Oil and gas storage tank safety risk monitoring and early warning method and system

Abstract

The embodiment of the invention provides an oil and gas storage tank safety risk monitoring and early warning method and system, and belongs to the technical field of risk monitoring. The method comprises the steps that state parameters of a target oil and gas storage tank in a current period are collected, and key parameters are recognized based on the state parameters; calculating the volatility of each key parameter based on an improved exponentially weighted moving average method; historical parameter information of each key parameter is correspondingly collected, and a risk early warning model is obtained through training based on the historical parameter information and the volatility; and determining a target oil and gas storage tank early warning scheme based on the risk early warning model, and outputting alarm information based on the determined early warning scheme. According to the scheme, the comprehensive risk early warning method combining the historical data and the real-time data of the key operation parameters of the oil and gas storage tank can objectively reflect the safety risk state and timely and effectively carry out risk early warning.

Description

Technical Field

[0001] The present invention relates to the technical field of risk monitoring, and particularly to a method and a system for monitoring and early warning of safety risks of oil and gas storage tanks. Background Art

[0002] In recent years, the petrochemical industry in China has developed rapidly and has entered a high-quality development cycle of high-end, diversified and in-depth development. To reduce the cost of oil and gas, increase the reserves of oil and gas products, and expand the production scale, oil and gas storage tanks are gradually developing towards large-scale, which brings greater pressure to the enterprise's safety risk management work. Therefore, it is urgent to establish a scientific and reasonable risk early warning method and system for oil and gas storage tanks, and use intelligent and digital means to assist enterprise safety management personnel in safety risk control work, realize rapid perception and early warning of risks, help safety management personnel to take preventive measures in advance, and prevent the occurrence of major safety accidents.

[0003] At present, oil and gas storage tanks usually use industrial control systems such as SCADA to monitor parameters such as temperature, pressure, and liquid level during storage and transportation, and alarm for process abnormalities, leakage abnormalities, equipment failures and accidents, etc. to ensure the safety of storage, transportation and other processes. In addition, the method of threshold early warning using multi-parameter real-time data is often used for risk early warning of oil and gas storage tanks. For example, CN111461333A proposes a method for dynamic risk early warning of LNG storage tanks, which comprehensively considers the influence of four real-time monitoring parameters, namely density difference, temperature difference, temperature, and liquid level, on the operation risk of large LNG storage tanks. Shao Hui et al. applied the method of fuzzy mathematics to analyze the safety influencing factors of the storage tank area of chemical enterprises from five aspects and established a safety evaluation model for the chemical storage tank area. However, although the method of threshold early warning using real-time monitoring data comprehensively considers multiple parameters, it does not utilize and analyze a large amount of historical data, and the accuracy of early warning depends on the rationality and scientificity of threshold setting; while the method of early warning through comprehensive safety evaluation weakens the influence of a single factor on safety because it needs to assign weights to multiple influencing factors, which may reduce the timeliness of early warning. In view of the problems of low accuracy and weak timeliness existing in the existing oil and gas storage tank monitoring methods, a new oil and gas storage tank safety risk monitoring scheme needs to be proposed. Summary of the Invention

[0004] The purpose of the embodiments of the present invention is to provide a method and a system for monitoring and early warning of safety risks of oil and gas storage tanks, so as to solve at least the problems of low accuracy and weak timeliness existing in the oil and gas storage tank monitoring methods.

[0005] To achieve the above object, a first aspect of the present invention provides a method for monitoring and warning of safety risks of oil and gas storage tanks, the method comprising: collecting state parameters of a target oil and gas storage tank in a current period, and identifying key parameters based on the state parameters; calculating the volatility of each key parameter based on an improved exponentially weighted moving average method; correspondingly collecting historical parameter information of each key parameter, and training a risk warning model based on the historical parameter information and the volatility; determining a warning scheme for the target oil and gas storage tank based on the risk warning model, and outputting an alarm message based on the determined warning scheme.

[0006] Optionally, the power consumption parameter information includes: the key parameters include one or more of the pressure, temperature and liquid level of the oil and gas storage tank.

[0007] Optionally, the calculating the volatility of each key parameter based on an improved exponentially weighted moving average method includes: starting from the current sampling moment, selecting key parameters at multiple historical sampling moments before and combining them with the key parameter at the current sampling moment to form an array; in the array, assigning weights to all key parameters; calculating the volatility of each key parameter based on the array and the corresponding assigned weights, and the calculation rule is:

[0008]

[0009] Wherein, is the volatility of a certain key parameter; λ i-1 w0 is the assigned weight of the i-th key parameter in the array; r i is the i-th key parameter in the array; is the average value of the key parameters in the array.

[0010] Optionally, the assigning weights to all key parameters in the array includes: presetting a weight assignment coefficient, and calculating an initial weight based on the number of key parameters in the array, and the calculation rule is:

[0011]

[0012] Wherein, w0 is the initial weight; λ is the weight assignment coefficient, 0 < λ < 1; n is the number of key parameters in the array.

[0013] Optionally, the training a risk warning model based on the historical parameter information and a preset neural network includes: calculating the alarm rate of each key parameter based on the historical parameter information; using the historical parameter information, the alarm rate and the volatility as training samples to perform model training to obtain a risk warning model.

[0014] Optionally, the risk warning model includes: an input layer, a fuzzification layer, a rule layer, a fuzzy inference layer, and an output layer; the fuzzification layer is used to fuzzify the alarm rate and the volatility into linguistic variables of multiple levels; the rule layer is used to traverse the pairwise combination relationships of the linguistic variables of the alarm rate and the volatility, and calibrate the risk value levels of each corresponding relationship; the fuzzy inference layer is used to perform operations on the volatility and the alarm rate based on the Mamdani algorithm to obtain the fuzzy sets corresponding to each fuzzy rule triggered by the input layer parameters and the fuzzy sets of the risk values of each key parameter.

[0015] Optionally, the risk value calculation rule of the risk warning model is:

[0016]

[0017] where N is the number of fuzzy sets; is the center of the i-th fuzzy set; ω i (y) is the maximum membership degree corresponding to the i-th fuzzy set.

[0018] Optionally, determining the warning plan for the target oil and gas storage tank based on the risk warning model includes: obtaining the risk value of the current period through training based on the risk warning model and the key parameters of the target oil and gas storage tank in the current period; determining the preset warning level risk value threshold interval where the risk value of the current period is located based on the warning level comparison table, and determining the warning level of the current period; determining the warning plan corresponding to the warning level of the current period.

[0019] The second aspect of the present invention provides an oil and gas storage tank safety risk monitoring and warning system, the system includes: a collection unit, configured to collect the state parameters of the target oil and gas storage tank in the current period, and identify key parameters based on the state parameters; a processing unit, configured to calculate the volatility of each key parameter based on an improved exponentially weighted moving average method; a training unit, configured to collect the historical parameter information of each key parameter correspondingly, and train to obtain a risk warning model based on the historical parameter information and the volatility; a warning unit, configured to determine the warning plan for the target oil and gas storage tank based on the risk warning model, and output an alarm message based on the determined warning plan.

[0020] Optionally, the key parameters include one or more of the pressure, temperature, and liquid level of the oil and gas storage tank.

[0021] The third aspect of the present invention provides a computer-readable storage medium, on which instructions are stored, and when running on a computer, the computer is caused to execute the above-mentioned oil and gas storage tank safety risk monitoring and warning method.

[0022] In a fourth aspect of the present invention, an electronic device is provided. The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above-mentioned safety risk monitoring and early warning method for oil and gas storage tanks is implemented.

[0023] Through the above technical solution, the present invention applies the comprehensive evaluation method of fuzzy mathematics to the safety evaluation of the complex multi-factor, multi-variable, and multi-level human-machine system in the chemical storage tank area. The influencing factors of the safety of the storage tank area of chemical enterprises are analyzed from multiple aspects, and a relatively reasonable safety evaluation index system is established through the mutual connection between various factors; and the analytic hierarchy process is used to give the weights of relevant safety factors, thereby establishing a safety evaluation model for the chemical storage tank area; finally, it is verified through examples, and the results are relatively consistent with the actual situation, with strong practical applicability.

[0024] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent specific embodiments section. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The drawings are used to provide a further understanding of the embodiments of the present invention, and constitute a part of the specification, and are used to explain the embodiments of the present invention together with the following specific embodiments, but do not constitute a limitation to the embodiments of the present invention. In the drawings:

[0026] Figure 1 is a flowchart of the steps of the safety risk monitoring and early warning method for oil and gas storage tanks provided by an embodiment of the present invention;

[0027] Figure 2 is a graph of the volatility membership function degree provided by an embodiment of the present invention;

[0028] Figure 3 is a graph of the alarm rate membership function degree provided by an embodiment of the present invention;

[0029] Figure 4 is a graph of the risk value membership function degree provided by an embodiment of the present invention;

[0030] Figure 5 is a system structure diagram of the safety risk monitoring and early warning system for oil and gas storage tanks provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] The following will describe the specific embodiments of the present invention in detail with reference to the drawings. It should be understood that the specific embodiments described herein are only for explaining and illustrating the present invention, and are not used to limit the present invention.

[0032] In recent years, the petrochemical industry in China has developed rapidly and has entered a high-quality development cycle of high-end, diversified, and in-depth development. To reduce the cost of oil and gas, increase the reserves of oil and gas products, and expand the production scale, oil and gas storage tanks are gradually developing towards large-scale, which brings greater pressure to the safety risk management work of enterprises. Therefore, it is urgent to establish a scientific and reasonable risk early warning method and system for oil and gas storage tanks, and use intelligent and digital means to assist enterprise safety management personnel in safety risk control work, realize rapid perception and early warning of risks, help safety management personnel to take preventive measures in advance, and prevent the occurrence of major safety accidents.

[0033] At present, oil and gas storage tanks usually use industrial control systems such as SCADA to monitor parameters such as temperature, pressure, and liquid level during storage and transportation, and alarm for process abnormalities, leakage abnormalities, equipment failures, and accidents to ensure the safety of storage, transportation, and other processes. In addition, the method of threshold early warning using multi-parameter real-time data is often used for risk early warning of oil and gas storage tanks. For example, CN111461333A proposes a dynamic risk early warning method for LNG storage tanks, which comprehensively considers the influence of four real-time monitoring parameters, namely density difference, temperature difference, temperature, and liquid level, on the operation risk of large LNG storage tanks. Shao Hui et al. applied the method of fuzzy mathematics to analyze the safety influencing factors of the storage tank area of chemical enterprises from five aspects and established a safety evaluation model for the chemical storage tank area. However, although the method of threshold early warning using real-time monitoring data comprehensively considers multiple parameters, it does not utilize and analyze a large amount of historical data, and the accuracy of early warning depends on the rationality and scientificity of threshold setting; while the method of early warning through comprehensive safety evaluation weakens the influence of a single factor on safety because it needs to assign weights to multiple influencing factors, which may reduce the timeliness of early warning.

[0034] Aiming at the problems of low accuracy and weak timeliness existing in the existing oil and gas storage tank monitoring methods, the present invention proposes a safety risk monitoring and early warning method for oil and gas storage tanks. The present invention applies the comprehensive evaluation method of fuzzy mathematics to the complex multi-factor, multi-variable, multi-level man-machine system of the chemical storage tank area for safety evaluation. Analyze the safety influencing factors of the storage tank area of chemical enterprises from multiple aspects, and establish a relatively reasonable safety evaluation index system through the mutual connection between various factors; and apply the analytic hierarchy process to give the weights of relevant safety factors, thereby establishing a safety evaluation model for the chemical storage tank area; finally, verify through examples, and the results are in good agreement with the actual situation, with strong practical applicability.

[0035] Figure 1 It is the method flow chart of the safety risk monitoring and early warning method for oil and gas storage tanks provided by an embodiment of the present invention. As Figure 1 shown, the embodiment of the present invention provides a safety risk monitoring and early warning method for oil and gas storage tanks, and the method includes:

[0036] Step S10: Collect the state parameters of the target oil and gas storage tank in the current cycle, and identify the key parameters based on the state parameters.

[0037] Specifically, to optimize the floor area and save infrastructure investment, oil and gas storage tanks are gradually developing towards large-scale, bringing greater pressure to the enterprise's safety risk management work. At present, methods such as threshold warning and safety comprehensive evaluation are commonly used for risk warning of oil and gas storage tanks. However, the method of using real-time monitoring data for threshold warning, although considering various parameters comprehensively, does not utilize and analyze a large amount of historical data, and the warning accuracy depends on the rationality and scientificity of the threshold setting; while the method of warning through safety comprehensive evaluation weakens the impact of a single factor on safety due to the need to assign weights to multiple influencing factors, which may reduce the timeliness of warning. Therefore, the solution of the present invention proposes a comprehensive warning method that can effectively combine historical monitoring data and real-time data and ensure the timeliness of warning to guide the risk management of oil and gas storage tanks.

[0038] Specifically, the key parameters include one or more of the pressure, temperature, and liquid level of the oil and gas storage tank.

[0039] In the embodiment of the present invention, the volatility of the key operating parameters of the temperature, pressure, and liquid level of the oil and gas storage tank can reflect the smoothness of the data in a certain cycle. The sudden fluctuation or the increase in the volatility of the monitoring parameters during a certain period indicates that there have been significant changes in the state of the storage tank in this cycle, and abnormalities may have occurred. Therefore, the volatility of a certain parameter of the storage tank can be used as an important reference factor for risk warning.

[0040] Step S20: Calculate the volatility of each key parameter based on the improved exponentially weighted moving average method.

[0041] Specifically, starting from the current sampling moment, select the key parameters of multiple historical sampling moments before and form an array with the key parameter of the current sampling moment; in the array, assign weights to all key parameters; based on the array and the corresponding assigned weights, calculate the volatility of each key parameter, and the calculation rule is:

[0042]

[0043] Among them, is the volatility of a certain key parameter; λ i-1 w0 is the assigned weight of the i-th key parameter in the array; r i is the i-th key parameter in the array; is the average value of the key parameters in the array.

[0044] In the embodiment of the present invention, n consecutive values are respectively selected forward from the current real-time values of each key operating parameter of the storage tank to form a data set X = (x1, x2, …, x n ), where x i represents the i-th historical value starting from the current real-time value. The fluctuation of conventional data can be represented by variance, that is

[0045]

[0046] where

[0047] Since it is considered that the data closer to the current moment has a greater impact, a larger weight should be set. Therefore, the weight of r1 is set to w0, the weight of r2 is set to λw0, and the weight of r n is set to λ n-1 w0, and 0 < λ < 1. That is

[0048]

[0049] Then the data volatility can be expressed as

[0050]

[0051] Starting from 0.05 and taking λ at intervals of 0.1, the average value of the 10 data volatilities obtained is the final volatility σ 2 of this parameter.

[0052] Furthermore, in the said array, weights are assigned to all key parameters, including:

[0053] Preset a weight distribution coefficient, calculate the initial weight based on the number of key parameters in the array, and the calculation rule is:

[0054]

[0055] where w0 is the initial weight;

[0056] λ is the weight distribution coefficient, 0 < λ < 1;

[0057] n is the number of key parameters in the array.

[0058] Step S30: Correspondingly collect the historical parameter information of each key parameter, and train a risk warning model based on the historical parameter information and the volatility.

[0059] Specifically, calculate the alarm rate of each key parameter based on the historical parameter information; use the historical parameter information, the alarm rate, and the volatility as training samples to perform model training to obtain a risk warning model.

[0060] Example 1:

[0061] The historical alarm data of the key monitoring parameters of temperature, pressure, and liquid level of an oil and gas storage tank can reflect the enterprise's safety control level of a certain risk of the storage tank or the safety and stability of the storage tank. Therefore, the historical alarm data of a certain key operating parameter of the storage tank can be used as an important reference factor for risk early warning.

[0062] For each of the key operating parameters of temperature, liquid level, and pressure, record the number of alarms of each parameter in the past month as N (an alarm lasting until the next day is regarded as two alarms), and the duration of the i-th alarm is S i , in hours, then the alarm rate

[0063]

[0064] Furthermore, the risk early warning model includes: an input layer, a fuzzification layer, a rule layer, a fuzzy inference layer, and an output layer; the fuzzification layer is used to fuzzify the alarm rate and the volatility into linguistic variables of multiple levels; the rule layer is used to traverse the pairwise combination relationship of the linguistic variables of the alarm rate and the volatility, and calibrate the risk value level of each corresponding relationship; the fuzzy inference layer is used to perform operations on the volatility and the alarm rate based on the Mamdani algorithm to obtain the fuzzy sets corresponding to each fuzzy rule triggered by the input layer parameters and the fuzzy sets of the risk values of each key parameter.

[0065] Example 2:

[0066] Construct an oil and gas storage tank risk early warning model through a fuzzy inference method, that is, use fuzzy logic to formulate the mapping process from the given input to the output, and form a fuzzy inference process from the input layer, fuzzification layer, rule layer, fuzzy inference layer to the output layer.

[0067] Input layer: This layer determines the input layer parameters, takes the volatility calculated from the current real-time values of the single key monitoring parameters of the temperature, pressure, and liquid level of the oil and gas storage tank by selecting the previous consecutive numerical values as one of the input parameters, and takes the alarm rate of each key operating parameter as another input parameter.

[0068] Fuzzification layer: Use the triangular membership function to fuzzify the input values into three linguistic variables of "high", "medium", and "low" respectively, and construct the volatility membership function, alarm rate membership function, and risk value membership function as Figure 2 , Figure 3 and Figure 4 shown, where the X-axis represents the influencing factor data, and the Y-axis is the membership degree of this data to the evaluation level. is a critical value of the volatility factor, which is set according to the expected sensitivity of the risk early warning to be achieved. The smaller this value is set, the more it means that a relatively small fluctuation is considered to bring a higher risk. The overlapping part of the images represents that the numerical values in this area belong to multiple ranges.

[0069] Rule layer: Fuzzy rules are established based on the historical alarms and the degree of risk representation of volatility to determine the fuzzy relationships among volatility, alarm rate, and risk value under different circumstances. For example, it is set that:

[0070] 1) If σ 2 is "low", and L is "low", then the risk value is "low";

[0071] 2) If σ 2 is "low", and L is "medium", then the risk value is "low";

[0072] 3) If σ 2 is "low", and L is "high", then the risk value is "medium";

[0073] 4) If σ 2 is "medium", and L is "low", then the risk value is "low";

[0074] 5) If σ 2 is "medium", and L is "medium", then the risk value is "medium";

[0075] 6) If σ 2 is "medium", and L is "high", then the risk value is "high";

[0076] 7) If σ 2 is "high", and L is "low", then the risk value is "medium";

[0077] 8) If σ 2 is "high", and L is "medium", then the risk value is "high";

[0078] 9) If σ 2 is "high", and L is "high", then the risk value is "high".

[0079] Fuzzy inference layer: Based on the Mamdani algorithm, operations are performed on volatility and alarm rate to obtain the fuzzy sets corresponding to each fuzzy rule triggered by the input layer parameters and the fuzzy set of the risk value of each key parameter.

[0080] Output layer: This layer mainly performs defuzzification, that is, determining an exact value that best represents each fuzzy set. The risk value y of each key parameter is calculated using the following formula * .

[0081]

[0082] where N represents the number of fuzzy sets, is the center of the i-th fuzzy set, and ω i (y) is the maximum membership degree corresponding to the i-th fuzzy set.

[0083] Step S40: Based on the risk early warning model, determine the early warning plan for the target oil and gas storage tank, and output an alarm message based on the determined early warning plan.

[0084] Specifically, train the risk value of the current period based on the risk early warning model and the key parameters of the target oil and gas storage tank in the current period; based on the early warning level comparison table, determine the preset early warning level risk value threshold interval where the risk value of the current period is located, and determine the early warning level of the current period; determine the early warning plan corresponding to the early warning level of the current period.

[0085] Embodiment III:

[0086] When the real-time monitoring value of the key operating parameters of the oil and gas storage tank reaches high alarm, low alarm, high-high alarm or low-low alarm, it means that the current exceeds the safety state critical value and there is a certain safety risk. At the same time, the greater the risk value of each key parameter obtained based on the volatility and alarm rate, the higher the risk. Therefore, the risk early warning level is set according to the numerical value and the current real-time monitoring value of the key parameters as shown in Table 1.

[0087]

[0088] Table 1 Risk Early Warning Rules

[0089] That is, when the real-time monitoring value of the key operating parameters of the oil and gas storage tank reaches high alarm or low alarm, a first-level early warning is carried out; when the real-time monitoring value of the key operating parameters of the storage tank reaches low alarm or low-low alarm, a second-level early warning is carried out; comprehensively considering the risk value y * , and taking the highest early warning level among the parameters of the oil and gas storage tank as the early warning level of the storage tank for early warning to ensure the timeliness of early warning.

[0090] In the embodiment of the present invention, for the problems that the existing early warning models mainly focus on single-factor early warning or use real-time monitoring data of multiple parameters for threshold early warning and comprehensive safety risk evaluation, and a large amount of historical data has not been effectively utilized and analyzed, the volatility and alarm rate are calculated using historical monitoring data and alarm data, and the volatility and alarm rate are used as input parameters. The risk situation of each parameter is calculated collaboratively through fuzzy reasoning, ensuring the robustness of the risk value calculation result under the condition of conforming to human cognition; at the same time, combined with the current real-time parameter threshold early warning situation, the risk value inferred by fuzzy reasoning and the highest early warning level among the monitoring parameters are used as the early warning level of the storage tank for early warning, ensuring the timeliness of early warning.

[0091] Furthermore, compared with the prior art, the solution of the present invention considers the historical data of multiple parameters of oil and gas storage tanks, and constructs a comprehensive risk early warning method combining the historical data and real-time data of the key operating parameters of oil and gas storage tanks based on the fuzzy inference method, which can objectively reflect the safety risk status, conduct risk early warning in a timely and effective manner, and help enterprise personnel carry out safety management work.

[0092] Example 4:

[0093] Taking a 50,000 cubic meter diesel storage tank (hereinafter referred to as T003 storage tank) of a certain company as the research object, the fuzzy risk early warning model is applied to evaluate its risk level. The changes of the key operating parameters of the storage tank in a certain 3 days are shown in Table 2, and there is a liquid level alarm once in the recent month, lasting for 22 hours and 50 minutes, and there is no temperature alarm.

[0094] Moment Liquid level Temperature Pressure 1 1.15m 24.37℃ 0.703 kpa 2 1.15m 24.37℃ 0.711 kpa 3 1.16m 24.38℃ 0.667 kpa 4 1.16m 24.39℃ 0.682 kpa 5 1.15m 24.40℃ 0.697 kpa 6 12.08m 24.40℃ 0.645 kpa 7 12.08m 24.41℃ 0.561 kpa 8 12.08m 24.41℃ 0.578 kpa 9 12.08m 24.43℃ 0.672 kpa 10 1.16m 24.43℃ 0.572 kpa

[0095] Table 2 Data of parameter changes

[0096] Select n consecutive values respectively from the current real-time values of each key operating parameter of the storage tank to form a data set X = (x1, x2,..., x n ), x i represents the i-th historical value starting from the current real-time value. The fluctuation of conventional data can be represented by variance, that is

[0097]

[0098] Among them,

[0099] Since it is considered that the data closer to the current moment has a greater impact, a larger weight should be set. Therefore, the weight of r1 is set as w0, the weight of r2 is set as λw0, and the weight of r n is set as λ n-1 w0, and 0 < λ < 1, That is

[0100]

[0101] Then the data volatility can be expressed as

[0102]

[0103] Starting from 0.05, taking λ at intervals of 0.1, and taking the average value of the 10 data volatilities obtained as the final volatility σ 2 of this parameter, and the data volatilities of each key operating parameter are shown in Table 3:

[0104]

[0105] Table 3 Numerical volatility

[0106] Then the volatility σ of liquid level, temperature, and pressure 2 takes values of 21.41340802, 0.000658403, and 0.002796788 respectively.

[0107] For each key operating parameter of temperature, liquid level, and pressure, record the number of alarms for each parameter in the past month as N (an alarm lasting until the next day is regarded as two alarms), and the duration of the i-th alarm is S i , in hours, then the alarm rate

[0108]

[0109] In the past month, the liquid level of T003 storage tank had one alarm, lasting for 22 hours and 50 minutes, and there was no alarm for temperature. Therefore, the alarm rates of liquid level, temperature, and pressure are 0.3171, 0, and 0 respectively.

[0110] Based on the calculation results of the volatility and alarm rate of each key operating parameter of liquid level, temperature, and pressure, determine the numerical values of the input layer parameters as shown in Table 4.

[0111] Serial number Monitoring parameter <![CDATA[Volatility σ 2 > Alarm rate L 1 Liquid level 21.41340802 0.3171 2 Temperature 0.000658403 0 3 Pressure 0.002796788 0

[0112] Table 4 Numerical Values of Input Layer Parameters

[0113] Use the triangular membership function to fuzzify the input values into three linguistic variables: "high", "medium", and "low" respectively, and construct the volatility membership function, alarm rate membership function, and risk value membership function. The lower limit of the volatility factor is 0, and there is no upper limit. The upper and lower limits of the alarm rate factor are 0 and 1 respectively. H is a critical value of the volatility factor, and this value is set to 1 in the calculation process of this case. And set the slope of each straight line of the image to or

[0114] Establish fuzzy rules according to the historical alarms and the degree of risk represented by the volatility, and determine the fuzzy relationship between the volatility, alarm rate, and risk value under different conditions. Based on the Mamdani algorithm, perform operations on the volatility and alarm rate to obtain the fuzzy sets corresponding to each fuzzy rule triggered by the input layer parameters and the fuzzy sets of the risk values of each key parameter.

[0115] Since the volatility and alarm rate of the liquid level in the case are 21.4134080 and 0.3171 respectively, rules 7) and 8) are triggered. The maximum membership degrees of rules 7) and 8) are 0.0487 and 0.4513 respectively. The fuzzy inference process of the risk values of the temperature and pressure key parameters is similar to that of the liquid level key parameter, and only rule 1) is triggered for both of them. The maximum membership degrees of rule 1) are 0.9993 and 0.9972 respectively.

[0116] Defuzzification is carried out, that is, an exact value that best represents each fuzzy set is determined. The risk value y of each key parameter is calculated using the following formula * .

[0117]

[0118] where N represents the number of fuzzy sets, is the center of the i-th fuzzy set, and ω i (y) is the maximum membership degree corresponding to the i-th fuzzy set. The risk value of the liquid level key parameter is obtained as 0.7476. The calculation processes of the risk values of the temperature and pressure key parameters are similar to that of the liquid level key parameter, and are 0.0003 and 0.0014 respectively.

[0119] When the real-time monitoring values of the key operating parameters of the oil and gas storage tank reach high alarm, low alarm, high-high alarm or low-low alarm, it means that the current exceeds the safety state critical value and there is a certain safety risk. At the same time, the greater the risk value of each key parameter obtained based on the volatility and alarm rate, the higher the risk.

[0120] When the real-time monitoring value of the key operating parameter of the oil and gas storage tank reaches high alarm or low alarm, a first-level early warning is carried out; when the real-time monitoring value of the key operating parameter of the tank reaches low alarm or low-low alarm, a second-level early warning is carried out; comprehensively considering the risk value y * , and taking the highest early warning level among the parameters of the oil and gas storage tank as the early warning level of the tank for early warning to ensure the timeliness of early warning.

[0121] The current real-time parameters of the liquid level, pressure and temperature of the storage tank do not reach high / low alarm, while the liquid level risk value is greater than 0.5, and the pressure and temperature risk values are less than 0.1. Therefore, based on the risk early warning conditions and the grade table, a second-level early warning is carried out for the liquid level key parameter, and no early warning is carried out for the temperature and pressure key parameters.

[0122] Figure 5 is the system structure diagram of the oil and gas storage tank safety risk monitoring and early warning system provided by an embodiment of the present invention. As Figure 5 shown, an embodiment of the present invention provides an oil and gas storage tank safety risk monitoring and early warning system, and the system includes: a collection unit for collecting the state parameters of the target oil and gas storage tank in the current period and identifying key parameters based on the state parameters; a processing unit for calculating the volatility of each key parameter based on an improved exponentially weighted moving average method; a training unit for correspondingly collecting the historical parameter information of each key parameter and training to obtain a risk early warning model based on the historical parameter information and the volatility; an early warning unit for determining the early warning plan of the target oil and gas storage tank based on the risk early warning model and outputting an alarm message based on the determined early warning plan.

[0123] Preferably, the acquisition unit uses data acquisition protocols such as OPC through the edge gateway to collect the liquid level, temperature, and pressure data of the storage tank in real time, and performs data cleaning and governance on the real-time key operation parameters, clarifying the effective range, control value, high and low alarm values, high-high and low-low alarm values, and other threshold information of each parameter, and simultaneously stores the processed data.

[0124] Preferably, the early warning unit combines the volatility rate and historical alarm rate of each key operation parameter of the oil and gas storage tank, and uses the fuzzy inference method to quantify the risk value of each key operation parameter. Combining the real-time monitoring value of each key operation parameter of the oil and gas storage tank and the risk value of the risk value calculation module, according to the risk early warning mechanism, judge the risk early warning level, and send a text message to the relevant person in charge.

[0125] The embodiment of the present invention also provides a computer-readable storage medium, on which instructions are stored, and when the instructions run on a computer, the computer is made to execute the above-mentioned oil and gas storage tank safety risk monitoring and early warning method.

[0126] Those skilled in the art can understand that all or part of the steps in the method of the above embodiment can be completed by instructing relevant hardware through a program. The program is stored in a storage medium, including several instructions to make a single-chip microcomputer, chip or processor execute all or part of the steps of the method described in each embodiment of the present invention. The foregoing storage medium includes: USB flash drive, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other various media that can store program codes.

[0127] The optional embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the embodiments of the present invention are not limited to the specific details in the above embodiments. Within the scope of the technical concept of the embodiments of the present invention, various simple modifications can be made to the technical solutions of the embodiments of the present invention, and these simple modifications all belong to the protection scope of the embodiments of the present invention. In addition, it should be noted that, in the above specific embodiments, the various specific technical features described can be combined in any appropriate manner without conflict. To avoid unnecessary repetition, the embodiments of the present invention will not separately describe various possible combination methods.

[0128] In addition, any combination can be made between various different embodiments of the present invention, as long as it does not violate the idea of the embodiments of the present invention, it should also be regarded as the content disclosed by the embodiments of the present invention.

Claims

1. A method for monitoring and warning of safety risks of oil and gas storage tanks, characterized in that, The method includes: Collecting the state parameters of the target oil and gas storage tank in the current period, and identifying key parameters based on the state parameters; Calculating the volatility of each key parameter based on the improved exponentially weighted moving average method; Collecting the historical parameter information of each key parameter correspondingly, and training a risk warning model based on the historical parameter information and the volatility; Based on the risk warning model, determining the warning scheme for the target oil and gas storage tank, and outputting an alarm message based on the determined warning scheme.

2. The method according to claim 1, wherein The key parameters include: One or more of the pressure, temperature and liquid level of the oil and gas storage tank.

3. The method according to claim 1, wherein The calculating the volatility of each key parameter based on the improved exponentially weighted moving average method includes: Starting from the current sampling moment, selecting the key parameters at multiple historical sampling moments forward and forming an array with the key parameter at the current sampling moment; Assigning weights to all key parameters in the array; Calculating the volatility of each key parameter based on the array and the corresponding assigned weights, and the calculation rule is: Among them, is the volatility of a certain key parameter; λ i-1 w0 is the assigned weight of the i-th key parameter in the array; r i is the i-th key parameter in the array; It is the mean of the key parameters of the array.

4. The method according to claim 3, characterized in that, The assigning weights to all key parameters in the array includes: Presetting a weight distribution coefficient, and calculating the initial weight based on the number of key parameters in the array, and the calculation rule is: Where w0 is the initial weight; λ is the weight distribution coefficient, 0 < λ < 1; n is the number of key parameters in the array.

5. The method according to claim 1, wherein The training the risk warning model based on the historical parameter information and a preset neural network includes: Calculating the alarm rate of each key parameter based on the historical parameter information; Using the historical parameter information, the alarm rate and the volatility as training samples to perform model training to obtain a risk warning model.

6. The method according to claim 5, wherein The risk warning model includes: An input layer, a fuzzification layer, a rule layer, a fuzzy inference layer and an output layer; The fuzzification layer is used to fuzzify the alarm rate and the volatility into language variables of multiple levels; The rule layer is used to traverse the pairwise combination relationship of the language variables of the alarm rate and the volatility, and calibrate the risk value level of each corresponding relationship; The fuzzy inference layer is used to perform operations on the volatility and the alarm rate based on the Mamdani algorithm to obtain the fuzzy sets corresponding to each fuzzy rule triggered by the input layer parameters and the fuzzy sets of the risk values of each key parameter.

7. The method according to claim 6, wherein The risk value calculation rule of the risk warning model is: Where N is the number of fuzzy sets; is the center of the i-th fuzzy set; ω i (y) is the maximum membership degree corresponding to the i-th fuzzy set.

8. The method according to claim 1, characterized in that The determining the warning scheme for the target oil and gas storage tank based on the risk warning model includes: Training based on the risk warning model and the key parameters of the target oil and gas storage tank in the current period to obtain the risk value of the current period; Based on the warning level comparison table, determining the preset warning level risk value threshold interval where the risk value of the current period is located, and determining the warning level of the current period; Determining the warning scheme corresponding to the warning level of the current period.

9. An oil and gas storage tank safety risk monitoring and early warning system, characterized in that, The system includes: A collection unit, configured to collect the state parameters of the target oil and gas storage tank in the current period, and identify key parameters based on the state parameters; A processing unit, configured to calculate the volatility of each key parameter based on the improved exponentially weighted moving average method; A training unit, configured to collect the historical parameter information of each key parameter correspondingly, and train a risk warning model based on the historical parameter information and the volatility; An early warning unit, configured to determine an early warning plan for a target oil and gas storage tank based on the risk early warning model, and output an alarm message based on the determined early warning plan.

10. The system according to claim 9, wherein The key parameters include: One or more of the pressure, temperature, and liquid level of the oil and gas storage tank.

11. A computer-readable storage medium, characterized in that, Instructions are stored on the computer-readable storage medium, which, when running on a computer, cause the computer to execute the oil and gas storage tank safety risk monitoring and early warning method according to any one of claims 1-8.

12. An electronic device, the electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the oil and gas storage tank safety risk monitoring and early warning method according to any one of claims 1-8.

Images