Storage tank foundation settlement intelligent algorithm driving early warning system and method

Abstract

The invention provides a storage tank foundation settlement intelligent algorithm driving early warning system and method, and the method comprises the steps: S1, obtaining settlement data of each monitoring point of a storage tank, carrying out the preprocessing of the settlement data, and carrying out the quantification after the preprocessing, and obtaining an actual settlement value of each monitoring point; s2, establishing a target function; s3, iteratively solving a mean value and a standard deviation after optimization by adopting a butterfly optimization algorithm, and recording the mean value and the standard deviation as mu optimal and sigma optimal; s4, setting a low-position abnormal threshold and a high-position abnormal threshold, comparing the settlement data actually measured by the subsequent monitoring point with the low-position abnormal threshold and the high-position abnormal threshold respectively, and judging whether the settlement data monitored by the monitoring point is abnormal data or not according to the comparison result; and judging whether the continuous multiple settlement data of the monitoring point under the monitoring frequency is abnormal data, if so, determining that the monitoring point settles, and if not, determining that the monitoring point does not settle. The method has high stability and strong anti-interference capability, can effectively reduce the false alarm rate, and can accurately and rapidly identify the settlement abnormity of the storage tank foundation.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of tank foundation settlement, and particularly relates to a tank foundation settlement intelligent algorithm driving early warning system and method. BACKGROUND

[0002] Steel tank structures have been widely used in many industrial fields such as petroleum, chemical industry, and electric power. With the rapid development of these fields, the number and capacity of large tanks are increasing, and the safety requirements for tanks are also increasing. However, foundation settlement may cause abnormal conditions of the tank body, and these abnormalities are often difficult to be found and handled in time, thereby increasing the potential disaster risk and threatening the safe and stable operation of the crude oil storage area. Therefore, establishing and improving the tank monitoring and early warning method can significantly reduce the disaster risk and is of great significance to improving industrial safety and reducing economic losses.

[0003] The prior art does not fit and optimize the real-time settlement data, thereby improving the accuracy of the monitoring data early warning. SUMMARY

[0004] To solve the above problems, the application provides a tank foundation settlement intelligent algorithm driving early warning system and method.

[0005] To achieve the above purpose, the application provides the following technical scheme:

[0006] A tank foundation settlement intelligent algorithm driving early warning method comprises the following steps,

[0007] S1: installing a fiber Bragg grating sensor at each of a plurality of monitoring points of a tank to obtain settlement data of each monitoring point, and pre-processing the settlement data to obtain actual settlement values of each monitoring point after pre-processing;

[0008] S2: establishing a target function;

[0009]

[0010] In the formula, S i represents the actual settlement value of the i th monitoring point; S(x i ) represents the Gaussian model calculation value of the i th monitoring point; and n represents the number of monitoring points.

[0011] In the formula, x i represents the monitoring value of the i th monitoring point.

[0012]

[0013] In the formula, x i represents the monitoring value of the i th monitoring point.

[0014] S(x i ) represents the Gaussian model calculation value of the i th monitoring point.

[0015] μ represents the mean of the settlement data of all monitoring points; and σ represents the standard deviation of the settlement data of all monitoring points;

[0016] S3: The butterfly optimization algorithm is used to iteratively solve the objective function in S2 to obtain the optimized mean and standard deviation, denoted as μ 优 and σ 优 ;

[0017] S4: The low abnormal threshold and the high abnormal threshold are set by μ 优 and σ 优 , and the measured settlement data of the subsequent monitoring points are compared with the low abnormal threshold and the high abnormal threshold respectively. According to the comparison result, it is judged whether the settlement data monitored by the monitoring point is abnormal data or not. It is also judged whether the settlement data of the monitoring point is abnormal data for multiple times under the monitoring frequency. If yes, the settlement of the monitoring point occurs, and if no, the settlement of the monitoring point does not occur.

[0018] S5: The Unity3D software is used to display the monitoring result of the settlement of the tank foundation.

[0019] Further, in step S4, the low abnormal threshold δ 低 is set as: δ 低 = μ 优 - 2σ 优 , and the high abnormal threshold δ 高 is set as: δ 高 = μ 优 + 2σ 优 . When the settlement data monitored by a monitoring point is greater than or equal to the high abnormal threshold δ 高 or less than or equal to the low abnormal threshold δ 低 , it is indicated that the settlement data monitored by the monitoring point is abnormal data, and preliminary alarm is given.

[0020] Further, in step S4, when the settlement data of a monitoring point for three times in succession are all abnormal data, the settlement of the monitoring point occurs, and continuous alarm is given.

[0021] Further, in step S3, the butterfly optimization algorithm includes the following steps:

[0022] S3.1: The population size, the maximum number of iterations, the perception factor, the stimulation intensity and the control constant are set;

[0023] S3.2: The "fragrance intensity" of each butterfly is calculated, that is, the objective function value of each butterfly is calculated;

[0024] S3.3: The "odor concentration" of each butterfly is calculated, that is, f = cI a wherein c, I and a respectively represent the perception factor, the stimulation intensity and the control constant.

[0025] S3.4: Update the position of each butterfly according to the "odor concentration" of the butterfly;

[0026]

[0027] wherein, and respectively represent the position of the i-th butterfly at the t+1th and tth iteration, r is a random factor; g* represents the current optimal butterfly position; and are the positions of other individuals randomly selected from the population, f i is the odor concentration of the i-th individual, and p is a threshold parameter;

[0028] S3.5: Iteratively update until the stopping condition is met; the stopping condition is that the maximum number of iterations is reached 20 times or the objective function value changes less than the threshold value;

[0029] S3.6: Output the optimal mean and standard deviation μ 优 and σ 优 .

[0030] Further, an intelligent algorithm-driven early warning system for tank foundation settlement includes the following modules:

[0031] Data acquisition module: used for collecting and displaying the settlement data of each monitoring point;

[0032] Simulation iteration module: used for iteratively calculating the optimal mean and standard deviation, i.e. μ 优 and σ 优 , by combining the settlement data of each monitoring point with the objective function;

[0033] Data analysis and alarm module: used for determining whether the settlement data of the monitoring point is abnormal data by setting a low abnormal threshold and a high abnormal threshold; when the settlement data is abnormal data, a preliminary alarm is issued; according to the monitoring frequency, it is constantly determined whether the monitoring data of the monitoring point is abnormal data for three times, and if so, the alarm is continuously issued;

[0034] Display module: based on C4D software to build a 3D model of the target tank, and based on Unity3D software to establish a monitoring system for tank foundation settlement, the real-time settlement data collected at each monitoring point of the target tank is displayed in the monitoring system for tank foundation settlement.

[0035] The above technical solutions can achieve the following beneficial effects:

[0036] The scheme fuses a Gaussian model and a butterfly optimization algorithm, optimizes parameters in the Gaussian model, realizes accurate monitoring of the settlement of the tank foundation, has high stability and strong anti-interference capability, and can effectively reduce the false alarm rate. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 is a module diagram of the method;

[0038] Figure 2 is a system architecture schematic diagram of the method;

[0039] Figure 3 is a flowchart of the method. DETAILED DESCRIPTION

[0040] To make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the present application will be described in detail below with reference to the drawings. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0041] Embodiment 1: As shown in Figure 1 , Figure 2 and Figure 3 , an intelligent algorithm driven early warning method for tank foundation settlement includes the following steps:

[0042] S1: Install fiber Bragg grating sensors at multiple monitoring points of the tank, obtain original settlement data of each monitoring point, and pretreat the settlement data, and convert the pretreated settlement data into settlement values of each monitoring point.

[0043] Preprocessing includes removing noise and filling missing values.

[0044] S2: Establish a target function.

[0045] Specifically, the formula is:

[0046]

[0047] In formula (1), S i represents the actual settlement value of the i-th monitoring point, which is obtained by step S1; S(x i ) represents the Gaussian model calculation value of the i-th monitoring point. n represents the number of monitoring points;

[0048] Wherein:

[0049]

[0050] In formula (2), x iS (x) represents the settlement data of the i-th monitoring point.

[0051] S (x) represents the settlement data of the i-th monitoring point. i S (x) represents the settlement data of the i-th monitoring point.

[0052] μ represents the mean of the settlement data of all monitoring points; σ represents the standard deviation of the settlement data of all monitoring points.

[0053] S3: Adopting the butterfly optimization algorithm, iteratively solving the objective function in S2, obtaining the optimized mean and standard deviation, denoted as μ 优 and σ 优 .

[0054] The butterfly optimization algorithm belongs to the existing algorithm.

[0055] It includes the following steps:

[0056] S3.1: Set the population size (50), the maximum number of iterations, the perception factor, the stimulation intensity and the control constant, etc.

[0057] S3.2: Calculate the "scent intensity" of each butterfly, i.e. calculate the objective function value of each butterfly.

[0058] S3.3: Calculate the "odor concentration" of each butterfly; that is, f = cI a Where c, I, a represent the perception factor, the stimulation intensity and the control constant, respectively.

[0059] S3.4: According to the "odor concentration" of each butterfly, update the butterfly position;

[0060]

[0061] In the formula, and represent the position of the i-th butterfly at the t+1th and tth iteration, respectively, and r is a random factor; g* represents the current optimal butterfly position; and are the positions of other individuals randomly selected from the population, f i is the odor concentration of the i-th individual, and p is the threshold parameter.

[0062] S3.5: Repeat S3.2-S3.3 for iterative updating until the stopping condition is met (the maximum number of iterations is reached 20 times or the objective function value changes less than the threshold value, such as 0.05).

[0063] S3.6: Output the mean and standard deviation μ 优 and σ 优 corresponding to the optimal butterfly individual.

[0064] S4: Set a low abnormal threshold and a high abnormal threshold, and determine whether the measured settlement data of the subsequent monitoring point is abnormal in combination with the low abnormal threshold and the high abnormal threshold, and when the measured settlement data is abnormal, a preliminary alarm information is sent out.

[0065] Low abnormal threshold: δ 低 = μ 优 - 2σ 优 ;

[0066] High abnormal threshold: δ 高 = μ 优 + 2σ 优 ;

[0067] The low abnormal threshold and the high abnormal threshold are set in this way because, based on the Gaussian probability model, when the settlement data monitored by a monitoring point is greater than the low abnormal threshold δ 低 and less than the high abnormal threshold δ 高 , it is indicated that the settlement data monitored by the monitoring point is non-abnormal data; when the settlement data monitored by a monitoring point is greater than or equal to the high abnormal threshold δ 高 or less than or equal to the low abnormal threshold δ 低 , it is indicated that the settlement data monitored by the monitoring point is abnormal data, and a preliminary alarm is sent out.

[0068] When the settlement data of a monitoring point is determined to be abnormal data for three times in succession at a monitoring frequency, settlement occurs at the monitoring point.

[0069] S5: Use Unity3D software to display the monitoring result of the settlement of the tank foundation. The Unity3D software has the characteristic of visualizing the settlement. First, a three-dimensional model of the tank is established by using the Unity3D software, and the positions of the monitoring points are set. The monitoring points of the abnormal data obtained in S1-S4 are intuitively embodied in the Unity3D software, and when settlement occurs, the monitoring points at the corresponding positions in the Unity3D software send out a continuous alarm.

[0070] Embodiment 2: A tank foundation settlement monitoring system, comprising:

[0071] A data acquisition module for acquiring and displaying the settlement data of each monitoring point;

[0072] A simulation iteration module for iteratively calculating the optimal mean and standard deviation, i.e., μ 优 and σ 优 , by using the settlement data of each monitoring point in combination with a target function.

[0073] The data analysis alarm module is used for judging whether the settlement data of the monitoring point is abnormal data through the set low abnormal threshold and high abnormal threshold; when the settlement data is abnormal data, a preliminary alarm is sent; according to the monitoring frequency, it is continuously judged whether the monitoring data of the monitoring point is abnormal data for three times, and if so, the alarm is continuously sent;

[0074] The display module: based on C4D software to build the target tank 3D model, based on Unity3D software to establish the monitoring system of the tank foundation settlement, and the real-time settlement data collected at each monitoring point of the target tank is displayed in the monitoring system of the tank foundation settlement.

[0075] Based on the above ideal embodiments according to the present application, through the above description, relevant personnel can make various changes and modifications without deviating from the technical idea of the present application. The technical scope of the present application is not limited to the contents of the specification, and must be determined according to the scope of the claims.

Claims

1. A smart algorithm-driven early warning method for tank foundation settlement, characterized in that, Includes the following steps, S1: Install fiber optic grating sensors at multiple monitoring points in the storage tank to obtain settlement data at each monitoring point, and preprocess the settlement data to obtain the actual settlement value at each monitoring point. S2: Establish the objective function; In the formula, S i S(x) represents the actual settlement value at the i-th monitoring point; i () represents the Gaussian model calculated value of the i-th monitoring point; n represents the number of monitoring points; in, In the formula, x i This represents the monitoring value of the i-th monitoring point; S(x i () represents the Gaussian model calculated value for the i-th monitoring point; μ represents the mean settlement data of all monitoring points; σ represents the standard deviation of settlement data of all monitoring points. S3: Using the butterfly optimization algorithm, iteratively solve the objective function in S2 to obtain the optimized mean and standard deviation, denoted as μ. 优 and σ 优 ; S4: via μ 优 and σ 优 Set low-level and high-level anomaly thresholds. Compare the actual settlement data measured at subsequent monitoring points with the low-level and high-level anomaly thresholds respectively. Based on the comparison results, determine whether the settlement data monitored at the monitoring point is abnormal. Determine whether the settlement data of the monitoring point at the monitoring frequency is abnormal. If it is, the monitoring point has settled; if not, the monitoring point has not settled. S5: Use Unity3D software to display the results of the tank foundation settlement monitoring.

2. The intelligent algorithm-driven early warning method for tank foundation settlement according to claim 1, characterized in that, In step S4, the low-order anomaly threshold δ 低 The formula is: δ 低 =μ 优 -2σ 优 High-level abnormal threshold δ 高 The formula is: δ 高 =μ 优 +2σ 优 ; When the settlement data monitored at a certain monitoring point is greater than or equal to the high-level anomaly threshold δ 高 Or less than or equal to the low-order anomaly threshold δ 低 If the value is not found, it indicates that the settlement data monitored at that monitoring point is abnormal, triggering a preliminary alarm.

3. The intelligent algorithm-driven early warning method for tank foundation settlement according to claim 2, characterized in that, In step S4, if the settlement data of a monitoring point are abnormal three times in a row, then the monitoring point has settled and will continuously trigger an alarm.

4. The intelligent algorithm-driven early warning method for tank foundation settlement according to claim 3, characterized in that, In step S3, the butterfly optimization algorithm includes the following steps: S3.1: Set the population size, maximum number of iterations, perceptual factor, stimulus intensity, and control constant; S3.2: Calculate the "fragrance intensity" of each butterfly, that is, calculate the objective function value of each butterfly; S3.3: Calculate the "odor concentration" of each butterfly; that is, f = cI a Where c, I, and a represent the perceptual factor, stimulus intensity, and control constant, respectively; S3.4: Update the butterfly positions based on the "scent concentration" of each butterfly; In the formula, and Let r represent the positions of the i-th butterfly at the (t+1)-th and t-th iterations, respectively, where r is a random factor; g* represents the current optimal butterfly position. and f is the position of another individual randomly selected from the population. i is the odor concentration of the i-th individual, and p is the threshold parameter; S3.5: Iterate and update until the stopping condition is met; the stopping condition is: reaching the maximum number of iterations of 20 or the change in the objective function value is less than the threshold. S3.6: Output the optimal mean and standard deviation μ 优 and σ 优 .

5. A smart algorithm-driven early warning system for tank foundation settlement, based on the smart algorithm-driven early warning method for tank foundation settlement as described in any one of claims 1-4, characterized in that, Includes the following modules: Data acquisition module: used to collect and display settlement data from various monitoring points; The simulation iteration module is used to iteratively calculate the optimal mean and standard deviation, i.e., μ, using settlement data from various monitoring points and an objective function. 优 and σ 优 ; Data analysis and alarm module: It is used to determine whether the settlement data of the monitoring point is abnormal by setting low-level and high-level abnormal thresholds; when the settlement data is abnormal, it issues an initial alarm; according to the monitoring frequency, it continuously determines whether the monitoring data of the monitoring point is abnormal for three consecutive times, and if so, it continuously issues alarms. The demonstration module uses C4D software to build a 3D model of the target storage tank and Unity3D software to build a monitoring system for the settlement of the tank foundation. Real-time settlement data collected at various monitoring points of the target storage tank is displayed in the monitoring system for the settlement of the tank foundation.

Images