Oil storage tank parameter monitoring system based on Internet of Things

Abstract

The invention discloses an oil storage tank parameter monitoring system based on Internet of Things, which relates to the technical field of parameter monitoring, and comprises the following steps: shooting the outer surface of an oil tank by using a thermal infrared imager to obtain a two-dimensional image, and obtaining two-dimensional temperature data in a heating process according to the two-dimensional image; converting the two-dimensional temperature data into three-dimensional temperature field data by using a Kriging interpolation method; extracting oil-water interface stripes and oil-gas interface stripes of the oil storage tank in the image, and calculating the height of oil in the oil storage tank according to the height difference between the oil-water interface stripes and the oil-gas interface stripes; analyzing the spectrum of the two-dimensional image, judging the stability of the two-dimensional image according to the spectrum analysis, and evaluating whether parameters of the oil storage tank need to be monitored and adjusted or not by using fuzzy logic. The problem that defects of oil products in the oil storage tank are detected through a traditional contact type measurement method is solved, the nondestructive detection technology is provided, and damage to the overall structure of the oil storage tank during detection is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of parameter monitoring, more particularly, the present application relates to an oil storage tank parameter monitoring system based on the Internet of Things. BACKGROUND

[0002] With the acceleration of the industrialization process of the world, the proportion of chemical fuel consumption is rapidly increasing, and the demand for chemical fuel in countries around the world is gradually increasing. Chemical fuels usually include oil, natural gas and coal, etc. Compared with coal, natural gas and other chemical fuels, oil has higher energy density and is more convenient to use. Its position in the energy field is unshakable, and it is also one of the most widely used fossil fuels and the most important strategic resources in the world. In fact, what is extracted from oil fields is not pure crude oil, but also contains a large amount of water, natural gas and other impurities. If you want to simply obtain crude oil, you need to separate water and natural gas and other impurities. The common method is to send the extracted crude oil into an oil storage tank for high-temperature storage. Through heating, crude oil, water and natural gas can be separated from each other, and form oil layer, water layer and gas layer in the oil storage tank.

[0003] Disadvantages of the prior art: When heating to separate crude oil, water and natural gas from each other, different temperatures will affect the physical state of crude oil, water and natural gas. Suitable heating temperature can promote effective separation of three phases and ensure that each component can be optimally separated under corresponding temperature conditions. Therefore, the heating temperature needs to be evaluated. In addition, detecting the liquid level and oil quantity of the oil storage tank can accurately separate crude oil and water, and effectively alleviate the problem of oil leakage caused by full tank, thereby reducing the risk of oil leakage polluting the environment. However, the oil in the oil storage tank is in a high-temperature environment, and the liquid stored in the tank has the characteristics of being flammable, explosive and corrosive. Detecting in such an environment not only seriously affects the service life of the sensor, resulting in increased cost, but also increases the risk of explosion and endangers people's safety. Therefore, the traditional contact type liquid level measurement method is not suitable for liquid level detection of oil in the oil storage tank. Infrared detection technology is a non-destructive detection technology that does not need to be installed inside the oil tank. It only needs to install an infrared thermal imager near the oil tank to take pictures of the outer surface of the oil tank to achieve non-contact monitoring of the interface state of the material in the tank, and will not damage the overall structure of the oil storage tank.

[0004] In view of the above problems, the present application provides a solution. SUMMARY

[0005] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide an oil storage tank parameter monitoring system based on the Internet of Things, which realizes non-contact monitoring of the interface state of the material in the tank to solve the problems raised in the background art.

[0006] To achieve the above object, the present application provides the following technical solutions: The oil storage tank parameter monitoring system based on the Internet of Things comprises a temperature acquisition module, an image analysis module, an error acquisition module and a monitoring evaluation module. The temperature acquisition module is used for simulating the heating process of the oil storage tank, obtaining a two-dimensional image by shooting the outer surface of the oil tank by using an infrared thermal imager, obtaining two-dimensional temperature data in the heating process according to the two-dimensional image, and converting the two-dimensional temperature data into three-dimensional temperature field data by using a Kriging interpolation method. The image analysis module is used for pre-processing the two-dimensional image, obtaining the edges of the tank top and tank bottom by using an edge contour detection algorithm, extracting the oil-water interface stripe and the oil-gas interface stripe of the oil storage tank in the image, and calculating the height of the oil in the oil storage tank according to the height difference between the oil-water interface stripe and the oil-gas interface stripe. The error acquisition module is used for obtaining the storage range of the oil in the oil storage tank according to the oil-water interface stripe and the oil-gas interface stripe of the oil storage tank in the image, obtaining the three-dimensional temperature field data in the oil storage range, obtaining the oil temperature by weightedly averaging the temperature field data, and performing error calculation on the oil temperature and the known wall temperature of the oil storage tank to obtain an oil temperature difference. The monitoring evaluation module is used for analyzing the frequency spectrum of the two-dimensional image, judging the stability of the two-dimensional image according to the change of the low-frequency component of the frequency spectrum, and using fuzzy logic to evaluate whether it is necessary to monitor and adjust the parameters of the oil storage tank according to the height of the oil and the oil temperature in combination with the stability of the two-dimensional image.

[0007] In a preferred embodiment, the specific process of converting the two-dimensional temperature data into three-dimensional temperature field data by using the Kriging interpolation method is as follows: obtaining the two-dimensional temperature data by using the infrared thermal imager, converting the two-dimensional temperature data into coordinate form; calculating the semi-variogram function between the two-dimensional temperature data points, determining the weight of the Kriging interpolation by using the least square method based on the semi-variogram function, so as to minimize the mean square error between the estimated value and the actual value; for three-dimensional interpolation, defining a three-dimensional grid, determining the coordinates of the points to be estimated, performing interpolation calculation by using the Kriging formula to obtain the temperature value of the points to be estimated, and repeating the interpolation process on the entire three-dimensional grid to generate a complete three-dimensional temperature field data.

[0008] In a preferred embodiment, the pre-processing of the two-dimensional image comprises Gaussian filtering processing of the two-dimensional image, and the specific process of the Gaussian filtering processing of the two-dimensional image is as follows: determining the standard deviation of the Gaussian filter according to the noise in the two-dimensional image; creating a Gaussian kernel in the form of a matrix, calculating the value of each element in the matrix according to the Gaussian function; normalizing the Gaussian kernel, so that the sum of all values in the kernel is 1; placing the Gaussian kernel on each pixel of the two-dimensional image, calculating the weighted average value of the current pixel and its adjacent pixels; assigning the calculated new value to the current pixel position, moving to the next pixel, and repeating the process until the entire two-dimensional image is processed, to generate a smoothed two-dimensional image.

[0009] In a preferred embodiment, the height of oil in the oil tank is obtained by using an edge contour detection algorithm to extract the oil-water interface stripe and the oil-gas interface stripe in the image. The two-dimensional image is traversed by row, and the number of pixel points with a gray value of 255 in each row is counted respectively. The two rows with the largest number of pixel points with a gray value of 255 are found, the higher one is recorded as the oil-gas interface stripe, and the lower one is recorded as the oil-water interface stripe. Under the premise of knowing the actual height of the crude oil tank, the height difference between the tank top and the tank bottom in the two-dimensional image is obtained by using an edge contour detection algorithm. The ratio between the pixel height and the actual height is established, the actual height of the crude oil tank represented by each row of pixel values in the image is calculated, and the measurement values of the oil-water interface and the oil-gas interface heights of the crude oil tank are obtained, which are recorded as the height of oil in the oil tank.

[0010] In a preferred embodiment, the oil temperature difference is obtained by defining the oil storage range according to the oil-water interface stripe and the oil-gas interface stripe in the image, obtaining the three-dimensional temperature field data in the oil storage range according to the oil storage range and the three-dimensional temperature field data of the oil tank, setting a weighting factor according to the volume distribution of the oil, and performing weighted averaging on the three-dimensional temperature field data in the oil storage range to obtain the oil temperature. The wall temperature of the oil tank is obtained, the absolute error between the wall temperature and the oil temperature is calculated, and the oil temperature difference is obtained.

[0011] In a preferred embodiment, the stability of the two-dimensional image is determined according to the change of the low-frequency component of the frequency spectrum analysis, and the specific process is as follows: Fourier transform is performed on the two-dimensional image to obtain a frequency spectrum, and the frequency spectrum is divided into low-frequency and high-frequency components by setting a threshold. The low-frequency component is reconstructed by inverse Fourier transform, and the difference between adjacent frames of low-frequency components is calculated. The change of the overall low-frequency component is calculated using the root mean square error, and the smaller the change of the low-frequency component, the more stable the two-dimensional image.

[0012] In a preferred embodiment, the stability of the two-dimensional image is determined according to the change of the low-frequency component of the frequency spectrum analysis, and the specific process is as follows: Fourier transform is performed on the two-dimensional image to obtain a frequency spectrum, and the frequency spectrum is divided into low-frequency and high-frequency components by setting a threshold. The low-frequency component is reconstructed by inverse Fourier transform, and the difference between adjacent frames of low-frequency components is calculated. The change of the overall low-frequency component is calculated using the root mean square error, and the smaller the change of the low-frequency component, the more stable the two-dimensional image.

[0013] In a preferred embodiment, the specific steps of monitoring and adjusting the parameters of the oil storage tank according to the height of the oil and the oil temperature are as follows: according to the height of the oil and the oil temperature, a monitoring evaluation coefficient is set, when the monitoring evaluation coefficient is greater than the preset evaluation threshold, it indicates that there is potential risk in the operating state, the alarm system is automatically triggered to notify the operator to pay attention, and the parameters of the oil storage tank are adjusted; when the monitoring evaluation coefficient is less than or equal to the preset evaluation threshold, it indicates that the operating state is within the safe range, the system is monitored in real time, and the oil temperature and oil height are checked regularly.

[0014] The technical effects and advantages of the oil storage tank parameter monitoring system based on the Internet of Things are as follows: 1. The present application can collect the temperature, liquid level and other parameters of the oil storage tank, ensure the comprehensive understanding of the oil storage state, set the threshold value, automatically trigger the alarm when it exceeds the safe range, reduce the risk of accidents; the non-contact monitoring equipment can be integrated with the Internet of Things system, data analysis platform and other systems to provide comprehensive monitoring and analysis, early warning of potential problems and optimization of operation strategy; it can quickly obtain data and make response, improve the operation efficiency, accurate interface monitoring helps to optimize the distribution and use of materials, and improve the overall production efficiency; the system design has good compatibility, and can be seamlessly integrated with existing equipment and systems.

[0015] 2. The present application avoids direct contact of personnel with the material in the tank through non-contact monitoring, reduces the potential safety risk such as explosion or harmful gas leakage, and does not pollute the material in the tank during the monitoring process, ensuring the accuracy of data and the purity of the material; the non-contact equipment usually has low maintenance frequency, does not need to be cleaned or replaced frequently, reducing the maintenance cost, the non-contact sensor is usually simple to install, does not need complex accessories or disassembly of the tank structure, the non-contact monitoring is not affected by the physical properties of the interface material; reduces the dependence on manual inspection, reduces the labor cost, reduces the operating cost in the long run by reducing the maintenance demand and improving the operation safety. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 The structure diagram of the oil storage tank parameter monitoring system based on the Internet of Things is shown in the figure. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present application will be described in detail below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0018] Embodiment 1, Figure 1The application provides a structure schematic diagram of an oil storage tank parameter monitoring system based on an Internet of Things.

[0019] A temperature collection module is configured to simulate an oil storage tank heating process, capture a two-dimensional image of the outer surface of the oil tank by using an infrared thermal imager, obtain two-dimensional temperature data in the heating process according to the two-dimensional image, and convert the two-dimensional temperature data into three-dimensional temperature field data by using a Kriging interpolation method. First, the specific process of simulating the oil storage tank heating process is as follows: A geometric model of the oil storage tank is created, a CAD software is used to design the geometric shape of the oil storage tank, including the tank body, the bottom and the top, material properties are defined, including thermal conductivity, specific heat capacity and density, to ensure accurate reflection of the physical characteristics of the oil storage tank; A heat source is set according to the heating method (such as electric heating, steam heating, etc.), the corresponding heat source distribution and intensity are set, the boundary conditions of the oil storage tank are set, such as the external environment temperature, the heat exchange coefficient and the wall temperature of the oil storage tank; The heating process is simulated, and the initial temperature and other parameters of the liquid in the oil storage tank are set.

[0020] The two-dimensional image is captured by using the infrared thermal imager, and the specific process is as follows: Important components of the infrared thermal imager include an optical system, a spectral filter, a detector assembly, an information processing assembly and an observer, and the most core part is the detector assembly for realizing photoelectric conversion; the front part of the infrared thermal imaging system is composed of an optical scanning assembly and a mechanical scanning assembly, which converges infrared radiation and scans in the vertical and horizontal directions to obtain two-dimensional observation image data; the rear part of the infrared thermal imaging system is composed of an electronic assembly, which converts the output of the detector assembly and performs some image processing operations.

[0021] The two-dimensional temperature data in the heating process is obtained according to the two-dimensional image, and the specific process is as follows: The captured infrared image is processed to extract specific temperature values, the image can be converted into a two-dimensional temperature data table containing temperature information of each pixel point; a two-dimensional temperature distribution map is generated by the extracted temperature data to intuitively show the temperature change of the surface of the oil tank.

[0022] The specific process of converting the two-dimensional temperature data into three-dimensional temperature field data by using the Kriging interpolation method is as follows: The temperature data is obtained by using the infrared thermal imager, which usually exists in the form of an image, including the temperature value and the coordinates of the pixel; The two-dimensional temperature data is converted into a coordinate form, usually a temperature value T corresponding to (x, y), and each pixel position can be represented as (x_i, y_i, T_i); The semi-variogram between two-dimensional temperature data points is calculated, which describes the spatial autocorrelation and represents the temperature value variation between different positions, and is usually calculated by the following formula: ; wherein, is the semi-variogram between data points, is the number of data pairs within the distance h, is the temperature value of the position , and is the temperature value of the position +h; Based on the semi-variogram, the weights of the Kriging interpolation are determined using the least squares method, the goal of which is to find a weight vector that minimizes the mean square error between the estimated value and the actual value; the weight formula is solved by a system of linear equations as follows: ; wherein, w is the weight vector, is the point to be estimated, is the semi-variogram between data points; For three-dimensional interpolation, a three-dimensional grid is first defined to determine the coordinates of the points to be estimated , The Kriging formula is used for interpolation to calculate the temperature value of the point, and the temperature value calculation formula is as follows: ; wherein, is the temperature value of the point to be estimated, N is the number of points participating in interpolation, is the weight, is the temperature of the point in the two-dimensional image; and The interpolation process is repeated on the entire three-dimensional grid to generate a complete three-dimensional temperature field data.

[0023] The image analysis module is used to preprocess the two-dimensional image, detect the edges of the tank top and bottom using the edge contour detection algorithm, and extract the oil-water interface stripe and the oil-gas interface stripe in the image, and calculate the height of the oil in the oil tank according to the height difference between the oil-water interface stripe and the oil-gas interface stripe. The preprocessing of the two-dimensional image includes Gaussian filtering of the two-dimensional image, which is a linear linear smoothing filtering method, and the original image is convolved using a filtering window, and the filtering window is a weighted average template based on the Gaussian function; there is a connection between different pixels in the two-dimensional image, and the closer the pixels are, the closer the connection is. The Gaussian filter generates a weighted average filtering window based on the Gaussian distribution, and the closer to the center of the window, the greater the weight. The image processed by this method can retain more overall detail information. The specific process of Gaussian filtering of the two-dimensional image is as follows: ​The standard deviation of the Gaussian filter is determined, which affects the smoothing degree of the filter. A small σ value results in less smoothing effect and retains more details. A large σ value results in stronger smoothing effect and may blur details. If there is more noise in the two-dimensional image (e.g., poor image quality), a larger σ value can be selected to more strongly smooth the noise. If the two-dimensional image is relatively clean, a smaller σ value can be selected to retain more details. A Gaussian kernel in matrix form is created, and the values of the matrix in the Gaussian kernel are calculated according to a Gaussian function. The specific calculation formula of the Gaussian function is as follows: In the formula, is the value of the Gaussian function, is the coordinate relative to the center of the kernel, and the coordinate of the center point is , is the standard deviation, which controls the width of the Gaussian function; The Gaussian kernel is normalized by adding all the values in the kernel to make the sum equal to 1, which can keep the brightness of the two-dimensional image unchanged. The Gaussian kernel is placed on each pixel of the two-dimensional image, and the weighted average value of the current pixel and its adjacent pixels is calculated. The new value calculated is assigned to the current pixel position, and the process is repeated for the next pixel until the entire two-dimensional image is processed to generate a smoothed two-dimensional image.

[0024] The height of the oil in the oil storage tank is obtained as follows: The oil-water interface stripe and the oil-gas interface stripe of the oil storage tank in the image are obtained, and the two-dimensional image is traversed by row to count the number of pixel points with a gray value of 255. If there are a large number of pixels with a gray value of 255 in a row of the two-dimensional image, the row is likely to be a horizontal interface stripe. The two rows with the most pixels with a gray value of 255 are found in the row with a large number of pixels with a gray value of 255, and the higher row is recorded as the oil-gas interface stripe, and the lower row is recorded as the oil-water interface stripe. Under the premise that the actual height of the crude oil storage tank is known, the height difference between the top and the bottom of the oil tank in the two-dimensional image is obtained by using an edge contour detection algorithm, a proportion between the pixel height and the actual height is established, and the actual height of the crude oil storage tank represented by each row of pixel values in the image can be calculated, so that the measurement values of the oil-water interface and the oil-gas interface heights of the crude oil storage tank are obtained.

[0025] An error acquisition module is configured to acquire the storage range of the oil in the oil storage tank according to the oil-water interface stripe and the oil-gas interface stripe of the oil storage tank in the image, acquire three-dimensional temperature field data in the oil storage range, and obtain the oil temperature by weighted averaging the temperature field data, and then calculate the error with the known wall temperature of the oil storage tank to obtain the oil temperature difference. According to the oil-water interface stripe and the oil-gas interface stripe in the image, the accurate positions of the oil-water interface and the oil-gas interface are determined, and then the oil storage range is defined; According to the oil storage range and the three-dimensional temperature field data of the oil tank, three-dimensional temperature field data in the oil storage range are obtained; According to the volume distribution of the oil, a weighting factor is set, and the three-dimensional temperature field data in the oil storage range are weighted and averaged to obtain the oil temperature; The wall temperature of the known oil tank is obtained, the absolute error between the wall temperature of the oil tank and the oil temperature is calculated, and the oil temperature difference is obtained. In order to prevent stress or material fatigue caused by excessive temperature difference of the tank body, it is usually necessary to design a reasonable difference between the oil temperature and the tank wall temperature.

[0026] The monitoring and evaluation module is used to analyze the frequency spectrum of the two-dimensional image, evaluate the change of the low-frequency component to judge the stability of the two-dimensional image, and use fuzzy logic to evaluate whether the oil height and oil temperature need to be used to monitor and adjust the parameters of the oil tank.

[0027] The specific process of analyzing the frequency spectrum of the two-dimensional image and evaluating the change of the low-frequency component to judge the stability of the image is as follows: The frequency spectrum is obtained by Fourier transform of the two-dimensional image, and the frequency spectrum is divided into low-frequency and high-frequency components by setting a threshold. The difference between the low-frequency components of adjacent frames is calculated by inverse Fourier transform. The change of the overall low-frequency component is calculated using the root mean square error. The smaller the change of the low-frequency component, the more stable the global structure of the two-dimensional image, and the more stable the two-dimensional image, which indicates that the image sequence is stable in vision.

[0028] The specific calculation formula of using the root mean square error to calculate the change of the overall low-frequency component is as follows: ; In the formula, is the root mean square error, which represents the change of the overall low-frequency component, N is the number of pixels in the two-dimensional image, k and g are the width and height of the image respectively, is the difference between the low-frequency components of adjacent frames; Step C1, define the oil temperature difference and the change of the overall low-frequency component as input variables, and divide them into different fuzzy sets respectively.

[0029] For example, "Low", "Medium", "High" for the oil temperature difference, "Low", "Medium", "High" for the change of the overall low-frequency component.

[0030] Step C2, whether the oil height and oil temperature need to be used to monitor and adjust the parameters of the oil tank is defined as an output variable, which is divided into a fuzzy set, for example, "Yes", "No", for whether the oil height and oil temperature need to be used to monitor and adjust the parameters of the oil tank.

[0031] Step C3, a set of fuzzy rules are made to describe the influence of different input variables on the output variable. The definition of the rules can be based on professional knowledge or obtained through data analysis and experiments. For example: Let the oil temperature difference be marked as B, the change of the overall low-frequency component be marked as , and whether it is necessary to monitor and adjust the parameters of the oil storage tank according to the height of the oil and the oil temperature be marked as P, then the rules can be defined as: Rule 1: IF (B is Low) AND ( is High) THEN (P is No ) Rule 2: IF (B is High) AND ( is Low) THEN (P is Yes ) ...

[0032] Step C4, fuzzy reasoning is performed according to the fuzzy rules to determine the monitoring and adjustment scheme of the parameters of the oil storage tank.

[0033] It should be noted that the division of the fuzzy set can be adjusted according to the actual situation. For example, although three fuzzy sets are taken as an example in this embodiment, the oil temperature difference and the change of the overall low-frequency component, whether it is necessary to monitor and adjust the parameters of the oil storage tank according to the height of the oil and the oil temperature, can be divided into more than three sets to facilitate better accurate identification.

[0034] Further, for the judgment of high, medium and low of the oil temperature difference and the change of the overall low-frequency component, threshold values can be set for judgment according to the actual situation; when the oil temperature difference is higher than 10℃, it is marked as "High"; when the change of the overall low-frequency component is higher than 0.7, it is marked as "High", and the like, which will not be repeated here.

[0035] If it is necessary to monitor and adjust the parameters of the oil storage tank according to the height of the oil and the oil temperature, a monitoring evaluation coefficient is set according to the height of the oil and the oil temperature, and the monitoring evaluation coefficient is obtained by weighted calculation of the height of the oil and the oil temperature. The specific calculation formula is as follows: ; in the formula, is the monitoring evaluation coefficient, is the weight corresponding to the oil temperature, is the weight corresponding to the height of the oil, and all are greater than 0, is the oil temperature, is the reference oil temperature, is the height of the oil, is the reference height of the oil; When the monitoring evaluation coefficient is greater than the preset evaluation threshold, it indicates that the operation state has potential risks, the alarm system is automatically triggered, the operator is notified, and the oil tank parameters are adjusted; when the monitoring evaluation coefficient is less than or equal to the preset evaluation threshold, it indicates that the operation state is within the safe range, the system is monitored in real time, and the oil temperature and oil height are checked regularly.

[0036] The above formulas are dimensionless values calculated, the formulas are obtained by collecting a large amount of data to simulate a formula of the most recent real situation, and the preset parameters in the formula are set by a person skilled in the art according to the actual situation.

[0037] The above embodiments can be realized wholly or partially by software, hardware, firmware or any combination thereof. When realized by software, the above embodiments can be realized wholly or partially in the form of a computer program product.

[0038] Those skilled in the art can realize that the modules and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solutions. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0039] In addition, the functional modules in each embodiment of the present application can be integrated in one processing module, or each module can exist physically, or two or more modules can be integrated in one module.

[0040] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0041] Finally, the above is only the preferred embodiment of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. An oil storage tank parameter monitoring system based on Internet of Things, characterized in that, The method comprises the following steps: A temperature acquisition module is used to simulate the heating process of the oil storage tank, to obtain a two-dimensional image by taking a picture of the outer surface of the oil tank using an infrared thermal imager, to obtain two-dimensional temperature data in the heating process according to the two-dimensional image, and to convert the two-dimensional temperature data into three-dimensional temperature field data using Kriging interpolation method; An image analysis module is used to pre-process the two-dimensional image, to obtain the edges of the tank top and tank bottom using an edge contour detection algorithm, to extract the oil-water interface stripe and the oil-gas interface stripe of the oil storage tank in the image, to calculate the height of the oil in the oil storage tank according to the height difference between the oil-water interface stripe and the oil-gas interface stripe, and to obtain the oil temperature by weighted average of the three-dimensional temperature field data in the oil storage range, and to calculate the error with the known wall temperature of the oil storage tank to obtain the oil temperature difference; A monitoring and evaluation module is used to analyze the frequency spectrum of the two-dimensional image, to judge the stability of the two-dimensional image according to the change of the low-frequency component of the frequency spectrum, to combine the stability of the two-dimensional image with the oil temperature difference to evaluate whether it is necessary to monitor and adjust the parameters of the oil storage tank according to the height of the oil and the oil temperature using fuzzy logic. The specific process of converting the two-dimensional temperature data into three-dimensional temperature field data using Kriging interpolation method is as follows:

2. The IoT based oil storage tank parameter monitoring system as claimed in claim 1 wherein, The two-dimensional temperature data is obtained by using an infrared thermal imager, and the two-dimensional temperature data is converted into coordinate form; The semi-variogram function between the two-dimensional temperature data points is calculated, and the weight of Kriging interpolation is determined based on the semi-variogram function using the least square method to minimize the mean square error between the estimated value and the actual value; For three-dimensional interpolation, a three-dimensional grid is defined, the coordinates of the points to be estimated are determined, the interpolation calculation is performed using the Kriging formula to obtain the temperature value of the points to be estimated, and the interpolation process is repeated on the entire three-dimensional grid to generate a complete three-dimensional temperature field data. The specific process of pre-processing the two-dimensional image includes Gaussian filtering of the two-dimensional image, which is as follows:

3. The IoT based oil storage tank parameter monitoring system as claimed in claim 2, wherein, The standard deviation of the Gaussian filter is determined according to the noise in the two-dimensional image; A Gaussian kernel in matrix form is created, and the values of each element in the matrix are calculated according to the Gaussian function; The Gaussian kernel is normalized so that the sum of all values in the kernel is 1; The Gaussian kernel is placed on each pixel of the two-dimensional image, and the weighted average value of the current pixel and its neighboring pixels is calculated; The new value calculated is assigned to the current pixel position, and the process is repeated until the entire two-dimensional image is processed to generate a smoothed two-dimensional image. The height of the oil in the oil storage tank is obtained as follows:

4. The Internet of Things based oil storage tank parameter monitoring system as claimed in claim 3, wherein, The edge contour detection algorithm is used to extract the oil-water interface stripe and the oil-gas interface stripe of the oil storage tank in the image; The two-dimensional image is traversed by row, and the number of pixel points with a gray value of 255 is counted for each row to find the two rows with the most pixel points with a gray value of 255, with the higher row being the oil-gas interface stripe and the lower row being the oil-water interface stripe; Under the premise of knowing the actual height of the crude oil storage tank, the height difference between the tank top and the tank bottom in the two-dimensional image is obtained using the edge contour detection algorithm; ​ A ratio between the pixel height and the actual height is established to calculate the actual height of the crude oil storage tank represented by each row of pixel values in the image, and the measurement value of the oil-water interface and the oil-gas interface height of the crude oil storage tank is obtained, denoted as the height of the oil in the oil storage tank.

5. The IoT based oil storage tank parameter monitoring system as claimed in claim 4, wherein, The oil temperature difference is obtained as follows: The oil storage range is defined according to the oil-water interface stripe and the oil-gas interface stripe of the oil storage tank in the image; The three-dimensional temperature field data in the oil storage range is obtained according to the oil storage range and the three-dimensional temperature field data of the oil storage tank; The oil temperature is obtained by weighted averaging the three-dimensional temperature field data in the oil storage range according to the volume distribution of the oil and the weighting factor; The wall temperature of the oil storage tank is obtained, and the absolute error between the wall temperature and the oil temperature is calculated to obtain the oil temperature difference.

6. The Internet of Things based oil storage tank parameter monitoring system as claimed in claim 5, wherein, The specific process of judging the stability of the two-dimensional image according to the change of the low-frequency component of the spectrum analysis is as follows: The frequency spectrum is obtained by Fourier transform of the two-dimensional image, and the frequency spectrum is divided into low-frequency and high-frequency components by setting a threshold; The low-frequency component is reconstructed by inverse Fourier transform, and the difference between adjacent frames of low-frequency components is calculated; The change of the overall low-frequency component is calculated using the root mean square error, and the smaller the change of the low-frequency component, the more stable the two-dimensional image.

7. The Internet of Things based oil storage tank parameter monitoring system as claimed in claim 6, wherein, The specific process of using fuzzy logic to evaluate whether the oil height and oil temperature need to be used to monitor and adjust the parameters of the oil storage tank according to the stability of the two-dimensional image and the oil temperature difference is as follows: The oil temperature difference and the change of the overall low-frequency component are defined as input variables, which are divided into different fuzzy sets respectively; Whether the oil height and oil temperature need to be used to monitor and adjust the parameters of the oil storage tank is defined as an output variable, which is divided into a fuzzy set; Fuzzy rules are developed to describe the influence of the oil temperature difference and the change of the overall low-frequency component on whether the oil height and oil temperature need to be used to monitor and adjust the parameters of the oil storage tank; Fuzzy reasoning is performed according to the fuzzy rules to evaluate whether the oil height and oil temperature need to be used to monitor and adjust the parameters of the oil storage tank.

8. The Internet of Things based oil storage tank parameter monitoring system as claimed in claim 7, wherein, The specific steps of monitoring and adjusting the parameters of the oil storage tank according to the oil height and oil temperature are as follows: The monitoring evaluation coefficient is set according to the oil height and oil temperature, and when the monitoring evaluation coefficient is greater than the preset evaluation threshold, it indicates that there is a potential risk in the operating state, and the alarm system is automatically triggered to notify the operator and adjust the parameters of the oil storage tank; When the monitoring evaluation coefficient is less than or equal to the preset evaluation threshold, it indicates that the operating state is within the safe range, and the system monitors in real time and checks the oil temperature and oil height regularly.

9. The Internet of Things based oil storage tank parameter monitoring system as claimed in claim 8, wherein, The semi-variogram formula between the two-dimensional temperature data points is as follows: ; wherein, is the semi-variogram between the data points, is the number of data pairs within the distance h, is the temperature value of the position , is the temperature value of the position +h.

10. The Internet of Things based oil storage tank parameter monitoring system as claimed in claim 9, wherein, The specific calculation formula for calculating the change of the overall low-frequency component using the root mean square error is as follows: ; in the formula, is the root mean square error, indicating the change of the overall low-frequency component, N is the number of pixels in the two-dimensional image, k and g are the width and height of the image respectively, is the difference of the low-frequency components of adjacent frames.

Images