuzzy value exceeds the threshold, an early warning should be issued even if the probability of the high-risk level is not the highest probability.
[0086] Step 34: Weight the various risk levels in the preliminary risk assessment results according to their respective risk level weights to obtain the safety risk assessment results.
[0087] In some embodiments, risk level weights can be set according to data type. For example, battery data, atomization data, chemical composition data, and usage behavior data all have corresponding risk level weights. The preliminary risk assessment results are then quantified, with different numerical values assigned to different risk levels. Finally, a weighted sum is calculated based on the weights and numerical values to obtain the final safety risk assessment result.
[0088] In some embodiments, referring to FIG4, step 34 may be the following process:
[0089] Step 341: Quantify at least two risk levels from the preliminary risk assessment results to obtain at least two quantified risk level values.
[0090] In some embodiments, a corresponding quantitative value can be set for each risk level, as follows:
[0091] Low risk: value 1; Medium risk: value 2; High risk: value 3.
[0092] Step 342: Perform a weighted summation according to the risk level weight corresponding to the quantitative value of each risk level to obtain the risk score.
[0093] In some embodiments, a risk score can be calculated based on the quantified value of each risk level and its corresponding risk level weight. The calculation formula is as follows:
[0094] Risk score = Quantitative value of risk level × Weight of risk level.
[0095] Step 343: Sum the risk scores corresponding to each risk level with weights to obtain the overall risk score and overall risk level, which will be used as the safety risk assessment result.
[0096] In some embodiments, the risk scores corresponding to each type of data are weighted and summed to obtain the total risk score of the input data. The total risk score is then mapped to a risk level to obtain the security risk assessment result. For example, the score ranges for risk levels are as follows:
[0097] Low risk: 1.0-1.6.
[0098] Medium risk: 1.7-2.4.
[0099] High risk: 2.5-3.0.
[0100] In some embodiments, after obtaining the security risk assessment results corresponding to several data points, the security risk assessment results can be displayed, and / or risk warnings corresponding to the security risk assessment results can be provided.
[0101] In this embodiment, by collecting several data points corresponding to the electronic atomizing device, including at least two of battery data, atomization data, chemical composition data, and usage behavior data; fuzzifying the data points to obtain a fuzzy value for each data point; and inputting the fuzzy values into a Bayesian network to obtain the safety risk assessment results corresponding to the data points, the safety risks in the use of the electronic atomizing device can be dynamically and in real time assessed, thereby providing users with scientific risk warnings and prevention and control suggestions.
[0102] In one application scenario, the overall implementation of this application is as follows:
[0103] 1. Data collection.
[0104] This electronic atomizing device is equipped with multiple sensors:
[0105] For battery data, IR thermal imaging technology can be used to measure the rate of temperature rise on the battery surface.
[0106] The voltage fluctuation coefficient is calculated by the number of battery charge-discharge cycles.
[0107] For atomization data, temperature sensors and PID control systems can be used to detect atomization temperature deviation. A laser particle size analyzer is used to monitor the particle size distribution of the atomized particles. Gas chromatography is used to measure aerosol generation efficiency.
[0108] For chemical composition data, HPLC-MS can be used to detect fluctuations in nicotine concentration. DNPH derivatization-HPLC can be used to analyze the content of aldehydes and ketones. ICP-MS can be used to detect the migration of heavy metals.
[0109] For usage behavior data, pressure sensors can be used to record the duration of a single suction session. Daily usage frequency can be analyzed through app data synchronization. Furthermore, machine learning-based behavioral analysis systems can be used to identify abnormal usage patterns.
[0110] 2. Blurring process.
[0111] Based on the definition of fuzzy sets and expert experience, the collected data is processed using fuzzy methods. For example, the battery surface temperature rise rate is "normal", the voltage fluctuation coefficient is "too large", the atomization temperature deviation is "dangerous", the nicotine concentration fluctuation is "harmful", the single inhalation time is "moderate", the average daily usage frequency is "medium", the atomized particle size distribution is "too small", and the aerosol generation efficiency is "low".
[0112] 3. Fuzzy reasoning
[0113] Based on the above fuzzy values, the fuzzy inference module applies predefined fuzzy rules to perform inference. For example:
[0114] Rule 1: If the voltage fluctuation coefficient is "relatively large" and the surface temperature rise rate is "normal", then the battery safety risk is "medium".
[0115] Rule 2: If the atomization temperature deviation value is "dangerous", then the atomization safety risk is "high".
[0116] Rule 3: If nicotine concentration fluctuations are "harmful", then the risk of chemical components is "high".
[0117] Rule 4: If the duration of a single suction session is "moderate" and the average daily usage frequency is "moderate", then the risk of the usage behavior is "moderate".
[0118] Based on these rules, the reasoning module concludes:
[0119] The battery safety risk level corresponding to the battery data is: Medium.
[0120] The atomization safety risk level corresponding to the atomization data is: High.
[0121] The chemical composition data corresponds to a high risk level for the chemical composition.
[0122] The risk level of the usage behavior data is: Medium.
[0123] The risk level derived from the above reasoning is a preliminary risk assessment result.
[0124] 4. Bayesian network inference.
[0125] Bayesian network modules represent the conditional dependencies between various risk factors by constructing conditional probability tables (CPTs). Assume the dependencies between nodes in a Bayesian network are as follows:
[0126] Battery safety risks depend on voltage fluctuation coefficient and surface temperature rise rate.
[0127] The safety risks of atomization depend on the atomization temperature deviation, the particle size distribution of atomized particles, and the aerosol generation efficiency.
[0128] The risks associated with chemical components depend on fluctuations in nicotine concentration, the content of aldehydes and ketones, and the migration of heavy metals.
[0129] Behavioral risks depend on the duration of a single suction session, the average daily frequency of use, and abnormal usage patterns.
[0130] Based on the CPT (Contingency Planning) method, the probability of occurrence of each sub-risk is inferred, and then the final overall risk level is derived. For example, the calculation result is:
[0131] The battery safety risk is "medium" (0.6 probability).
[0132] The safety risk of atomization is "high" (0.8 probability).
[0133] The risk associated with the chemical components is "high" (probability 0.75).
[0134] The risk of using this behavior is "medium" (probability of 0.5).
[0135] 5. Comprehensive Assessment
[0136] Based on the risk level and weight of each module (battery safety: 0.4, atomization safety: 0.2, chemical composition: 0.3, usage behavior: 0.1), the system calculates the risk score for each module and sums them up with weights.
[0137] Battery safety risk score = 2 (medium) × 0.4 = 0.8.
[0138] Atomization safety risk score = 3 (high) × 0.2 = 0.6.
[0139] Chemical composition risk score = 3 (high) × 0.3 = 0.9.
[0140] Use the behavioral risk score = 2 (medium) × 0.1 = 0.2.
[0141] Overall risk score = 0.8 + 0.6 + 0.9 + 0.2 = 2.5.
[0142] Based on the preset risk level range:
[0143] Low risk: 1.0-1.6.
[0144] Medium risk: 1.7-2.4.
[0145] High risk: 2.5-3.0.
[0146] The overall risk level is "high".
[0147] 6. Issue early warnings.
[0148] Finally, the output module displays the evaluation results to the user in chart form, along with corresponding risk warnings. For example, it displays: "Battery safety risk: Medium," "Atomization safety risk: High," "Chemical composition risk: High," and "Usage behavior risk: Medium." It also provides suggestions: "Please pay attention to the battery status and avoid overuse; check the atomization temperature and particle distribution, and maintain an appropriate usage frequency."
[0149] Referring to Figure 5, Figure 5 is a schematic structural diagram of an embodiment of the safety risk assessment system for an electronic atomizing device provided in this application. The safety risk assessment system 50 includes a memory 51 and a processor 52 coupled to the memory 51. The memory 51 stores a computer program, which, when executed by the processor 52, is used to implement the following methods:
[0150] Collect several data points corresponding to the electronic atomization device; wherein, the several data points include at least two of the following: battery data, atomization data, chemical composition data, and usage behavior data; perform fuzzification processing on the several data points to obtain a fuzzy value corresponding to each data point; input the several fuzzy values into a Bayesian network to perform risk calculation and obtain the safety risk assessment results corresponding to the several data points.
[0151] In some embodiments, when the computer program is executed by the processor 52, it is also used to implement the following method: using a Bayesian network to infer several fuzzy values to obtain a preliminary risk assessment result, wherein the preliminary risk assessment result includes at least two of the following: battery safety risk level, atomization safety risk level, chemical composition risk level, and usage behavior risk level; and weighting and summing the various risk levels in the preliminary risk assessment result according to the risk level weights to obtain a safety risk assessment result.
[0152] In some embodiments, when the computer program is executed by the processor 52, it is also used to implement the following method: using a conditional probability table in a Bayesian network to reason about several fuzzy values to obtain a preliminary risk assessment result, wherein the conditional probability table includes the probability of each type of risk level corresponding to each fuzzy value.
[0153] In some embodiments, when the computer program is executed by the processor 52, it is further configured to implement the following methods: selecting the risk level with the highest probability from the probabilities of various risk levels corresponding to each fuzzy value as the preliminary risk assessment result; when there are two identical and highest probabilities among the probabilities of various risk levels corresponding to each fuzzy value, taking the risk level with the higher probability as the preliminary risk assessment result; and / or, when the probability of the highest risk level corresponding to each fuzzy value exceeds a threshold, issuing an early warning.
[0154] In some embodiments, when the computer program is executed by the processor 52, it is further configured to implement the following method: quantify at least two risk levels in the preliminary risk assessment results to obtain at least two risk level quantification values; perform weighted summation according to the risk level weight corresponding to each risk level quantification value to obtain a risk score; and perform weighted summation of the risk scores corresponding to each risk level to obtain an overall risk score and an overall risk level, which are used as the safety risk assessment results.
[0155] In some embodiments, when the computer program is executed by the processor 52, it is further configured to implement the following method: according to the type of each data, obtain a fuzzy set of the corresponding type, wherein the fuzzy set includes different fuzzy values; and determine the corresponding fuzzy value from the fuzzy set according to the value of each data.
[0156] In some embodiments, after inputting several fuzzy values into a Bayesian network to obtain security risk assessment results corresponding to several data, the computer program, when executed by the processor 52, is further used to implement the following methods: displaying the security risk assessment results; and / or providing risk warnings corresponding to the security risk assessment results.
[0157] In some embodiments, battery data includes at least one of voltage fluctuation coefficient and surface temperature rise rate; atomization data includes at least one of atomization temperature deviation value, atomization particle size distribution and aerosol generation efficiency; chemical composition data includes at least one of nicotine concentration fluctuation, aldehyde and ketone content and heavy metal migration; and usage behavior data includes at least one of single suction duration, average daily usage frequency and abnormal usage pattern recognition.
[0158] It is understood that when the computer program is executed by the processor 52, it is also used to implement the methods of any of the above embodiments.
[0159] Referring to Figure 6, which is a schematic diagram of an embodiment of the security risk assessment device provided in this application, the security risk assessment device 60 includes: a data acquisition module 61, a fuzzy processing module 62, and a risk calculation module 63.
[0160] The acquisition module 61 is used to acquire several data from the electronic atomization device; wherein the several data include at least two of the following: battery data, atomization data, chemical composition data, and usage behavior data.
[0161] The fuzzing module 62 is used to fuzzify the data to obtain a fuzzy value for each data.
[0162] The risk calculation module 63 is used to input several of the fuzzy values into a Bayesian network to perform risk calculation and obtain a security risk assessment result.
[0163] In some embodiments, the risk calculation module 63 is further configured to use a Bayesian network to infer several fuzzy values to obtain a preliminary risk assessment result, wherein the preliminary risk assessment result includes at least two of the following: battery safety risk level, atomization safety risk level, chemical composition risk level, and usage behavior risk level; and to obtain a safety risk assessment result by weighted summation of the various risk levels in the preliminary risk assessment result according to the risk level weight.
[0164] In some embodiments, the risk calculation module 63 is further configured to use a conditional probability table in a Bayesian network to reason about several fuzzy values to obtain a preliminary risk assessment result, wherein the conditional probability table includes the probability of each type of risk level corresponding to each fuzzy value.
[0165] In some embodiments, the risk calculation module 63 is further configured to select the risk level with the highest probability from the probabilities of various risk levels corresponding to each fuzzy value as the preliminary risk assessment result; when there are two identical and highest probabilities among the probabilities of various risk levels corresponding to each fuzzy value, the risk level with the higher probability among the two probabilities is used as the preliminary risk assessment result; and / or, when the probability of the highest risk level corresponding to each fuzzy value exceeds a threshold, an early warning is issued.
[0166] In some embodiments, the risk calculation module 63 is further used to quantify at least two risk levels in the preliminary risk assessment results to obtain at least two risk level quantification values; to perform weighted summation according to the risk level weight corresponding to each risk level quantification value to obtain a risk score; and to perform weighted summation of the risk scores corresponding to each risk level to obtain the overall risk score and the overall risk level, which are used as the safety risk assessment results.
[0167] In some embodiments, the fuzzy processing module 62 is further configured to obtain a fuzzy set of a corresponding type according to the type of each data, wherein the fuzzy set includes different fuzzy values; and determine the corresponding fuzzy value from the fuzzy set according to the value of each data.
[0168] In some embodiments, the security risk assessment device 60 may further include a display module. After inputting several fuzzy values into a Bayesian network to perform risk calculation and obtain the security risk assessment result, the display module is used to display the security risk assessment result; and / or to provide risk warnings corresponding to the security risk assessment result.
[0169] In some embodiments, the battery data mentioned above includes at least one of voltage fluctuation coefficient and surface temperature rise rate; the atomization data includes at least one of atomization temperature deviation value, atomization particle size distribution and aerosol generation efficiency; the chemical composition data includes at least one of nicotine concentration fluctuation, aldehyde and ketone content and heavy metal migration; and the usage behavior data includes at least one of single suction duration, average daily usage frequency and abnormal usage pattern recognition.
[0170] It is understood that the acquisition module 61, the fuzzy processing module 62, and the risk calculation module 63 work together to implement the method of any of the above embodiments.
[0171] Referring to Figure 7, which is a schematic diagram of an embodiment of a computer-readable storage medium provided in this application, the computer-readable storage medium 70 is used to store a computer program 71. When executed by a processor, the computer program 71 is used to implement the following method:
[0172] Collect several data points corresponding to the electronic atomization device; wherein, the several data points include at least two of the following: battery data, atomization data, chemical composition data, and usage behavior data; perform fuzzification processing on the several data points to obtain a fuzzy value corresponding to each data point; input the several fuzzy values into a Bayesian network to perform risk calculation and obtain the safety risk assessment results corresponding to the several data points.
[0173] In some embodiments, when the computer program 71 is executed by the processor, it is further configured to implement the following method: using a Bayesian network to infer several fuzzy values to obtain a preliminary risk assessment result, wherein the preliminary risk assessment result includes at least two of the following: battery safety risk level, atomization safety risk level, chemical composition risk level, and usage behavior risk level; and weighting and summing the various risk levels in the preliminary risk assessment result according to the risk level weights to obtain a safety risk assessment result.
[0174] In some embodiments, when the computer program 71 is executed by the processor, it is also used to implement the following method: using a conditional probability table in a Bayesian network to reason about several fuzzy values to obtain a preliminary risk assessment result, wherein the conditional probability table includes the probability of each type of risk level corresponding to each fuzzy value.
[0175] In some embodiments, when the computer program 71 is executed by the processor, it is further configured to implement the following methods: selecting the risk level with the highest probability from the probabilities of various risk levels corresponding to each fuzzy value as the preliminary risk assessment result; when there are two identical and highest probabilities among the probabilities of various risk levels corresponding to each fuzzy value, taking the risk level with the higher probability as the preliminary risk assessment result; and / or, when the probability of the highest risk level corresponding to each fuzzy value exceeds a threshold, issuing an early warning.
[0176] In some embodiments, when the computer program 71 is executed by the processor, it is further configured to implement the following method: quantify at least two risk levels in the preliminary risk assessment results to obtain at least two risk level quantification values; perform weighted summation according to the risk level weight corresponding to each risk level quantification value to obtain a risk score; and perform weighted summation of the risk scores corresponding to each risk level to obtain an overall risk score and an overall risk level, which serve as the safety risk assessment result.
[0177] In some embodiments, when the computer program 71 is executed by the processor, it is further configured to implement the following method: according to the type of each data, obtain a fuzzy set of the corresponding type, wherein the fuzzy set includes different fuzzy values; and determine the corresponding fuzzy value from the fuzzy set according to the value of each data.
[0178] In some embodiments, after inputting several fuzzy values into a Bayesian network to perform risk calculation and obtain a security risk assessment result, the computer program 71, when executed by a processor, is also used to implement the following methods: displaying the security risk assessment result; and / or providing risk warnings corresponding to the security risk assessment result.
[0179] In some embodiments, battery data includes at least one of voltage fluctuation coefficient and surface temperature rise rate; atomization data includes at least one of atomization temperature deviation value, atomization particle size distribution and aerosol generation efficiency; chemical composition data includes at least one of nicotine concentration fluctuation, aldehyde and ketone content and heavy metal migration; and usage behavior data includes at least one of single suction duration, average daily usage frequency and abnormal usage pattern recognition.
[0180] It is understood that when computer program 71 is executed by a processor, it is also used to implement the methods of any of the above embodiments.
[0181] In summary, the safety risk assessment method, system, and readable storage medium for electronic atomization devices provided in this application dynamically and in real time assess the safety risks of using electronic atomization devices by collecting several data points corresponding to the electronic atomization device, including at least two of battery data, atomization data, chemical composition data, and usage behavior data; fuzzifying the data points to obtain fuzzy values for each data point; and inputting the fuzzy values into a Bayesian network to obtain the safety risk assessment results corresponding to the data points. This approach can provide users with scientific risk warnings and prevention and control suggestions.
[0182] In the several embodiments provided in this application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of circuits or units is merely a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed.
[0183] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.
[0184] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0185] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural or procedural changes made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.
Claims
1. A method for assessing the safety risks of an electronic atomizing device, characterized in that, The security risk assessment method includes: Collect several data points from the electronic atomizing device; wherein, the several data points include at least two of the following: battery data, atomization data, chemical composition data, and usage behavior data; The data are fuzzified to obtain a fuzzy value for each data point. Several fuzzy values are input into a Bayesian network to calculate the risk and obtain the security risk assessment result.
2. The safety risk assessment method according to claim 1, characterized in that, The step of inputting several fuzzy values into a Bayesian network for risk calculation to obtain a security risk assessment result includes: The Bayesian network is used to infer several fuzzy values to obtain a preliminary risk assessment result, wherein the preliminary risk assessment result includes at least two of the following: battery safety risk level, atomization safety risk level, chemical composition risk level, and usage behavior risk level. The safety risk assessment result is obtained by weighting and summing the various risk levels in the preliminary risk assessment result according to their respective risk level weights.
3. The safety risk assessment method according to claim 2, characterized in that, The step of using the Bayesian network to infer from several fuzzy values to obtain preliminary risk assessment results includes: The preliminary risk assessment result is obtained by inferring about several fuzzy values using the conditional probability table in the Bayesian network, wherein the conditional probability table includes the probability of each type of risk level corresponding to each fuzzy value.
4. The safety risk assessment method according to claim 3, characterized in that, The step of using the conditional probability table in the Bayesian network to infer several fuzzy values to obtain the preliminary risk assessment result includes: The risk level with the highest probability among the various risk levels corresponding to each fuzzy value is selected as the preliminary risk assessment result. When there are two identical and largest probabilities among the probabilities of various risk levels corresponding to each fuzzy value, the risk level with the higher probability is taken as the preliminary risk assessment result. And / or, When the probability of the highest risk level corresponding to each of the aforementioned fuzzy values exceeds the threshold, an early warning will be issued.
5. The safety risk assessment method according to claim 2, characterized in that, The safety risk assessment result is obtained by weighting and summing the various risk levels in the preliminary risk assessment result according to their respective risk level weights, including: Quantify at least two risk levels in the preliminary risk assessment results to obtain at least two quantitative values for the risk levels; The risk score is obtained by weighting and summing the values corresponding to the risk level quantification values of each risk level. The risk scores corresponding to each risk level are weighted and summed to obtain the overall risk score and overall risk level, which are used as the safety risk assessment results.
6. The safety risk assessment method according to any one of claims 1 to 5, characterized in that, The process of fuzzifying the data to obtain a fuzzy value for each data point includes: According to the type of each data item, obtain the corresponding fuzzy set, wherein the fuzzy set includes different fuzzy values; Based on the numerical value of each of the data, the corresponding fuzzy value is determined from the fuzzy set.
7. The safety risk assessment method according to any one of claims 1 to 5, characterized in that, After inputting several fuzzy values into a Bayesian network for risk calculation and obtaining a security risk assessment result, the process includes: Present the results of the security risk assessment; And / or, provide risk warnings corresponding to the security risk assessment results.
8. The safety risk assessment method according to any one of claims 1 to 5, characterized in that, The battery data includes at least one of voltage fluctuation coefficient and surface temperature rise rate; the atomization data includes at least one of atomization temperature deviation value, atomization particle size distribution and aerosol generation efficiency; the chemical composition data includes at least one of nicotine concentration fluctuation, aldehyde and ketone content and heavy metal migration; the usage behavior data includes at least one of single inhalation duration, average daily usage frequency and abnormal usage pattern identification.
9. A safety risk assessment device, characterized in that, include: A data acquisition module is used to acquire several data points from the electronic atomization device; wherein, the several data points include at least two of the following: battery data, atomization data, chemical composition data, and usage behavior data; The fuzzing module is used to fuzzify the data to obtain a fuzzy value for each data point. The risk calculation module is used to input several fuzzy values into a Bayesian network to calculate the risk and obtain the security risk assessment result.
10. A safety risk assessment system for an electronic atomizing device, characterized in that, The security risk assessment system includes a memory and a processor coupled to the memory. The memory is used to store a computer program, which, when executed by the processor, is used to implement the security risk assessment method as described in any one of claims 1-8.
