Methods for optimizing the process parameters and / or structure of oil tank descaling

Abstract

This invention discloses a method for optimizing the process parameters and / or structure of oil tank cleaning, comprising: selecting independent variables affecting the cleaning evaluation index; designing combinations of independent variables using methods such as orthogonal experimental tables to determine the experimental scheme for the cleaning process; selecting the dependent variable of the cleaning evaluation index; conducting experiments on various experimental methods; and calculating the evaluation index. This invention defines the economic cost of cleaning, the distribution of residual oil scale in the tank, and the particle size of oil scale in wastewater as dependent variables affecting the cleaning effect. It calculates and records the values ​​of the independent and dependent variables for each experimental group, and uses a neural network algorithm to calculate the functional relationship between the independent and dependent variables based on the current data. Based on this functional relationship, the optimal cleaning process and cleaning structure are obtained. This invention has significant practical implications for obtaining optimal cleaning process parameters and cleaning structures.

Description

Technical Field

[0001] This invention relates to the technical field of jet flushing to remove oil deposits inside oil storage tanks, specifically to a method for optimizing the process parameters and structure of oil storage tank cleaning. Background Technology

[0002] Petroleum contains a significant amount of sulfur, waxes, and other substances, which easily lead to the formation of scale during storage. This scale reduces the effective storage volume of oil storage tanks and can even block inlet and outlet ports, hindering the normal oil receiving and dispatching process. Scale removal is a necessary measure to ensure the safe storage and transportation of oil in storage tanks. However, the costly, difficult, and time-consuming nature of scale removal results in substantial losses in economical storage. Therefore, finding an economical and efficient method for optimizing scale removal process parameters and structure is essential.

[0003] Currently, the main method for simulating the scaling process of oil storage tanks is through on-site testing, recording, analyzing, and optimizing data from the tanks undergoing scaling operations. However, on-site testing cannot systematically test the scaling effect under multiple process parameters, nor does it provide a quantitative description of the scaling effect. In conclusion, there is currently no mature method for optimizing the process parameters and scaling structure of oil storage tanks.

[0004] CN201720406379.5 discloses an indoor simulation device for targeted descaling of barium strontium sulfate scale, which analyzes and studies the scaling rate of barium strontium sulfate. CN201620563145.7 discloses a simulation device for removing inorganic composite scale, which simulates the temperature and medium environment downhole, and analyzes the scale quality before and after simulated dissolution of the scale sample downhole by adding a descaling agent solution, mixing and stirring, reacting and dissolving the scale, and drying, thereby simulating the downhole descaling efficiency.

[0005] The aforementioned simulation devices are designed for specific application scenarios to simulate scale removal. However, the simulated processes, conditions, and application scenarios are limited. They cannot simulate jet scouring for tank cleaning, nor do they offer any optimization measures. Therefore, it is essential to design a system that can simulate the scaling and cleaning process of oil storage tanks, conduct experiments on the cleaning process under various process parameters, and quantitatively characterize the cleaning efficiency for optimizing the cleaning process parameters and structure. Summary of the Invention

[0006] This invention provides a method for optimizing the process parameters and / or structure of oil tank cleaning. This method for optimizing the process parameters and structure of oil tank cleaning is used to solve the problem that there is currently no effective method for optimizing and evaluating the oil tank cleaning process.

[0007] The present invention includes the following aspects: the optimization of the descaling process parameters and the descaling structure for oil storage tanks.

[0008] 1. Using an experimental apparatus that simulates scaling and descaling in oil storage tanks, the scaling and descaling process under different conditions is simulated;

[0009] II. Select independent variables affecting the descaling evaluation indicators: including independent variables of descaling process parameters and / or independent variables of descaling structure; among which, the independent variables of descaling process parameters should include at least one or more of the following: temperature, flow rate, and type of added chemical agents of the descaling medium in the jet descaling process; among which, the independent variables of descaling structure should include at least one or two of the following: nozzle type, jet descaling motion mode; use the orthogonal experimental table method to design the combination of independent variables and determine the experimental scheme of the descaling process;

[0010] III. Select the dependent variable for the scale removal evaluation index: at least one or more of the following should be included: scale removal economic cost, distribution of residual oil scale in the storage tank, and oil scale particle size in wastewater; the scale removal economic cost should have a functional relationship with liquid consumption, energy consumption, and scale removal time; conduct experiments on each experimental method and calculate the evaluation index.

[0011] Fourth, a neural network algorithm is used to input the independent and dependent variables of all experimental schemes into the neural network, train all the data, and calculate the functional relationship between the independent and dependent variables under the current data.

[0012] Fifth, based on the functional relationship between the independent and dependent variables, by comparing the coefficients of the independent variables, we can obtain the degree and pattern of influence of each independent variable on the dependent variable. This allows us to determine the optimal value of the independent variable within the current range.

[0013] In the method of this invention, the economic cost of descaling and its functional relationship with liquid consumption, energy consumption, and descaling time are determined as follows:

[0014] (a) Cleaning time t

[0015] Before the descaling process, a laser camera is used to photograph the inside of the storage tank. Based on the tank dimensions, the volume V of existing oil residue inside the entire tank is calculated. When the descaling process removes oil residue to a set percentage of the total oil residue volume (e.g., 85%–98%, preferably 92%–97%), the descaling is considered optimal, the descaling process ends, and the descaling time is recorded as t.

[0016] To determine the volume of oil residue removed, a laser camera can be used to capture real-time images of the oil residue inside the storage tank during the cleaning process. Simultaneously, the wastewater collected in the wastewater collection tank is filtered and solid-liquid separated. The volume of precipitated oil residue is measured to assist the laser camera's calculations and facilitate the determination of the cleaning time t.

[0017] (ii) Liquid consumption

[0018] The flow rate consumed during the cleaning process due to jet cleaning is defined as the liquid consumption A. The formula for liquid consumption is:

[0019] A = Qt (1)

[0020] Where Q represents the flow rate of the cleaning medium, and t represents the cleaning time.

[0021] (III) Energy Consumption

[0022] The energy consumed by the cleaning medium during the cleaning process, after heating the cleaning medium from ambient temperature T′ to a predetermined temperature T, is defined as the energy consumed by the liquid during the entire cleaning process.

[0023] E = c p A(TT′)=c p Qt(TT′) (2)

[0024] Among them, c p This indicates the specific heat of the cleaning medium.

[0025] In summary, the economic cost of cleaning is determined as follows:

[0026] The cost incurred during the descaling process is defined as the economic cost of descaling, S. The economic cost of descaling is proportional to the fluid consumption (A), energy consumption (E), and descaling time (t).

[0027] S=aA+bE=aQt+bc p Qt(TT′) (3)

[0028] Where 'a' represents the cost of cleaning media per unit volume, and 'b' represents the cost of energy consumption per unit volume (such as electricity).

[0029] In the method of the present invention, the dependent variable of the scale removal evaluation index can also be selected as one or two of the following: the distribution of residual oil scale in the storage tank and the particle size of oil scale in the sewage.

[0030] The distribution of residual oil sludge inside the storage tank is determined according to the following scheme:

[0031] When the cleaning process ends after the designed volume of scale has been removed, a small amount of oil scale still remains in the tank. Analyzing this remaining oil scale using a laser camera can effectively enhance the cleaning capability. The distribution of oil scale is obtained using a threshold segmentation method, and the specific operation steps are as follows:

[0032] A laser camera was used to photograph the outline of the inner wall of the storage tank to obtain an image of the oil stains.

[0033] The maximum entropy threshold segmentation algorithm is used to convert the image from the color mode (RGB mode) to the bitmap mode. According to the actual gray scale distribution of each region, the maximum entropy threshold segmentation algorithm is used to calculate the corresponding optimal threshold for threshold segmentation to obtain a binary image.

[0034] Suppose the size of the selected image or its sub-region I is M×N, the gray level is L, and the gray scale range is {0, 1, 2,..., L-1}. Then the probability that the gray level i (0 < i < L-1) appears in the image is shown as follows:

[0035]

[0036] Among them, h(i) represents the number of pixels with the gray level of i in the image.

[0037] Assume that the image threshold is T (0 < T < L-1). The pixel points with gray levels less than T form the target region (Region A), and the pixel points with gray levels greater than T form the background region (Region B). The probability of each gray level in the two types of regions is shown as follows:

[0038]

[0039] Among them P A (T) and P B (T) respectively represent the cumulative probabilities of Region A and Region B pixels when T is the threshold.

[0040] According to the entropy solution formula The entropy functions corresponding to the target region and the background region under the threshold T are shown as follows:

[0041]

[0042] At this time, the total entropy of the image I is H I (T) = H A (T) + H B (T).

[0043] T * = arg max[H I (T)] (7)

[0044] Among all the segmentation thresholds, the maximum value of the total entropy H I (T) of the image I is the maximum entropy, and the threshold corresponding to this maximum entropy is the optimal threshold T * .

[0045] Based on the above calculation, the distribution of the participating oil scale in the oil storage tank is obtained, and the uniformity of the participating oil scale distribution is calculated based on the uniformity theory. In the n-dimensional space, let be the total exclusive sphere volume of the point set, A νLet L represent the volume of the region, and let L be the uniformity of the point set X.

[0046]

[0047]

[0048] The particle size of oil residue in wastewater is determined according to the following scheme:

[0049] During the descaling process, the oil sludge inside the storage tank is washed away and melted into fine particles and sludge. Large-diameter particles can damage the centrifugal pump, while the finer the sludge after washing, the easier it is to be transported to the wastewater collection tank by the centrifugal pump. Therefore, the particle size of the oil sludge in the wastewater is also an indicator for evaluating the descaling process.

[0050] The wastewater inside the experimental storage tank was photographed using a laser camera. The area S of the oil particles in the wastewater images was calculated using a threshold segmentation method, yielding the oil particle size d in the wastewater:

[0051]

[0052] In the method of this invention, the specific method for using a laser camera to photograph the inside of the storage tank and calculating the existing oil sludge volume V inside the entire storage tank, in conjunction with the tank dimensions, is as follows:

[0053] Given the diameter D and height h of the experimental storage tank, calculate the actual storage volume of the experimental storage tank. for:

[0054]

[0055] A laser camera is used to take real-time images of the interior of the storage tank. Based on the principle of laser ranging, the effective storage space V′ inside the tank can be determined. Therefore, the total volume V of oil residue inside the entire storage tank is:

[0056]

[0057] In the method of this invention, the specific method for calculating the volume of oil sludge removed by using a laser camera to capture real-time images of the inside of the storage tank during the desludge removal process is as follows:

[0058] The laser camera rotates at high speed during the shooting process to obtain real-time images of the inside of the storage tank. As the descaling process proceeds, based on the principle of laser ranging, the laser camera can obtain the effective storage space V′ inside the storage tank in real time. The effective storage space V′ is related to the descaling time t.

[0059] During the descaling process, the effective storage space V′ is observed and calculated. When the volume of descaling removed reaches a set proportion of the total volume of descaling, the descaling process ends, and the descaling time is recorded as t.

[0060] Using laser cameras to measure the effective storage space V′ inside a storage tank is a mature technology. For example, the LJ-X8000 series laser probe developed by KEYENCE and the Intel RealSense LiDAR camera developed by Intel are equipped with professional real-sense dimension measurement software, which can provide accurate and fast volume measurement based on the principle of laser ranging.

[0061] In the method of this invention, the experimental apparatus for simulating scaling and descaling in oil storage tanks is as follows:

[0062] The experimental setup includes:

[0063] Storage tank, including tank bottom, tank body, and optional tank top;

[0064] Optional tank temperature control unit, which regulates the temperature at least one location on the tank bottom and the cylinder;

[0065] The jet flushing unit includes a robotic arm and a jet pipe mounted on the robotic arm. One end of the jet pipe is connected to a first delivery pump, and the other end of the jet pipe is equipped with a jet nozzle. The robotic arm extends into the storage tank from the top of the tank, and the robotic arm drive and control system is located outside the storage tank.

[0066] The visualization unit includes a small laser camera mounted on a robotic arm. The camera transmits the captured images to a display device and a data processing device via a data transmission system.

[0067] The working condition simulation control unit includes a jet material temperature control system, a jet material flow control system, and a robotic arm motion control system;

[0068] The discharge unit includes a suction pipe, a second conveying pump, and a discharge storage tank connected in sequence.

[0069] In the aforementioned device, the tank bottom and cylinder are preferably designed as a sandwich structure, with a temperature regulating medium flowing through the sandwich to simulate different actual operating temperatures of the tank. The sandwich structure can be integrally connected or divided into separate zones to facilitate regional temperature control. Preferably, the tank bottom has an independently temperature-controlled sandwich structure, and the cylinder has 2 to 4 independently temperature-controlled sandwich structures from top to bottom. The tank temperature regulating unit can be a water bath structure (the temperature regulating medium can be water, water with additives, or substances with low freezing points such as oil), and the temperature can be set from -50 to 80℃ to simulate the ambient temperature under actual operating conditions. Multiple temperature regulating units can be set to control the temperature of different sandwich zones. Additionally, a sandwich structure connected to the temperature regulating unit can also be set at the tank top. Alternatively, the tank temperature regulating unit can also employ electric heating, with the tank exterior covered by heating cables or other heating facilities connected to the temperature control system.

[0070] In the aforementioned device, the tank top can employ a floating roof structure. The floating roof is movable and connected to the tank wall via a sealing structure; the floating roof and the tank wall are made of the same material. A through-hole test hole is located at the center of the floating roof, through which a robotic arm extends into the storage tank. A drain port can be installed at the bottom of the tank to drain wastewater and other materials from the storage tank.

[0071] In the aforementioned device, the jet nozzle in the jet flushing unit is detachable, facilitating the replacement of nozzles with different structures to test the impact of nozzle shape on the cleaning effect. The jet tube is made of a corrosion-resistant, high-temperature-resistant, and easily ductile material to avoid the influence of chemicals present in the jet material on the jet tube. Simultaneously, the good ductility of the jet tube allows it to be fixed to the robotic arm and move with it. The robotic arm can be an existing programmable robotic arm with functions such as lifting, translation, rotation, and extension, preferably driven by electric or hydraulic power. The jet flushing unit includes a jet material storage device connected to the inlet of the first delivery pump. The storage device has a temperature regulation function, such as using a heating box, to regulate the temperature of the jet material during use.

[0072] In the aforementioned device, the laser camera in the visualization unit has real-time photo taking, video recording, and laser ranging functions. Images captured by the small laser camera must clearly show the morphology and structure of the oil sludge particles, with distinct edge contours. The data acquired by the small laser camera is transmitted in real-time to a display terminal or data storage and processing device.

[0073] In the aforementioned device, the working condition simulation control unit includes a chemical agent addition unit for the jet material to test the flushing effect of the chemical agent on the oil stains; it also includes a jet material temperature and flow control unit to simulate different cleaning working conditions; the robotic arm control system is computer-controlled, which controls the movement of the robotic arm through computer programming. The movement of the robotic arm includes, but is not limited to, horizontal and vertical movement, circular movement along the central axis, rotation, lifting, and extension, to simulate the cleaning effect of different jet flushing methods.

[0074] In the above-mentioned device, the suction pipe of the discharge unit is also characterized by corrosion resistance and good ductility; the second conveying pump is selected to adapt to the working conditions of small solid oil particles and highly viscous sewage mixtures; the discharge storage tank has a volume that meets the requirements for sewage recycling and the tank body is corrosion resistant.

[0075] The simulation platform device for the scaling and descaling process of the aforementioned storage tank is used in the following specific methods for studying the scaling and descaling process of oil storage tanks:

[0076] 1. Install nozzles of the predetermined type on the jet pipe and suction pipe;

[0077] 2. Use a water bath (oil bath) to heat the water bath medium to a predetermined temperature and then transport it to the experimental storage tank body, tank bottom, or tank top jacket;

[0078] 3. The experimental medium, prepared to the predetermined properties, is placed into the experimental storage tank to simulate the settling process, and the scaling is observed.

[0079] 4. Use a water bath to heat the water medium to a predetermined temperature, and use a heating box to heat the heating medium to a predetermined temperature;

[0080] 5. Open the first valve, and use the first conveying device to flush the oil stains in the heating box through the temperature sensor and flow sensor via the jet pipe. Use a small laser camera to record the changes in the oil stains during the cleaning process.

[0081] 6. Open the second valve and use the second conveying device to collect the wastewater in the experimental storage tank into the wastewater collection box via the temperature sensor and flow sensor;

[0082] 7. Drain the remaining wastewater in the experimental storage tank through the drain hole.

[0083] By employing experimental media with different parameters and adjusting the temperature of the water bath medium, the scaling conditions of different types of oils during the settling process under varying ambient temperatures can be simulated. By installing different types of jet nozzles, setting different movement patterns of the robotic arm, and adjusting parameters such as the flow rate and temperature of the heating medium within the jet tube, the scaling removal process can be simulated. Furthermore, by studying different jet nozzle structures, heating medium flow rates, temperatures, and nozzle movement patterns using camera data, the influence of various factors on the scaling removal effect can be investigated. These studies can provide sufficient support for industrial storage tank scaling solutions, improving the scaling effect and efficiency of industrial storage tanks and reducing pollutant emissions.

[0084] The method of the present invention can achieve the following beneficial effects:

[0085] 1. This invention uses temperature and flow sensors and a laser camera to monitor oil stain data in real time. The nozzle type, temperature and flow rate of the cleaning medium during the jet cleaning process, the type of added chemical agents, and the movement mode of the robotic arm are used as independent variables affecting the cleaning effect. Orthogonal experimental design methods are employed to design and conduct experiments on these independent variables. The economic cost of cleaning, the distribution of remaining oil stains in the tank, and the particle size of oil stains in the wastewater are defined as dependent variables affecting the cleaning effect. The values ​​of the independent and dependent variables for each experimental group are calculated and recorded. A neural network algorithm is used to calculate the functional relationship between the independent and dependent variables based on the current data. Based on this functional relationship, the optimal cleaning process and cleaning structure are obtained. This method has significant practical implications for obtaining optimal cleaning process parameters and cleaning structures.

[0086] 2. This invention is applicable to optimizing the process parameters and cleaning structure of jet cleaning processes.

[0087] 3. This method can be used to conduct experiments on different simulated oils at different water bath temperatures to obtain a wider range of data. Attached Figure Description

[0088] Figure 1 : This is a schematic diagram of the experimental platform device used in the method of the present invention.

[0089] Figure 2 : Diagrams showing nozzle structures of different sizes and forms.

[0090] Figure 3 : Overall flowchart of the method of the present invention.

[0091] 1. Small water bath 1; 2. Small water bath 2; 3. Small water bath 3; 4. Small water bath 4; 5. Test port; 6. Small laser camera; 7. Sludge suction nozzle; 8. Jet nozzle; 9. Drainage port; 10. Data cable; 11. Sludge suction pipe; 12. Jet pipe; 13. Controller cable; 14. Robotic arm; 15. Computer; 16. Controller; 17. Second centrifugal pump; 18. First centrifugal pump; 19. Second temperature sensor; 20. Second flow sensor; 21. First temperature sensor; 22. First flow sensor; 23. First valve; 24. Second valve; 25. Heating chamber; 26. Sewage collection tank; 27. Experimental storage tank; 28. Floating roof. Detailed Implementation

[0092] The invention will be further described below with reference to the accompanying drawings:

[0093] like Figure 1As shown, this experimental setup for studying the scaling and descaling process of oil storage tanks includes a cylindrical experimental tank 27, with a freely movable floating roof 28 connected to the tank wall via a sealed structure. Small water baths 1, 2, 3, and 4, along with the tank wall and bottom of the experimental tank 27, are used to simulate the scaling process under various ambient temperatures. A through-hole test hole 5 is located at the center of the floating roof 28. A robotic arm 14 passes through the test hole 5 into the experimental tank. A jet pipe 12, a suction pipe 11, and a data cable 10 are mounted side-by-side on the robotic arm 14 to facilitate the simulation of descaling and suction processes. Corresponding jet nozzles 8 and suction nozzles 7 are installed at the ends of the jet pipe 12 and suction pipe 11, and a small laser camera 6 is installed at the end of the data cable 10. The other end of the jet pipe 12 is connected to the first centrifugal pump 18, which is connected to the heating box 25 via a conduit. The other end of the suction pipe 11 is connected to the second centrifugal pump 17, which is connected to the sewage collection tank 26 via a conduit. The robotic arm 14 is connected to the controller 16 via the controller cable 13, and the data cable 10 is connected to the computer 15. The experimental storage tank 27 and the floating roof 28 are both made of organic transparent glass for easy observation. The small laser camera 6, the data cable 10, and the computer 15 constitute an oil stain data acquisition system, which can calculate the total oil stain volume before cleaning and record and collect cleaning data in real time during the cleaning process, facilitating comparison of cleaning effects under different working conditions. The robotic arm 14, the controller cable 13, and the controller 16 form an operating robotic arm system, which can be programmed to control different movement modes of the robotic arm.

[0094] like Figure 3 As shown, the method for optimizing the process parameters and cleaning structure of oil storage tanks is as follows:

[0095] 1. Determine the independent variables affecting the scale removal evaluation indicators: The nozzle type, temperature and flow rate of the scale removal medium during the jet scale removal process, the type of added chemical agents, and the movement mode of the robotic arm are used as the independent variables affecting the scale removal evaluation indicators.

[0096] 2. Determine the dependent variables for evaluating the scale removal indicators: The economic cost of scale removal, the distribution of residual oil in the storage tank, and the particle size of oil in the wastewater are used as the dependent variables for evaluating the scale removal indicators.

[0097] 3. Design combinations of independent variables using orthogonal experimental tables, conduct experiments on each combination, record and calculate the values ​​of the dependent variable for each experiment, and calculate the economic cost of descaling, the distribution of residual oil scale in the storage tank, and the particle size of oil scale in the wastewater. The economic cost of descaling is related to the degree of scale buildup, liquid consumption, and energy consumption.

[0098] 4. Using a neural network algorithm, input the independent and dependent variables of all experimental schemes into the neural network, train all the data, and calculate the functional relationship between the independent and dependent variables under the current data.

[0099] 5. Based on the functional relationship between the independent and dependent variables, by comparing the coefficients of the independent variables, we can obtain the degree and pattern of influence of each independent variable on the dependent variable. This allows us to determine the optimal value of the independent variable within the current range.

[0100] Example:

[0101] This invention provides a method for optimizing the process parameters and structure of oil tank cleaning. This method optimizes the process parameters and structure coefficients of cleaning to obtain the optimal cleaning method.

[0102] Table 1 Initial experimental scheme for independent variables affecting the scale removal evaluation index

[0103]

[0104] Table 2 Experimental schemes designed using orthogonal experimental methods

[0105]

[0106]

[0107] Experiments were conducted on the above 25 sets of working conditions. First, the actual storage volume of the experimental storage tank was calculated based on the diameter, height, and other parameters of the tank. Measuring the effective storage space V′ inside a storage tank using a small laser camera is a relatively mature technology. This experiment employed the LJ-X8000 series laser sensor developed by KEYENCE, whose ranging accuracy reaches the millimeter level. The effective storage space V′ inside the tank was calculated using the built-in software. The actual storage volume of the tank was then determined through the experiment. Given the effective storage space V′ inside the tank, calculate the total volume V of existing oil stains inside the entire tank.

[0108] The cleaning volume was 95% V, the cleaning degree was excellent, and the cleaning process was completed. First, the liquid consumption, energy consumption, and cleaning time during the cleaning process were calculated, and the economic cost of cleaning was calculated according to relevant formulas. The distribution of remaining oil residue in the storage tank and the particle size of oil residue in the wastewater were calculated using a threshold segmentation algorithm.

[0109] The experimental data samples were used as the training set for the neural network. The neural network was trained, and the functional relationship between the impact and the evaluation index of scale removal based on the current data was calculated.

[0110] Based on the current functional relationship, the independent variable that has the greatest impact on the dependent variable is calculated, and the optimal descaling process and descaling structure are obtained based on this functional relationship.

[0111] For example, conduct an experiment on the following two sets of data.

[0112]

[0113] Data on independent variables were collected during the experiment, and corresponding dependent variable data were calculated. These data were then input into a neural network. After training, coefficients a1, a2, a3, a4, a5 for the independent variables and coefficients b1, b2, b3 for the dependent variable were obtained. In this experiment, the variable was the flow rate of the cleaning medium, corresponding to coefficient a3. If a3 ≥ 0, the higher the flow rate of the cleaning medium, the larger the value of the dependent variable. The coefficients b1, b2, b3 can be used to analyze the economic cost of cleaning, the distribution of residual oil in the tank, and the changing trend of oil particle size in the wastewater. If a3 ≤ 0, the calculation method remains the same.

[0114] This method can also be used to conduct experiments on different types of simulated oils and at different ambient temperatures, and obtain relevant data for analysis, thus expanding the experimental scope.

Claims

1. A method for optimizing the process parameters and / or structure of descaling in oil storage tanks, comprising the following steps:

1. Using an experimental apparatus that simulates scaling and descaling in oil storage tanks, the scaling and descaling process under different conditions is simulated; II. Select independent variables affecting the descaling evaluation indicators: including independent variables of descaling process parameters and / or independent variables of descaling structure; among which, the independent variables of descaling process parameters should include at least one or more of the following: temperature, flow rate, and type of added chemical agent of the descaling medium in the jet descaling process; among which, the independent variables of descaling structure should include at least one or two of the following: nozzle type and jet descaling motion mode; use the orthogonal experimental table method to design the combination of independent variables and determine the experimental scheme of the descaling process; III. Select the dependent variable for the scale removal evaluation index: at least one or more of the following should be included: scale removal economic cost, distribution of residual oil in the storage tank and oil particle size in wastewater. The economic cost of descaling is calculated and its functional relationship with fluid consumption, energy consumption, and descaling time is as follows: Before the cleaning process, the inside of the storage tank is photographed by a laser camera, and the existing oil stain volume in the whole storage tank is calculated according to the size of the storage tank When the cleaning process cleans the oil stain volume to reach 85% to 98% of the total volume of the oil stain, the cleaning process is ended, and the cleaning time is recorded as ; The flow rate consumed during the cleaning process due to jet cleaning is recorded as the liquid consumption. The formula for liquid consumption is: ； in, Indicates the flow rate of the descaling medium. Indicates the cleaning time; During the descaling process, the descaling medium is heated from ambient temperature. Heat to the predetermined temperature The energy consumed by the entire cleaning process is the energy required to process the liquid. ; ； in, Indicates the specific heat of the descaling medium; In summary, the economic cost of cleaning is determined as follows: The cost incurred during the descaling process is called the descaling economic cost. The economic cost of descaling is directly proportional to the amount of liquid consumed. Energy consumption Cleaning time ; ； in, This indicates the cost per unit volume of descaling media. The cost required to consume a unit of energy; Experiments were conducted on each experimental method and evaluation indicators were calculated. Fourth, a neural network algorithm is used to input the independent and dependent variables of all experimental schemes into the neural network, train all the data, and calculate the functional relationship between the independent and dependent variables under the current data.

5. Based on the functional relationship between the independent and dependent variables, by comparing the coefficients of the independent variables, we can obtain the degree and pattern of influence of each independent variable on the dependent variable, and determine the optimal value of the independent variable within the current range of independent variables; The distribution of residual oil sludge inside the storage tank is as determined: When the cleaning process ends after the designed volume of scale has been removed, a small amount of oil scale still remains in the tank. The remaining oil scale is analyzed using a laser camera. The distribution of oil scale is obtained by threshold segmentation, and the uniformity of the oil scale distribution is calculated based on uniformity theory. The particle size of oil residue in wastewater is determined according to the following scheme: Wastewater inside the experimental storage tank was photographed using a laser camera, and the area of ​​oil particles in the wastewater images was segmented using a thresholding method. Calculations were performed to obtain the particle size of oil residue in the wastewater. for: 。 2. The method according to claim 1, characterized in that: The distribution of oil stains is obtained using a threshold segmentation method. The specific operation steps are as follows: A laser camera was used to photograph the outline of the inner wall of the storage tank to obtain an image of the oil stains. The maximum entropy thresholding algorithm is used to convert the image from color mode to bitmap mode. Based on the actual grayscale distribution of each region, the maximum entropy thresholding algorithm is used to calculate the corresponding optimal threshold for thresholding to obtain a binarized image. Select the image or its sub-region I The size is grayscale level is grayscale range is Then the gray levels in the image , The probability of occurrence is shown in the following formula: ； in, This indicates that the image contains gray levels of 1000. The number of pixels; Assuming the image threshold is T, 0 < T <L-1, gray level less than The pixels constitute the target area A, with a gray level greater than 1. The background region B of the pixels; the probability of each gray level in the two regions is shown in the following formula: ； in , and They represent The cumulative probability of pixels in region A and B when the threshold is reached; According to the formula for calculating entropy The target region and the background region are at the threshold The entropy function corresponding to the following condition is shown in the following equation: ； At this time, the image The total entropy is ; ； Images at all segmentation thresholds I Total entropy The maximum value is the maximum entropy, and the threshold corresponding to this maximum entropy is the optimal threshold. ; Based on the above calculations, the distribution of oil sludge within the oil storage tank is obtained, and the uniformity of this distribution is calculated based on uniformity theory; let... The total exclusive sphere volume of the point set. Represents the volume of a region, called For point set Uniformity; ； 。 3. The method according to claim 1, characterized in that: The interior of the storage tank was photographed using a laser camera, and the volume of existing oil sludge inside the tank was calculated based on the tank's dimensions. The specific method is as follows: Experimental storage tank diameter ,high The actual storage volume of the experimental storage tank is... for: ； A laser camera is used to capture real-time images of the interior of the storage tank. Based on the principle of laser ranging, the laser camera determines the effective storage space within the tank. The total volume of oil sludge inside the entire storage tank is... for: 。 4. The method according to claim 1, characterized in that: To determine the volume of oil residue removed, a laser camera was used to capture real-time images of the oil residue inside the storage tank during the cleaning process. Simultaneously, wastewater collected in the wastewater collection tank was filtered and subjected to solid-liquid separation. The volume of precipitated oil residue was measured to aid the laser camera's calculations and facilitate the determination of the cleaning time. ; The specific method for calculating the volume of oil residue removed using a laser camera by taking real-time images of the tank interior during the cleaning process is as follows: The laser camera rotates at high speed during the shooting process to obtain real-time images of the inside of the storage tank. As the descaling process proceeds, the laser camera can determine the effective storage space inside the tank in real time based on the principle of laser ranging. Effective storage space Related to cleaning time; During the cleaning process, the effective storage space is utilized. Observation and calculation are performed. The cleaning process ends when the volume of cleaned grease reaches a set proportion of the total grease volume. The cleaning time is recorded as follows: .

5. The method according to claim 1, characterized in that: The experimental setup for simulating scaling and descaling in oil storage tanks includes: Storage tank, including tank bottom, tank body, and tank top; The storage tank temperature control unit regulates the temperature at least one location, including the tank bottom and the tank body. The jet flushing unit includes a robotic arm and a jet pipe mounted on the robotic arm. One end of the jet pipe is connected to a first delivery pump, and the other end of the jet pipe is equipped with a jet nozzle. The robotic arm extends into the storage tank from the top of the tank, and the robotic arm drive and control system is located outside the storage tank. The visualization unit includes a small laser camera mounted on a robotic arm. The camera transmits the captured images to a display device and a data processing device via a data transmission system. The working condition simulation control unit includes a jet material temperature control system, a jet material flow control system, and a robotic arm motion control system; The discharge unit includes a suction pipe, a second conveying pump, and a discharge storage tank connected in sequence.

6. The method according to claim 5, characterized in that: The temperature regulation unit of the storage tank is set to -50~80℃.

7. The method according to claim 5, characterized in that: A robotic arm is a programmable robotic arm with lifting, translation, rotation, and extension functions.

8. The method according to claim 1, characterized in that: The cleaning process ends when the volume of grease removed reaches 92% to 97% of the total grease volume. The cleaning time is recorded as follows: .

Images