A data acquisition and control method and system for pipe interlayer insulation foam filling
By setting temperature and pressure detection points on the outer surface of the outer protective tube, and combining dual threshold criteria and fitting line intersection correction, the problem of real-time monitoring of the sandwich foaming process is solved, achieving uniformity of insulation layer density and consistency of quality, reducing material consumption and pressure, and possessing self-adaptive capabilities.
Patent Information
- Application Number
- CN202611384645.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-08
- Publication Date
- 2026-11-17
AI Technical Summary
In the current production process of direct-buried insulated pipes, the foaming process of the interlayer cannot be monitored in real time, resulting in fluctuations in material expansion capacity, changes in friction resistance, lack of basis for rotation and exhaust control, and lag in the outer wall temperature signal, making online observation impossible. Furthermore, the parameters lack adaptive updates, leading to large quality dispersion.
Temperature and pressure measuring points are set along the axial direction on the outer surface of the outer protective tube to collect data synchronously. The arrival time is corrected by using dual threshold criteria of temperature rise amplitude and rate and the intersection of fitted straight lines. The equivalent frontal propulsion density and circumferential flow are calculated to control the injection and rotation, and an adaptive update mechanism is set.
It enables online monitoring of the sandwich foaming process, improves the density uniformity of the insulation layer, reduces material consumption and internal pressure, reduces voids and bulges, has self-adaptive capabilities, and ensures consistent product quality.
Smart Images

Figure CN123113754A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of data acquisition and production process control technology, and in particular relates to a data acquisition and control method and system for pipe interlayer insulation foam filling. Background Technology
[0002] Direct-buried insulated pipes consist of a working steel pipe for transporting the medium, a rigid polyurethane foam insulation layer, and a high-density polyethylene outer protective pipe. They are widely used in media transportation networks in fields such as district heating, district cooling, and petrochemicals. The insulation layer is typically manufactured using a one-step injection foaming process within a pipe. First, a surface-treated working steel pipe is placed centrally inside the outer protective pipe, and both ends are sealed with end plugs containing injection holes and vent holes. The pipe body is then supported on rollers at the foaming station, maintaining a certain axial angle. A high-pressure foaming machine then mixes the isocyanate and a polyether mixture containing a physical foaming agent under high pressure in a mixing head, injecting the mixture into the annular interlayer between the steel pipe and the outer protective pipe through the injection port. Within the interlayer, the material undergoes four stages: milky whitening, expansion and climbing, gelation, and curing, ultimately filling the interlayer and bonding with the steel pipe and outer protective pipe to form the insulation layer. The foam density, density uniformity, porosity, compressive strength, and thermal conductivity of the insulation layer are key indicators determining the pipeline's insulation performance and service life.
[0003] In existing production lines, the process control method for the above-mentioned steps involves pre-calculating the injection volume based on the geometric volume of the interlayer and the target foam density, and then stopping the foaming machine once the set volume is filled. The vent holes on the end plugs open and close based on empirical time, or only have normally open holes of a fixed diameter. During foaming, the tube rotates continuously at a fixed speed; the amount and timing of rotation are not determined based on measurement results. After foaming and curing, tube sections are sampled in batches for offline measurement of foam density, porosity, compressive strength, and thermal conductivity. While some production lines have added surface temperature measurement for the outer protective tube or monitoring of the material tank temperature, these signals are only used for over-temperature alarms or for process record keeping and do not participate in the adjustment of injection volume, vent opening, or tube posture.
[0004] Several technical problems exist in the current technology: First, the annular interlayer is surrounded by an opaque outer protective tube and sealed at both ends. The foaming reaction takes place in a closed environment. The actual expansion capacity of the material fluctuates with the batch of polyether, the content of the foaming agent, the material temperature, the preheating temperature of the tube body, and the ambient temperature of the workshop. The injection volume given according to the fixed value cannot detect the actual expansion capacity of the tube at that time. When the material activity is too high, the peak pressure inside the cavity rises and causes local bulging of the outer protective tube. When the material activity is too low, the expansion is insufficient and leaves voids and low-density areas with poor wall adhesion.
[0005] Secondly, the foam flows a long distance within the annular narrow slit. The frictional resistance and heat dissipation along the way cause the expansion ratio to change along the axial direction. The existing process cannot know the actual filling state of the foam along the pipe length, so it can only rely on overall overfilling to ensure a minimum, which increases both material consumption and internal pressure.
[0006] Third, under the action of gravity, the material preferentially flows along the bottom of the interlayer. The arrival order of the leading edge in the circumferential direction is different. The side that arrives first gels first, while the side that arrives later leaves poor adhesion. Continuous rotation at a fixed speed lacks measurement basis, and the turning angle and timing are both blind.
[0007] Fourth, the venting opening and the timing of stopping injection also lack a basis. Insufficient venting will cause the pressure inside the cavity to exceed the pressure-bearing capacity of the outer protective tube, while excessive venting will result in insufficient overfilling and end collapse. The two constitute a process contradiction that is difficult to balance.
[0008] Fifth, the practice of embedding thermocouples or pressure probes in the interlayer to obtain information about the interior of the interlayer not only contaminates the foam structure and makes it impossible to recover the components, but also requires drilling holes in the outer protective tube and subsequent hot-melt welding repair, making it difficult to achieve online observation of the foaming process for a long time.
[0009] Sixth, the outer sheath is made of a low thermal conductivity material and has a certain wall thickness. The temperature rise transmitted through the wall due to the exothermic reaction lags behind and is broadened relative to the actual process within the interlayer. The lag varies with the wall thickness deviation. When the temperature rise exceeding a certain given level is used as the arrival criterion, the judgment result also depends on the threshold value, and the calibration value becomes invalid once adjusted. The extruded outer sheath has circumferential wall thickness eccentricity, and the lag in each direction of the same cross-section is unequal. The resulting pseudo-time difference is on the same order of magnitude as the actual circumferential flow deviation. The combination of these factors causes the outer wall temperature signal to remain at the level of post-event recording for a long time.
[0010] In addition, the parameters of the existing process are given once based on experience, and there is no adaptive update mechanism based on the actual test results of the root canal and the batch. It is not able to adapt to the fluctuations of material batches and environmental conditions, and the quality dispersion between batches is too large. Summary of the Invention
[0011] To solve the above problems, the technical solution of the present invention is: a data acquisition and control method for pipe interlayer insulation foam filling, wherein the interlayer is an annular interlayer between a steel pipe and an outer protective pipe, comprising: S1. Several measuring sections are set along the axial direction on the outer surface of the outer protective tube. Three temperature measuring points are set in the circumferential direction of each section. Pressure detection points are set at the sealed ends. Mass flow rate detection points are set in the two material paths. The thermal hysteresis value of the outer protective tube wall is calibrated point by point by the calibration tube, and the core density conversion factor, gel duration and total action delay are calibrated. S2. From the start of material injection, the outer wall temperature, interlayer pressure and material mass flow rate are collected synchronously using a unified clock. The average temperature before material injection is taken as the baseline temperature, and the material injection mass is assigned according to the opening interval of the injection port. S3. Filter the temperature channel and estimate the temperature rise rate. Determine the candidate arrival interval based on the dual threshold of temperature rise amplitude and rate. When the fit is valid, take the intersection of the baseline segment and the temperature rise segment as the arrival time of the front of the measuring point. Otherwise, backtrack or mark the invalid value. S4. The arrival time is corrected point by point using the thermal hysteresis value of the outer protective pipe wall. The difference in arrival time between adjacent sections, the section spacing, the mass flow rate in the corresponding time window and the interlayer cross-sectional area are used to obtain the equivalent forward propulsion density of each pipe section. S5. For each new effective measurement section, the circumferential deviation amplitude and underfill azimuth angle are analyzed by rolling the correction arrival time of the continuous section. When the deviation amplitude exceeds the threshold, the azimuth angle range does not exceed the set value, and the sum of the local material age and the total delay of the action is less than the gel duration, the injection is paused and the tube is driven to rotate around the axis to turn the underfill azimuth to the lowest point of gravity before the injection is resumed. S6. Determine the non-negative additional injection mass of each injection port from the compaction stage based on the equivalent advance density and the target average compaction density of the pipe section under its jurisdiction. Normal injection is stopped when the mass reaches the target and all sections under its jurisdiction have reached the target. Separate pressure safety cut-off and overtime protection are set up. After injection is stopped, maintain controlled venting until the interlayer pressure drops below the closing threshold before closing the valve. S7. Determine whether to enable real-time correction of the quality of additional material injection in this tube based on the remaining executable time after deducting the total delay of the action, so as to cut and check to update the core density conversion coefficient and the material temperature setting value of the next tube, and mark the suspected abnormal section according to the calibrated target.
[0012] Furthermore, step S1 includes: S11, along the axial direction at intervals Several measurement sections are set up, and each injection port is located at the corresponding measurement section; each measurement section has a temperature measuring point at the 0°, 120° and 240° circumferential positions, and the lowest point of tube weight is taken at the 0° position; the temperature sensor of each measuring point is detachably attached to the outer surface of the outer protective tube without penetrating the interlayer. S12. A pressure detection point is set on each of the two sealing end plugs. An electric proportional exhaust valve is set on the sealing end plug away from the first injection port. An electric shut-off valve is set on each injection port. A mass flow rate detection point is set on each of the isocyanate material path and the combined polyether material path of the foaming machine. S13. Take a calibration tube with the same outer sheath material, wall thickness, and sensor installation method as the product tube. Drill test point holes at each measurement section and insert armored thermocouples. Measure the arrival time of the temperature rise on the inner wall of the outer sheath. Subtract this time from the arrival time of the leading edge determined according to S3 at the same test point to obtain the thermal hysteresis value of the outer sheath wall at that test point. , i is the measurement section number, j is the circumferential orientation number; S14. The foam core density is measured by axially cutting the calibration tube and taking samples. The ratio of this measured foam core density to the average compacted density calculated from the total mass of the injected material and the interlayer volume is used to obtain the core density conversion factor. ; S15. Perform cup foaming tests on the same batch of materials at no less than three material temperature points, and measure the gel duration at each material temperature point using the fiber drawing method. The equivalent advance density at each material temperature point was measured according to S4, and a linear regression was performed between it and the material temperature to obtain the slope of the material temperature sensitivity. ; S16. The total action delay is calculated by summing the time taken from temperature sampling to rotation into position, including filtering, regression, judgment, valve switching, and rotation execution. Calculate the cross-sectional area of the annular interlayer using the following formula: ; In the formula, The cross-sectional area of the annular interlayer; The inner diameter of the outer protective tube; This refers to the outer diameter of the steel pipe, including the thickness of the anti-corrosion coating.
[0013] Furthermore, step S2 includes: S21. All temperature, pressure and mass flow rate channels are sampled synchronously with a sampling period of no more than 0.2s, and the data of each channel are timestamped according to the same clock. S22. Apply a moving average filter with a window length of 2 seconds to each temperature channel to obtain the filtered temperature sequence. The baseline temperature is the arithmetic mean of the sequence within 5 seconds before injection. ; S23. Interlock the electric shut-off valves of each injection port so that only one injection port is in the injection state at any given time. S24. The cumulative injection mass is obtained by integrating the sum of the mass flow rates of the two material paths over time, and the pipeline retention delay after shifting the opening interval of each injection port backward is taken as the belonging interval of that port; the pipeline retention delay is the ratio of the pipeline volume from the mixing head to that port to the material volume flow rate. S25. When the ratio of the mass flow rates of the two material paths deviates from the set ratio by more than the set value, stop feeding and issue an alarm.
[0014] Furthermore, step S3 includes: S31. For each temperature channel, perform a univariate linear regression on the filtered temperature and time within a sliding window of 3 seconds, and take the slope as the temperature rise rate at that time. ; S32, when the difference between the filtered temperature and the baseline temperature is not less than the temperature rise amplitude threshold. ,and Not less than the temperature rise rate threshold and maintain continuously When the time is right, the channel is determined to have entered the candidate arrival interval, and this time is recorded as the determination point; S33. Take the temperature sequence 5 seconds before the decision point and perform a univariate linear regression to obtain the baseline straight line, the slope of which is... The intercept is A univariate linear regression was performed on the temperature sequence 3 seconds after the decision point to obtain the straight line of the temperature rise segment, the slope of which is... The intercept is ; S34, when and When the absolute value of the difference is not less than the slope difference threshold, and the determination coefficients of both regressions are not less than the goodness-of-fit threshold, the fit is considered valid. The abscissa of the intersection of the two lines is taken as the arrival time of the front at that measurement point according to the following formula: ; In the formula, This is the arrival time of the forward at this measuring point; The intercept of the baseline segment is the line segment intercept. This is the intercept of the straight line during the temperature rise segment; The slope of the baseline segment is denoted as . The slope of the straight line during the temperature rise segment; S35. If the fitting is invalid or the intersection point falls outside the interval of 5 seconds before to 3 seconds after the judgment point, take the judgment point as the arrival time of the leading edge of the measurement point and set the retreat flag; if there are fewer than two valid measurement points on the same measurement section, mark the section as invalid and do not participate in subsequent calculations.
[0015] Furthermore, step S4 includes: S41, the first The arrival time of the leading edge at each effective measuring point of each measurement section is subtracted from the thermal hysteresis value of the outer protective pipe wall to obtain the corrected arrival time of each measuring point. The arithmetic mean of these corrected arrival times is taken as the corrected arrival time of the section. ; S42, calculate the following formula: The measurement section and the first Axial velocity of the pipe front between the measurement sections: ; In the formula, This represents the axial velocity of the leading edge of the pipe section; This represents the axial distance between adjacent measurement sections; For the first The calibration arrival time of each measurement section; For the first The calibration arrival time of each measurement section; S43, Time Window The arithmetic mean of the mass flow rate at the current injection port. The equivalent forward thrust density of this pipe segment is calculated using the following formula: ; In the formula, This is the equivalent forward thrust density for this pipe section; The mass flow rate of the material belonging to the current injection port within this time window; The cross-sectional area of the annular interlayer; This represents the axial velocity of the leading edge of the pipe section; S44 When the effective velocity range determined by the calibration tube is exceeded, the equivalent forward propulsion density of that tube segment is marked as invalid; take the first... The arithmetic mean of all effective values within the pipe section under the jurisdiction of each injection port is used as the equivalent forward propulsion density of that pipe section. .
[0016] Furthermore, step S5 includes: S51, regarding the first There are three effective measurement sections, and the correction arrival times of the three azimuth measuring points at 0°, 120°, and 240° are recorded as follows: , , The first harmonic cosine coefficient of this cross section is calculated using the following formula: ; In the formula, The first harmonic cosine coefficient of this cross section; , , These are the correction arrival times of the measuring points at the 0°, 120°, and 240° azimuth positions of the cross section, respectively. S52. Calculate the first harmonic sine coefficient of the measured cross section using the following formula: ; In the formula, The first harmonic sine coefficient of this cross section; S53. For each newly added valid measurement section, take the nearest measurement section including that section. One continuous effective measurement section, No less than 3; take its The arithmetic mean is , The arithmetic mean is The amplitude of the circumferential deflection is calculated using the following formula: ; In the formula, This refers to the amplitude of the circumferential deflection flow; For the The mean value of the first harmonic cosine coefficient of each cross section; It is the mean value of its first harmonic sine coefficient; S54. Calculate the underfilled azimuth angle using the following formula: ; In the formula, The underfilled azimuth angle is calculated from the 0° azimuth and ranges from −180° to 180°; atan2 is the four-quadrant arctangent function, with the first independent variable being the sine coefficient and the second independent variable being the cosine coefficient; S55. The difference between the current time and the opening time of the feed inlet under its jurisdiction is taken as the local material age; when Greater than the circumferential deflection amplitude threshold The above The range of the underfilled azimuth angles calculated by S54 for each cross section is no greater than the azimuth consistency threshold. Furthermore, the local material age is related to the total delay of the aforementioned action. The sum is less than the gel duration. When the current filling port's electric shut-off valve is closed, filling is paused. The rotary drive mechanism drives the tube body to rotate around the axis. The rotation angle is the negative value of the underfill azimuth angle, and the absolute value is not greater than 180°. Filling resumes after feedback from the tube body angle detection unit. If any condition is not met, rotation is stopped and the result is recorded. and .
[0017] Furthermore, step S6 includes: S61. During the advancement phase, the electric shut-off valves of the injection ports are opened sequentially from the injection end. When the leading edge reaches the measurement section of the next injection port, the current port is closed and the next port is opened. When the leading edge reaches the end measurement section, the advancement phase is determined to be over and the compaction phase begins. S62. Determine the number according to the following formula. Additional material mass at each injection port, calculated from the compaction stage: ; In the formula, For the first Additional injection mass from each injection port starting from the compaction stage; max is the maximum value calculated. The cross-sectional area of the annular interlayer; This refers to the length of the pipe section under the jurisdiction of that port. Target foam core density; This is the core density conversion factor; This is the equivalent forward advance density for the section under the jurisdiction of this port; S63. Take the larger of the measured values from the two pressure detection points as the interlayer pressure control value, and adjust it by comparing it with the exhaust pressure value. The difference is the deviation. The opening of the electric proportional exhaust valve is adjusted according to the incremental proportional-integral control law, and the opening is not lower than the set minimum opening. An alarm is triggered when the difference between the two measured values exceeds the set value. S64. Calculate the upper limit of the allowable interlayer pressure for the circumferential stress of the outer protective pipe using the following formula: ; In the formula, This is the upper limit of the interlayer pressure allowed for the circumferential stress of the outer protective tube; The outer protective pipe wall thickness; For the short-term allowable circumferential stress of the outer protective tube at the current wall temperature; The inner diameter of the outer protective tube; The product of the allowable pressure obtained from the pressure resistance test of the assembly consisting of the outer protective tube and the sealing end plug, and a safety factor not exceeding 0.8, is taken as follows: The smaller of the two is the safe shut-off pressure. ,and Greater than ; S65, Normal suspension of betting is: (The following is a continuation of the previous sentence) The cumulative injection mass of each injection port, calculated from the compaction stage, reaches Furthermore, when the arrival of the leading edge has been determined at all measuring sections under the jurisdiction of the port, the electric shut-off valve of the port shall be closed; the pressure protection is as follows: the interlayer pressure control value shall not be less than Immediately stop the pump and fully open the exhaust valve, then record the abnormality; the timeout protection is as follows: the local material age is not less than Stop injecting and mark the pipe section immediately; S66. After injection is stopped, the electric proportional exhaust valve maintains the closed-loop regulation of S63 until the interlayer pressure control value is lower than the closing threshold. And then turn it off after the set duration. Less than .
[0018] Furthermore, step S7 includes: S71, Subtract the current calculation completion time from the estimated total injection time of this pipe, and then subtract... The remaining executable time is used; if it is greater than zero, the additional injection quality of this pipe that has not yet been executed is corrected in real time according to the equivalent forward advance density of S4; otherwise, the additional injection quality allocation result of the previous pipe is used and only the data is recorded. S72. The density of the foam core is measured by cutting and sampling the first piece of pipe and the sampled pipe in each batch. The average compaction density is calculated from the total mass of the injected material and the volume of the interlayer. Update the core density conversion factor using the following formula: ; In the formula, This is the updated core density conversion factor; The conversion factor for core density before the update; The relaxation factor is set to 0.2–0.5. The density of the foam core was measured by cutting and sampling. This represents the average compaction density. S73. Take the arithmetic mean of the effective equivalent forward propulsion density of each tube. Correct the material temperature setpoint of the next tube using the following formula: ; In the formula, Set the material temperature setting for the next tube; This is the setpoint for the material temperature of this tube; The material temperature correction factor is taken as the slope of the linear relationship between material temperature and equivalent leading advance density obtained from calibration at no less than three material temperature points. The reciprocal, Since it is a negative value, It is a negative value; The equivalent forward propulsion density of the calibration tube at the reference material temperature; The value is the arithmetic mean of the effective equivalent forward propulsion density of each tube; the absolute value of the correction is not greater than 3°C, and the corrected value does not exceed the calibrated material temperature range; S74. Archive the equivalent front advance density, injection quality at each port, circumferential flow amplitude, underfill azimuth angle, rotation angle and interlayer pressure curve of each pipe section according to pipe number, and mark the pipe sections whose equivalent front advance density exceeds the target zone indicated by the above-mentioned cross-sectional sampling results as suspected abnormal sections according to axial coordinates.
[0019] This invention also designs a data acquisition and control system for pipe interlayer insulation foam filling, comprising: A non-invasive temperature sensing array unit that is detachably attached to the outer surface of the outer protective tube, distributed axially at multiple measurement sections, with a temperature sensor at each of the three circumferential directions along each section; A sandwich pressure detection unit consisting of two pressure transmitters located at the sealed end plugs at both ends of the pipe body; A material mass metering unit consisting of two mass flow meters located on the two material paths of the foaming machine; Multiple injection ports, each equipped with an electric shut-off valve and interlocked with each other, are spaced apart along the axial direction on the outer protective pipe. An electric proportional exhaust valve is installed on the sealed end plug at the end furthest from the first injection port; A rotary drive mechanism that drives the tube body to rotate around its own axis, and a tube body angle detection unit that directly detects the rotation angle of the tube body; A multi-channel synchronous data acquisition unit that is connected to the non-invasive temperature sensing array unit, the interlayer pressure detection unit, and the material mass metering unit, and samples each channel synchronously according to a unified clock. A leading edge identification module that is connected to the multi-channel synchronous data acquisition unit and outputs the arrival time of the leading edge of each measuring point according to the dual threshold continuous judgment criterion and fitting validity condition; A propulsion density estimation module that is signal-connected to the forward identification module and outputs the equivalent forward propulsion density of each pipe segment; A flow analysis module that is signal-connected to the forward identification module and continuously outputs the circumferential flow amplitude and underfilled azimuth angle; The input end is connected to the deflection analysis module and the pipe angle detection unit, and the output end is connected to the rotation compensation control module of the rotation drive mechanism and the electric shut-off valve. The input terminal is connected to the propulsion density estimation module and the material mass metering unit, and the output terminal is connected to the segmented injection control module of the electric shut-off valve; The input end is connected to the interlayer pressure detection unit, the output end is connected to the electric proportional exhaust valve and the foaming machine pump stop control end, and the exhaust pressure adjustment value and the safety stop pressure are respectively set in the exhaust and safety interlock module. The module includes an adaptive update and anomaly marking module whose input is connected to the propulsion density estimation module and whose output is connected to the segmented injection control module and the foaming machine material temperature control module.
[0020] Furthermore, the non-invasive temperature sensing array unit includes a wireless temperature acquisition node that rotates synchronously with the pipe body and is wirelessly connected to the multi-channel synchronous data acquisition unit; the pipe body angle detection unit includes an angle mark on the outer surface of the outer protective pipe and a photoelectric sensor fixed to the frame, the output end of the photoelectric sensor being connected to the feedback input end of the rotation compensation control module; the rotation drive mechanism includes a frequency conversion drive roller, the angle detection output end of which is also connected to the rotation compensation control module, forming an anti-slip dual feedback with the photoelectric sensor; each of the injection ports is located at the corresponding measurement section and connected to the foaming machine mixing head via a rotary joint; the output ends of the two pressure transmitters are both connected to the exhaust and safety interlock module, which takes the larger of the two as the control value and performs redundancy diagnosis on the difference between them.
[0021] The beneficial effects of this invention are as follows: This invention arranges temperature measuring points detachably on the outer surface of the outer protective tube, without embedding any components within the annular interlayer. Combined with a dual-threshold continuous criterion for temperature rise amplitude and rate, refinement of the intersection of the baseline and temperature rise segments' fitted straight lines, a fitting validity criterion and backoff rules, and the calibration and deduction of the outer protective tube wall thermal hysteresis value point by point, the foaming process within the sealed interlayer becomes an online observable process without altering the product structure. The dual-threshold continuous criterion suppresses misjudgments caused by slow ambient temperature drift and instantaneous interference. The intersection of the two fitted straight lines is extrapolated from the overall trend of the two data segments, and the resulting arrival time is almost independent of the threshold value. The same set of calibration values can be used under different pipe diameters and signal-to-noise ratios. Along the axial direction, adjacent measurement sections have similar wall thicknesses, and the hysteresis portion is eliminated as a common modulus when the arrival times are subtracted. In the circumferential direction, due to extrusion eccentricity, the wall thickness difference is larger, and the hysteresis differential modulus portion is deducted by point-by-point correction. The combination of these two methods enables the time difference to reach an accuracy suitable for subsequent analysis, transforming the traditionally unusable outer wall temperature signal for process control into a reliable process measurement. The fitting validity criterion avoids numerical amplification in intersection point calculations when the slopes of two segments are close, and the backoff rule ensures that the control flow still has a definite output even when the signal quality deteriorates.
[0022] This invention uses the relationship between material mass flow rate, annular interlayer cross-sectional area, and forward axial velocity to obtain the equivalent forward propulsion density of each pipe segment online. All input quantities are directly measurable or calculable based on drawings, without fitting coefficients. It transforms the actual expansion capacity of the material in a given cycle into a process index obtainable segmentally along the pipe length. During the compaction stage, the additional injection mass at each injection port is determined by the difference between the equivalent forward propulsion density of the pipe segment and the target average compaction density. Pipe segments with weaker expansion capacity automatically receive more material, while those with stronger expansion capacity automatically receive less. The injection method changes from a uniform open-loop quantitative filling and overall overfilling to segment-by-segment on-demand replenishment. The density uniformity of the insulation layer along the axial direction is improved, voids are reduced, and material consumption and peak pressure within the cavity decrease simultaneously. The additional injection mass is calculated non-negatively; when the measured expansion capacity has reached or exceeded the target, a zero additional injection amount (non-negative value) is given. The control command can be executed under any input conditions.
[0023] This invention performs first-order harmonic decomposition on the correction arrival times of multiple circumferential orientations of the same measurement section to obtain the circumferential flow amplitude and underfill azimuth angle, transforming the circumferential non-uniformity, which was originally only judged empirically, into a quantitatively expressible amplitude and orientation. The analysis uses a rolling window; for each newly added valid measurement section, a judgment is given based on the currently reached consecutive measurement sections. The judgment time matches the material state of the controlled pipe section. The rotation action is completed during the period when the material is still fluid and there is still material injection, thus avoiding situations where the action fails. The rotation angle is taken as the negative value of the underfill azimuth angle, rotating the underfill orientation to the point of lowest gravity. Subsequently injected liquid phase material preferentially flows to the originally underfilled side under the influence of gravity, thus converging the circumferential density difference and reducing poor wall adhesion. The orientation consistency criterion requires that the underfill orientations given by each measurement section participating in the analysis be close to each other. No action is executed when the flow direction of each section is discrete, avoiding the reverse effect of a single overall rotation of the rigid pipe on a part of the pipe section. The gel window criterion compares the sum of the local material age and the total action delay with the gel duration to ensure that the instructed action can indeed be executed in time.
[0024] This invention separates normal quality endpoints, pressure safety cutoff, and timeout protection into three independent conditions. The exhaust pressure adjustment value no longer serves as the injection stop threshold. The pressure closed-loop maintains its regulatory function throughout the compaction stage, ensuring sufficient additional material injection and suppressing underfilling. The safety cutoff pressure is the smaller of the component's pressure resistance test allowable pressure and the upper limit of the outer sheath's circumferential stress allowable, providing a verifiable basis for the safety boundary. After injection stoppage, the exhaust valve maintains closed-loop regulation until the interlayer pressure drops and stabilizes, releasing the secondary pressure rise caused by the material's continued reaction, exothermic expansion, and suppressing bulging of the outer sheath. This invention also incorporates two levels of adaptive design: real-time in-pipe correction adjusts the quality of unexecuted additional material injection based on the process data of the current pipe; inter-pipe batch updates adjust the core density conversion coefficient and the material temperature setpoint of the next pipe based on the results of cross-sectional sampling. The direction of the material temperature correction matches the sign of the material temperature sensitivity slope, creating a negative feedback loop. The process is adaptive to fluctuations in material batches and environmental conditions, reducing the quality dispersion between batches. The segmented density records and coordinates of suspected abnormal sections archived in each tube provide a basis for positioning in subsequent non-destructive testing or cross-section verification, ensuring product quality traceability. Attached Figure Description
[0025] Figure 1 The graphs show the thermal hysteresis characteristics and arrival time error of the outer protective tube wall in Examples 1 to 4 and Comparative Example 6. (a) is a curve showing the change of the arrival time determination value of the front with the thickness of the outer protective tube wall, and (b) is a bar chart comparing the standard deviation of the arrival time error in 5 cases.
[0026] Figure 2The graph shows the threshold sensitivity comparison of the arrival time of the leading edge in Examples 1 to 3 and Comparative Example 2, and the graph shows the change of arrival time with the threshold of temperature rise amplitude with error bars.
[0027] Figure 3 The diagrams show the rotational feasibility and circumferential density uniformity of Example 1, Comparative Example 3, and Comparative Example 4. (a) is a bar chart comparing the rotational action feasibility margin and the remaining compaction injection time after rotation, and (b) is a curve comparison chart showing the circumferential foam core density range at each measurement section along the axial direction.
[0028] Figure 4 The graphs show the comprehensive performance comparison of the insulation layers of Examples 1 to 4 and Comparative Examples 1 to 7. (a) is a bar chart of the coefficient of variation of axial core density, (b) is a bar chart of the range of circumferential core density, (c) is a bar chart of porosity, and (d) is a bar chart of the standard deviation of the density of 20 consecutive tube cores.
[0029] Figure 5 This is a schematic diagram of the data acquisition and control system for the pipe interlayer insulation foam filling of the present invention. Detailed Implementation
[0030] The technical solutions of this invention will be clearly and completely described below. In addition, the forms of the various structures described in the following embodiments are merely illustrative. This invention is not limited to the structures described in the following embodiments. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.
[0031] The specific embodiments of the present invention are further described below. The present invention targets the one-step injection foaming process of prefabricated direct-buried insulated pipes. After shot blasting and rust removal, the working steel pipe is centrally placed inside a high-density polyethylene outer protective pipe. Both ends of the pipe are sealed with end plugs. The pipe is supported on rollers at the foaming station and kept at an axial inclination angle. A high-pressure foaming machine mixes the isocyanate and polyether components under high pressure in a mixing head, and then injects the mixture into the annular interlayer between the steel pipe and the outer protective pipe through an injection port located on the outer protective pipe. Within the interlayer, the material sequentially undergoes milky whitening, expansion and climbing, gelation, and solidification, ultimately filling the interlayer and bonding with the steel pipe and the outer protective pipe to form an insulation layer. The annular interlayer is surrounded by an opaque outer protective pipe and sealed at both ends. The foaming process is carried out enclosed within this annular interlayer. Therefore, the core of the present invention lies in obtaining process information within the interlayer without intruding into it, and using this information to drive the injection, rotation, and venting actions.
[0032] The present invention provides a data acquisition and control method for pipe interlayer insulation foam filling, comprising steps S1 to S7, which are described in detail below.
[0033] Step S1 is sensor arrangement and calibration, including sub-steps S11 to S16. In sub-step S11, several measurement sections are set at intervals Ls along the axial direction of the tube on the outer surface of the outer sheath. Each injection port is located at the corresponding measurement section, so that the moment when the leading edge reaches the injection port can be directly determined by the temperature signal of the same section without the need for interpolation calculation from adjacent sections. Each measurement section has one temperature measuring point at each of the three circumferential directions: 0°, 120°, and 240°. The 0° position is the point with the lowest weight of the tube. The three positions are evenly distributed to provide orthogonal sampling conditions for subsequent first-order harmonic decomposition. The temperature sensors at each measuring point are detachably attached to the outer surface of the outer sheath, separated from the interlayer by the complete outer sheath wall, without penetrating the interlayer or remaining inside the product. The sensor is preferably a thin-film platinum resistance thermometer, with thermally conductive silicone grease bonded to the outer surface of the outer sheath to reduce contact thermal resistance. It is covered with an elastic thermal insulation pad and secured with a clamp. This reduces heat dissipation from the measuring point to the environment, thus increasing the temperature rise signal amplitude, and also ensures consistent installation thermal resistance across multiple assembly and disassembly cycles. The spacing Ls between the measuring sections is preferably 0.8m to 1.5m, and the number of measuring sections is preferably no less than 8. The first measuring section is preferably 0.3m to 0.6m from the injection end.
[0034] In sub-step S12, a pressure detection point and a pressure transmitter are installed at each of the sealed end plugs at both ends of the pipe body. An electric proportional exhaust valve is installed at the sealed end plug furthest from the first injection port. An electric shut-off valve is installed at each injection port. A mass flow rate detection point and a Coriolis mass flow meter are installed at each of the isocyanate feed path and the combined polyether feed path of the foaming machine. The pressure signal has no transmission lag and reflects the overall state within the interlayer. The temperature signal has wall-penetrating lag but can pinpoint the specific cross-section and orientation. These two types of signals are responsible for determining the safety boundary and analyzing the spatial distribution in subsequent steps, respectively.
[0035] In sub-step S13, a calibration tube is selected, whose outer sheath material, wall thickness, and sensor installation method are the same as those of the product tube. Measurement point holes are drilled on the outer sheath at each measurement section, and armored thermocouples are inserted, with the thermocouple tips flush with the inner wall of the outer sheath. The arrival time of the temperature rise on the inner wall of the outer sheath is then measured. The arrival time of the inner wall temperature rise is subtracted from the arrival time of the leading edge determined in step S3 for the same measurement point; the difference is the thermal hysteresis value of the outer sheath wall at that measurement point. Where i is the measurement section number and j is the circumferential orientation number. Calibration is performed point-by-point instead of using a single constant because thermal hysteresis increases with the square of the wall thickness. However, the extruded outer sheath has circumferential wall thickness eccentricity, and the wall thickness difference between the three orientations of the same section is often greater than the difference between adjacent sections. If a single constant were used, the circumferential wall thickness difference would enter the subsequent circumferential flow analysis in the form of arrival time differences, and be misjudged as material flow deviation. Point-by-point calibration absorbs these differences into the calibration value itself, ensuring that the corrected arrival time reflects the actual material propulsion within the interlayer. The calibration tube is not shipped as a product after calibration.
[0036] In sub-step S14, the calibration tube is cut axially to take a sample, and the foam core density is measured according to the current test method for the apparent density of rigid foam plastics. The compacted average density is calculated from the total mass of the injected material and the interlayer volume. The ratio of the two is the core density conversion factor. The foam has a dense outer layer with a high density near the steel pipe and outer protective pipe, while the core density is lower than the average density of the entire cross-section. The value used to characterize the above differences is preferably between 0.88 and 0.95.
[0037] In sub-step S15, cup foaming tests are performed on the same batch of material at no less than three material temperature points, and the gel duration at each material temperature point is measured by the fiber drawing method. This refers to the duration from the mixing of the two components until the foam loses its fluidity. The duration is defined as the duration rather than the absolute time starting from the injection point of the entire pipe because the mixing starting point of materials in different pipe sections differs when injecting materials sequentially from multiple injection ports. Only by combining the duration with the mixing starting point of each pipe section can the gelation period truly correspond to the controlled pipe section be obtained. Simultaneously, the equivalent leading density at each material temperature point is measured according to step S4. A univariate linear regression is performed on the equivalent leading density and material temperature to obtain the material temperature sensitivity slope s. Increased material temperature increases the reaction rate and the degree of foaming agent vaporization, leading to a larger expansion ratio and a decrease in the equivalent leading density; therefore, s is negative. Using at least three regression points instead of a two-point line allows for the determination of goodness of fit and the definition of the effective material temperature range.
[0038] In sub-step S16, the total action delay is obtained by measuring the sum of the filtering, regression, judgment, valve switching, and rotation execution times from the time of temperature sampling to the time of rotation to the final position. This is then used to determine whether the rotational action can be performed within the gelation period. The cross-sectional area of the annular interlayer is calculated using the following formula. ; In the formula, A is the cross-sectional area of the annular interlayer (m²). 2 ); Pi is the mathematical constant; Di is the inner diameter of the outer protective tube (m); do is the outer diameter of the steel pipe (m), including the thickness of the anti-corrosion layer.
[0039] Step S2 involves synchronous data acquisition and quality attribution, comprising sub-steps S21 to S25. In sub-step S21, all temperature channels, the two pressure detection point channels, and the mass flow rate channels of the two material paths are synchronously sampled and timestamped using a unified clock. The sampling period is preferably no greater than 0.2s. A unified timestamp is a prerequisite for all subsequent differential operations; if the clocks of different channels are from different sources, the difference in arrival times between cross-sections and directions will be mixed with clock deviations. In sub-step S22, a moving average filter is applied to each temperature channel to suppress power frequency interference and random noise. The filter window length is preferably 2s, resulting in a filtered temperature sequence T(t). The arithmetic mean of this sequence within a certain period before material injection is taken as the baseline temperature Tb, used to eliminate the difference between the workshop ambient temperature and the pipe preheating temperature. In sub-step S23, the electric shut-off valves of each injection port are interlocked, ensuring that only one injection port is in the injection state at any given time. This allows the sum of the mass flow rates of the two material paths to be uniquely attributed to the currently open injection port. In sub-step S24, the cumulative injection mass is obtained by integrating the sum of the mass flow rates of the two feed paths over time. The opening interval of each injection port is then shifted backward by the pipeline retention delay to determine the assigned interval of the injection port. The pipeline retention delay is the ratio of the pipeline volume from the mixing head to the injection port to the material volumetric flow rate. Since there is a pipeline volume between the mixing head and the injection port, and there is a lag between the valve opening time and the actual entry of material into the interlayer, if no shift is made, the injection mass assignments of two adjacent injection ports before and after the switch will be interfering with each other. In sub-step S25, when the ratio of the mass flow rates of the two feed paths deviates from the set ratio by more than a set value, injection is stopped and an alarm is triggered to prevent mismatched material from entering the interlayer and causing subsequent density estimation to lose its basis.
[0040] Step S3 is the identification of the arrival time of the leading edge, including sub-steps S31 to S35. The formation reaction of rigid polyurethane is a strongly exothermic reaction. The exothermic reaction of the foam wall layer causes the inner wall of the outer sheath to heat up. The heat is transferred through the outer sheath wall to the outer surface, forming a measurable temperature rise on the outer surface. The outer sheath is a low thermal conductivity material with a certain wall thickness. The temperature rise on the outer surface lags behind the temperature rise on the inner wall by tens of seconds and is broadened. It is generally believed in the art that the temperature measurement of the outer surface can only be used for post-event recording and not for process control. The present invention is based on the understanding that the wall thickness of the outer sheath at each measuring point of the same pipe is similar to the heat dissipation boundary conditions. The main part of the lag is a common modulus for each measuring point. Quantities with the difference in arrival time as input, including the axial leading edge velocity and the circumferential arrival time distribution, are not affected by the common modulus. Therefore, spatial distribution information can be obtained without solving the inverse thermal conductivity problem. The differential modulus part of the lag, which varies with the wall thickness deviation, is calibrated point by point by sub-step S13. In step S4, the measurement points are subtracted one by one. The wall thickness of adjacent measurement sections along the axial direction is close, and common mode cancellation reduces the hysteresis effect in the axial time difference by about an order of magnitude; the wall thickness difference in the circumferential direction is large due to extrusion eccentricity, and the effect of common mode cancellation is limited. It must be eliminated by measurement point correction. Together, the axial and circumferential time differences reach the accuracy that can be used for velocity analysis and flow deviation analysis.
[0041] In sub-step S31, for each temperature channel, a univariate linear regression is performed on the filtered temperature and time within a sliding window. The slope of the regression line is taken as the temperature rise rate r at that time, and the regression window length is preferably 3s. Using the regression slope instead of the difference between two adjacent points can suppress the amplification of noise on the rate estimation without sacrificing time resolution. In sub-step S32, when the difference between the filtered temperature and the baseline temperature Tb is not less than the temperature rise amplitude threshold... Furthermore, the temperature rise rate r is not less than the temperature rise rate threshold kth, and the criterion duration is maintained continuously. When the channel enters the candidate arrival interval, the time is recorded as the determination point. Using both amplitude and rate thresholds can eliminate two types of false judgments: slow drift in ambient temperature and instantaneous electromagnetic interference. The duration requirement further eliminates single-point outliers. The preferred values are 0.3K to 1.0K, and kth is preferably 0.02K / s to 0.06K / s. The preferred value is between 1.0s and 3.0s, with the specific value determined by the measured temperature rise curve of the calibration tube, so that the threshold is more than 3 times higher than the filtered noise but less than 1 / 3 of the actual temperature rise rate.
[0042] In sub-step S33, a baseline segment is obtained by performing a univariate linear regression on the temperature sequence before the decision point, with a slope of r1 and an intercept of b1. A temperature rise segment is obtained by performing a univariate linear regression on the temperature sequence after the decision point, with a slope of r2 and an intercept of b2. The preferred sampling lengths for the two segments are 5s and 3s, respectively. In sub-step S34, when the absolute value of the difference between r2 and r1 is not less than the slope difference threshold, and the determination coefficients of both regressions are not less than the goodness-of-fit threshold, the fit is deemed valid, and the x-coordinate of the intersection of the two lines is taken as the arrival time of the front of the measuring point according to the following formula. ; In the formula, tarr is the arrival time of the front at the measuring point (s); b1 is the intercept of the baseline segment (°C); b2 is the intercept of the temperature rise segment (°C); r1 is the slope of the baseline segment (K / s); and r2 is the slope of the temperature rise segment (K / s). The above formula is derived from the equations of the two lines: r1·t + b1 = r2·t + b2, which can be rearranged to obtain t·(r1−r2) = b2−b1. Dividing both sides by r1−r2 gives the result. Using the intersection of the two lines as the starting point is a common practice in thermal analysis to determine the transition initiation temperature. Its significance lies in the fact that the obtained time is determined by the overall trend of the two data segments, rather than by the sampling point that exactly crosses the threshold. Therefore, it is more effective for determining the transition initiation temperature. The value of kth is insensitive, ensuring comparability between different batches. The validity criterion serves to prevent the intersection value from being amplified when the denominator r1−r2 approaches zero.
[0043] In sub-step S35, when the fitting is invalid, or the intersection point falls outside the time interval of the two data segments taken before and after the decision point, the decision point is taken as the arrival time of the leading edge of the measurement point, and a backoff flag is set, so that the control flow still has a definite output without interruption when the signal quality is poor. When there are fewer than two valid measurement points on the same measurement section, the measurement section is marked as invalid and does not participate in subsequent calculations.
[0044] Step S4 involves estimating the segmented velocity and equivalent forward propulsion density, comprising substeps S41 to S44. In substep S41, the arrival time of the forward at each effective measuring point of the i-th measuring section is subtracted from the thermal hysteresis value of the outer sheath wall. The corrected arrival times of each measuring point are obtained, and their arithmetic mean is taken as the corrected arrival time *ti* of the cross section. After correction, *ti* corresponds to the actual time when the material in the interlayer passes through the cross section, which is consistent with the timing reference of the mass flow rate. In sub-step S42, the axial velocity of the pipe front between the i-th measuring cross section and the (i+1)-th measuring cross section is calculated using the following formula. ; In the formula, vi is the axial velocity of the pipe segment (m / s); Ls is the axial distance between adjacent measurement sections (m); ti is the correction arrival time of the i-th measurement section (s); ti+1 is the correction arrival time of the (i+1)-th measurement section (s). Without correction, both ti and ti+1 contain approximately equal thermal hysteresis of the outer protective pipe wall; when they are subtracted, the common mode cancels out. This is then subtracted point by point in sub-step S41. Subsequently, the differential mode caused by wall thickness deviation was also eliminated, and the accuracy of vi was further improved.
[0045] In sub-step S43, the time window is taken from ti to ti+1, and the arithmetic mean qi of the mass flow rate belonging to the current injection port within the time window is taken. Then, the equivalent frontal thrust density of the pipe segment is calculated according to the following formula. ; In the formula, The equivalent forward propulsion density of the pipe segment (kg / m) 3 qi is the mass flow rate (kg / s) of the material belonging to the current injection port within the time window; A is the cross-sectional area of the annular sandwich (m²). 2 ); vi is the axial velocity (m / s) of the leading edge of the pipe segment. The above formula is derived from the fact that, within any small time interval of the propulsion phase, the mass of material injected into the interlayer is the product of qi and the time interval length, the distance the leading edge travels axially is the product of vi and the time interval length, and the volume of the interlayer swept is the product of A, vi, and the time interval length. Dividing the two gives the mass of material per unit volume, and the time interval length is approximately 0. Dimensionally, kg·s -1 Divide by m 2 With m·s -1 The product equals kg·m -3 Both sides are consistent. Since ti and ti+1 have already deducted the thermal hysteresis of the outer protective pipe wall in sub-step S41, the time window taken corresponds to the time period during which the material in the interlayer actually passes through the pipe section, and the mass and volume correspond within the same time period.
[0046] The equivalent forward density is obtained by dividing the mass flow rate by the forward volume fraction. It reflects the actual expansion capacity of the material in the current tube under the confined conditions of the interlayer. Its value is affected by the reactivity, material temperature, ambient temperature, and flow resistance. The correspondence between it and the final foam core density is determined by sectional sampling. Therefore, it is used as a process indicator in this invention and is not directly equivalent to the material density. In sub-step S44, when vi exceeds the effective velocity range determined by the calibration tube, the equivalent forward density of the tube segment is marked as invalid to eliminate abnormal values caused by measurement point failure or air cavity crossflow. The arithmetic mean of the effective values in the tube segment under the jurisdiction of the k-th injection port is taken as the equivalent forward density of the tube segment under the jurisdiction of the injection port. .
[0047] Step S5 involves the rolling analysis and rotational compensation of the circumferential flow deviation, comprising sub-steps S51 to S55. Under gravity, the material preferentially flows along the bottom of the interlayer. The arrival times of the leading edge differ along the circumferential direction; foam arriving earlier on one side gels first, while foam arriving later on the other side is prone to poor adhesion and localized low density. To quantify this non-uniformity, it is noted that the arrival times at the three azimuth positions of the same cross-section are approximately distributed according to the first harmonic along the circumferential direction, i.e., the arrival time equals the mean plus the sum of a cosine component and a sine component. In sub-step S51, for the i-th effective measurement cross-section, the corrected arrival times at the three azimuth measuring points at 0°, 120°, and 240° are respectively... , , The first harmonic cosine coefficient of the cross section is calculated using the following formula: ; In the formula, The first harmonic cosine coefficient (s) of the cross section is given. , , The corrected arrival times (in seconds) for the 0°, 120°, and 240° azimuth measuring points of the cross section are respectively. The above formula is derived from discrete orthogonality. When the three azimuths are equally spaced, the sum of the cosine values of the three azimuths is zero, the sum of the squares of the cosine values of the three azimuths is 3 / 2, and the sum of the products of the cosine and sine values of the three azimuths is zero. Therefore, by multiplying the arrival time by the cosine value of the corresponding azimuth and then summing the results, the result is equal to the product of the cosine coefficient and 3 / 2. Multiplying both sides by 2 / 3 yields the above formula. In sub-step S52, the first harmonic sine coefficient of the cross section is calculated similarly using the following formula. ; In the formula, Let be the first harmonic sine coefficient (s) of the cross section. The above formula is obtained by simplifying the sum of 2 / 3 multiplied by the product of the sine values of the two azimuth angles of 120° and 240° and the corresponding arrival time. The sine value of 120° is √3 divided by 2, and the sine value of 240° is negative √3 divided by 2. Substituting these values and combining like terms yields the result.
[0048] In sub-step S53, for each newly added valid measurement cross section, the nearest nw consecutive valid measurement cross sections, including the newly added cross section, are taken, where nw is preferably not less than 3, and the arithmetic mean of the cosine coefficients is taken as... The arithmetic mean of the sine coefficients is And calculate the circumferential deflection amplitude according to the following formula. ; In the formula, U is the amplitude of the circumferential deflection (s); The mean value (s) of the first harmonic cosine coefficients of the nw cross sections; The average value (s) of the first harmonic sine coefficients of the nw cross sections is used. A rolling window is employed instead of waiting for all pipe sections to be completed before calculation because rotational compensation is only meaningful when the material has not yet gelled. If a leading signal is waited for at the end section, the previously injected material will have approached or exceeded its gelation period, making the control action impossible to execute in sequence. The rolling window allows each newly added effective cross section to trigger a judgment, advancing the control action to a time matching the material age of the controlled pipe section. Using the average of multiple cross sections instead of a single cross section suppresses measurement errors at single cross section points. Furthermore, since the pipe is a rigid body, a single rotation can only achieve one angle, and using the average value aligns with the degrees of freedom of the actuator. U is directly expressed in units of time without dimensionless normalization, preventing the threshold from being diluted with pipe length and propulsion time.
[0049] In sub-step S54, the underfilled azimuth angle is calculated using the following formula. ; In the formula, is the underfilled azimuth angle (°), calculated from 0° azimuth, with a value range of −180° to 180°; atan2 is the four-quadrant arctangent function, with its first independent variable being the sine coefficient and its second independent variable being the cosine coefficient; and The meaning is the same as before. Since the distribution of arrival times along the circumferential direction can be written as the sum of the mean and the cosine and sine components, the azimuth angle at which it reaches its maximum value is the direction of the composite vector determined by the cosine and sine coefficients, which is the direction where the material arrives latest and is most prone to underfilling. Using the four-quadrant arctangent function instead of the ordinary arctangent function can give the correct quadrant when the cosine coefficient is negative.
[0050] In sub-step S55, the difference between the current time and the opening time of the subordinate inlet is taken as the local material age. When U is greater than the circumferential deflection amplitude threshold Uth, the range of the underfill azimuth angles calculated by the nw cross-sections is not greater than the azimuth consistency threshold. Furthermore, the age of local materials and the total delay of the action. The sum is less than the gel duration When the current injection port's electric shut-off valve is closed, injection is paused. The rotary drive mechanism rotates the pipe around its own axis. The rotation angle is the negative value of the underfill azimuth angle, with an absolute value not exceeding 180°. Injection resumes after feedback from the pipe angle detection unit. The negative value of the underfill azimuth angle rotates the underfill position to the point of lowest gravity, causing subsequently injected liquid material to preferentially flow towards the originally underfilled side under gravity. The azimuth consistency criterion is used because when the underfill azimuths of different sections are discrete, overall rotation is beneficial to some pipe sections but detrimental to others; in this case, it is not advisable to operate. Pausing injection before rotation avoids continued material injection during rotation, preventing new flow deviations, and also facilitates the follow-up movement of the injection hose and sensor cable. Uth is preferably the value determined by calibration that corresponds to half the allowable fluctuation of the circumferential density difference according to the relevant product standard. The rotation speed is preferably not greater than 3 r / min, and the angle positioning accuracy is preferably better than ±5°. If any of the above conditions are not met, rotation is not performed; only U and... .
[0051] Step S6 involves segmented injection and venting pressure control, comprising sub-steps S61 to S66. In sub-step S61, during the advancement phase, the electrically operated shut-off valves of the injection ports are opened sequentially from the injection end. When the leading edge reaches the measurement section of the next injection port, the current injection port is closed and the next injection port is opened. When the leading edge reaches the end measurement section, the advancement phase is considered complete, and the process transitions to the compaction phase. In sub-step S62, the additional injection mass at the k-th injection port, calculated from the compaction phase, is determined using the following formula. ; In the formula, mk is the additional injection mass (kg) from the k-th injection port starting from the compaction stage; max is the maximum value calculation; A is the cross-sectional area of the annular interlayer (m²). 2 Lk is the length (m) of the pipe section under the jurisdiction of the injection port. Target foam core density (kg / m³) 3 ); This is the core density conversion factor; The equivalent forward propulsion density (kg / m³) of the pipe section under the jurisdiction of the injection port. 3 The above formula is derived from the fact that, in order to achieve the core density measured by cross-section... The average compaction density of the entire cross section should reach and The volume of the section under its jurisdiction is the product of A and Lk, and the mass deposited in the section during the advancement phase is... The product of A and Lk, and the difference between the two, represents the additional mass required during the compaction stage. The purpose of the maximum value calculation is to provide a zero additional quantity (not a negative value) when the measured expansion capacity has reached or exceeded the target, ensuring that the control command can be executed under any input. All input quantities in the above formula are either actual measurements or calibrated quantities, excluding fitting coefficients. Pipe sections with weak expansion capacity automatically receive more additional material, while pipe sections with strong expansion capacity automatically receive less. The practice of uniform overfilling throughout the pipe is thus replaced by segment-by-segment, on-demand material replenishment, resulting in a simultaneous decrease in material consumption and peak pressure within the cavity.
[0052] In sub-step S63, the larger of the measured values from the two pressure detection points is taken as the interlayer pressure control value. The difference between this value and the exhaust pressure adjustment value `preg` is used as the deviation. The opening of the electric proportional exhaust valve is adjusted according to the incremental proportional-integral control law, and the opening is not lower than the set minimum opening to ensure continuous gas discharge. An alarm is triggered when the difference between the measured values from the two pressure detection points exceeds the set value. The larger value is taken as the control value because bulging always occurs first at the end with higher pressure; the pressure difference between the two ends can be used to diagnose end-sealing failure or exhaust port blockage. In sub-step S64, the upper limit of the allowable interlayer pressure for the circumferential stress of the outer protective tube is calculated according to the following formula. ; In the formula, pmax is the upper limit of the interlayer pressure allowed by the circumferential stress of the outer sheath (Pa); e is the wall thickness of the outer sheath (m). Let be the short-term allowable circumferential stress (Pa) of the outer sheath at the current wall temperature; and Di be the inner diameter (m) of the outer sheath. The above formula is derived from the circumferential stress relationship of a thin-walled cylinder under internal pressure in mechanics of materials. This relationship states that the circumferential stress equals the product of the internal pressure and the diameter divided by twice the wall thickness. This is obtained by setting the circumferential stress to the allowable value and then solving for the internal pressure. The smaller of the product of the allowable pressure and the safety factor measured by the pressure resistance test of the assembly consisting of the outer sheath and the sealing end plug, and pmax, is taken as the safe cut-off pressure pcut. The safety factor is preferably no greater than 0.8, and pcut is greater than preg. The smaller of the experimental value and the analytical upper limit is chosen because the thin-walled relationship does not take into account the ellipticity, openings, end constraints, and softening due to temperature rise of the outer sheath. Using these factors alone is insufficient to determine the safety boundary, while the pressure resistance test value directly corresponds to the actual assembly.
[0053] In sub-step S65, three independent stopping conditions are set. Normal stopping occurs when the cumulative injected mass from the compaction stage at the k-th injection port reaches mk, and the leading edge has been determined to have arrived at all measuring sections under the jurisdiction of the injection port. The electric shut-off valve of the injection port is closed. Pressure protection occurs when the interlayer pressure control value is not less than pcut, the pump is immediately stopped, the electric proportional exhaust valve is fully opened, and an anomaly is recorded. Local material age is not less than... The injection is stopped and the pipe section is marked; this is a timeout protection measure. The three conditions belong to the quality endpoint, safety boundary, and time boundary, respectively, corresponding to different physical meanings and different subsequent treatments. If they are combined into a single condition that stops injection upon the achievement of any one of them, the exhaust pressure regulation value simultaneously becomes the injection stop threshold. The pressure closed loop is terminated and loses its regulatory function when the regulation target is reached, and the temperature signal at the far end will cause unrelated pipe sections to stop injection prematurely. In sub-step S66, after injection stoppage, the electric proportional exhaust valve maintains the closed-loop regulation of sub-step S63 until the interlayer pressure control value is lower than the closing threshold poff and remains there for a set time before closing; poff is less than preg. After injection stoppage, the material continues to react, releasing heat and expanding. The interlayer pressure does not immediately disappear with pump stoppage; controlled exhaust is maintained until the pressure drops automatically, preventing a secondary pressure rise after the exhaust channel is closed.
[0054] Step S7 involves two levels of adaptive update and file output, comprising sub-steps S71 to S74. In sub-step S71, the estimated total injection time for this root canal is subtracted from the current calculation completion time, and then the total action delay is subtracted. The remaining executable time. If the remaining executable time is greater than zero, the quality of the additional injection material that has not yet been executed in this root canal is corrected in real time according to the equivalent forward advance density obtained in step S4; otherwise, the additional injection material quality allocation result of the previous root canal is used and only the data is recorded. The remaining executable time is calculated from the current calculation completion time and the total action delay is deducted to ensure that the activated correction has sufficient time to be executed.
[0055] In sub-step S72, the foam core density is measured by cutting and sampling the first tube and the sampled tubes of each batch. The average compaction density is calculated from the total mass of the injected material and the volume of the interlayer. Update the core density conversion factor according to the following formula: ; In the formula, This is the updated core density conversion factor; The conversion factor for core density before the update; The relaxation factor is preferably between 0.2 and 0.5; The density of the foam core (kg / m³) was measured by cutting and sampling. 3 ); The average compaction density (kg / m³) 3 The above formula is a first-order exponential weighted average. The smaller the relaxation factor, the smoother the response to the results of a single sampling, and the stronger the ability to resist measurement fluctuations. Updates are only made when there are actual test results available, which is different from additional real-time quality corrections based solely on process data. The former updates on a batch basis, while the latter updates on a single tube basis.
[0056] In substep S73, the arithmetic mean of the effective equivalent forward advance densities of each root canal is taken. Adjust the material temperature setpoint for the next pipe using the following formula: ; In the formula, Set the material temperature setting (°C) for the next pipe; Set the material temperature (°C) for this root canal. Material temperature correction factor (K·m) 3 / kg), take the reciprocal of the material temperature sensitivity slope s obtained in sub-step S15; The equivalent forward propulsion density (kg / m³) of the calibration tube at the reference material temperature. 3 ); This is the arithmetic mean of the effective equivalent forward advance density of each root canal (kg / m²). 3 The above formula is derived from a local linear relationship. The change in the equivalent leading density is equal to the product of the temperature sensitivity slope s and the temperature change. To make the equivalent leading density of the next pipe... Back The required density change is minus Dividing both ends by s gives the required change in material temperature, and the reciprocal of s is... Because the expansion ratio increases with rising material temperature, the equivalent forward density decreases, resulting in a negative s value. Consequently, the value becomes negative. Therefore, when the equivalent leading edge density of this root canal is higher than the reference value, i.e., the material activity is low, the above formula gives a positive temperature increment, causing the material temperature to rise. The control direction is consistent with the physical process. The absolute value of the correction is preferably no greater than 3K, and the corrected set value does not exceed the calibration material temperature range determined in sub-step S15, so as to avoid extrapolation outside the calibration relationship.
[0057] In sub-step S74, the equivalent leading edge advance density, injection mass at each injection port, circumferential flow amplitude, underfill azimuth angle, rotation angle, and interlayer pressure curve for each pipe segment are archived according to pipe number. Pipe segments with equivalent leading edge advance density exceeding the target zone indicated by the section inspection results are marked as suspected abnormal sections according to their axial coordinates. Because the equivalent leading edge advance density is a process equivalent quantity, the marking results serve as the location basis for subsequent non-destructive testing or section verification, rather than being directly determined as defects.
[0058] This invention relates to a data acquisition and control system for pipe interlayer insulation foam filling, such as... Figure 5As shown, it includes a non-invasive temperature sensing array unit, a sandwich pressure detection unit, a material mass metering unit, multiple injection ports with electric shut-off valves, an electric proportional exhaust valve, a rotary drive mechanism, a pipe angle detection unit, a multi-channel synchronous data acquisition unit, a leading edge identification module, a propulsion density estimation module, a flow deviation analysis module, a rotation compensation control module, a segmented injection control module, an exhaust and safety interlock module, and an adaptive update and anomaly marking module.
[0059] The non-invasive temperature sensor array unit is detachably mounted on the outer surface of the outer sheath and distributed axially across multiple measurement sections. Each measurement section has a temperature sensor at three circumferential positions, its function being to pick up the temperature rise signal transmitted through the wall during the exothermic reaction without entering the interlayer. The interlayer pressure detection unit consists of two pressure transmitters located at the sealed end plugs at both ends of the tube body. Both transmitters' outputs are connected to an exhaust and safety interlock module, providing lag-free input for pressure regulation and safety shut-off and supporting differential value diagnosis between the two ends. The material mass metering unit consists of two mass flow meters located on the two material paths of the foaming machine, providing direct measurement values for mass attribution and density estimation. Multiple injection ports are axially spaced on the outer sheath and located at corresponding measurement sections, each equipped with an electric shut-off valve and interlocked with each other, ensuring that only one injection port supplies material to the interlayer at any given time. An electric proportional exhaust valve is located on the sealed end plug furthest from the first injection port, forming the only controlled exhaust channel for the interlayer gas phase. The rotary drive mechanism drives the tube to rotate around its own axis. The tube angle detection unit directly detects the tube angle and feeds it back to the rotary compensation control module. Its function is to avoid angle distortion caused by slippage when the roller angle is used to replace the tube angle.
[0060] The multi-channel synchronous data acquisition unit is connected to the non-invasive temperature sensing array unit, the interlayer pressure detection unit, and the material mass metering unit. It synchronously samples each channel according to a unified clock, providing a common time base for all differential operations. The leading edge identification module is signal-connected to the multi-channel synchronous data acquisition unit, outputting the arrival time of the leading edge at each measuring point according to the dual-threshold continuous criterion and fitting validity condition. The propulsion density estimation module is signal-connected to the leading edge identification module, outputting the equivalent leading edge propulsion density for each pipe segment. The eccentric flow analysis module is signal-connected to the leading edge identification module, continuously outputting the circumferential eccentric flow amplitude and underfill azimuth angle. The input of the rotation compensation control module is connected to the eccentric flow analysis module and the pipe angle detection unit, and the output is connected to the rotation drive mechanism and the electric shut-off valve, thus organizing the three actions of pausing injection, rotating to position, and resuming injection into a coherent action sequence. The input of the segmented injection control module is connected to the propulsion density estimation module and the material mass metering unit, and the output is connected to the electric shut-off valve, realizing the sequential switching of the injection port and the execution of additional injection mass. The input of the exhaust and safety interlock module is connected to the interlayer pressure detection unit, and the output is connected to the electric proportional exhaust valve and the foaming machine pump stop control terminal. It has internally set exhaust pressure adjustment values and safety cut-off pressures to ensure that normal adjustment and safety protection do not interfere with each other, and to continue exhausting after injection stops until the pressure drops. The input of the adaptive update and anomaly marking module is connected to the propulsion density estimation module, and the output is connected to the segmented injection control module and the foaming machine material temperature control terminal. It is responsible for updating the core density conversion coefficient and material temperature setpoint, as well as marking and archiving suspected abnormal sections.
[0061] Example 1: Example 1 uses a DN300 prefabricated direct-buried insulated pipe for heating. The working steel pipe is a φ325mm×7mm Q235B steel pipe, and the outer protective pipe is a φ450mm×7.4mm PE80 high-density polyethylene pipe. The inner diameter of the outer protective pipe is... The outer diameter of the steel pipe is 0.4352m. The diameter is 0.325m and the pipe length is 12m. The material is a combination of polyether containing cyclopentane physical foaming agent and polymethylene polyphenyl polyisocyanate, with a mass ratio of 1:1.10. The temperature of the combined material is set at 22℃ and the preheating temperature of the pipe body is 40℃.
[0062] Arrange the sensing and actuating components according to step S1 and complete the calibration. Following sub-step S11, arrange the components along the axial direction at intervals... Twelve measurement sections are set up for a 1.0m section. The first measurement section is 0.4m from the injection end. There are three injection ports, located at the 1st, 5th, and 9th measurement sections respectively. The lengths of the pipe sections under their jurisdiction are 4.0m, 4.0m, and 3.8m respectively. At each measurement section, one measuring point is set at each of the three circumferential angles: 0°, 120°, and 240°. The temperature sensor is a Class A thin-film platinum resistance thermometer, which is bonded to the outer surface of the outer protective tube with thermally conductive silicone grease and clamped tightly. According to sub-step S12, a pressure transmitter is installed at each of the two sealed end plugs, an electric proportional exhaust valve is installed at the end sealed end plug, an electric shut-off valve is installed at each injection port, and a Coriolis mass flow meter is installed on each of the two material paths. According to sub-step S13, the thermal hysteresis value of the outer protective tube wall is calibrated at each measuring point by inserting a calibration tube with a drilled hole into the armored thermocouple. The outer protective tube wall thickness is measured at each measuring point. According to sub-step S14, the core density conversion factor is calibrated by cutting and sampling. The value was 0.92. Following sub-step S15, cup foaming tests were conducted at three material temperature points, and the gel duration at 22°C was measured using the fiber drawing method. The time was 45 seconds, and the slope of the material temperature sensitivity was obtained by regression. =−0.60kg·m -3 ·K -1 According to sub-step S16, the total measured motion delay was... The value is 6s, and the cross-sectional area of the annular interlayer is calculated. It is 0.0658m 2 Target foam core density Take 62kg / m 3 .
[0063] Data is collected according to step S2. Sub-step S21 sets the sampling period to 0.1s. Sub-step S22 sets the filter window length to 2s. Sub-step S23 interlocks the electric shut-off valves of the three injection ports. Sub-steps S24 and S25 assign the cumulative injection mass to each injection port after being shifted according to pipeline retention delay. Injection is stopped and an alarm is triggered when the ratio deviation between the two material paths exceeds the limit.
[0064] Identify the arrival time of the leading edge in step S3. In sub-step S31, set the regression window length to 3 seconds. In sub-step S32, set the temperature rise threshold. Take 0.5K as the temperature rise rate threshold. Take 0.03 K / s as the criterion duration. Take 1.5s. Following sub-steps S33 to S35, perform regression analysis on the baseline segment of 5s and the temperature rise segment of 3s respectively, and take the intersection of the two straight lines, in conjunction with the slope difference threshold, the coefficient of determination threshold, and the backtracking rule. The measured thermal hysteresis values of the outer protective pipe wall at each measuring point are between 13.4s and 16.0s, with an average of 14.6s.
[0065] The segmented quantities are obtained according to step S4. According to sub-step S41, the arrival time of each measuring point is subtracted from the thermal hysteresis value of its outer protective pipe wall, and the average value is taken to obtain the corrected arrival time of each measuring section. According to sub-steps S42 to S44, the forward axial velocity is approximately 0.48 m / s, and the equivalent forward propulsion density of each pipe segment is approximately 38 kg / m³. 3 .
[0066] Perform flow analysis and rotation compensation in step S5. Then, adjust the number of rolling window sections according to sub-steps S51 to S53. Take 3. Following sub-steps S54 and S55, determine the threshold value for the circumferential deflection amplitude. Set the azimuth consistency threshold to 0.35s. The rotation angle is set to 40°, the rotation speed is set to 2 r / min, and the angle positioning accuracy is better than ±5°. The initial rotation judgment output is at 18.8s, and the rotation action is completed in 24.8s.
[0067] Perform segmented material injection and venting control according to step S6. According to sub-step S61, the advancement phase ends at 22.9 seconds. According to sub-step S62, the target average compaction density is 67.4 kg / m³. 3 The additional feed masses at the three feed ports are 7.74 kg, 7.74 kg, and 7.35 kg respectively, totaling 22.8 kg. The exhaust pressure is adjusted according to sub-steps S63 and S64. Taking 0.045 MPa as the allowable pressure for the component's pressure resistance test, and a safety factor of 0.75, the calculated upper limit of the allowable interlayer pressure for the circumferential stress of the outer protective tube is 0.102 MPa, and the safety cut-off pressure is... Set the pressure to 0.075 MPa. According to sub-steps S65 and S66, set the three types of injection stop conditions respectively, stopping injection at 41.9 s, and then maintaining controlled venting until the pressure is below the shut-off threshold. .
[0068] Perform two-level updates and file output according to step S7. Determine the remaining executable time and enable real-time correction for this root canal according to sub-step S71. Adjust the relaxation factor according to sub-step S72. Take 0.35. According to sub-step S73, the material temperature correction factor... Take −1.67 K·m 3 / kg, with a correction limit of ±3K. Output the segmented density profile and coordinates of suspected anomalous sections according to sub-step S74.
[0069] Example 2 uses a DN200 insulated pipe. The working steel pipe is φ219mm×6mm, the outer protective pipe is φ315mm×5.8mm, and the annular sandwich cross-sectional area is... It is 0.0346m 2There are two injection ports. All parameters are taken as the lower limit of the range defined in this invention, and the measurement cross-sectional spacing is... A measurement depth of 0.8m was used, with 14 measurement sections, and the temperature rise threshold was set. Take 0.3K as the temperature rise rate threshold. Take 0.02 K / s as the criterion duration. Take 1.0s, number of scroll window sections Take 3, relaxation factor The value is set to 0.2, and the safety factor is set to 0.7. The average thermal hysteresis value of the outer protective pipe wall is 9.0 s, and the gelation duration is... The total delay is 40 seconds. The time is 5 seconds, the forward axial velocity is 0.456 m / s, the first rotation is completed in 17.5 seconds, and the executable margin is 22.5 seconds. The rest is the same as in Example 1.
[0070] Example 3 uses a DN600 insulated pipe. The working steel pipe is φ630mm×9mm, the outer protective pipe is φ800mm×8.5mm, and the annular sandwich cross-sectional area is... It is 0.1698m 2 There are four injection ports. All parameters are taken at the upper limit of the range defined in this invention, and the measurement cross-sectional spacing is... Use a 1.5m diameter, measure 8 cross-sections, and set a temperature rise threshold. Take 1.0K as the temperature rise rate threshold. Take 0.06 K / s as the criterion duration. Take 3.0s, number of scroll window sections Take 5 as the relaxation factor. A value of 0.5 is used, with a safety factor of 0.8. Due to the large pipe diameter and long flow distance, the gel duration is used instead. The formula is slow-acting with a 65s time limit. The average thermal hysteresis value of the outer protective tube wall is 19.3s, and the total action delay is... The time is 8s, the forward axial velocity is 0.573m / s, the first rotation is completed in 37.8s, and the executable margin is 27.2s. The rest is the same as in Example 1.
[0071] Example 4 uses a DN400 insulated pipe. The working steel pipe is φ426mm×7mm, the outer protective pipe is φ560mm×9mm, and the annular sandwich cross-sectional area is... It is 0.0882m 2 There are 3 injection ports. All parameters are taken as average values, and the distance between the measured sections is considered. Use a 1.2m diameter, measure 10 cross-sections, and set a temperature rise threshold. Take 0.6K as the temperature rise rate threshold. Take 0.04 K / s as the criterion duration. Take 2.0s, number of scroll window sections Take 4, relaxation factor A value of 0.35 was used, with a safety factor of 0.75. A thin-film thermocouple was used to replace the temperature sensor, verifying the substitutability of the temperature sensing element. The average thermal hysteresis value of the outer sheath wall was 21.6 s, and the gel duration... The total delay is 50 seconds. The time is 6s, the forward axial velocity is 0.537m / s, the first rotation is completed in 34.3s, and the executable margin is 15.7s. The rest is the same as in Example 1.
[0072] The pipe type, materials, proportions, material temperature, preheating temperature, foaming machine and operators in the following comparative examples are the same as in Example 1. Anything not mentioned is the same as in Example 1.
[0073] Comparative Example 1 serves as a blank control for existing technologies. It does not have an external wall temperature sensing array. The total mass of the injection material is given at one time according to the geometric volume of the interlayer and the target density. The vent holes are opened and closed in a fixed sequence. The tube rotates continuously at a fixed speed during foaming without determining the rotation angle based on the measurement results.
[0074] Comparative Example 2: The double-line intersection point refinement and fitting validity criterion in sub-steps S33 to S35 are cancelled. The judgment point in sub-step S32 is directly used as the arrival time of the forward, which is used to isolate the role of intersection point refinement in step S3 of the present invention.
[0075] Comparative Example 3, the rolling window in sub-step S53 is changed to wait for all measurement sections to arrive before taking the average value of the entire pipe and performing a deflection analysis and rotation determination again, in order to isolate the timing effect of the rolling analysis in step S5 of the present invention.
[0076] Comparative Example 4: The rotation compensation action in sub-step S55 is cancelled, and only the circumferential deflection amplitude and underfill azimuth angle are recorded to isolate the effect of the rotation action in step S5 of the present invention.
[0077] Comparative Example 5: The three types of injection stop conditions in sub-step S65 are combined into one where injection stops immediately upon any successful injection, and the venting pressure is adjusted accordingly. Simultaneously, it serves as a threshold for stopping injection, and the electric proportional exhaust valve is immediately closed after injection stops, thus isolating the boundary separation and controlled exhaust after injection stops in step S6 of this invention.
[0078] Comparative Example 6 uses a single constant of 14.6s for the outer protective tube wall as the thermal hysteresis value, without performing point-by-point calibration and correction, to isolate the effects of sub-steps S13 and S41 of the present invention.
[0079] Comparative Example 7: The material temperature correction in sub-step S73 is formally rewritten as the average value of this root canal minus the reference value, so that the sign of the correction direction does not match the slope of the material temperature sensitivity, which is used to isolate the effect of the material temperature feedforward direction in step S7 of this invention.
[0080] Experimental Example 1: Experimental Example 1 uses the methods of Examples 1 to 4 and Comparative Example 6 to investigate the sources of thermal hysteresis in the outer protective pipe wall and its impact on the arrival time of the leading edge. The experimental results are as follows: Figure 1 As shown.
[0081] The outer protective tube is made of high-density polyethylene with a thermal conductivity of 0.40 W / (m·K) and a density of 950 kg / m³. 3 With a specific heat capacity of 1900 J / (kg·K), the thermal diffusivity is obtained. 2.216×10 -7 m 2 / s. The exothermic effect of the foam wall layer causes the inner wall temperature of the outer sheath to rise by approximately 60K. The one-dimensional transient thermal conductivity within the outer sheath wall depends only on the combination of wall thickness, thermal diffusivity, and time. Therefore, the time corresponding to any characteristic point on the temperature rise curve is proportional to the square of the wall thickness and is independent of the selection of the characteristic point. The wall thickness of the outer sheath was measured at each measuring point. The wall thickness of the finished outer sheath ranges from 7.0mm to 8.2mm. The measured standard deviation of the wall thickness between two adjacent points 1.0m along the axial direction of the same outer sheath was 0.03mm, and the measured standard deviation of the wall thickness between three locations on the same measuring section was 0.145mm.
[0082] Figure 1 In the middle (a), the value for determining the arrival time of the front varies with the thickness of the outer protective tube wall. The solid line represents the double-straight-line intersection method of sub-steps S33 to S35 of the present invention. The four dashed lines represent the results of the pure threshold method used in Comparative Example 2 when the temperature rise amplitude threshold is 0.3K, 0.5K, 0.6K and 1.0K. The four marked points on the solid line are the working points of Examples 1 to 4, and their outer protective tube wall thermal hysteresis values are 14.6s, 9.0s, 19.3s and 21.6s, respectively. Figure 1 In (b), the standard deviation of the arrival time or arrival time difference is given in five cases. The first item is the deviation of the arrival time of a single measuring point relative to the actual arrival time in the interlayer when no difference is made and no correction is made. The second and third items are the axial time difference residuals and circumferential time difference residuals when only adjacent cross-sections are differentiated in Comparative Example 6 without point-by-point correction. The fourth and fifth items are the axial time difference residuals and circumferential time difference residuals after point-by-point correction in Example 1.
[0083] Depend on Figure 1As can be seen in (a), the threshold arrival time increases with the square of the outer sheath wall thickness, from approximately 13.1s to approximately 17.9s in the wall thickness range of 7.0mm to 8.2mm, and from 9.0s to 21.6s in the wall thickness range of 5.8mm to 9.0mm in Examples 1 to 4, a span of 12.6s. The four dashed lines are separated from each other and move upward as the threshold increases, indicating that the threshold method determines the timing based on both the wall thickness and the threshold value. Although Examples 1 to 4 used four different temperature rise threshold values of 0.5K, 0.3K, 1.0K, and 0.6K, respectively, their operating points all fall on the same solid line.
[0084] Depend on Figure 1 As can be seen from (a), the thermal hysteresis of the outer protective pipe wall must be calibrated point by point. Taking a single constant will inevitably introduce a systematic deviation that is proportional to the wall thickness deviation. After adopting the double-straight-line intersection method, the thermal hysteresis calibration value and the threshold value are decoupled, and the same set of calibration values can be used universally within the entire threshold range defined in this invention.
[0085] The measurement point calibration defined in sub-step S13 and the intersection point refinement defined in sub-steps S33 to S35 of this invention have self-similarity in the solution of one-dimensional transient heat conduction. The shape of the temperature rise curve is uniquely determined by the combination of wall thickness, thermal diffusivity and time. Therefore, when the wall thickness increases, the entire curve is stretched along the time axis in a square relationship, and the thermal hysteresis increases squarely accordingly. The pure threshold method takes the point where the curve crosses a certain given level. When the level is raised, the point position moves backward along the stretched curve, and the judgment time changes with both the wall thickness and the threshold. The double straight line intersection method takes the intersection of the baseline segment and the temperature rise segment two regression lines. Its position is extrapolated from the overall trend of the temperature rise segment and is basically independent of which point on the curve is taken as the judgment starting point. Therefore, it only changes with the wall thickness and not with the threshold.
[0086] Depend on Figure 1 As can be seen in (b), the standard deviation of the arrival time deviation without any processing is 1.37s. In Comparative Example 6, when only adjacent cross-sections are differentiated without point-by-point correction, the axial time difference residual is reduced to 0.17s, a decrease of 88% relative to 1.37s, while the circumferential time difference residual is still 0.81s, a decrease of only 41%. After point-by-point correction in Example 1, both the axial and circumferential time difference residuals are reduced to 0.07s, a further reduction of 59% and 91% respectively compared to Comparative Example 6.
[0087] Figure 1(b) indicates that common mode cancellation and point-by-point correction have different effects in the axial and circumferential directions, and the two cannot be substituted for each other and must be used simultaneously. The adjacent section difference based on step S4 of this invention and the point-by-point correction defined in sub-step S41 are based on the same outer sheath being continuously extruded from the same mold. The wall thickness of two adjacent points 1.0m along the axial direction is extremely close, and their thermal hysteresis is approximately equal. When the arrival times of adjacent sections are subtracted, it is eliminated as a common modulus, so the axial time difference residual is greatly reduced. However, the extruded outer sheath has eccentricity in the circumferential direction. The standard deviation of the wall thickness between the three directions of the same measurement section is about 4.8 times that of the axial direction. Their thermal hysteresis is not equal and constitutes a differential modulus. The differential operation cannot eliminate it. Only by subtracting the calibration value of each measurement point can it be eliminated. This difference is particularly important for the present invention because the actual circumferential arrival time difference in Example 1 is on the order of 0.3s to 1.5s. If no point-by-point correction is performed, the 0.81s pseudo-time difference of wall thickness is on the same order of magnitude as the actual flow deviation. The first harmonic analysis of sub-steps S51 to S54 of the present invention will misjudge the wall thickness eccentricity as material flow deviation, and thus give an incorrect rotation direction.
[0088] Experimental Example 2, using the methods of Examples 1, 2, 3 and Comparative Example 2, examines the sensitivity and repeatability of identifying the arrival time of the leading edge to the threshold value of the temperature rise. The experimental results are as follows: Figure 2 As shown.
[0089] Under the outer tube wall thickness of Example 1, the outer surface temperature rise curve was synthesized according to the heat transfer parameters of Experimental Example 1. Gaussian white noise equivalent to the measured noise level was superimposed, and the noise standard deviation was taken as the measured value after filtering, which was 0.01℃. For each threshold value, 30 Monte Carlo calculations were performed. Six values of temperature rise amplitude threshold were taken between 0.3K and 1.0K. The arrival time of the front was determined by the pure threshold method of Comparative Example 2 and the double straight line intersection method of Example 1, respectively. The index was the mean and standard deviation of the arrival time calculated by 30 times.
[0090] Figure 2 The horizontal axis represents the temperature rise threshold, and the vertical axis represents the arrival time of the front. The two curves with error bars represent the pure threshold method of Comparative Example 2 and the double straight line intersection method of Example 1, respectively. The half length of the error bar is the standard deviation of 30 calculations. The threshold working points used in Example 2, Example 1 and Example 3 are marked with additional markers in the figure.
[0091] Depend on Figure 2It can be seen that the arrival time of Comparative Example 2 monotonically increases from 15.68s to 21.56s, drifting by 5.88s with the threshold value. Its standard deviation of 30 calculations is between 0.24s and 0.30s, and decreases slightly as the threshold increases. The arrival time of Example 1 is between 14.55s and 14.75s, drifting by only 0.20s with the threshold value, which is 3.4% of the drift of Comparative Example 2. Its standard deviation is between 0.04s and 0.06s, which is about 82% lower than that of Comparative Example 2. The three threshold operating points of Examples 2, 1, and 3 fall on the same nearly horizontal curve, and the difference in their arrival times does not exceed 0.20s.
[0092] Depend on Figure 2 It can be concluded that the arrival time obtained by the pure threshold method is not the objective arrival time of the material, but a function of the threshold value. If the threshold is adjusted under different pipe diameters or different signal-to-noise ratios, the calibration value becomes invalid. The arrival time obtained by the double-line intersection method is almost independent of the threshold value, and the random repeatability is also improved by about 5 times. Within the threshold range defined by different pipe diameters, the same set of thermal hysteresis calibration values can be used. In the piecewise regression and intersection refinement defined by sub-steps S33 to S35 of the present invention, the pure threshold method takes a single sampling point where the temperature rise curve crosses a certain given level. If the threshold is raised, the intersection point will inevitably shift backward along the curve. At the same time, the slope at the crossing point is limited, and the noise is directly converted into time error by dividing the noise amplitude by the slope at this point. The quantization error of the sampling period also cannot be averaged. The double-line intersection method first performs least-squares regression on dozens of sampling points in both the baseline segment and the temperature rise segment. Noise is reduced by the square root of the number of sampling points. Then, the intersection point is calculated from the two regression lines. The position of the intersection point is determined by the overall trend of the two data segments. Changes in the position of the decision point only alter the starting point of the temperature rise segment, while the overall slope and intercept of the temperature rise segment within its sampling length change very little, thus the intersection point position is stable. Furthermore, the slope difference threshold and determination coefficient threshold defined in sub-step S34 of this invention ensure that the denominator of the intersection formula does not approach zero, and the backoff rule defined in sub-step S35 ensures that there is still a definite output when the fitting fails. Together, these two factors ensure that the above-mentioned accuracy can be consistently obtained under engineering conditions.
[0093] Experimental Example 3, using the methods of Example 1, Comparative Example 3, and Comparative Example 4, investigated the feasibility of rotation compensation within the gelation period and its effect on circumferential density uniformity. The experimental results are as follows: Figure 3 As shown.
[0094] In Example 1, the forward axial velocity was 0.48 m / s, the calibration arrival times of the 12 measurement sections ranged from 0 s to 22.9 s, the mixing start times of the three injection ports were 0 s, 8.3 s, and 16.7 s respectively, the average thermal hysteresis value of the outer protective tube wall was 14.6 s, the total action delay was 6 s, and the gel duration was 45 s. The executable margin of the rotation action was defined as the gel duration minus the time from the mixing start time of this injection port to the completion time of the rotation action. A positive margin indicates that the action can be completed before the material loses its fluidity. The propulsion phase ended at 22.9 s, and the total additional injection mass of the three injection ports was 22.8 kg. Injection stopped at 41.9 s at a rate of 1.2 kg / s. The remaining compaction injection time after rotation was defined as the injection stop time minus the rotation completion time. The circumferential foam core density range was measured by sampling at three circumferential positions along each measurement section. A total of 36 points were taken from 12 sections for each pipe, and the range of three points was taken as the circumferential core density range of this measurement section.
[0095] Figure 3 In Figure (a), four time indicators are compared. The first three indicators are the rotational margin of the pipe sections under the jurisdiction of the three injection ports, and the fourth indicator is the remaining compaction injection time after rotation. The two sets of columns are Example 1 and Comparative Example 3, respectively. The zero reference line is marked with a dashed line in the figure. Figure 3 In the middle (b), the circumferential foam core density range of 12 measurement sections along the axial direction is shown. The three curves are Example 1, Comparative Example 3 and Comparative Example 4, respectively.
[0096] Depend on Figure 3 As can be seen in Figure (a), in Example 1, the executable margin for the pipe sections under the jurisdiction of the three injection ports is 20.2s, the rotation action is completed in 24.8s, and there is still 17.1s of compaction injection time afterward. The executable margins for Comparative Example 3 are 1.5s, 9.8s, and 18.1s respectively. The margin for the pipe section under the jurisdiction of the first injection port is only 7% of that in Example 1, the rotation action is completed as late as 43.5s, and the remaining compaction injection time after rotation is −1.6s.
[0097] Depend on Figure 3 As can be seen in Figure (b), the circumferential core density range of Comparative Example 4 starts from 3.1 kg / m³. 3 The weight increased to 7.4 kg / m along the way. 3 The average value for the entire pipe is 5.2 kg / m. 3 The average weight of the entire pipe in Comparative Example 3 was 4.9 kg / m. 3 Only the four measuring sections at the end showed a slight decrease. The circumferential core density range of Example 1 was 2.6 kg / m³. 3 Up to 3.8kg / m 3 Between, the average value for the entire pipe was 3.2 kg / m. 3 Compared to comparative example 4, it decreased by 38%, and compared to comparative example 3, it decreased by 34%.
[0098] Depend on Figure 3 It can be concluded that the rolling analysis advanced the initial executable rotation determination by 18.7 seconds, increasing the executable margin for all pipe sections from barely positive to 20.2 seconds, thus providing an engineering margin for the action. Although the rotation action in Comparative Example 3 was barely completed within the gelation period of the first segment, its completion time was after the injection stop time, resulting in a negative remaining compaction injection time. After the rotation, no more liquid phase material was injected, and the action did not actually produce a filling effect. The rolling window analysis defined in step S5 and sub-step S53 of this invention, and the gelation period criterion and rotation action defined in sub-step S55, cause the outer wall judgment time of each measuring point to lag behind the actual arrival time inside the interlayer by about 14.6s. If, as in Comparative Example 3, the judgment is made after the 12th measuring section arrives, the judgment output time is delayed to 37.5s. After adding the total delay of 6s, the rotation is completed at 43.5s. At this time, the compaction injection has already ended at 41.9s. Even if the underfilled position is rotated to the lowest gravity point, there is no liquid material to flow to this side, and the action fails. After adopting the rolling window, a judgment can be given for the three most recent arrived measuring sections. The first judgment output is advanced to 18.8s, and the rotation is completed at 24.8s. In the following 17.1s, the continuously injected compacted material preferentially flows to the underfilled position that has been rotated to the bottom under the action of gravity, and the circumferential density difference converges. Although Comparative Example 4 also obtained the circumferential flow amplitude and underfill azimuth angle, without performing rotation, the circumferential difference accumulated along the process and could not be eliminated. Its range monotonically increased along the axial direction, indicating that the analytical results must be combined with the execution of actions to be transformed into improvements in product indicators.
[0099] Experimental Example 4 uses the methods of Examples 1 to 4 and Comparative Examples 1 to 7 to verify the comprehensive performance of the insulation layer. The experimental results are as follows: Figure 4 As shown. Each group produces 20 tubes, with all conditions remaining the same except for the technical features to be compared.
[0100] The axial core density was measured at 1.0 m intervals along the pipe length, with 3 points taken from each interval, for a total of 36 points per pipe. The coefficient of variation for the axial core density was calculated by dividing the standard deviation of the density at all measurement points by its mean. The sampling and calculation method for the circumferential core density range was the same as in Experiment Example 3. The standard deviation of the core density for 20 consecutive pipes was calculated by taking the mean of the 36 densities for each pipe and then calculating the standard deviation for all 20 pipes. This was used to characterize the stability between batches. The test environment temperature was 23℃±2℃, and the relative humidity was 50%±10%. Each indicator was repeated 3 times, and the average value was taken.
[0101] Figure 4 In the middle (a), the coefficient of variation of axial core density for each group is represented. Figure 4 (b) shows the range of circumferential core density for each group. Figure 4 (c) represents the porosity of each group. Figure 4In the figure (d), the standard deviation of the core density of 20 consecutive tubes in each group is represented. In each figure, the examples and comparative examples are represented by bars of different colors.
[0102] Depend on Figure 4 As can be seen in (a), the coefficients of variation of axial core density in the four embodiments range from 4.3% to 4.9%, with Comparative Example 1 at 12.4%, Comparative Example 5 at 7.8%, Comparative Example 6 at 7.1%, Comparative Example 2 at 6.4%, Comparative Example 7 at 5.6%, and Comparative Examples 3 and 4 at 4.5% and 4.4% respectively, which are comparable to Example 1. This indicates that the uniformity of axial density is determined by the quality of the segmented additional injection, the accuracy of the arrival time identification, and the correction at each measurement point. Rolling analysis and rotation compensation have virtually no effect on the axial density.
[0103] In sub-steps S62, S33 to S35, and S41 of this invention, the additional injection mass is determined by the difference between the measured equivalent forward propulsion density and the target average compaction density of the pipe section under the jurisdiction of each injection port. Pipe sections with weak expansion capacity automatically receive more material, while pipe sections with strong expansion capacity automatically receive less material. The uniform overfilling of the entire pipe is thus replaced by segment-by-segment on-demand material replenishment. The equivalent forward propulsion density is calculated from the forward axial velocity, which is in turn calculated from the difference in arrival time of adjacent sections. Therefore, the accuracy of arrival time identification and the accuracy of step-by-step measurement point correction directly determine the accuracy of the additional injection mass. Comparative Examples 2 and 6 are inaccurate at these two points, and their segmented material replenishment amount is also inaccurate, resulting in increased axial density dispersion. The high axial variation coefficient of Comparative Example 5 has another reason: its injection stop condition is combined with the exhaust pressure adjustment value. Injection stops as soon as the pressure reaches the adjustment target, and the additional injection mass is often not yet sufficient.
[0104] Depend on Figure 4 As can be seen in (b), the circumferential core density range in the four embodiments is 3.2 kg / m³. 3 Up to 3.6 kg / m 3 Between these two values, Comparative Example 1 is 6.8 kg / m³. 3 Comparative Example 6 is 5.6 kg / m 3 Comparative Example 4 is 5.2 kg / m³ 3 Comparative Example 3 is 4.9 kg / m³ 3 Comparative Example 2 is 4.1 kg / m 3 Comparative Example 5 and Comparative Example 7 were 3.4 kg / m³. 3 With 3.3kg / m 3 This is similar to the example. It shows that circumferential uniformity is determined by three factors: point-by-point correction, rolling analysis, and rotational motion, while pressure boundary separation and material temperature feedforward have virtually no effect on the circumferential direction.
[0105] In the present invention, sub-steps S41, S53 and S55, the circumferential range of Comparative Example 6 is higher than that of Comparative Example 3 and Comparative Example 4. The reason is that after the missing measurement point correction, the wall thickness eccentricity is misjudged as material flow deviation. The rotation action is not only useless, but may also turn the already well filled side to the bottom. Its harm is greater than not rotating.
[0106] Depend on Figure 4 As can be seen in (c), the void ratios of the four embodiments range from 2.5% to 2.9%, compared to 6.8% for Comparative Example 1, 5.6% for Comparative Example 5, and between 3.4% and 4.4% for the remaining comparative examples. The void ratio of Comparative Example 5 is significantly higher, and its measured outer casing bulge rate is 3.1%, while the rate for all embodiments is no higher than 0.5%. This indicates that separating the quality endpoint, pressure regulation target, and safety protection into three independent conditions, and maintaining controlled venting after injection cessation, is essential for simultaneously suppressing underfill and bulging.
[0107] This invention designs sub-steps S63 to S66. The exhaust pressure adjustment value is the normal operating point that the exhaust valve closed loop needs to maintain. If it is also used as the injection stop threshold, once the pressure is adjusted to the target value, injection stop is triggered, and the pressure closed loop immediately loses its adjustment target. The added material mass is not yet sufficient, resulting in insufficient compaction and an increase in porosity. After injection stop, the material continues to react, release heat, and expand. The pressure inside the cavity does not immediately disappear with the pump stopping. If the exhaust valve is closed immediately at this time, the gas phase has nowhere to escape, the pressure rises again, and exceeds the pressure-bearing capacity of the outer protective tube at the current wall temperature, causing local bulging of the outer protective tube. This invention uses the mass endpoint for normal injection stop, uses a safety stop pressure higher than the adjustment value for independent protection, and maintains the exhaust valve closed loop after injection stop until the pressure is lower than the closing threshold. The three boundaries do not interfere with each other, so underfilling and bulging can be suppressed simultaneously.
[0108] Depend on Figure 4 As can be seen in (d), the standard deviation of the core density of 20 consecutive tubes in the four examples is 1.1 kg / m³. 3 Up to 1.3kg / m 3 Between these, the ratio 7 reached as high as 6.9 kg / m 3 Comparative Example 1: 4.2 kg / m 3 Comparative Example 5 is 2.6 kg / m³ 3 The remaining comparative proportions are at 1.3 kg / m². 3 Up to 1.9 kg / m 3 The interval indicates that the correction direction of the material temperature feedforward must match the sign of the slope of the material temperature sensitivity; otherwise, the batch stability will be worse than that of the existing process without feedforward.
[0109] The modification form defined in sub-step S73 of this invention has a negative slope relationship between the equivalent leading density and the material temperature. Increased material temperature increases the reaction rate and the degree of foaming agent vaporization, leading to a larger expansion ratio and a decrease in the equivalent leading density. To bring the equivalent leading density of the next tube back to the baseline value, the required change in material temperature is equal to the difference between the baseline value and the average value of this tube, divided by the material temperature sensitivity slope. Therefore, the modification form takes the baseline value minus the average value of this tube, multiplied by the reciprocal of the slope. Since the slope is negative, the correction coefficient is also negative. After multiplying the two negative signs, a positive temperature increment is obtained when the equivalent leading density is too high. As the material temperature increases, the deviation is offset, and the iteration converges tube by tube. Comparative Example 7 swaps the two terms of the difference while keeping the correction coefficient unchanged, which is equivalent to changing the negative feedback of the closed loop to positive feedback. The deviation is amplified tube by tube, and the material temperature quickly reaches the ±3K limiting boundary and oscillates between the two boundaries, causing the core density to fluctuate significantly between batches.
[0110] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.
Claims
1. A data acquisition and control method for pipe interlayer insulation foam filling, wherein the interlayer is an annular interlayer between a steel pipe and an outer protective pipe, characterized in that, include: S1. Several measuring sections are set along the axial direction on the outer surface of the outer protective tube. Three temperature measuring points are set in the circumferential direction of each section. Pressure detection points are set at the sealed ends. Mass flow rate detection points are set in the two material paths. The thermal hysteresis value of the outer protective tube wall is calibrated point by point by the calibration tube, and the core density conversion factor, gel duration and total action delay are calibrated. S2. From the start of material injection, the outer wall temperature, interlayer pressure and material mass flow rate are collected synchronously using a unified clock. The average temperature before material injection is taken as the baseline temperature, and the material injection mass is assigned according to the opening interval of the injection port. S3. Filter the temperature channel and estimate the temperature rise rate. Determine the candidate arrival interval based on the dual threshold of temperature rise amplitude and rate. When the fit is valid, take the intersection of the baseline segment and the temperature rise segment as the arrival time of the front of the measuring point. Otherwise, backtrack or mark the invalid value. S4. The arrival time is corrected point by point using the thermal hysteresis value of the outer protective pipe wall. The difference in arrival time between adjacent sections, the section spacing, the mass flow rate in the corresponding time window and the interlayer cross-sectional area are used to obtain the equivalent forward propulsion density of each pipe section. S5. For each new effective measurement section, the circumferential deviation amplitude and underfill azimuth angle are analyzed by rolling the correction arrival time of the continuous section. When the deviation amplitude exceeds the threshold, the azimuth angle range does not exceed the set value, and the sum of the local material age and the total delay of the action is less than the gel duration, the injection is paused and the tube is driven to rotate around the axis to turn the underfill azimuth to the lowest point of gravity before the injection is resumed. S6. Based on the equivalent advance density and target compaction average density of the pipe section under the jurisdiction of each injection port, determine the non-negative additional injection mass of the port from the compaction stage. When the mass reaches the target and all sections under its jurisdiction have reached the target, the injection is considered to be stopped normally. Separate pressure safety cut-off and overtime protection are set up. After the injection is stopped, maintain controlled venting until the interlayer pressure drops below the closing threshold before closing the valve. S7. Determine whether to enable real-time correction of the quality of additional material injection in this tube based on the remaining executable time after deducting the total delay of the action, so as to cut and check to update the core density conversion coefficient and the material temperature setting value of the next tube, and mark the suspected abnormal section according to the calibrated target.
2. The data acquisition and control method according to claim 1, characterized in that, Step S1 includes: S11, along the axial direction at intervals Several measurement sections are set up, and each injection port is located at the corresponding measurement section; each measurement section has a temperature measuring point at the 0°, 120° and 240° circumferential positions, and the lowest point of tube weight is taken at the 0° position; the temperature sensor of each measuring point is detachably attached to the outer surface of the outer protective tube without penetrating the interlayer. S12. A pressure detection point is set on each of the two sealing end plugs. An electric proportional exhaust valve is set on the sealing end plug away from the first injection port. An electric shut-off valve is set on each injection port. A mass flow rate detection point is set on each of the isocyanate material path and the combined polyether material path of the foaming machine. S13. Take a calibration tube with the same outer sheath material, wall thickness, and sensor installation method as the product tube. Drill test point holes at each measurement section and insert armored thermocouples. Measure the arrival time of the temperature rise on the inner wall of the outer sheath. Subtract this time from the arrival time of the leading edge determined according to S3 at the same test point to obtain the thermal hysteresis value of the outer sheath wall at that test point. , i is the measurement section number, j is the circumferential orientation number; S14. The foam core density is measured by axially cutting the calibration tube and taking samples. The ratio of this measured density to the average compacted density calculated from the total mass of the injected material and the interlayer volume is used to obtain the core density conversion factor. ; S15. Perform cup foaming tests on the same batch of materials at no less than three material temperature points, and measure the gel duration at each material temperature point using the fiber drawing method. The equivalent advance density at each material temperature point was measured according to S4, and a linear regression was performed between it and the material temperature to obtain the slope of the material temperature sensitivity. ; S16. The total action delay is calculated by summing the time taken from temperature sampling to rotation into position, including filtering, regression, judgment, valve switching, and rotation execution. Calculate the cross-sectional area of the annular interlayer using the following formula: ; In the formula, The cross-sectional area of the annular interlayer; The inner diameter of the outer protective tube; This refers to the outer diameter of the steel pipe, including the thickness of the anti-corrosion coating.
3. The data acquisition and control method according to claim 1, characterized in that, Step S2 includes: S21. All temperature, pressure and mass flow rate channels are sampled synchronously with a sampling period of no more than 0.2s, and the data of each channel are timestamped according to the same clock. S22. Apply a moving average filter with a window length of 2 seconds to each temperature channel to obtain the filtered temperature sequence. The baseline temperature is the arithmetic mean of the sequence within 5 seconds before injection. ; S23. Interlock the electric shut-off valves of each injection port so that only one injection port is in the injection state at any given time. S24. The cumulative injection mass is obtained by integrating the sum of the mass flow rates of the two material paths over time, and the pipeline retention delay after shifting the opening interval of each injection port backward is taken as the belonging interval of that port; the pipeline retention delay is the ratio of the pipeline volume from the mixing head to that port to the material volume flow rate. S25. When the ratio of the mass flow rates of the two material paths deviates from the set ratio by more than the set value, stop feeding and issue an alarm.
4. The data acquisition and control method according to claim 1, characterized in that, Step S3 includes: S31. For each temperature channel, perform a univariate linear regression on the filtered temperature and time within a sliding window of 3 seconds, and take the slope as the temperature rise rate at that time. ; S32, when the difference between the filtered temperature and the baseline temperature is not less than the temperature rise threshold. ,and Not less than the temperature rise rate threshold and maintain continuously When the time is right, the channel is determined to have entered the candidate arrival interval, and this time is recorded as the determination point; S33. Take the temperature sequence 5 seconds before the decision point and perform a univariate linear regression to obtain the baseline straight line, the slope of which is... The intercept is A univariate linear regression was performed on the temperature sequence 3 seconds after the decision point to obtain the straight line of the temperature rise segment, the slope of which is... The intercept is ; S34, when and When the absolute value of the difference is not less than the slope difference threshold, and the determination coefficients of both regressions are not less than the goodness-of-fit threshold, the fit is considered valid. The abscissa of the intersection of the two lines is taken as the arrival time of the front at that measurement point according to the following formula: ; In the formula, This is the arrival time of the forward at this measuring point; The intercept of the baseline segment is the line segment intercept. This is the intercept of the straight line during the temperature rise segment; The slope of the baseline segment is denoted as . The slope of the straight line during the temperature rise segment; S35. If the fitting is invalid or the intersection point falls outside the interval of 5 seconds before to 3 seconds after the judgment point, take the judgment point as the arrival time of the leading edge of the measurement point and set the retreat flag; if there are fewer than two valid measurement points on the same measurement section, mark the section as invalid and do not participate in subsequent calculations.
5. The data acquisition and control method according to claim 4, characterized in that, Step S4 includes: S41, the first The arrival time of the leading edge at each effective measuring point of each measurement section is subtracted from the thermal hysteresis value of the outer protective pipe wall to obtain the corrected arrival time of each measuring point. The arithmetic mean of these corrected arrival times is taken as the corrected arrival time of the section. ; S42, calculate the following formula: The measurement section and the first Axial velocity of the pipe front between the measurement sections: ; In the formula, This represents the axial velocity of the leading edge of the pipe section; This represents the axial distance between adjacent measurement sections; For the first The calibration arrival time of each measurement section; For the first The calibration arrival time of each measurement section; S43, Time Window The arithmetic mean of the mass flow rate at the current injection port. The equivalent forward thrust density of this pipe segment is calculated using the following formula: ; In the formula, This is the equivalent forward thrust density for this pipe section; The mass flow rate of the material belonging to the current injection port within this time window; The cross-sectional area of the annular interlayer; This represents the axial velocity of the leading edge of the pipe section; S44 When the effective velocity range determined by the calibration tube is exceeded, the equivalent forward propulsion density of that tube segment is marked as invalid; take the first... The arithmetic mean of all effective values within the pipe section under the jurisdiction of each injection port is used as the equivalent forward propulsion density of that pipe section. .
6. The data acquisition and control method according to claim 5, characterized in that, Step S5 includes: S51, regarding the first There are three effective measurement sections, and the correction arrival times of the three azimuth measuring points at 0°, 120°, and 240° are recorded as follows: , , The first harmonic cosine coefficient of this cross section is calculated using the following formula: ; In the formula, The first harmonic cosine coefficient of this cross section; , , These are the correction arrival times of the measuring points at the 0°, 120°, and 240° azimuth positions of the cross section, respectively. S52. Calculate the first harmonic sine coefficient of the measured cross section using the following formula: ; In the formula, The first harmonic sine coefficient of this cross section; S53. For each newly added valid measurement section, take the nearest measurement section including that section. One continuous effective measurement section, No less than 3; take its The arithmetic mean is , The arithmetic mean is The amplitude of the circumferential deflection is calculated using the following formula: ; In the formula, This refers to the amplitude of the circumferential deflection flow; For the The mean value of the first harmonic cosine coefficient of each cross section; It is the mean value of its first harmonic sine coefficient; S54. Calculate the underfilled azimuth angle using the following formula: ; In the formula, The underfilled azimuth angle is calculated from the 0° azimuth and ranges from −180° to 180°; atan2 is the four-quadrant arctangent function, with the first independent variable being the sine coefficient and the second independent variable being the cosine coefficient; S55. The difference between the current time and the opening time of the feed inlet under its jurisdiction is taken as the local material age; when Greater than the circumferential deflection amplitude threshold The above The range of the underfilled azimuth angles calculated by S54 for each cross section is no greater than the azimuth consistency threshold. Furthermore, the local material age is related to the total delay of the aforementioned action. The sum is less than the gel duration. When the current filling port's electric shut-off valve is closed, filling is paused. The rotary drive mechanism drives the tube body to rotate around the axis. The rotation angle is the negative value of the underfill azimuth angle, and the absolute value is not greater than 180°. Filling resumes after feedback from the tube body angle detection unit. If any condition is not met, rotation is stopped and the result is recorded. and .
7. The data acquisition and control method according to claim 6, characterized in that, Step S6 includes: S61. During the advancement phase, the electric shut-off valves of the injection ports are opened sequentially from the injection end. When the leading edge reaches the measurement section of the next injection port, the current port is closed and the next port is opened. When the leading edge reaches the end measurement section, the advancement phase is determined to be over and the compaction phase begins. S62. Determine the number according to the following formula. Additional material mass at each injection port, calculated from the compaction stage: ; In the formula, For the first Additional injection mass from each injection port starting from the compaction stage; max is the maximum value calculated. The cross-sectional area of the annular interlayer; This refers to the length of the pipe section under the jurisdiction of that port. Target foam core density; This is the core density conversion factor; This is the equivalent forward advance density for the section under the jurisdiction of this port; S63. Take the larger of the measured values from the two pressure detection points as the interlayer pressure control value, and adjust it by comparing it with the exhaust pressure value. The difference is the deviation. The opening of the electric proportional exhaust valve is adjusted according to the incremental proportional-integral control law, and the opening is not lower than the set minimum opening. An alarm is triggered when the difference between the two measured values exceeds the set value. S64. Calculate the upper limit of allowable interlayer pressure for the circumferential stress of the outer protective pipe using the following formula: ; In the formula, This is the upper limit of the interlayer pressure allowed for the circumferential stress of the outer protective tube; The outer protective pipe wall thickness; For the short-term allowable circumferential stress of the outer protective tube at the current wall temperature; The inner diameter of the outer protective tube; The product of the allowable pressure obtained from the pressure resistance test of the assembly consisting of the outer protective tube and the sealing end plug, and a safety factor not exceeding 0.8, is taken as follows: The smaller of the two is the safe shut-off pressure. ,and Greater than ; S65, Normal suspension of betting is: (The following is a continuation of the previous sentence) The cumulative injection mass of each injection port, calculated from the compaction stage, reaches Furthermore, when the arrival of the leading edge has been determined at all measuring sections under the jurisdiction of the port, the electric shut-off valve of the port shall be closed; the pressure protection is as follows: the interlayer pressure control value shall not be less than Immediately stop the pump and fully open the exhaust valve, then record the abnormality; the timeout protection is as follows: the local material age is not less than Stop injecting and mark the pipe section immediately; S66. After injection is stopped, the electric proportional exhaust valve maintains the closed-loop regulation of S63 until the interlayer pressure control value is lower than the closing threshold. And then turn it off after the set duration. Less than .
8. The data acquisition and control method according to claim 7, characterized in that, Step S7 includes: S71, Subtract the current calculation completion time from the estimated total injection time of this pipe, and then subtract... The remaining executable time is used; if it is greater than zero, the additional injection quality of this pipe that has not yet been executed is corrected in real time according to the equivalent forward advance density of S4; otherwise, the additional injection quality allocation result of the previous pipe is used and only the data is recorded. S72. The density of the foam core is measured by cutting and sampling the first piece of pipe and the sampled pipe in each batch. The average compaction density is calculated from the total mass of the injected material and the volume of the interlayer. Update the core density conversion factor using the following formula: ; In the formula, This is the updated core density conversion factor; The conversion factor for core density before the update; The relaxation factor is set to 0.2–0.
5. The density of the foam core was measured by cutting and sampling. This represents the average compaction density. S73. Take the arithmetic mean of the effective equivalent forward propulsion density of each tube. Correct the material temperature setpoint of the next tube using the following formula: ; In the formula, Set the material temperature setting for the next tube; This is the setpoint for the material temperature of this tube; The material temperature correction factor is taken as the slope of the linear relationship between material temperature and equivalent leading advance density obtained from calibration at no less than three material temperature points. The reciprocal, Since it is a negative value, It is a negative value; The equivalent forward propulsion density of the calibration tube at the reference material temperature; The value is the arithmetic mean of the effective equivalent forward propulsion density of each tube; the absolute value of the correction is not greater than 3°C, and the corrected value does not exceed the calibrated material temperature range; S74. Archive the equivalent front advance density, injection quality at each port, circumferential flow amplitude, underfill azimuth angle, rotation angle and interlayer pressure curve of each pipe section according to pipe number, and mark the pipe sections whose equivalent front advance density exceeds the target zone indicated by the above-mentioned cross-sectional sampling results as suspected abnormal sections according to axial coordinates.
9. A data acquisition and control system for pipe interlayer insulation foam filling, characterized in that, include: A non-invasive temperature sensing array unit that is detachably attached to the outer surface of the outer protective tube, distributed axially at multiple measurement sections, with a temperature sensor at each of the three circumferential directions along each section; A sandwich pressure detection unit consisting of two pressure transmitters located at the sealed end plugs at both ends of the pipe body; A material mass metering unit consisting of two mass flow meters located on two separate material paths of the foaming machine; Multiple injection ports are spaced along the axial direction on the outer protective tube, each equipped with an electric shut-off valve and interlocked with each other; An electric proportional exhaust valve is installed on the sealed end plug at the end furthest from the first injection port; A rotary drive mechanism that drives the tube body to rotate around its own axis, and a tube body angle detection unit that directly detects the rotation angle of the tube body; A multi-channel synchronous data acquisition unit that is connected to the non-invasive temperature sensing array unit, the interlayer pressure detection unit, and the material mass metering unit, and samples each channel synchronously according to a unified clock. A leading edge identification module that is connected to the multi-channel synchronous data acquisition unit and outputs the arrival time of the leading edge of each measuring point according to the dual threshold continuous judgment criterion and fitting validity condition; A propulsion density estimation module that is signal-connected to the forward identification module and outputs the equivalent forward propulsion density of each pipe segment; A flow analysis module that is signal-connected to the forward identification module and continuously outputs the circumferential flow amplitude and underfilled azimuth angle; The input end is connected to the deflection analysis module and the pipe angle detection unit, and the output end is connected to the rotation compensation control module of the rotation drive mechanism and the electric shut-off valve. The input terminal is connected to the propulsion density estimation module and the material mass metering unit, and the output terminal is connected to the segmented injection control module of the electric shut-off valve; The input end is connected to the interlayer pressure detection unit, the output end is connected to the electric proportional exhaust valve and the foaming machine pump stop control end, and the exhaust pressure adjustment value and the safety stop pressure are respectively set in the exhaust and safety interlock module. And an adaptive update and anomaly marking module whose input end is connected to the propulsion density estimation module and whose output end is connected to the segmented injection control module and the foaming machine material temperature control end.
10. The data acquisition and control system according to claim 9, characterized in that, The non-invasive temperature sensing array unit includes a wireless temperature acquisition node that rotates synchronously with the pipe body and is wirelessly connected to the multi-channel synchronous data acquisition unit; the pipe body angle detection unit includes an angle mark on the outer surface of the outer protective pipe and a photoelectric sensor fixed to the frame, the output of the photoelectric sensor being connected to the feedback input of the rotation compensation control module; the rotation drive mechanism includes a frequency conversion drive roller, the angle detection output of which is also connected to the rotation compensation control module, forming an anti-slip dual feedback with the photoelectric sensor; each of the injection ports is located at the corresponding measurement section and connected to the foaming machine mixing head via a rotary joint; the outputs of the two pressure transmitters are both connected to the exhaust and safety interlock module, which takes the larger of the two as the control value and performs redundancy diagnosis on the difference between them.
