PVDC production device pipeline cleaning method

Abstract

The invention provides a PVDC production device pipeline cleaning method, and belongs to the technical field of PVDC production device pipeline cleaning. A pickling solution containing nitric acid and hydrofluoric acid is adopted for pretreatment cleaning, a self-made blind plate is used for constructing a temporary cleaning system to achieve circulating cleaning, the cleaning process of a DMF or DMSO solvent at different temperature levels is accurately controlled, and the cleaning efficiency is improved. Impact cleaning is performed by combining a pulse type high-pressure water jet cleaning technology and a pressure fluctuation parameter sequence optimized by a traveling salesman problem algorithm, desalted water is used for high-pressure flushing, thorough flushing is ensured through conductivity monitoring, compressed air provided with a filter is used for purging and drying, and finally, the deerskin-imitated flannelette is used for performing cleanliness inspection. In the whole process, collaborative optimization of the cleaning efficiency function and the damage degree function is achieved through a double-layer game optimization system, and the technical problem that in the PVDC production device pipeline cleaning process, balance optimization is difficult to achieve between the cleaning efficiency and pipeline damage control is solved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of pipeline cleaning of PVDC production device, and in particular, relates to a pipeline cleaning method of PVDC production device. BACKGROUND

[0002] Pipeline cleaning in PVDC (polyvinylidene chloride) production device is a key link to ensure product quality and production safety. The traditional pipeline cleaning technology mainly adopts chemical pickling combined with mechanical cleaning method, which soaks the pipeline by configuring pickling solution containing nitric acid and hydrofluoric acid, and then uses high-pressure water to remove residual substances. This method is widely used in pipeline maintenance of petrochemical, pharmaceutical, food processing and other industries, which can effectively remove polymer residues and various pollutants on the inner wall of the pipeline. However, in the current pipeline cleaning practice of PVDC production device, due to the high chemical stability and strong adhesion of PVDC polymer, the traditional cleaning method often needs to use high-concentration chemical solvents and strong mechanical impact to meet the cleanliness requirements, which causes corrosion damage to the inner wall of the pipeline during cleaning. At the same time, the setting of cleaning parameters lacks systematic optimization guidance, and the operator often adjusts the cleaning intensity according to experience, which may affect the product quality due to insufficient cleaning, or shorten the service life of the equipment due to excessive cleaning. The selection of cleaning process parameters in the traditional technology mainly depends on single target optimization, which cannot achieve the best balance between cleaning efficiency and equipment protection, especially when dealing with complex pipeline systems, there is a lack of precise parameter regulation mechanism for different pipeline specifications and pollution levels, and it is difficult to achieve the coordination and unity of cleaning effect and equipment safety. SUMMARY

[0003] Therefore, the present application provides a pipeline cleaning method of PVDC production device, which can solve the technical problem that the cleaning efficiency and pipeline damage control are difficult to balance and optimize in the pipeline cleaning process of PVDC production device in the prior art.

[0004] The application is implemented in the following manner: the application provides a pipeline cleaning method for a PVDC production device, comprising the following steps: using a SmartPlant3D modeling technology to perform three-dimensional modeling on pipelines in the PVDC production device that have cleanliness requirements; placing prefabricated stainless steel pipelines in a cleaning tank, and soaking and cleaning the pipelines by proportionally configuring pickling solution containing nitric acid and hydrofluoric acid; for pipelines that cannot be cleaned in the cleaning tank, connecting the pipelines at the head and tail by using a self-made blind plate to form a temporary cleaning system for circulating soaking and cleaning; using a multi-stage temperature control solvent cleaning technology, calculating temperature level parameters by using a multi-stage temperature control equation group, using DMF or DMSO solvent to clean the pipelines at the calculated temperature level, using a pulse high-pressure water jet cleaning technology, calculating a pressure fluctuation parameter sequence by using a traveling salesman problem algorithm, and making the pressure pulse change according to the pressure fluctuation parameter sequence; using desalted water with a conductivity of less than or equal to 5.00 μs / cm to perform high-pressure water flushing on the pipelines; using an air compressor equipped with a 10 μm filter to perform compressed air blowing and drying on the pipelines cleaned by flushing; using imitation deer suede cloth to perform pipeline cleanliness inspection; establishing a game model composed of an upper model with the goal of maximizing cleaning efficiency and a lower model with the goal of minimizing damage to the inner wall of the pipeline and internal equipment, and achieving balance optimization between cleaning efficiency and pipeline damage control through double-layer game optimization.

[0005] The step of the SmartPlant3D modeling technology specifically comprises adding a flange mounting position according to the pipeline length and engineering requirements in the pipeline model, counting the flange specifications and quantity, and marking the cleanliness requirement of the pipeline.

[0006] The step of soaking and cleaning the pickling solution specifically comprises making the liquid level of the pickling solution completely submerge the pipelines, soaking and cleaning for 8 to 12 hours, regularly sampling and analyzing the concentration of the pickling solution, and supplementing solutes.

[0007] The step of circulating soaking and cleaning by the temporary cleaning system specifically comprises punching the configured pickling solution into the temporary cleaning system by a temporary pump to perform circulating soaking and cleaning, controlling the cleaning liquid flow rate in the pipeline to be greater than 1.5 m / s, and circulating for 8 to 12 hours.

[0008] The multi-stage temperature control equation group comprises a temperature grading calculation equation and a temperature optimization adjustment equation, the temperature grading calculation equation is used to calculate the cleaning temperature of each stage according to the pipeline diameter, wall thickness, pollution degree, and solvent type, and the temperature optimization adjustment equation is used to adjust the cleaning temperature of the next stage according to the cleaning effect of the previous stage and the pipeline material parameters.

[0009] The DMF refers to dimethyl formamide, chemical formula C3H7NO, which is a high-polarity organic solvent and has good solubility for PVDC polymer. The DMSO refers to dimethyl sulfoxide, chemical formula C2H6OS, which is another high-polarity organic solvent and is often used to dissolve high-molecular polymer.

[0010] The pulse high-pressure water jet cleaning technology refers to generating shock wave effect by controlling the periodic change of water flow pressure, which is used to remove stubborn contaminants and polymer residues on the inner wall of the pipeline.

[0011] The traveling salesman problem algorithm refers to solving the optimal pressure switching path by establishing a pressure node graph and calculating the switching cost between nodes. The pressure demand of different cleaning areas of the pipeline is taken as a node, and the energy consumption cost and time cost of pressure switching are taken as edge weights. The shortest Hamilton circuit is solved by dynamic programming method, and the optimal pressure fluctuation parameter sequence is output.

[0012] The pressure fluctuation parameter sequence includes pressure value sequence, pressure duration sequence and pressure switching interval sequence.

[0013] The high-pressure water flushing step specifically monitors the conductivity of the inlet and outlet water during the flushing process. When the conductivity of the inlet and outlet water is consistent, the flushing process is ended.

[0014] The compressed air purging and drying step specifically slowly introduces compressed air from the purging side until no water flows out from the discharge side, then opens the purging valve to continue purging until the pipeline is completely dry.

[0015] The step of using imitation deerskin cloth to inspect the cleanliness of the pipeline specifically binds the imitation deerskin cloth with a hemp rope and pulls it from one end of the pipeline to the other end repeatedly several times. After the imitation deerskin cloth is found to have no contamination marks, the pipeline opening is wrapped with PE plastic film for protection and storage.

[0016] The method for calculating the pressure fluctuation parameter sequence adopts the traveling salesman problem algorithm. After the calculation is completed, the method further includes the step of optimizing the cleaning parameters using a game model. Specifically, a game model is established with the upper model aiming to maximize the cleaning efficiency and the lower model aiming to minimize the damage to the inner wall of the pipeline and internal equipment. Through double-layer game optimization, the balance optimization between cleaning efficiency and pipeline damage control is achieved.

[0017] The objective function of the upper model refers to maximizing the cleaning efficiency function, which includes the linear term of cleaning compliance rate, the logarithmic function term of cleaning time parameter, the square root function term of solvent concentration parameter, the exponential function term of pressure fluctuation intensity parameter, and the product term of cleaning temperature parameter and cleaning time parameter.

[0018] The upper layer model constraint condition includes that a cleaning compliance rate is greater than or equal to 95%, a cleaning time parameter is greater than or equal to 8 hours and less than or equal to 24 hours, a solvent concentration parameter is greater than or equal to a minimum effective concentration value, a pressure fluctuation intensity parameter is greater than or equal to 5 MPa and less than or equal to 15 MPa, and a cleaning temperature parameter is greater than or equal to 60 DEG C and less than or equal to 80 DEG C.

[0019] The lower layer model target function refers to minimizing a damage degree function, and includes a quadratic function term of a pressure impact intensity parameter, a linear term of a chemical corrosion degree parameter, a square root function term of a temperature stress coefficient parameter, a logarithmic function term of a cleaning frequency parameter and a coupling term of the pressure fluctuation intensity parameter and the cleaning time parameter.

[0020] The lower layer model constraint condition includes that the pressure impact intensity parameter is less than or equal to a pipeline pressure limit value, the chemical corrosion degree parameter is less than or equal to a material corrosion threshold value, the temperature stress coefficient parameter is less than or equal to a thermal stress safety coefficient, the cleaning frequency parameter is greater than or equal to 1 time and less than or equal to 3 times, and a damage accumulation coefficient is less than or equal to a safety damage limit value.

[0021] The double-layer game optimization system is established by the upper layer model for maximizing cleaning efficiency and the lower layer model for minimizing pipeline damage, accurate modeling and cleaning path planning of the pipeline system are realized by combining with the SmartPlant3D three-dimensional modeling technology, and the synergetic effect of the multi-stage temperature control solvent cleaning technology and the pulse high-pressure water jet cleaning technology is adopted, so that the problem that the cleaning efficiency and equipment protection are difficult to balance in the traditional cleaning method is solved. The double-layer game model optimizes key parameters such as cleaning compliance rate, cleaning time, solvent concentration, pressure fluctuation intensity and cleaning temperature through the cleaning efficiency function of the upper layer model, and controls damage factors such as pressure impact intensity, chemical corrosion degree and temperature stress coefficient through the damage degree function of the lower layer model, the coordination and optimization between parameters are realized through the coupling term, the local optimal solution caused by single target optimization is avoided, and the equipment damage is minimized while the predetermined cleaning effect is achieved. The optimal pressure fluctuation parameter sequence is calculated by the traveling salesman problem algorithm, so that the pulse pressure change follows a scientific optimization path, and the hierarchical regulation of the solvent cleaning temperature is realized by combining with the multi-stage temperature control equation set. This systematic parameter optimization method fundamentally changes the traditional empirical operation mode, realizes the accurate and intelligent control of the cleaning process, and solves the technical problem that the cleaning efficiency and pipeline damage control are difficult to balance and optimize in the pipeline cleaning process of the PVDC production device in the background technology. BRIEF DESCRIPTION OF DRAWINGS

[0022] Figure 1 The flowchart of the method of the present application.

[0023] Figure 2 Modeling diagram for process piping in Example 2.

[0024] Figure 3 Flange addition diagram for cleanliness-required pipeline in Example 2.

[0025] Figure 4 Physical diagram for pipe circulation (filling) soaking cleaning in Example 2.

[0026] Figure 5 Process diagram for double-target optimization of cleaning efficiency and damage degree in Example 2.

[0027] Figure 6 Interface diagram for intelligent cleaning monitoring system in Example 2. DETAILED DESCRIPTION

[0028] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application.

[0029] As shown in Figure 1 is a flowchart of a PVDC production device pipe cleaning method provided by the present application, and the method comprises the following steps:

[0030] S01, using SmartPlant3D modeling technology to perform three-dimensional modeling on the pipeline with cleanliness requirements in the PVDC production device, adding flange installation positions in the pipeline model according to the pipeline length and engineering requirements, counting the flange specifications and quantity and marking the cleanliness-required pipeline;

[0031] S02, placing the prefabricated stainless steel pipeline in the cleaning tank, configuring the pickling solution containing nitric acid and hydrofluoric acid in proportion, so that the pickling solution completely covers the pipeline, and soaking and cleaning for 8 to 12 hours, regularly sampling and analyzing the pickling solution concentration and supplementing the solute;

[0032] S03, for long pipelines or large-diameter pipelines that cannot be cleaned in the cleaning tank, using self-made blind plates to connect the pipelines end to end to form a temporary cleaning system, and using a temporary pump to pour the configured pickling solution into the temporary cleaning system for circulating soaking and cleaning, controlling the cleaning liquid flow rate in the pipeline to be above 1.5 m / s, and the circulating time to be 8 to 12 hours;

[0033] S04, using multi-stage temperature control solvent cleaning technology, calculating temperature level parameters through a multi-stage temperature control equation group, using DMF or DMSO solvent to clean the pipeline at the calculated temperature level, using pulse high-pressure water jet cleaning technology, calculating a pressure fluctuation parameter sequence through a traveling salesman problem algorithm, and making the pressure pulse change according to the pressure fluctuation parameter sequence;

[0034] S05, high pressure water flushing is performed on the pipeline using desalted water with conductivity less than or equal to 5.00 μs / cm, and the conductivity of the inlet water and outlet water is continuously monitored during the flushing process, and the flushing process is ended when the conductivity of the inlet water and outlet water is consistent;

[0035] S06, compressed air is used to dry the flushed clean pipeline, and the compressed air is slowly introduced from the blowing side until no water flows out from the discharge side, and then the blowing valve is opened to continue blowing until the pipeline is completely dry;

[0036] S07, the pipeline cleanliness is tested using imitation deer suede cloth, the imitation deer suede cloth is bound with hemp rope and pulled from one end of the pipeline to the other end repeatedly several times, and after the imitation deer suede cloth is found to have no any pollution marks, the pipeline opening is wrapped with PE plastic film for protection and storage.

[0037] The SmartPlant3D modeling technology is a three-dimensional piping design software technology used to create accurate three-dimensional models of industrial piping, enabling visual design and management of piping systems. The self-made blind plate refers to a circular or square metal plate customized according to the specifications of the pipe to close the pipe port, used to form a closed cleaning circulation system during the cleaning process. The multi-stage temperature control solvent cleaning technology is a technology for cleaning pipes by controlling the temperature of the solvent in stages, which can gradually dissolve polymer contaminants at different temperature levels. The multi-stage temperature control equation set includes a temperature staging calculation equation and a temperature optimization adjustment equation. The temperature staging calculation equation is used to calculate the cleaning temperature at each stage according to the pipe diameter, wall thickness, pollution level, and solvent type. The input includes pipe diameter parameters, pipe wall thickness parameters, pollution level coefficients, and solvent heat capacity coefficients, and the output is the numerical value of the cleaning temperature at each stage. The temperature optimization adjustment equation is used to adjust the cleaning temperature at the next stage according to the cleaning effect of the previous stage and the pipe material parameters. The input includes the previous stage cleaning efficiency value, the pipe material thermal expansion coefficient, the solvent evaporation rate, and the cleaning time parameters, and the output is the optimized cleaning temperature value at the next stage. DMF is dimethylformamide, with the chemical formula C3H7NO, which is a highly polar organic solvent with good solubility for PVDC polymers. DMSO is dimethyl sulfoxide, with the chemical formula C2H6OS, which is another highly polar organic solvent commonly used to dissolve high molecular polymers. The pulse high-pressure water jet cleaning technology is used to remove stubborn contaminants and polymer residues on the inner wall of the pipe by controlling the periodic change of water pressure to produce shock wave effect. The traveling salesman problem algorithm is a classic combinatorial optimization algorithm that solves the optimal pressure switching path by establishing a pressure node graph and calculating the switching cost between nodes. The pressure requirements of different cleaning areas of the pipe are taken as nodes, and the energy consumption cost and time cost of pressure switching are taken as edge weights. The shortest Hamilton circuit is solved by dynamic programming method, and the optimal pressure fluctuation parameter sequence is output. The pressure fluctuation parameter sequence includes pressure value sequence, pressure duration sequence and pressure switching interval sequence. The imitation deer suede cloth is a soft fiber material with a similar deer skin texture, with a smooth surface and no shedding, suitable for testing the cleanliness of the inner wall of the pipe.

[0038] Further, in the S04 step, a step of optimizing the parameters of the multi-stage temperature-controlled solvent cleaning and the pulsed high-pressure water jet cleaning by using a game model is further included, specifically: a double-layer game optimization model is established, the upper-layer model takes maximizing the cleaning efficiency as the target, and by optimizing the cleaning temperature parameter, the solvent concentration parameter, the pressure fluctuation intensity parameter and the cleaning time parameter, the comprehensive function of the cleaning compliance rate, the cleaning time, the solvent concentration and the pressure fluctuation intensity is maximized; the lower-layer model takes minimizing the pipeline damage as the target, and by controlling the pressure impact intensity parameter, the chemical corrosion degree parameter, the temperature stress coefficient parameter and the cleaning frequency parameter, the damage degree function is minimized. When the multi-stage temperature-controlled solvent cleaning is performed, the game model solves the optimal cleaning temperature of each stage through a temperature grading calculation equation and a temperature optimization adjustment equation; when the pulsed high-pressure water jet cleaning is performed, the game model further optimizes the pulse pressure range, the pulse frequency and the pulse duration by combining the pressure fluctuation parameter sequence solved by the traveling salesman problem algorithm, so as to ensure that the cleaning efficiency requirement is met while the damage to the pipeline is minimized. The game model is described in detail below.

[0039] The game model includes an upper-layer model taking maximizing the cleaning efficiency as the target and a lower-layer model taking minimizing the damage to the inner wall of the pipeline and the internal equipment as the target.

[0040] The upper-layer model target function: maximizing the cleaning efficiency function, including a linear term of the cleaning compliance rate, a logarithmic function term of the cleaning time parameter, a square root function term of the solvent concentration parameter, an exponential function term of the pressure fluctuation intensity parameter, and a product term of the cleaning temperature parameter and the cleaning time parameter.

[0041] The upper-layer model constraint condition: the cleaning compliance rate is greater than or equal to 95%; the cleaning time parameter is greater than or equal to 8 hours and less than or equal to 24 hours; the solvent concentration parameter is greater than or equal to the minimum effective concentration value; the pressure fluctuation intensity parameter is greater than or equal to 5 MPa and less than or equal to 15 MPa; the cleaning temperature parameter is greater than or equal to 60 DEG C and less than or equal to 80 DEG C.

[0042] The lower-layer model target function: minimizing the damage degree function, including a quadratic function term of the pressure impact intensity parameter, a linear term of the chemical corrosion degree parameter, a square root function term of the temperature stress coefficient parameter, a logarithmic function term of the cleaning frequency parameter, and a coupling term of the pressure fluctuation intensity parameter and the cleaning time parameter.

[0043] The lower-layer model constraint condition: the pressure impact intensity parameter is less than or equal to the pipeline pressure limit value; the chemical corrosion degree parameter is less than or equal to the material corrosion threshold value; the temperature stress coefficient parameter is less than or equal to the thermal stress safety coefficient; the cleaning frequency parameter is greater than or equal to 1 time and less than or equal to 3 times; the damage cumulative coefficient is less than or equal to the safe damage limit value.

[0044] The cleaning efficiency function is used to evaluate the comprehensive efficiency of the PVDC pipeline cleaning process, the inputs include the cleaning compliance rate, cleaning time parameter, solvent concentration parameter, pressure fluctuation intensity parameter and cleaning temperature parameter, and the output is a cleaning efficiency evaluation value, which is used for target optimization of the upper model and adjustment of cleaning process parameters. The damage degree function is used to calculate the damage degree caused by the cleaning process to the inner wall of the pipeline and the internal equipment, the inputs include the pressure impact intensity parameter, chemical corrosion degree parameter, temperature stress coefficient parameter, cleaning frequency parameter and pressure fluctuation intensity parameter, and the output is a damage degree evaluation value, which is used for damage control and equipment safety evaluation of the lower model. The coupling term represents the common influence of the interaction between the pressure fluctuation intensity parameter and the cleaning time parameter on the cleaning efficiency and the damage degree, and is used to coordinate the balance relationship between the two targets.

[0045] The specific implementation of the above steps is described in detail below. The specific implementation of step S01 is to first start the SmartPlant3D software platform, establish a three-dimensional coordinate system for the pipeline system of the PVDC production device, and accurately define the geometry of the pipeline using parameterized modeling technology. This modeling technology is based on the B-spline surface reconstruction principle and realizes accurate description of the pipeline profile through control points and weight factors. According to the pipeline diameter, wall thickness, material and other parameters, the pipeline entity model is established, and the thermal expansion coefficient and mechanical strength parameters of the pipeline are also considered. In the pipeline model, the flange mounting position is added according to the engineering design specification, the reference value of the flange spacing is 3-5 meters, the flange specification is selected according to the national standard, and the commonly used specifications include DN15, DN25, DN50, DN80, DN100, etc. The number of flanges of various specifications is counted and marked in the model, and the pipelines with cleanliness requirements are coded or labeled with color, and the cleanliness level reference value is level 5 or level 6 in the ISO14644 standard. The purpose of this step is to provide accurate pipeline geometry information and cleaning path planning basis for subsequent cleaning process.

[0046] The specific implementation of step S02 is to place the prefabricated stainless steel pipes according to length and diameter in a cleaning tank made of polytetrafluoroethylene or stainless steel 316L. The pickling solution is prepared according to the volume ratio, with the concentration of nitric acid controlled at 10-15% and the concentration of hydrofluoric acid controlled at 2-5%. During the preparation process, deionized water is added first and then the acidic solvent is slowly added to avoid violent reaction. The liquid level of the pickling solution is 10-20 cm higher than the highest point of the pipe to ensure complete immersion of the pipe. The soaking and cleaning time is controlled at 8-12 hours, which is determined based on the kinetics of polymer contamination dissolution, considering the breaking rate and diffusion mass transfer rate of the acidic solvent on the PVDC polymer chain. The concentration of nitric acid and hydrofluoric acid in the pickling solution is sampled and analyzed every 2 hours, and when the concentration drops by more than 20% of the initial value, the corresponding solute is supplemented in time. The purpose of this step is to remove the polymer residues and organic contaminants on the inner wall of the pipe through chemical dissolution.

[0047] The specific implementation of step S03 is to connect the pipes end to end to form a closed temporary cleaning system using blind plates customized and processed according to the pipe specifications for pipes longer than 6 meters or with a diameter greater than 300 mm. The blind plate is made of stainless steel 316L, with a thickness determined according to the pipe pressure rating, generally 8-15 mm, and a sealing ring made of polytetrafluoroethylene. The prepared pickling solution is pumped into the temporary cleaning system through a centrifugal pump, with the pump flow calculated according to the pipe inner diameter to ensure that the cleaning liquid flow rate in the pipe reaches 1.5 meters per second or more. This flow rate parameter is determined based on the Reynolds number theory and mass transfer boundary layer theory, which can effectively destroy the contamination adhesion layer on the inner wall of the pipe. The circulating cleaning time is controlled at 8-12 hours, during which the pH value and contamination concentration of the cleaning liquid are monitored every hour. The purpose of this step is to implement forced circulation cleaning for large pipes, and to improve the cleaning effect through the synergistic effect of fluid shear force and chemical dissolution.

[0048] The specific implementation of step S04 is to first use a multi-stage temperature-controlled solvent cleaning technology. The temperature of each stage is calculated by a temperature grading calculation equation according to the pipe diameter, wall thickness, pollution degree coefficient and solvent heat capacity coefficient. Generally, it is divided into 3-5 temperature levels. The primary temperature is controlled at 40-50°C, the intermediate temperature is controlled at 60-70°C, and the high-level temperature is controlled at 75-80°C. The temperature optimization adjustment equation adjusts the cleaning temperature of the next stage according to the previous stage cleaning efficiency value, pipe material thermal expansion coefficient, solvent evaporation rate and cleaning time parameters. The adjustment range is generally ±5°C. DMF or DMSO solvent is used to clean the pipe at the calculated temperature level. The boiling point of DMF is 153°C, and the boiling point of DMSO is 189°C. At the set temperature, PVDC polymer can be effectively dissolved without solvent decomposition. Then, a pulse high-pressure water jet cleaning technology is used. A pressure node diagram is established by a traveling salesman problem algorithm. The pressure requirements of different cleaning areas of the pipe are taken as nodes. The energy consumption cost and time cost of pressure switching are taken as edge weights. The shortest Hamilton circuit is solved by dynamic programming method. The optimal pressure fluctuation parameter sequence including pressure value sequence, pressure duration sequence and pressure switching interval sequence is output. The pulse pressure range is controlled at 5-15 MPa, the pulse frequency is controlled at 5-20 Hz, and the pulse duration is controlled at 0.1-0.5 s. The purpose of this step is to completely remove the stubborn contaminants on the inner wall of the pipe through the dual action of solvent dissolution and high-pressure pulse impact.

[0049] The specific implementation of step S05 is to use desalted water with a conductivity of less than or equal to 5.00 μS / cm to flush the pipe with high-pressure water. The flushing pressure is controlled at 2-5 MPa, and the flow rate is controlled at 3-8 m / s. During the flushing process, online conductivity detectors are installed at the inlet and outlet of the pipe. The detection accuracy is required to be ±0.1 μS / cm, and the sampling frequency is set to 1 Hz. The conductivity values of the inlet and outlet water are monitored in real time. When the difference between the two is less than 0.5 μS / cm and remains stable for more than 5 minutes, it is determined that the flushing is complete. This determination criterion is based on the principle of ion mass transfer balance. When the inlet and outlet water conductivity tends to be consistent, it indicates that the residual ionic contaminants in the pipe have been thoroughly washed. The flushing time is generally controlled at 30-120 minutes, and the specific time is determined according to the pipe length and the degree of pollution. The purpose of this step is to remove the residual acidic solvent and soluble ionic contaminants in the previous steps to ensure that the inner wall of the pipe meets the electrochemical cleaning standard.

[0050] The specific implementation of step S06 is to use a screw air compressor equipped with a 10 μm filter to dry the flushed and cleaned pipe with compressed air. The output pressure of the air compressor is controlled at 0.6-0.8 MPa, and the flow rate is calculated according to 5-10 times the pipe volume per hour. The filter uses a high-efficiency particulate air filter with a filtration efficiency of more than 99.97% to ensure that the particulate matter concentration in the compressed air is less than 105 The purging process is divided into two stages, the first stage slowly passes compressed air from the purging side, the inlet pressure gradually increases to 50% of the working pressure, the duration is the length of the pipeline divided by 100 minutes, the purpose of this stage is to avoid the impact of water hammer effect on the pipeline damage. The second stage opens the purge valve to full open state, the compressed air pressure is increased to the working pressure, and the purging is continued until the discharge side is continuously 5 minutes without water droplets. The criterion for judging the completion of the purging is that the dew point temperature of the outlet gas is lower than -40℃. The purpose of this step is to thoroughly remove the residual moisture in the pipeline to prevent corrosion or microbial contamination during subsequent use.

[0051] The specific implementation of step S07 is to select a simulated deerskin cloth with a fiber diameter less than 10 microns and a surface smoothness less than 0.2 microns for pipeline cleanliness inspection. The size of the simulated deerskin cloth is determined according to 80% of the inner diameter of the pipeline, and the length is controlled within 50-100 cm. The simulated deerskin cloth is bundled using a clean hemp rope, and the spiral winding method is used for bundling, with a spacing of 5-10 cm to ensure that the cloth is in full contact with the inner wall of the pipeline during dragging. The cloth is dragged from one end of the pipeline to the other end at a speed of 0.5-1.0 m / s, and the surface of the cloth is checked for contamination marks after 3-5 times of dragging. The test standard is that the cloth surface has no visible particles, no discoloration area, no fiber shedding, and the detection limit of pollutants is less than 10 particles per square centimeter. After the test is passed, immediately wrap the openings of the pipeline with a polyethylene plastic film with a thickness of 0.1-0.2 mm, and use heat sealing or adhesive tape sealing method for protection and storage. The sealed pipeline should be stored in a clean environment, and the storage time should not exceed 6 months. The purpose of this step is to verify whether the cleanliness of the inner wall of the pipeline meets the process requirements, and to maintain the cleanliness by effective storage until installation and use.

[0052] The key technical ideas of the application are analyzed as follows. The first key technical idea is the three-dimensional modeling and path optimization technology based on SmartPlant3D. This technology establishes an accurate geometric model of the pipeline through parameterized modeling and B-spline surface reconstruction principles. Compared with the traditional two-dimensional drawing design method, it can provide three-dimensional pipeline system information, effectively avoiding blind spots and missed areas in the cleaning process. The three-dimensional modeling technology combined with the flange position optimization algorithm can reasonably plan the cleaning path according to the geometric characteristics of the pipeline and the cleaning process requirements, significantly improving the systematicness and integrity of the cleaning. The second key technical idea is the synergistic cleaning technology of multi-stage temperature-controlled solvent cleaning and pulsed high-pressure jet. This technology combines chemical dissolution and physical impact two cleaning mechanisms, realizes accurate control of solvent cleaning temperature through temperature grading calculation equation and temperature optimization adjustment equation, avoids the problems of pipeline thermal stress damage and solvent decomposition caused by too high temperature. The pulsed high-pressure water jet technology uses the traveling salesman problem algorithm to optimize the pressure fluctuation sequence, which can produce stronger shock wave effect compared with the traditional constant pressure cleaning method, effectively breaking and peeling off stubborn polymer contamination layer. The third key technical idea is the double-layer optimization model based on game theory. This model coordinates the dual objectives of maximizing cleaning efficiency and minimizing damage degree in the lower layer, solving the problem of balancing efficiency and safety in traditional cleaning process. The game model can dynamically adjust the cleaning parameters according to the pipeline material, pollution degree and process requirements, realizing the intelligentization and self-adaptation of the cleaning process. The fourth key technical idea is the endpoint determination and quality control technology based on conductivity monitoring. This technology monitors the change of water conductivity difference in and out in real time, establishes an objective and accurate cleaning endpoint determination standard, avoiding the problems of insufficient or excessive cleaning caused by experience-based judgment in traditional methods. The synergistic effect of these technical ideas is reflected in the following aspects. The three-dimensional modeling technology provides accurate geometric parameters and spatial information for multi-stage temperature control and pulse cleaning, making the calculation and optimization of cleaning process parameters more accurate. The game theory optimization model considers the parameter configuration of each cleaning link, realizing the whole process optimization from modeling design to cleaning execution. The conductivity monitoring technology provides real-time feedback information for the game model, so that the cleaning parameters can be dynamically adjusted according to the actual cleaning effect. Compared with the traditional experience-based cleaning method, this synergistic technology system has the advantages of strong systematization, high precision, good safety and strong adaptability, and can realize the standardized, intelligent and fine management of PVDC production device pipeline cleaning.

[0053] Specifically, the principle of the present application is that the present application can solve the technical problem that the cleaning efficiency and the pipeline damage control are difficult to balance and optimize during the pipeline cleaning process of the PVDC production device, and the core principle is to establish a mathematical balance relationship between the maximum cleaning efficiency and the minimum damage through a double-layer game optimization model, and to realize the collaborative optimization control of multiple cleaning parameters. The cleaning efficiency function of the upper model can accurately describe the combined influence of the cleaning compliance rate, cleaning time, solvent concentration, pressure fluctuation intensity and cleaning temperature on the cleaning effect through the combination of linear terms, logarithmic function terms, square root function terms, exponential function terms and product terms, wherein the linear terms reflect the direct contribution of the cleaning compliance rate, the logarithmic function terms reflect the diminishing marginal effect law of the cleaning time, the square root function terms describe the nonlinear action mechanism of the solvent concentration, the exponential function terms capture the sensitivity influence of the pressure fluctuation intensity, and the product terms reflect the synergistic effect of temperature and time. The damage degree function of the lower model is constructed by a quadratic function term, a linear term, a square root function term, a logarithmic function term and a coupling term, which accurately quantifies the comprehensive effect of pressure impact intensity, chemical corrosion degree, temperature stress coefficient and cleaning frequency on pipeline damage, wherein the quadratic function term emphasizes the nonlinear damage mechanism of pressure impact, the linear term embodies the cumulative effect of chemical corrosion, the square root function term describes the progressive influence of temperature stress, the logarithmic function term reflects the marginal damage characteristics of cleaning frequency, and the coupling term captures the interaction of pressure fluctuation and cleaning time. The double-layer model ensures that the cleaning parameters change within the safe and feasible range through their respective constraints, the constraints of the upper model ensure that the cleaning effect meets the process requirements, and the constraints of the lower model ensure that the equipment is safe and not damaged, and the global optimal solution is achieved through iterative solution of the two models, avoiding the local optimal dilemma that the traditional single-objective optimization may fall into. The multi-stage temperature control solvent cleaning technology realizes the dynamic adjustment of the cleaning temperature of each stage according to the pipeline diameter, wall thickness, pollution degree and solvent type through the joint action of the temperature grading calculation equation and the temperature optimization adjustment equation, so that the PVDC polymer is gradually dissolved by the solvent without causing thermal stress damage to the pipeline due to excessive temperature. The pulse high-pressure water jet cleaning technology combines the pressure fluctuation parameter sequence optimized by the traveling salesman problem algorithm, and through the shock wave effect generated by the periodic pressure change, it can effectively remove stubborn contaminants while avoiding excessive impact on the inner wall of the pipeline due to sustained high pressure. The optimized sequence of pressure fluctuations ensures efficient use of cleaning energy and minimization of equipment damage.

[0054] A specific embodiment 1 of the present application is provided below, and the specific implementation of each step in embodiment 1 is described in detail as follows.

[0055] The specific implementation of step S01 is to establish a pipeline three-dimensional model using SmartPlant3D modeling technology, and to perform parameterized modeling based on the B-spline surface reconstruction principle. The pipeline surface equation is specifically represented as follows:

[0056]

[0057] where P(u, v) is the three-dimensional coordinates of any point on the pipe surface; u, v are parameter coordinates, taking values in the range [0, 1]; N i，p (u) and N j，q (v) are B-spline basis functions; p, q are the degrees of B-spline; w ij is the weight factor; P ij is the control point coordinates; m, n are the maximum indices of the control point grid. The flange installation position optimization formula is:

[0058]

[0059] where L flange is the flange spacing; k stress is the stress concentration coefficient, taking values in the range 1.2-1.8; D pipe is the pipe outer diameter; t wall is the pipe wall thickness; σ allow is the allowable stress; L thermal is the thermal expansion compensation length. Among them, D pipe and t wall are obtained through the pipe design drawings, σ allow is obtained according to the pipe material table, and L thermal is obtained by calculating the thermal expansion coefficient. Among them, the B-spline surface reconstruction formula is based on the numerical geometry theory, and the precise mathematical description of the pipe geometry is realized through the double-parameter surface interpolation. The basis functions N i，p (u) and N j，q (v) in the formula provide local support characteristics, and the weight factor w ij realizes the flexible control of the non-uniform rational B-spline, so that any point on the pipe surface can be accurately calculated through the parametric equation. Compared with the traditional polygonal grid modeling method, this mathematical model can realize the continuous and smooth surface representation, avoid the influence of discretization error on the subsequent cleaning path planning, ensure the geometric precision of the cleaning process parameter calculation, and thus improve the reliability and consistency of the entire cleaning system. Among them, the flange installation position optimization formula is based on the stress analysis theory in materials mechanics and the thermal expansion theory. The first term of the formula reflects the stress distribution law of the pipe under internal pressure, according to the thin-walled cylinder theory, the hoop stress is proportional to the pipe diameter and inversely proportional to the wall thickness, and the square root relationship reflects the nonlinear characteristics of the stress concentration effect. The second term L thermal considers the thermal expansion compensation demand caused by temperature change. Compared with the traditional empirical flange arrangement method, this mathematical model realizes accurate calculation based on the mechanics principle, effectively avoids the problems of pipe deformation caused by too large flange spacing and stress concentration caused by too small flange spacing, and ensures the structural stability and safety of the pipe system in the cleaning process.

[0060] The specific implementation of step S02 is to prepare pickling solution and perform immersion cleaning. The formula for calculating the concentration of the pickling solution is:

[0061]

[0062] Where C acid is the concentration of pickling solution; V acid，pure is the volume of pure acid; V total is the total solution volume. The pickling reaction kinetic equation is:

[0063]

[0064] Where C polymer is the concentration of polymer pollutants; t react is the reaction time; k react is the reaction rate constant; α acid is the reaction order of the pickling solution, ranging from 0.5 to 1.5; β polymer is the polymer reaction order, ranging from 0.8 to 1.2; E a is the reaction activation energy; R is the gas constant, which is 8.314 joules per mole per Kelvin; T react is the reaction temperature. react The results were obtained through experimental determination, and the experimental steps included: Step 1: preparing standard acid washing solutions of different concentrations; Step 2: measuring the amount of polymer dissolved at different time points under constant temperature conditions; Step 3: obtaining the reaction rate constant by linear regression fitting. a The range is 45 to 65 kilojoules per mole. The pickling reaction kinetic equation is based on the chemical reaction kinetics theory and the Arrhenius equation. This equation describes the change of polymer contaminant concentration over time, where the power function term Reflects the effect of reactant concentration on reaction rate, the exponential term This model describes the exponential effect of temperature on the reaction rate constant. Through the derivation of the Arrhenius equation, it is shown that as temperature increases, molecular kinetic energy increases, the effective collision frequency increases, and the activation energy barrier for the reaction is more easily overcome, resulting in an exponential increase in the reaction rate. Compared to traditional qualitative cleaning methods, this kinetic model can quantitatively predict the relationship between cleaning effect and time, temperature, and concentration, enabling precise optimization of cleaning parameters, avoiding over- or under-cleaning, and significantly improving cleaning efficiency and consistency.

[0065] The specific implementation method of step S03 is to perform cyclic cleaning on the large pipeline, and the flow rate in the pipeline is calculated as follows:

[0066]

[0067] wherein v flow is the flow rate in the pipe; Q is the volumetric flow rate; D inner is the pipe diameter. The Reynolds number is calculated by the formula:

[0068]

[0069] wherein Re is the Reynolds number; p is the density of the cleaning liquid; and m is the dynamic viscosity of the cleaning liquid. The Sherwood number is calculated by the formula:

[0070] Sh = 0.023Re 0.8 Sc 0.33 ;

[0071] wherein Sh is the Sherwood number; Sc is the Schmidt number, The mass transfer coefficient is calculated by the formula:

[0072]

[0073] wherein k m is the mass transfer coefficient; D diff is the diffusion coefficient. Q is measured by a flow meter, p and m are determined by experiments, D diff is obtained by a diffusion experiment, the diffusion experiment comprising the following steps: Step 1: preparing a tracer solution with a standard concentration; Step 2: measuring the diffusion rate of the tracer in the cleaning liquid under constant temperature conditions; and Step 3: calculating the diffusion coefficient by Fick's law.

[0074] The specific implementation of Step S04 is to use multi-stage temperature control solvent cleaning and pulse high-pressure jet cleaning, and the temperature grading calculation equation is:

[0075]

[0076] wherein T i is the cleaning temperature of the i-th stage; T base is the base temperature, which is 40°C; AT j is the temperature increment of the j-th stage; g pollution is the pollution degree coefficient, which is 0.1-1.0; C solvent is the heat capacity coefficient of the solvent; f(D pipe , t wall , g pollution , C solvent ) is a temperature correction function, and the specific form is wherein D ref is the reference pipe diameter, which is 100 mm, and t ref is the reference wall thickness, which is 5 mm. The temperature optimization adjustment equation is:

[0077]

[0078] wherein T i+1 is the temperature of the next stage of optimization; k adjust is the adjustment coefficient, taking a value of 0.1-0.5; η target is the target cleaning efficiency; η i is the cleaning efficiency of the previous stage; α material is the thermal expansion coefficient of the pipe material; is the partial derivative of the control temperature with respect to the processing time, indicating the rate of temperature change. The objective function in the traveling salesman problem algorithm is:

[0079]

[0080] wherein c ij is the switching cost from pressure node i to node j; x ij is the decision variable, taking a value of 0 or 1; n is the total number of pressure nodes. The constraint condition is:

[0081] and The pressure fluctuation parameter sequence matrix is represented as:

[0082]

[0083] wherein P k is the kth pressure value; t duration，k is the kth pressure duration; τ k is the kth pressure switching interval. Wherein γ pollution is obtained by measuring the thickness of the contaminants on the inner wall of the pipe, η i is obtained by calculating the concentration of contaminants before and after cleaning, c ij is obtained by energy consumption testing and time testing, T control and t process are obtained by real-time monitoring of the temperature control system.

[0084] The specific implementation of step S05 is high-pressure water flushing based on conductivity monitoring, and the conductivity difference determination formula is:

[0085] Δσ=|σ in -σ out |;

[0086] wherein Δσ is the conductivity difference between the inlet and outlet water; σ in is the inlet water conductivity; σ out is the outlet water conductivity. The cleaning endpoint determination condition is:

[0087] Δσ≤σ threshold and

[0088] wherein σthreshold is a conductivity difference threshold value, taking a value of 0.5 microsiemens per centimeter; ε stability is a stability determination parameter, taking a value of 0.1 microsiemens per centimeter per minute. Wherein, σ in and σ out is obtained by real-time measurement through an online conductivity meter.

[0089] The specific implementation of steps S06-S07 is the same as the foregoing, and will not be described in detail here.

[0090] Wherein, the upper model objective function of the game model is:

[0091]

[0092] In the formula, F efficiency is a cleaning efficiency function; a1, a2, a3, a4, a5 are weight coefficients; η pass is a cleaning compliance rate; t clean is a cleaning time parameter; P pulse is a pressure fluctuation intensity parameter; P0 is a reference pressure, taking a value of 1 megapascal; T clean is a cleaning temperature parameter. The upper model constraint condition matrix is:

[0093] A upper x upper ≤b upper ;

[0094] In the formula, A upper is a constraint coefficient matrix; x upper is a decision variable vector; b upper is a constraint right end vector. The lower model objective function is:

[0095]

[0096] In the formula, F damage is a damage degree function; b1, b2, b3, b4, b5 are weight coefficients; P impact is a pressure impact intensity parameter; γ corrosion is a chemical corrosion degree parameter; α stress is a temperature stress coefficient parameter; n frequency is a cleaning frequency parameter. The lower model constraint condition is:

[0097]

[0098] In the formula, P limit is a pipeline pressure bearing limit value; γ threshold is a material corrosion threshold value; α safety is a thermal stress safety coefficient; δ i is a damage accumulation of the i-th cleaning; δlimit is the safety damage limit; m is the total number of cleaning times. Among them, the weight coefficients a1 to a5 and b1 to b5 are determined by multi-objective optimization experiments, and the experimental steps include: step 1: design an orthogonal experiment scheme; step 2: perform cleaning experiments under different parameter combinations; step 3: measure the cleaning efficiency and damage degree; step 4: determine the weight coefficients by using the analytic hierarchy process. limit obtained by pipeline pressure test, γ threshold obtained by material corrosion test, α safety is in the range of 1.5-2.5.

[0099] It should be noted that the Reynolds number formula and the Sherwood number formula Sh = 0.023Re 0.8 Sc 0.33 Based on the dimensionless analysis theory and mass transfer theory in fluid mechanics. The Reynolds number represents the ratio of fluid inertial force to viscous force. When the Reynolds number exceeds the critical value, the flow changes from laminar flow to turbulent flow, and the mixing effect in the turbulent flow state is significantly enhanced. The Re 0.8 term in the Sherwood number formula reflects the promotion of flow intensity to mass transfer, and the Sc 0.33 term reflects the influence of material diffusion properties. The mass transfer coefficient formula converts the dimensionless mass transfer parameter into the mass transfer coefficient in engineering applications. Compared with the traditional constant flow rate cleaning method, this mass transfer model can accurately calculate the optimal flow rate according to the pipeline geometric parameters and fluid properties, ensure the full renewal of the boundary layer on the inner wall of the pipeline, improve the mass transfer removal efficiency of pollutants, and achieve more thorough cleaning effect.

[0100] Temperature staging calculation equation Based on the theory of heat transfer and the principle of polymer dissolution kinetics. This equation realizes the step-by-step increase of temperature through the cumulative form, and the term in the correction function f reflects the influence of pipe diameter on temperature distribution, (t wall / t ref ) 0.5 The term γ pollution reflects the influence of pollution degree on the required temperature. Temperature optimization adjustment equation Based on feedback control theory, the efficiency difference (η target -η i ) is used for deviation correction, the thermal expansion coefficient α material term considers the material thermal stress constraint, and the partial derivative term reflects the dynamic characteristics of temperature change. Compared with the traditional single temperature cleaning method, this staged temperature control model can dynamically adjust the temperature according to the pipeline characteristics and cleaning process, ensure the cleaning effect while avoiding thermal stress damage, and achieve the optimal balance between cleaning efficiency and safety.

[0101] Traveling salesman problem algorithm objective function Based on the theory of combinatorial optimization, the optimal pressure switching path is solved by constructing a pressure node graph and calculating the switching cost between nodes. The cleaning pressure demand of different areas of the pipeline is abstracted as a node in graph theory, and the energy consumption and time cost of pressure switching are taken as edge weights. The global optimization of pressure sequence is realized by solving the shortest Hamiltonian circuit. Compared with the traditional fixed pressure or simple pressure change mode, this optimization algorithm can design the optimal pressure fluctuation sequence according to the geometric characteristics and pollution distribution characteristics of the pipeline, maximize the pulse impact effect and minimize the energy cost, significantly improving the efficiency and economy of high-pressure jet cleaning.

[0102] Conductivity difference determination formula Δσ = |σ in -σ out | and the cleaning endpoint determination condition is based on ion mass transfer balance theory and steady-state detection principle. When the residual ions in the pipeline are fully washed, the conductivity of the inlet and outlet water tends to be consistent, and the difference tends to zero, and the stability condition ensures that the system reaches a true dynamic balance. Compared with the traditional time control or subjective judgment method, the conductivity monitoring model provides an objective and accurate cleaning endpoint determination standard, avoids the problems of insufficient cleaning or over-cleaning, and realizes quantitative control and consistency guarantee of cleaning quality.

[0103] Game model upper objective function Based on the theory of multi-objective optimization, each item reflects the contribution of different factors to cleaning efficiency. Linear term a1η pass directly reflects the importance of the compliance rate, logarithmic term a2ln(t clean reflects the marginal decreasing effect of cleaning time, square root term reflects the nonlinear effect of solvent concentration, exponential term emphasizes the exponential enhancement effect of pressure fluctuation, and product term a5T clean ·t clean considers the synergistic effect of temperature and time. The quadratic term in the lower objective function reflects the square relationship of pressure impact damage, and the coupling term b5P pulse ·t clean reflects the cumulative damage effect of pressure and time. Compared with the traditional single objective optimization method, the game model realizes the coordination of the dual objectives of maximizing cleaning efficiency and minimizing damage, and through the game solution of the upper and lower models, it ensures that the best cleaning effect is obtained while ensuring the safety of the equipment, and realizes the intelligent and fine management of the cleaning process.

[0104] For better understanding and implementation of the present application, the following provides an embodiment 2 of a specific application scenario of the present application: a technical team receives a task of thoroughly cleaning the pipeline system of a polyvinylidene chloride (PVDC) production device, which includes a main pipeline of stainless steel 316L with a diameter of 150 millimeters, a wall thickness of 8 millimeters, and a total length of 328 meters, and multiple branch pipelines with diameters of 50-100 millimeters. There is a serious PVDC polymer contamination on the inner wall of the pipeline, with a contamination layer thickness of about 2.3 millimeters, and the cleanliness requirement reaches level 5 of the ISO14644 standard.

[0105] The technical team first organizes relevant personnel to conduct a preliminary review of drawings, and according to the design drawings, pipeline engineering regulations, engineering quantity, and technical difficulties, a detailed pipeline cleaning construction scheme of the PVDC production device is prepared. After receiving the process pipeline plan layout drawing, the project chief engineer leads to start the SmartPlant3D software for three-dimensional modeling. According to the B-spline surface reconstruction formula, the value range of the parameter coordinates u and v is set to [0, 1], the B-spline order p and q are both set to 3, and an accurate pipeline model containing 1247 control points is established, as shown in the modeling model schematic diagram of the process pipeline in embodiment 2. Figure 2 Through three core modules of pipeline division, support design, and pipe group assembly, the pipeline is further decomposed and transformed into different parts by using the SmartPlant3D software, to realize the accurate splitting of the pipeline with cleanliness requirements. Through flange installation position optimization calculation, the stress concentration coefficient k stress is 1.5, the allowable stress σ allow is 200 megapascals, the thermal expansion compensation length L thermal is 0.8 meters, and the calculated flange spacing L flange is 4.2 meters. According to the pipeline length and engineering requirements, a total of 78 flange connection points are set on the main pipeline, and 156 flange connection points are set on the branch pipeline, as shown in the flange addition schematic diagram of the pipeline with cleanliness requirements in embodiment 2. Figure 3 The number of each specification flange is counted to report the material demand plan.

[0106] In the construction preparation stage, the technical team prepares the cleaning tank according to the process pipeline specifications and lengths obtained from the SmartPlant3D model, ensures that the prefabricated pipeline passes the non-destructive testing, and that the desalted water for flushing meets the requirements. The safety and technical briefing is conducted for the on-site construction personnel, the construction personnel qualification certificates are checked, and the tools, construction machinery, and labor protection articles are prepared. The technical personnel and operators participating in the construction carefully read the design drawings and technical specification documents, and clearly understand the design intent and requirements.

[0107] For the branch pipelines with smaller diameters, the technical team adopts the immersion cleaning method. When the pickling solution is proportionally configured in the cleaning tank, the volume V acid,pure of nitric acid is 120 liters, and the total solution volume Vtotal 1000 liters, the calculated pickling solution concentration C acid 12%. At the same time, a mixed pickling solution with a hydrofluoric acid concentration of 3.5% is configured. According to the pickling reaction kinetics equation, the reaction rate constant k react 0.025 per hour, the pickling solution reaction order a acid 1.2, the polymer reaction order β polymer 1.0, the reaction activation energy E a 52 kilojoules per mole, the reaction temperature T react 45°C. The prefabricated pipeline that has been welded and inspected is completely immersed in the cleaning tank for 10 hours of immersion cleaning, and samples are taken for analysis at regular intervals during the cleaning process to check the pickling solution concentration and supplement the solute.

[0108] For the main pipeline system, the technical team adopts a circulating cleaning method. For the length of the pipeline that cannot be cleaned in the cleaning tank, a circulating immersion cleaning method is adopted. The pipeline to be cleaned is connected end to end using a self-made blind plate to form a temporary cleaning system, such as Figure 4 as shown in the pipeline circulating (full) immersion cleaning physical diagram in Example 2. The prepared pickling solution is pumped from the configuration tank to the temporary cleaning system using a centrifugal pump, and the volumetric flow rate Q is set to 285 liters per minute. According to the flow rate calculation formula, the pipeline flow rate v flow 2.7 meters per second. The cleaning liquid density p is 1050 kilograms per cubic meter, and the dynamic viscosity μ is 0.0012 Pa·s. The calculated Reynolds number Re is 318750, which belongs to a fully turbulent flow state. In order to control the corrosion rate, a hanging piece and a monitoring pipe section of the same material are placed in the system. The diffusion coefficient D diff 2.3×10 -9 square meters per second, the Schmidt number Sc is 0.51, and the Sherwood number Sh is 1247. The mass transfer coefficient k m 1.9×10 -5 meters per second. The circulating cleaning lasts for 11 hours, during which the pickling solution concentration is analyzed at regular intervals.

[0109] Next, multi-stage temperature-controlled solvent cleaning is carried out. The basic temperature T base 40°C, the pollution degree coefficient γ pollution 0.85 according to the measured pollution layer thickness, the solvent heat capacity coefficient C solvent 4.2 kilojoules per kilogram per kelvin. The reference pipe diameter D ref 100 millimeters, the reference wall thickness t ref5mm, the first stage cleaning temperature T1 is 47℃, the second stage is 58℃, the third stage is 69℃, and the fourth stage is 76℃. The DMF solvent is used to clean at each temperature level for 2.5 hours. The technical team monitors the cleaning efficiency η i of the previous stage in real time target , the target cleaning efficiency η adjust is set to 95%, the adjustment coefficient k material is 0.3, the pipe material thermal expansion coefficient α -5 is 1.6×10 threshold per Kelvin, and the temperature optimization adjustment equation is used to dynamically adjust the temperature of each stage.

[0110] The pulse high-pressure jet cleaning uses the traveling salesman problem algorithm to optimize the pressure sequence. The pipeline is divided into 15 cleaning areas as pressure nodes, and the switching cost matrix between nodes is determined by energy consumption test and time test. The total number of pressure nodes n is 15, the shortest Hamilton circuit is solved by dynamic programming, and the optimal pressure fluctuation parameter sequence is obtained. The pulse pressure range is 8-14 MPa, the pulse frequency is 12 Hz, the pressure duration is 0.3 seconds, and the switching interval is 0.8 seconds. The whole pulse cleaning process lasts for 4 hours, and the key cleaning parameter configuration is shown in Table 1.

[0111] Table 1 Multi-stage temperature control pulse cleaning parameter configuration table

[0112] Cleaning level Temperature (°C) Pressure (MPa) Duration (hours) Solvent type First level 47 8.5 2.5 DMF Second level 58 10.2 2.5 DMF Third level 69 12.8 2.5 DMSO Fourth level 76 14.0 2.5 DMSO

[0113] In the high-pressure water flushing stage, the technical team uses temporary pipelines to lead the on-site desalted water to the pipeline cleaning site, and uses clean water tanks for storage. The cleaned pipeline is transported to the high-pressure water flushing site and arranged, and the high-pressure cleaning machine is used to flush the pipeline within 3 hours. The desalted water with a conductivity of 3.2 μS / cm is used, the conductivity meter is used before flushing to ensure that the desalted water conductivity is less than or equal to 5.00 μS / cm, the flushing pressure is 3.5 MPa, and the flow rate is 5.8 m / s. The online conductivity detector with an accuracy of ±0.1 μS / cm is installed at the inlet and outlet of the pipeline, and the sampling frequency is 1 times per second. The conductivity difference threshold σ threshold is set to 0.5 μS / cm, and the stability determination parameter ε stability is 0.1 μS / cm per minute. After 75 minutes of flushing, the pipeline is visually clean, the inlet and outlet water conductivity difference Δσ is stable below 0.3 μS / cm, and the high-pressure water flushing is completed when the inlet and outlet water conductivity is consistent.

[0114] The compressed air purge adopts a screw air compressor equipped with a 10-micron high-efficiency filter to ensure the cleanliness of the compressed air, with an output pressure of 0.7 MPa and a flow rate of 8 times the pipe volume per hour. Slowly introduce compressed air from the purge side, slowly pressurize to 50% of the working pressure in the first stage, and continue for 25 minutes until no water flows out of the discharge side. In the second stage, open the compressed air purge valve to full open, fully pressurize until the outlet gas dew point temperature drops to-45℃, and continue to purge until the pipeline is completely dry, with a total purge time of 180 minutes.

[0115] The cleanliness test uses a simulated suede cloth to wipe and test, using a simulated suede cloth with a fiber diameter of 8 microns and a surface smoothness of 0.15 microns, with a size of 80% of the pipe diameter and a length of 80 cm. The simulated suede cloth is bundled with a cleaning hemp rope with a spacing of 8 cm, and is pulled from one end of the pipe to the other end at a speed of 0.8 meters per second, with 4 rounds of back and forth pulling. After several repeated pulls, the suede cloth is tested, and the test results show that the surface of the suede cloth is free of visible particles, no discoloration area, and no any contamination trace, with a detection limit of pollutants below 5 particles per square centimeter, meeting the expected cleanliness requirements.

[0116] After the cleaning acceptance, immediately proceed to the protective storage. The pipe opening is wrapped with a 0.15 mm thick polyethylene plastic film heat-sealed to prevent secondary pollution during installation, and is properly stored in a clean environment. The waste liquid generated after pickling and passivation is strongly acidic, containing a large amount of iron ions, oil stains and other impurities, and the flushing water is weakly acidic, containing trace amounts of welding slag, iron ions and other impurities. After collection, it is stored in a temporary collection tank on site, and is treated with acid-base neutralization to make the pH value reach 6-9, and is transported to a treatment plant with relevant qualifications for treatment.

[0117] In the optimization process of the game model, the upper model weight coefficient is set as a1=0.35, a2=0.15, a3=0.20, a4=0.18, a5=0.12, and the reference pressure P0 is 1 MPa. The lower model weight coefficient is set as b1=0.28, b2=0.22, b3=0.20, b4=0.15, b5=0.15. The pipe pressure limit value P limit is 25 MPa, the material corrosion threshold γ threshold is 0.05 mm per year, the thermal stress safety factor α safety is 2.0, the cleaning frequency n frequency is set to 2 times, the safety damage limit δ limit is 0.15. Through the game model solution, the final cleaning compliance rate η pass reaches 97.8%, the cleaning time t clean is 28.5 hours, the pressure fluctuation intensity P pulse is 12.6 MPa, and the cleaning temperature T clean68℃, the game model optimization results are shown in Table 2.

[0118] Table 2 Game model double-layer optimization results table

[0119]

[0120] During the whole cleaning process, the technical team uses the integrated monitoring system to track the changes of various parameters in real time, and the system interface displays temperature curves, pressure waveforms, conductivity changes and cleaning efficiency evaluation and other key information. The cleaning efficiency function F efficiency From the initial value 1.23 to the final value 2.87, the damage degree function F damage From 2.15 to 0.86, the efficiency and safety are realized.

[0121] Compared with the traditional single temperature constant pressure cleaning method, the present application realizes significant progress through a plurality of technical innovations. First, the three-dimensional modeling technology based on SmartPlant3D breaks through the limitations of traditional two-dimensional design, realizes the mathematical accurate description and visual management of the geometric shape of the pipeline through the B-spline surface reconstruction principle and the three core modules of pipeline division, support design and pipe group assembly, provides a reliable geometric basis for the subsequent cleaning process, and avoids the cleaning blind spots and inconsistency problems caused by experience operation. Secondly, the multi-stage temperature control solvent cleaning technology realizes the step-by-step control of temperature according to the polymer dissolution kinetics principle, which can avoid thermal stress damage while ensuring the cleaning effect compared with the traditional single temperature method, and realizes the balance of cleaning efficiency and equipment safety. Thirdly, the pulse high pressure jet combined with the traveling salesman problem algorithm optimizes the pressure fluctuation sequence, which produces stronger shock wave effect compared with the traditional constant pressure cleaning, and significantly improves the removal ability of stubborn contaminants. Fourthly, the double-layer optimization model based on game theory solves the contradiction between the maximum cleaning efficiency and the minimum damage through mathematical modeling, which realizes the intelligent solution of the global optimal solution compared with the traditional experience parameter adjustment. Fifthly, the conductivity real-time monitoring technology establishes an objective and accurate cleaning endpoint judgment standard, which eliminates the influence of human factors compared with the traditional time control or subjective judgment method, and ensures the consistency and controllability of the cleaning quality. Sixthly, the standardized construction process and waste liquid treatment measures ensure the standardized execution of the cleaning operation and the environmental protection requirements. The synergistic effect of these technical innovations makes the whole cleaning system have the characteristics of systematization, accuracy, intelligence and reliability, and provides a scientific and complete technical solution for the PVDC production device pipeline cleaning.

[0122] It should be noted that the variables involved in the present application are explained in detail as shown in the following table 3.

[0123] Table 3 Variable explanation table

[0124]

[0125]

[0126] The above descriptions are only specific embodiments of the present application, and the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application.

Claims

1. A method for cleaning a pipe of a PVDC production apparatus, characterized by, The application relates to a pipeline cleaning method for a PVDC production device. The pipeline with cleanliness requirements in the PVDC production device is modeled by using a SmartPlant3D modeling technology; the prefabricated stainless steel pipeline is placed in a cleaning tank, and pickling liquid containing nitric acid and hydrofluoric acid is proportionally configured to soak and clean the pipeline; for the pipeline that cannot be cleaned in the cleaning tank, a self-made blind plate is used to connect the pipeline head to tail to form a temporary cleaning system for circulating soaking and cleaning; a multi-stage temperature control solvent cleaning technology is used, a temperature level parameter is calculated through a multi-stage temperature control equation group, DMF or DMSO solvent is used to clean the pipeline at the calculated temperature level, a pulse high-pressure water jet cleaning technology is used, a pressure fluctuation parameter sequence is calculated, and the pressure is pulsed according to the pressure fluctuation parameter sequence; desalted water with an electric conductivity of less than or equal to 5.00 mu s / cm is used to clean the pipeline by high-pressure water flushing; a compressed air blowing and drying process is carried out on the cleaned pipeline by using an air compressor equipped with a 10-micron filter; and a pipeline cleanliness test is carried out by using a simulated deerskin cloth.

2. The PVDC production apparatus pipe cleaning method according to claim 1, characterized by, The step of the SmartPlant3D modeling technology adopts a traveling salesman problem algorithm to calculate the pressure fluctuation parameter sequence, and after the calculation is completed, the step of optimizing the cleaning parameters by using a game model is further included.

3. The PVDC production apparatus pipe cleaning method according to claim 2, characterized by, The step of soaking and cleaning the pickling liquid is specifically that the pickling liquid liquid surface completely covers the pipeline, and the soaking and cleaning lasts for 8 to 12 hours, the pickling liquid concentration is sampled and analyzed at regular time intervals, and the solute is supplemented.

4. The PVDC production apparatus pipe cleaning method according to claim 3, characterized by, The step of circulating soaking and cleaning the temporary cleaning system is specifically that the prepared pickling liquid is pumped into the temporary cleaning system by a temporary pump to perform the circulating soaking and cleaning, the cleaning liquid flow velocity in the pipeline is controlled to be higher than 1.5 m / s, and the circulating time lasts for 8 to 12 hours.

5. The PVDC production apparatus pipe cleaning method according to claim 4, characterized by, The multi-stage temperature control equation group includes a temperature grading calculation equation and a temperature optimization adjustment equation, the temperature grading calculation equation is used to calculate the cleaning temperature of each stage according to the pipeline diameter, wall thickness, pollution degree and solvent type, and the temperature optimization adjustment equation is used to adjust the cleaning temperature of the next stage according to the cleaning effect of the previous stage and the pipeline material parameter.

6. The PVDC production apparatus pipe cleaning method according to claim 5, characterized by, The DMF refers to dimethylformamide, and the chemical formula is C3H7NO; the DMSO refers to dimethyl sulfoxide, and the chemical formula is C2H6OS.

7. The PVDC production apparatus pipe cleaning method according to claim 6, characterized by, The pulse high-pressure water jet cleaning technology refers to generating an impact wave effect through the periodic change of the water flow pressure, and is used for removing stubborn pollutants and polymer residues on the inner wall of the pipeline.

8. The PVDC production apparatus pipe cleaning method according to claim 7, characterized by, The traveling salesman problem algorithm refers to solving an optimal pressure switching path by establishing a pressure node graph and calculating the switching cost between nodes, taking the pressure demand of different cleaning areas of the pipeline as nodes, taking the energy consumption cost and time cost of pressure switching as edge weight, solving the shortest Hamilton circuit by a dynamic programming method, and outputting the optimal pressure fluctuation parameter sequence.

9. The PVDC production apparatus pipe cleaning method according to claim 8, characterized by, The pressure fluctuation parameter sequence includes a pressure value sequence, a pressure duration sequence and a pressure switching interval sequence.

10. The PVDC production apparatus pipe cleaning method according to claim 9, characterized by, The step of high-pressure water flushing is specifically that the electric conductivity of the inlet water and outlet water is continuously monitored during the flushing process, and the flushing process is ended when the electric conductivity of the inlet water and outlet water is consistent.

Images