Industrial boiler explosion assessment method and device

Abstract

The invention relates to the field of boiler safety protection, and particularly discloses an industrial boiler explosion assessment method and device. According to the method, the fuzzy number and the accident tree are introduced into analysis of the industrial boiler explosion accident, subjective judgment factors of experts are added for quantitative analysis, basic events capable of causing the industrial boiler explosion accident are listed, and a fault probability is added to each basic event; the influence of each basic event on the occurrence risk of boiler explosion is quantitatively reflected, the risk causing boiler explosion can be traced and tracked by an operator conveniently, danger elimination can be carried out in time, various sensors are arranged on the industrial boiler, data are obtained through the sensors, and operation of the boiler operator is facilitated. The fault probability of the corresponding basic event is adjusted in real time, and the accuracy of the occurrence probability of the event on the top of the accident tree is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of boiler safety protection, and in particular relates to an industrial boiler explosion assessment method and device. Background Art

[0002] Industrial boilers operate for a long time under adverse working conditions of high temperature and high pressure. Once the operation is improper or there are structural defects in the boiler equipment, it will cause overpressure or overheating of the boiler and cause blasting or explosion accidents. In addition, industrial boilers have many internal structures, large volumes, and are equipped with multiple complex systems such as steam, water, air, and smoke. The causes of boiler explosions are complex and varied.

[0003] In the Chinese patent with publication number CN118485023A, a hazard source analysis and evaluation system and method based on data analysis are mentioned, which involves the field of safety management technology, including a data collection module, a simulation explosion module, an analysis and evaluation module and a first available numerical database. The present invention collects basic data and historical data of industrial boiler accidents; uses CFD to establish a full process model of industrial boiler explosion, including an accumulation model and a release model; simulates the explosion process of industrial boilers; analyzes and compares the differences between actual operation data and simulation results; constructs fault trees and event trees according to the analysis results; analyzes the hazard sources in the operation process of industrial boilers in combination with fault trees and event trees, and rates the hazard sources. The present invention realizes quantitative evaluation of the hazard of industrial boilers, provides technical support for early warning and emergency rescue of industrial boiler explosion accidents; inverts and reproduces the entire process of industrial boiler explosion accidents, determines the accurate cause of the accident, and thus better traces the accident.

[0004] During the application process, the applicant discovered that the above application had defects during use. The application extracted features related to sudden internal pressure and release from historical industrial boiler explosion data, analyzed these features to obtain usable data, established an industrial boiler model based on the available data, and simulated the boiler explosion risks and causes through the model. However, the output result of the model is only a third-level risk warning, and the risk source is not traced, nor is the risk probability of the risk source assessed, which makes it impossible for operators to understand the source of boiler explosion risk, and thus makes it impossible for operators to eliminate the risk source in time to avoid the boiler explosion risk. Summary of the invention

[0005] The object of the present invention is to provide an industrial boiler explosion assessment method and device to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] An industrial boiler explosion assessment method specifically comprises the following steps:

[0008] S1. Collect basic data and historical data of industrial boiler accidents, including structural design parameters, structural changes, operating pressure and temperature, operating personnel operation, gas transportation and boiler operation and maintenance records before the industrial boiler explosion, and establish an accident tree for industrial boiler explosion accidents based on the collected data;

[0009] S2. Quantitatively analyze the industrial boiler explosion accident tree, evaluate the probability of fuzzy events in the accident tree, and obtain the actual failure probability, critical importance and probability importance through aggregate calculation, and determine the probability of occurrence of the top event based on the aggregate calculation results;

[0010] S3. Collect the actual operation data and historical maintenance data of industrial boilers, correct the probability of occurrence of overhead events, and make early warnings and emergency responses based on the corrected probability of occurrence of overhead events.

[0011] Preferably, the top event of the accident tree of the industrial boiler explosion accident in S1 is the boiler explosion T1, M1~M6 are respectively configured as boiler overpressure, boiler water shortage, boiler full of water, boiler steam-water azeotropy, boiler pipe burst and gas leakage, the intermediate events J1~J6 are respectively configured as air pressure rise, safety valve failure, overpressure alarm failure, water level drop, abnormal water supply and water level rise, X1~X39 are configured as basic events, and the three minimum path sets of the accident tree are determined, which are P1={X1, X2, X9,…, X39}, P2={X3, X5, X9,…, X39} and P3={X6, X7, X8,…, X39}.

[0012] Preferably, when the industrial boiler explosion accident tree is quantitatively analyzed in S2, three experts are selected as expert one, expert two and expert three, respectively, and the weights are configured as 0.37, 0.33 and 0.30 respectively, and the fuzzy linguistic variables are configured as A, B, C, D, and E from high to low, and the fuzzy linguistic variables evaluated by the experts are expressed by trapezoidal fuzzy numbers: A (0.7, 0.8, 0.8, 0.9), B (0.5, 0.6, 0.7, 0.8), C (0.4, 0.5, 0.5, 0.6), D (0.2, 0.3, 0.4, 0.5), and E (0.1, 0.2, 0.2, 0.3).

[0013] Preferably, in S2, the expert scores are converted into trapezoidal fuzzy numbers, and the specific process includes:

[0014] S21. The two existing trapezoidal fuzzy numbers are represented by Ai = (a1, a2, a3, a4) and Aj = (b1, b2, b3, b4). The similarity between Ai and Aj is calculated for each expert, and the similarity function is:

[0015]

[0016] like S(A i ,A j ) is larger, the greater the similarity of the trapezoidal fuzzy numbers Ai and Aj is;

[0017] S22. Calculate the average similarity of each expert. The formula is:

[0018]

[0019] S23. Calculate the relative similarity of each expert:

[0020]

[0021] S24. Estimation of the expert's similarity coefficient:

[0022] C(E i )=β·W(E i )+(1-β)·RA i

[0023] Where: β——relaxation factor, 0≤β≤1, take β=0.5,

[0024] W(E i )——The weight value of each expert;

[0025] S25. Calculate the aggregate fuzzy number of basic events:

[0026] R AG =C(E1)·R1+C(E2)·R1…C(E i )·R i

[0027] Where: R i ——Expert scoring value;

[0028] S26. Perform defuzzification processing to obtain an exact value, that is, the probability of brittle failure, which is expressed as:

[0029]

[0030] S27. The probability of occurrence of each basic event is represented by a fuzzy number, and it is concluded that the probability of brittle failure is a fuzzy probability. Converting it into an actual failure probability value, we can get:

[0031]

[0032] After calculation, the actual failure probability, critical importance and probability importance of each basic event are obtained.

[0033] Preferably, in S2, the probability calculation formula of the top event is expressed as:

[0034]

[0035] Among them, q n is the failure probability of the basic event.

[0036] Preferably, in S3, the actual operation data and historical maintenance data of the industrial boiler are collected, including boiler structure data, real-time image of the boiler structure, boiler operating pressure, operating temperature, boiler water level, boiler inlet and outlet flow, water quality detection and maintenance personnel operation log, and information corresponding to the basic event is obtained. It is judged whether the basic event is resolved based on the acquired information. If it is resolved, the failure probability of the corresponding basic event is adjusted to zero. If it cannot be resolved, it is judged whether the basic event is a periodic event. If it is not a periodic event, the failure probability of the basic event is maintained unchanged. If it is a periodic event, a periodic change parameter is introduced:

[0037]

[0038] Among them, t 时长 is the elapsed time of a cycle, T 周期 The duration of a cycle.

[0039] Preferably, in S3, the user can set a warning threshold. When the probability of the top event occurring is greater than the warning threshold, an early warning is initiated. The early warning modes include sound warning and light flashing warning, and a solution corresponding to the basic event causing the early warning is output.

[0040] Preferably, in said S3, when collecting the surface structure data of the industrial boiler, a visual inspection algorithm is selected to perform automatic inspection of oxidation, wear and cracks of the industrial boiler structure, and the specific steps include:

[0041] S31. Collect images of industrial boilers using a camera or scanner;

[0042] S32, performing denoising, grayscale and contrast enhancement operations on the collected image to improve the accuracy of subsequent processing;

[0043] S33, using image processing algorithms to extract features in the image related to oxidation, wear and cracks in the boiler structure;

[0044] S34. Use deep learning algorithms to detect and classify the extracted image features to identify whether there are oxidation, wear and crack features in the boiler structure;

[0045] S35. Mark the characteristic locations of oxidation, wear and cracks of the boiler structure on the image.

[0046] An industrial boiler explosion assessment device, using an industrial boiler explosion assessment method, comprising:

[0047] A control center module, which is composed of a data collection unit, a data analysis and processing unit, an early warning unit, and a response processing unit;

[0048] A data acquisition module, the data acquisition module includes a visual sensor, a pressure sensor, a temperature sensor, a water level sensor, a flow sensor, a sound sensor and a water quality sensor installed on the boiler;

[0049] The data acquisition module collects information and transmits it to the data collection unit. The data analysis and processing unit analyzes and processes the internal information of the data acquisition unit. The early warning unit is used to issue early warnings through light alarms and sound alarms. The response processing unit is used to store and output industrial boiler explosion risk solutions.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] The present invention introduces fuzzy numbers and accident trees into the analysis of industrial boiler explosion accidents, and at the same time adds subjective judgment factors of experts for quantitative analysis, lists basic events that can cause industrial boiler explosion accidents, and adds failure probability to each basic event, quantitatively reflects the impact of each basic event on the risk of boiler explosion, and is convenient for operators to trace the risk of boiler explosion so as to eliminate the risk in time. In addition, by arranging multiple sensors on industrial boilers, obtaining data through sensors and the operation of boiler operators, the failure probability of corresponding basic events is adjusted in real time, thereby improving the accuracy of the probability of occurrence of events at the top of the accident tree. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is a block diagram of the evaluation method of the present invention;

[0053] Figure 2 is a structural block diagram of the evaluation device of the present invention;

[0054] Figure 3 It is the accident tree structure diagram of the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] Example:

[0057] See also Figure 1 - Figure 3As shown, a method for evaluating industrial boiler explosion specifically comprises the following steps:

[0058] S1. Collect basic data and historical data of industrial boiler accidents, including structural design parameters, structural changes, operating pressure and temperature, operating personnel operation, gas transportation and boiler operation and maintenance records before the industrial boiler explosion, and establish an accident tree for industrial boiler explosion accidents based on the collected data;

[0059] S2. Quantitatively analyze the industrial boiler explosion accident tree, evaluate the probability of fuzzy events in the accident tree, and obtain the actual failure probability, critical importance and probability importance through aggregate calculation, and determine the probability of occurrence of the top event based on the aggregate calculation results;

[0060] S3. Collect the actual operation data and historical maintenance data of industrial boilers, correct the probability of occurrence of overhead events, and make early warnings and emergency responses based on the corrected probability of occurrence of overhead events.

[0061] By introducing accident tree and fuzzy number into boiler explosion accident for analysis, a failure probability is given to each basic event that can cause boiler explosion. Finally, the risk probability of boiler explosion is comprehensively judged by the failure probability of each event. At the same time, operators can intuitively and clearly understand the basic events and their probabilities that cause boiler explosion risk through accident tree and failure probability, which is convenient for users to carry out targeted processing of basic failure events so as to eliminate risks in time.

[0062] The top event of the accident tree of the industrial boiler explosion accident in S1 is the boiler explosion T1, M1~M6 are respectively configured as boiler overpressure, boiler water shortage, boiler full of water, boiler steam-water azeotropy, boiler pipe burst and gas leakage, the intermediate events J1~J6 are respectively configured as pressure rise, safety valve failure, overpressure alarm failure, water level drop, abnormal water supply and water level rise, X1~X39 are configured as basic events, and the three minimum path sets of the accident tree are determined, which are P1={X1, X2, X9,…, X39}, P2={X3, X5, X9,…, X39}.} and P3 = {X6, X7, X8, …, X39}, X1 and X2 are connected under J1, X3-X5 are connected under J2, X6-X8 are connected under J3, X9-X11 are connected under J4, X12-X14 are connected under J5, X15-X16 are connected under J6, X17-X18 are connected under M3, X19-X23 are connected under M4, X24-X32 are connected under M5, X33-X39 are connected under M6, among which M1 is connected with J1, J2 and J3 through AND gate, and the rest of the connections are all OR gate connections.

[0063] When the industrial boiler explosion accident tree is quantitatively analyzed in S2, three experts are selected as expert one, expert two and expert three, and the weights are configured as 0.37, 0.33 and 0.30 respectively. The fuzzy linguistic variables are configured as A, B, C, D, and E from high to low, and the fuzzy linguistic variables evaluated by the experts are expressed by trapezoidal fuzzy numbers: A (0.7, 0.8, 0.8, 0.9), B (0.5, 0.6, 0.7, 0.8), C (0.4, 0.5, 0.5, 0.6), D (0.2, 0.3, 0.4, 0.5), and E (0.1, 0.2, 0.2, 0.3).

[0064] In S2, the expert scores are converted into trapezoidal fuzzy numbers. The specific process includes:

[0065] S21. The two existing trapezoidal fuzzy numbers are represented by Ai = (a1, a2, a3, a4) and Aj = (b1, b2, b3, b4). The similarity between Ai and Aj is calculated for each expert, and the similarity function is:

[0066]

[0067] like S(A i ,A j ) is larger, the greater the similarity of the trapezoidal fuzzy numbers Ai and Aj is;

[0068] S22. Calculate the average similarity of each expert. The formula is:

[0069]

[0070] S23. Calculate the relative similarity of each expert:

[0071]

[0072] S24. Estimation of the expert's similarity coefficient:

[0073] C(E i )=β·W(E i )+(1-β)·RA i

[0074] Where: β——relaxation factor, 0≤β≤1, take β=0.5,

[0075] W(E i )——The weight value of each expert;

[0076] S25. Calculate the aggregate fuzzy number of basic events:

[0077] R AG=C(E1)·R1+C(E2)·R1…C(E i )·R i

[0078] Where: R i ——Expert scoring value;

[0079] S26. Perform defuzzification processing to obtain an exact value, that is, the probability of brittle failure, which is expressed as:

[0080]

[0081] S27. The probability of occurrence of each basic event is represented by a fuzzy number, and it is concluded that the probability of brittle failure is a fuzzy probability. Converting it into an actual failure probability value, we can get:

[0082]

[0083] After calculation, the actual failure probability, critical importance and probability importance of each basic event are obtained.

[0084] In S2, the probability calculation formula of the top event is expressed as:

[0085]

[0086] Among them, q n is the failure probability of the basic event.

[0087] After calculation, the expert score, failure probability, critical importance and probability importance of each basic event are obtained, refer to the following table:

[0088]

[0089]

[0090]

[0091] In the process of calculating the probability of boiler explosion, the basic failure rate of each basic event can use the failure probability in the above table.

[0092] In S3, the actual operation data and historical maintenance data of the industrial boiler are collected, including boiler structure data, real-time image of boiler structure, boiler operation pressure, operation temperature, boiler water level, boiler inlet and outlet flow, water quality detection and maintenance personnel operation log, and information corresponding to the basic event is obtained. According to the obtained information, it is judged whether the basic event is solved. If it is solved, the failure probability of the corresponding basic event is adjusted to zero. If it cannot be solved, it is judged whether the basic event is a periodic event. If it is not a periodic event, the failure probability of the basic event is maintained unchanged. If it is a periodic event, a periodic change parameter is introduced:

[0093]

[0094] Among them, t 时长 is the elapsed time of a cycle, T 周期 The duration of a cycle.

[0095] From the above, we can know whether X1~X4 occurs through the pressure gauge and pressure gauge inspection. The existing operation data and inspection log can be used to determine whether X1~X4 are resolved, and then the probability of X1~X4 is adjusted to zero. X5 is a periodic factor, so it can be judged as a periodic event. The safety valve inspection cycle is taken as T 周期 , the time from the last check point is taken as t 时长 , so a periodic variation parameter μ can be obtained. At this time, the fault probability of X5 is multiplied by μ as the corrected fault probability of X5. Based on this, for the basic events X6-X39, X1, X2, X3, X4, X6, X9, X10, X14, X18, X19, X20, X21, X22, X23, X24, X26, X29, X31, X32, X33, X34, X35, X36, and X37 can be solved by operating data and operator inspections, thereby The failure probability is adjusted to zero, and the original failure probabilities of X7, X8, X11, X12, X13, X15, X16, X17, X18, X19, X25, X27, X28, X30, X38, and X39 remain unchanged. X5, X23, X35, and X31 are taken as periodic events. The periodic change parameters are introduced into the failure probabilities of these periodic events to correct the failure probabilities of these periodic events, and then the occurrence probabilities of the top events are adjusted to make the occurrence probabilities of the top events more accurate.

[0096] In S3, the user can set a warning threshold. When the probability of the top event is greater than the warning threshold, the warning is activated. The warning methods include sound warning and light flashing warning, and the solution corresponding to the basic event that caused the warning is output. For example, the user can set the warning threshold to 50%. After the equipment adjusts the failure probability of the basic event according to the boiler operating parameters and the operator's inspection log, the adjusted boiler explosion probability is calculated according to the top event probability calculation formula.

[0097] Some basic events cannot be eliminated in time by relying solely on regular inspections by operators, because operators cannot effectively and timely obtain changes in the boiler structure with their naked eyes, such as cracks on the boiler surface, local oxidation or damage. These boiler structural characteristics involve basic events such as X22 and X32. Therefore, visual algorithms can be used instead of human eyes to obtain some key structural data.

[0098] In S3, when collecting the surface structure data of the industrial boiler, a visual inspection algorithm is selected to perform automatic inspection of oxidation, wear and cracks of the industrial boiler structure. The specific steps include:

[0099] S31. Collect images of industrial boilers using a camera or scanner;

[0100] S32, performing denoising, grayscale and contrast enhancement operations on the collected image to improve the accuracy of subsequent processing;

[0101] S33, using image processing algorithms to extract features in the image related to oxidation, wear and cracks in the boiler structure;

[0102] S34. Use deep learning algorithms to detect and classify the extracted image features to identify whether there are oxidation, wear and crack features in the boiler structure;

[0103] S35. Mark the characteristic locations of oxidation, wear and cracks of the boiler structure on the image.

[0104] Through the above algorithm, some oxidation, wear and crack characteristics on the boiler surface can be accurately detected. These characteristics can be used to assist in judging whether the failure probability of some basic events can be reset to zero. In addition, visual algorithms can also be used to assist in judging some items that require manual inspection by operators. For example, when the main gas valve is opened during X20 boiler connection, the main steam valve closing position can be determined with the help of visual algorithms. When it is detected that the main steam valve is not in the closed position, it can be judged that the main steam valve is open. At this time, the X20 event has a basic failure probability, avoiding the failure probability of the X20 event being reset to zero and affecting the boiler explosion probability.

[0105] An industrial boiler explosion assessment device, using an industrial boiler explosion assessment method, comprising:

[0106] A control center module, which is composed of a data collection unit, a data analysis and processing unit, an early warning unit, and a response processing unit;

[0107] A data acquisition module, which includes a visual sensor, a pressure sensor, a temperature sensor, a water level sensor, a flow sensor, a sound sensor and a water quality sensor installed on the boiler, and is used to collect visual image information of the boiler structure, pressure value, temperature, water level, water intake, steam discharge, abnormal noise of boiler operation and boiler water quality, etc.;

[0108] The data acquisition module collects information and transmits it to the data collection unit. The data analysis and processing unit analyzes and processes the internal information of the data acquisition unit, and uses these collected data to judge the operating status of the boiler, and then makes a judgment on the basic events and adjusts the probability of occurrence of the basic events. The early warning unit is used to issue early warnings through light alarms and sound alarms. The response processing unit is used to store and output industrial boiler explosion risk solutions. Users can input basic event solutions in the response processing unit in advance. When the basic event becomes one of the risks causing the boiler explosion event, the response processing unit outputs the solution corresponding to the basic event to facilitate operators to quickly eliminate risks.

[0109] When the method is used to predict the explosion of an industrial boiler, the various sensors on the boiler system are connected to the data acquisition module, and the data acquisition module transmits the summary information to the data analysis and processing unit. At the same time, the operator also transmits the inspection log content to the data analysis and processing unit. The data analysis and processing unit adjusts the basic event failure rate of the accident tree according to the collected data and the inspection log content, and then calculates the probability of the top event based on the adjusted basic event failure rate, thereby evaluating the risk of boiler explosion. When the risk reaches the alarm threshold, the early warning unit issues a warning, and the response processing unit outputs the solution of the corresponding basic event to the outside for the operator to execute to eliminate the risk.

[0110] In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other.

Claims

1. An industrial boiler explosion assessment method, characterized in that: The specific steps include: S1. Collect basic data and historical data of industrial boiler accidents, including structural design parameters, structural changes, operating pressure and temperature, operating personnel operation, gas transportation and boiler operation and maintenance records before the industrial boiler explosion, and establish an accident tree for industrial boiler explosion accidents based on the collected data; S2. Quantitatively analyze the industrial boiler explosion accident tree, evaluate the probability of fuzzy events in the accident tree, and obtain the actual failure probability, critical importance and probability importance through aggregate calculation, and determine the probability of occurrence of the top event based on the aggregate calculation results; S3. Collect the actual operation data and historical maintenance data of industrial boilers, correct the probability of occurrence of overhead events, and make early warnings and emergency responses based on the corrected probability of occurrence of overhead events.

2. An industrial boiler explosion assessment method according to claim 1, characterized in that: The top event of the accident tree of the industrial boiler explosion accident in S1 is the boiler explosion T1, M1~M6 are respectively configured as boiler overpressure, boiler water shortage, boiler full of water, boiler steam-water azeotropy, boiler pipe burst and gas leakage, the intermediate events J1~J6 are respectively configured as air pressure rise, safety valve failure, overpressure alarm failure, water level drop, abnormal water supply and water level rise, X1~X39 are configured as basic events, and the three minimum path sets of the accident tree are determined, which are P1={X1, X2, X9,…, X39}, P2={X3, X5, X9,…, X39} and P3={X6, X7, X8,…, X39}.

3. An industrial boiler explosion assessment method according to claim 2, characterized in that: When the industrial boiler explosion accident tree is quantitatively analyzed in S2, three experts are selected as expert one, expert two and expert three, and the weights are configured as 0.37, 0.33 and 0.30 respectively. The fuzzy linguistic variables are configured as A, B, C, D, and E from high to low, and the fuzzy linguistic variables evaluated by the experts are expressed by trapezoidal fuzzy numbers: A (0.7, 0.8, 0.8, 0.9), B (0.5, 0.6, 0.7, 0.8), C (0.4, 0.5, 0.5, 0.6), D (0.2, 0.3, 0.4, 0.5), and E (0.1, 0.2, 0.2, 0.3).

4. An industrial boiler explosion assessment method according to claim 3, characterized in that: In S2, the expert scores are converted into trapezoidal fuzzy numbers. The specific process includes: S21. The two existing trapezoidal fuzzy numbers are represented by Ai = (a1, a2, a3, a4) and Aj = (b1, b2, b3, b4). The similarity between Ai and Aj is calculated for each expert, and the similarity function is: like S(A i ,A j ) is larger, the greater the similarity of the trapezoidal fuzzy numbers Ai and Aj is; S22. Calculate the average similarity of each expert. The formula is: S23. Calculate the relative similarity of each expert: S24. Estimation of the expert's similarity coefficient: C(E i )=β·W(E i )+(1-β)·RA i Where: β——relaxation factor, 0≤β≤1, take β=0.5, W(E i )——The weight value of each expert; S25. Calculate the aggregate fuzzy number of basic events: R AG =C(E1)·R1+C(E2)·R1…C(E i )·R i Where: R i ——Expert scoring value; S26. Perform defuzzification processing to obtain an exact value, that is, the probability of brittle failure, which is expressed as: S27. The probability of occurrence of each basic event is represented by a fuzzy number, and it is concluded that the probability of brittle failure is a fuzzy probability. Converting it into an actual failure probability value, we can get: After calculation, the actual failure probability, critical importance and probability importance of each basic event are obtained.

5. An industrial boiler explosion assessment method according to claim 4, characterized in that: In S2, the probability calculation formula of the top event is expressed as: Among them, q n is the failure probability of the basic event.

6. An industrial boiler explosion assessment method according to claim 5, characterized in that: In S3, the actual operation data and historical maintenance data of the industrial boiler are collected, including boiler structure data, real-time image of boiler structure, boiler operation pressure, operation temperature, boiler water level, boiler inlet and outlet flow, water quality detection and maintenance personnel operation log, and information corresponding to the basic event is obtained. According to the obtained information, it is judged whether the basic event is solved. If it is solved, the failure probability of the corresponding basic event is adjusted to zero. If it cannot be solved, it is judged whether the basic event is a periodic event. If it is not a periodic event, the failure probability of the basic event is maintained unchanged. If it is a periodic event, a periodic change parameter is introduced: Among them, t 时长 is the elapsed time of a cycle, T 周期 The duration of a cycle.

7. An industrial boiler explosion assessment method according to claim 6, characterized in that: In S3, the user can set a warning threshold. When the probability of the top event occurring is greater than the warning threshold, an early warning is initiated. The early warning modes include sound warning and light flashing warning, and a solution corresponding to the basic event causing the early warning is output.

8. An industrial boiler explosion assessment method according to claim 7, characterized in that: In S3, when collecting the surface structure data of the industrial boiler, a visual inspection algorithm is selected to perform automatic inspection of oxidation, wear and cracks of the industrial boiler structure. The specific steps include: S31. Collect images of industrial boilers using a camera or scanner; S32, performing denoising, grayscale and contrast enhancement operations on the collected image to improve the accuracy of subsequent processing; S33, using image processing algorithms to extract features in the image related to oxidation, wear and cracks in the boiler structure; S34. Use deep learning algorithms to detect and classify the extracted image features to identify whether there are oxidation, wear and crack features in the boiler structure; S35. Mark the characteristic locations of oxidation, wear and cracks of the boiler structure on the image.

9. An industrial boiler explosion assessment device, using an industrial boiler explosion assessment method according to any one of claims 1 to 8, characterized in that: include: A control center module, which is composed of a data collection unit, a data analysis and processing unit, an early warning unit, and a response processing unit; A data acquisition module, the data acquisition module includes a visual sensor, a pressure sensor, a temperature sensor, a water level sensor, a flow sensor, a sound sensor and a water quality sensor installed on the boiler; The data acquisition module collects information and transmits it to the data collection unit, the data analysis and processing unit analyzes and processes the internal information of the data acquisition unit, the early warning unit is used to issue early warnings through light alarms and sound alarms, and the response processing unit is used to store and output industrial boiler explosion risk solutions.

Images