Flywheel energy storage vacuum system seal control method

Abstract

A method for controlling the sealing performance of a flywheel energy storage vacuum system includes: placing an O-ring in the sealing groove of the flange sealing surface, placing bolts on the flange sealing surface, and tightening the bolts sequentially using a digital torque wrench, with each bolt tightened in three stages; welding the vacuum chamber and piping connections, performing leak detection after welding cooling, and recording the current leak rate as the initial leak rate; installing a molecular pump and adjusting the connection piping layout between the molecular pump and the vacuum chamber, then installing a vacuum gauge to measure the real-time vacuum level at the center of the vacuum chamber and calculating the real-time leak rate; conducting a flywheel trial run, monitoring and recording the real-time leak rate, and comparing it with the initial leak rate; when the real-time leak rate is greater than the initial leak rate, simultaneously adjusting the corresponding structure to make the real-time leak rate less than the initial leak rate, until the flywheel completes the trial run. This design forms a closed-loop leak rate control through flange sealing control, weld quality control, and pump assembly installation optimization control, resulting in high stability.

Description

Technical Field

[0001] This invention relates to the field of flywheel energy storage technology, and in particular to a method for controlling the sealing performance of a flywheel energy storage vacuum system. Background Technology

[0002] Currently, flywheel energy storage systems need to operate in a vacuum environment to reduce wind resistance losses; for every order of magnitude decrease in vacuum level, wind resistance losses increase by 30%-50%. However, existing flywheel vacuum systems still face the following technical bottlenecks during the construction and installation phase:

[0003] Existing technologies mostly use rubber sealing rings for compression sealing, but the lack of quantitative control over the cleanliness of the sealing surface, the alignment of the O-rings, and the sequence of bolt torque during installation leads to excessive initial leakage rates or leakage due to thermal stress relaxation after operation. Currently, vacuum chamber welding is mostly a single-step process, and welding defects (sand holes, incomplete penetration) are exposed during vacuuming, making subsequent repair welding difficult and costly. Vacuum pump installation lacks precision standards: molecular pumps are extremely sensitive to installation levelness, and existing construction specifications do not clearly define leveling tolerances and quantitative indicators for flow conductance matching, resulting in severe pumping speed losses and failure to meet ultimate vacuum standards.

[0004] For example, Chinese patent CN103527786A discloses a flywheel cryogenic welding vacuum sealing device, which uses a resealable sealing structure to seal and protect the flywheel energy storage vacuum. However, the patent focuses on the resealable structure design and does not involve the control of installation process parameters. Meanwhile, Chinese patent CN118757368A discloses a flywheel energy storage vacuum system and its control method. In the normal operation mode of the flywheel, it switches to the operation of a water ring Roots pump group, and the frequency conversion command of the Roots pump is calculated and output by a PID controller, which changes with the system vacuum degree. This can control the vacuum degree within a suitable and expected range. However, this patent mainly focuses on the redundant switching and frequency conversion control of the vacuum pump group, which is an optimization in the operation stage and does not solve the inherent leakage rate problem in the installation stage. Summary of the Invention

[0005] The purpose of this invention is to overcome the defects and problems of poor leakage control rate during the installation of flywheel energy storage systems in the prior art, and to provide a method for controlling the sealing performance of flywheel energy storage vacuum systems with better leakage control rate.

[0006] To achieve the above objectives, the technical solution of the present invention is: a method for controlling the sealing performance of a flywheel energy storage vacuum system, comprising:

[0007] Place the O-ring in the sealing groove of the flange sealing surface of the vacuum chamber, and place the bolts on the flange sealing surface. Use a digital torque wrench to tighten the bolts in sequence, and tighten each bolt in three stages.

[0008] Welding is performed to connect the vacuum chamber and pipelines. After the welding cools, leak detection is performed and leak repair is carried out. The current leak rate of the vacuum chamber is recorded as the initial leak rate.

[0009] Install a molecular pump and adjust the layout of the connecting pipeline between the molecular pump and the vacuum chamber. Then install a vacuum gauge to measure the real-time vacuum level at the center of the vacuum chamber and calculate the real-time leakage rate.

[0010] The flywheel undergoes trial operation, and the real-time leakage rate is monitored and recorded. At the same time, it is compared with the initial leakage rate. When the real-time leakage rate is greater than the initial leakage rate, the corresponding structure is adjusted synchronously to make the real-time leakage rate less than the initial leakage rate until the flywheel completes the trial operation.

[0011] The step of aligning the O-ring and placing it in the sealing groove of the flange sealing surface of the vacuum cavity includes:

[0012] Before placement, use a lint-free cloth dampened with anhydrous ethanol to clean the sealing groove of the flange sealing surface, and apply a layer of vacuum grease evenly inside the sealing groove, and then place the O-ring into the sealing groove.

[0013] Start the laser alignment instrument, fix the laser emitter to one side of the flange, align the receiver or vision probe with the circumference of the O-ring, and measure the radial and angular deviations of the center of the O-ring relative to the center of the sealing groove.

[0014] According to the radial deviation, use a nylon tool to move the O-ring until the distance and angle deviation between the center of the O-ring and the center of the sealing groove reach the set value. At this time, the center of the O-ring coincides with the center of the sealing groove, and the gap between the O-ring and the wall of the sealing groove is uniform.

[0015] Inspect the installation status of the O-rings using an industrial endoscope or reflector to ensure that the O-rings are not flipped or twisted.

[0016] The tightening sequence of the bolts is as follows:

[0017] Number all bolts 1, 2, 3, ..., n in a clockwise direction starting from the 12 o'clock position;

[0018] When n=4k, k∈N, the bolts are divided into several groups of four after numbering. The four bolts in each group are distributed in a cross-shaped symmetrical pattern on the circumference. The bolts in the first group are tightened in order of increasing number, and then the bolts in the second group are tightened in order, until all groups of bolts are tightened.

[0019] When n=4k-2, k∈N, the bolts after the number are divided into several groups of two. The two bolts in each group are symmetrically distributed at 180 degrees on the circumference. The first group of bolts is tightened in order of increasing number, and then the second group of bolts is tightened in order, until all groups of bolts are tightened.

[0020] When n≠4k and n≠4k-2, k∈N, the optimal loading sequence of the bolts is determined by the finite element simulation optimization method, and each bolt is tightened sequentially according to the optimal loading sequence.

[0021] The determination of the optimal loading sequence for the bolts using the finite element simulation optimization method includes:

[0022] Establish a parametric finite element model;

[0023] Encode the bolt tightening sequence as an arrangement variable. ,in, This is the number of the first bolt that was tightened. This is the number for the second tightening bolt. For the first The number of the tightening bolt;

[0024] Determine the minimum weighting :

[0025] ;

[0026] in, For the first The offset of the center of the resultant force corresponding to each bolt This represents the maximum offset of the center of force during the process of tightening the first bolt to the last bolt. This represents the final flange warpage deformation. The standard deviation of O-ring contact stress. , , These are the weighting coefficients. This is the reference maximum value for the offset of the resultant force center. This represents the maximum allowable value for flange warping deformation. This is the reference maximum value for the standard deviation of O-ring contact stress;

[0027] Based on minimizing weighted Genetic algorithms are used to search for the optimal permutation of variables. ;

[0028] Output the optimal permutation of variables. The optimal arrangement of variables was evaluated through physical experiments and simulation comparisons. Reliability.

[0029] Each bolt is tightened in three stages, including:

[0030] The target torque value of the bolt is calculated using the following formula:

[0031] ;

[0032] in, For the target torque value, For coefficients, For preload, The nominal diameter of the bolt thread;

[0033] In the first stage, a tightening torque of 30% of the target torque value is applied to the bolts, while the initial compression rate of the O-ring is 15%-20%.

[0034] In the second stage, a tightening torque is applied to the bolts, with the tightening torque value being 60% of the target torque value;

[0035] In the third stage, a tightening torque is applied to the bolt, which is 100% of the target torque value.

[0036] Before the first stage, place the reference plane of the depth micrometer on the flange sealing surface, contact the top of the O-ring with the measuring rod and read the value. Measure at 4-6 points evenly distributed around the circumference and take the average value to obtain the initial height of the O-ring above the flange sealing surface. ;

[0037] After the first stage is completed, use a feeler gauge or laser displacement sensor to measure the gap between the upper and lower flanges. The initial compression rate of the O-ring for:

[0038] ;

[0039] Determine the initial compression ratio If the initial compression rate is within 15%-20%, and exceeds this range, it indicates an error in the tightening torque. Adjust the tightening torque to achieve the correct initial compression rate. It falls within the range of 15%-20%.

[0040] The leak detection and repair process includes:

[0041] After the welding cools down, all load-bearing welds are subjected to ultrasonic phased array testing. When the ultrasonic phased array testing finds that the weld defect level is greater than II or that there are cracks, lack of fusion, or incomplete penetration, the defect location is determined and preheating, repair welding, and post-weld treatment are carried out.

[0042] The weld repair area was subjected to ultrasonic phased array flaw detection again. At the same time, the system was roughly evacuated to ≤10Pa, and helium was sprayed point by point on the weld and flange joint for rough inspection. If the leak rate was greater than the rough inspection leak rate threshold, the leak point was repaired by welding and the rough inspection was repeated until the entire area was qualified.

[0043] Then the system is evacuated to ≤0.01Pa and fine inspection is performed by scanning with a suction gun. If the leak rate is greater than the fine inspection leak rate threshold, the leak points are repaired by welding and then fine inspection is performed again until the entire area is qualified.

[0044] The simultaneous comparison with the initial leak rate, and when the real-time leak rate is greater than the initial leak rate, synchronously adjusting the corresponding structure to make the real-time leak rate less than the initial leak rate, includes:

[0045] Leakage rate values ​​are collected in real time, and a function of leakage rate versus time is obtained. ;

[0046] Linear regression was performed on the leak rate-time series within the window to obtain the slope. Simultaneously, the root mean square error between the actual leakage rate and the fitted straight line is calculated to determine if there is a sudden change point. If a sudden change point exists, it is determined to be a sudden leakage; if there is no sudden change point, the function expression is used. middle A number of consecutive sampling points were obtained Calculate a series of consecutive leakage rate values. Average leakage rate of consecutive leakage rate values ;

[0047] when >0.8 × the precision detection miss rate threshold and If the value is >0.05 and persists for 3 consecutive windows, it is determined to be a seal degradation. Replace the O-ring and retighten the bolts.

[0048] The vibration frequency of the flywheel is acquired in real time to obtain the time-domain signal. The time-domain signal is transformed by FFT. Convert to frequency domain signal The natural frequency of the bolted connection structure is set as , the natural frequency Substitute into the frequency domain signal The amplitude is obtained from Calculate the change in leakage rate With amplitude correlation coefficient ,like If the value is greater than 0.7, the bolt is considered loose, and it should be tightened again using a digital torque wrench.

[0049] The correlation coefficient The calculation formula is:

[0050] ;

[0051] in, No. The change in leak rate at each sampling point for The sample mean, For the first The vibration amplitude at a specific frequency at each sampling point for The sample mean.

[0052] After tightening the bolts, use a marker to draw a continuous vertical line between the hexagonal head of the bolt and the surface of the flange base.

[0053] Wait for the marking paint to dry completely, forming the zero-position baseline;

[0054] During the later inspection, observe the zero-position baseline. When the displacement arc length of the zero-position baseline on the bolt head relative to the flange in the circumferential direction exceeds 1mm, it is judged as "suspected loosening". Then, use a digital torque wrench to conduct spot checks and tighten it again.

[0055] When the zero-position baseline is completely broken into two independent lines with obvious misalignment, it is not necessary to use instruments to measure it. It is directly judged as "significantly loose". Tighten the bolts with a digital torque wrench and assess whether plastic deformation has occurred. At the same time, open the flange, check the sealing surface, O-ring and bolt condition, and replace the damaged parts.

[0056] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0057] 1. In the present invention, a method for controlling the sealing performance of a flywheel energy storage vacuum system, three major processes—flange installation, weld treatment and inspection, and molecular pump installation and leveling—are logically linked together to form a closed-loop control of leakage rate from component installation to system operation. This differs from existing technologies that focus on only a single stage. Flange installation sealing is quantitatively controlled, providing a stable sealing environment for subsequent weld treatment and inspection. By inspecting and treating the welds, it is ensured that all parameters of the flywheel meet standards before trial operation. By leveling the molecular pump and adjusting the connecting pipeline layout, the effective pumping speed of the molecular pump reaches the set pumping speed, thus making vacuum degree detection more accurate. During the flywheel trial operation, corresponding structures can be adjusted to ensure that the leakage rate meets design specifications, forming a closed-loop control of leakage rate during the operation of the flywheel energy storage system, resulting in high stability. Therefore, the present invention has a better leakage rate control rate.

[0058] 2. In the method for controlling the sealing performance of a flywheel energy storage vacuum system according to the present invention, optical alignment is used to ensure that the O-ring is not twisted within the groove, avoiding contact width loss due to initial position offset. Simultaneously, graded loading of 30%, 60%, and 100% is employed to gradually transition the O-ring from elastic deformation to steady-state compression, avoiding stress concentration and local overload caused by single-stage tightening. When tightening bolts, a symmetrical tightening sequence and finite element simulation optimization method are used to determine the optimal tightening sequence. Whether tightening symmetrical or asymmetrical bolts, the resultant force center of the bolts always coincides with the geometric center of the flange, allowing for stable control of flange warping deformation. Therefore, the working process of this invention is stable and can be quantitatively controlled.

[0059] 3. In the method for controlling the sealing of a flywheel energy storage vacuum system according to the present invention, the defect types detected by the ultrasonic phased array are fed back and recorded in real time. During subsequent welding, high-risk parameter ranges can be automatically avoided. Simultaneously, helium injection is used for inspection. Taking advantage of the physical property that helium molecules are much smaller than air molecules, coarse inspection can detect millimeter-level leaks, while fine inspection can detect micron-level grain boundary channels. By addressing the defects accordingly, the leakage rate can be effectively controlled. Therefore, the leakage rate control rate of the present invention is relatively high.

[0060] 4. In the flywheel energy storage vacuum system sealing control method of this invention, during the flywheel trial operation phase, the leakage rate change trend is monitored in real time, and corresponding leakage repair measures are taken according to different judgment results, thereby achieving dynamic leakage rate monitoring and compensation control with fast processing efficiency. By combining visual positioning and displacement quantification, bolt loosening is quantitatively judged, allowing for separate handling of different bolt loosening conditions, avoiding misjudgments, and enabling rapid bolt processing with high efficiency. Therefore, this invention has high accuracy and fast processing efficiency. Attached Figure Description

[0061] Figure 1 This is a flowchart of a method for controlling the sealing performance of a flywheel energy storage vacuum system according to the present invention.

[0062] Figure 2 This is a diagram showing the compression stress relaxation curves of fluororubber, ethylene acrylate rubber, and nitrile rubber in this invention. Detailed Implementation

[0063] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0064] Example 1:

[0065] See Figure 1 A method for controlling the sealing performance of a flywheel energy storage vacuum system, comprising:

[0066] Place the O-ring in the sealing groove of the flange sealing surface of the vacuum chamber, and place the bolts on the flange sealing surface. Use a digital torque wrench to tighten the bolts in sequence, and tighten each bolt in three stages.

[0067] Welding is performed to connect the vacuum chamber and pipelines. After the welding cools, leak detection is performed and leak repair is carried out. The current leak rate of the vacuum chamber is recorded as the initial leak rate.

[0068] Install a molecular pump and adjust the layout of the connecting pipeline between the molecular pump and the vacuum chamber. Then install a vacuum gauge to measure the real-time vacuum level at the center of the vacuum chamber and calculate the real-time leakage rate.

[0069] The flywheel undergoes trial operation, and the real-time leakage rate is monitored and recorded. At the same time, it is compared with the initial leakage rate. When the real-time leakage rate is greater than the initial leakage rate, the corresponding structure is adjusted synchronously to make the real-time leakage rate less than the initial leakage rate until the flywheel completes the trial operation.

[0070] In this embodiment, the welding of the vacuum chamber and pipeline connection refers to the welding of the splicing weld of the vacuum chamber and the welding of the pipeline connection. After leakage detection, if the leakage rate exceeds the standard, leak repair is performed, and leakage detection continues until the current leakage rate is less than the preset leakage rate. When installing the molecular pump, the deviation of the molecular pump mounting base is measured and the molecular pump is leveled. At the same time, the layout of the connecting pipeline between the molecular pump and the vacuum chamber is adjusted so that the effective pumping speed of the molecular pump reaches the set pumping speed. By logically linking the three major processes of flange installation, weld treatment and detection, and molecular pump installation and leveling, a leakage rate is formed from component installation to system operation. Closed-loop control, unlike existing technologies that only focus on a single link, uses quantitative control of flange installation sealing to provide a stable sealing environment for subsequent weld treatment and inspection. By inspecting and treating the welds, it is ensured that all parameters of the flywheel meet the standards before trial operation. By leveling the molecular pump and adjusting the layout of the connecting pipeline, the effective pumping speed of the molecular pump reaches the set pumping speed, thereby making the vacuum degree detection more accurate. During the flywheel trial operation, the corresponding structure can be adjusted to ensure that the leakage rate meets the design specifications, forming a closed-loop control of the leakage rate during the operation of the flywheel energy storage system, which has high stability.

[0071] In subsequent trial operations, if an increase in the leakage rate is detected, the fault type will be automatically determined and the corresponding compensation strategy will be triggered by comparing the current leakage rate data with the historical "leak fingerprint database".

[0072] When an existing leak expands, the system determines that it is caused by material fatigue or creep, and the compensation mechanism is "repair compensation": automatically tighten the associated bolts, and compensate by 5%-10%; if ineffective, replace the O-ring with a more aging-resistant material such as a perfluoroether rubber seal; update the fingerprint database after repair and record the number of repairs;

[0073] When a new leak occurs, the system determines it as a sudden structural damage, and the compensation mechanism is "emergency compensation": immediately start the backup vacuum pump to increase the effective pumping speed and counteract the leaked gas; actively suppress specific frequency vibrations through an adaptive notch filter to slow down crack propagation; if the leak rate gets out of control, trigger the vacuum glue injection device for temporary sealing; execute the emergency shutdown procedure and send a precise alarm to the maintenance personnel.

[0074] Example 2:

[0075] The basic content is the same as in Example 1, except that:

[0076] The step of aligning the O-ring and placing it in the sealing groove of the flange sealing surface of the vacuum cavity includes:

[0077] Before placement, use a lint-free cloth dampened with anhydrous ethanol to clean the sealing groove of the flange sealing surface, and apply a layer of vacuum grease evenly inside the sealing groove, and then place the O-ring into the sealing groove.

[0078] Start the laser alignment instrument, fix the laser emitter to one side of the flange, align the receiver or vision probe with the circumference of the O-ring, and measure the radial and angular deviations of the center of the O-ring relative to the center of the sealing groove.

[0079] According to the radial deviation, use a nylon tool to move the O-ring until the distance and angle deviation between the center of the O-ring and the center of the sealing groove reach the set value. At this time, the center of the O-ring coincides with the center of the sealing groove, and the gap between the O-ring and the wall of the sealing groove is uniform.

[0080] Inspect the installation status of the O-rings using an industrial endoscope or reflector to ensure that the O-rings are not flipped or twisted.

[0081] In this embodiment, Krytox GPL-206 fluoroether vacuum grease can be used. The function of the vacuum grease is to temporarily fix the O-ring and reduce friction and prevent twisting in the initial compression stage. When the O-ring is placed in the sealing groove, it ensures that it does not contact any parts outside the groove. The laser alignment instrument is an Easy-Laser E710 dual-axis laser alignment instrument. The laser alignment instrument includes a laser transmitter, a receiver, and a display screen. After the laser alignment instrument is started, the screen will display the radial deviation between the circle of the O-ring and the sealing groove in real time. Radial deviation refers to the deviation of the O-ring relative to the sealing groove along the X and Y axes. The set value of the distance between the center of the O-ring and the center of the sealing groove is 0.1mm. That is, when the distance between the center of the O-ring and the center of the sealing groove is less than or equal to 0.1mm and the angular deviation approaches zero, the center of the O-ring coincides with the center of the sealing groove. The installation error of the O-ring is obtained in real time by a laser alignment instrument, which can ensure the installation accuracy of the O-ring and avoid the initial leakage rate exceeding the standard or leakage due to thermal stress relaxation after the flywheel runs in subsequent tests.

[0082] Example 3:

[0083] The basic content is the same as in Example 1, except that:

[0084] The tightening sequence of the bolts is as follows:

[0085] Number all bolts 1, 2, 3, ..., n in a clockwise direction starting from the 12 o'clock position;

[0086] When n=4k, k∈N, the bolts are divided into several groups of four after numbering. The four bolts in each group are distributed in a cross-shaped symmetrical pattern on the circumference. The bolts in the first group are tightened in order of increasing number, and then the bolts in the second group are tightened in order, until all groups of bolts are tightened.

[0087] When n=4k-2, k∈N, the bolts after the number are divided into several groups of two. The two bolts in each group are symmetrically distributed at 180 degrees on the circumference. The first group of bolts is tightened in order of increasing number, and then the second group of bolts is tightened in order, until all groups of bolts are tightened.

[0088] When n≠4k and n≠4k-2, k∈N, the optimal loading sequence of the bolts is determined by the finite element simulation optimization method, and each bolt is tightened sequentially according to the optimal loading sequence.

[0089] In this embodiment, a 180-degree dual symmetrical tightening method is adopted, treating two bolts with an included angle of 180° as a dual pair, and tightening them sequentially according to the dual pairs. For multiples of 4 such as 8 bolts and 12 bolts, adjacent dual pairs can be combined into a four-bolt group to simplify the operation; for non-multiples of 4 such as 6 bolts and 10 bolts, they are tightened one by one according to the dual pair sequence. For flanges with non-uniform distribution, a finite element simulation optimization method is used to ensure that the center of the resultant force of each loading always coincides with the geometric center of the flange, thereby controlling the warpage deformation of the flange to the micrometer level.

[0090] Example 4:

[0091] The basic content is the same as Example 3, except that:

[0092] The determination of the optimal loading sequence for the bolts using the finite element simulation optimization method includes:

[0093] Establish a parametric finite element model;

[0094] Encode the bolt tightening sequence as an arrangement variable. ,in, This is the number of the first bolt that was tightened. This is the number for the second tightening bolt. For the first The number of the tightening bolt;

[0095] Determine the minimum weighting :

[0096] ;

[0097] in, For the first The offset of the center of the resultant force corresponding to each bolt This represents the maximum offset of the center of force during the process of tightening the first bolt to the last bolt. This represents the final flange warpage deformation. The standard deviation of O-ring contact stress. , , These are the weighting coefficients. This is the reference maximum value for the offset of the resultant force center. This represents the maximum allowable value for flange warping deformation. This is the reference maximum value for the standard deviation of O-ring contact stress;

[0098] Based on minimizing weighted Genetic algorithms are used to search for the optimal permutation of variables. ;

[0099] Output the optimal permutation of variables. The optimal arrangement of variables was evaluated through physical experiments and simulation comparisons. Reliability.

[0100] In this embodiment, physical tests such as measuring flange strain with resistance strain gauges are used to verify and compare the results with simulations, taking a flange with 5 bolts and a non-uniform distribution as an example:

[0101] The first bolt is located at 0°, the second at 60°, the third at 120°, the fourth at 200°, and the fifth at 280°. The bolt distribution circle radius R = 150 mm. The O-rings are made of fluororubber (A70) with a cross-sectional diameter of 3.53 mm. The target bolt torque is T = 50 Nm, the torque coefficient K = 0.2, the bolt diameter d = 10 mm, and the preload F is 25 kN.

[0102] Using the ANSYS Workbench finite element model, element type SOLID186 (hexahedron), contact settings: friction coefficient between bolt and flange is 0.15, friction coefficient between O-ring and flange is 0.2, apply bolt preload in stages (using "BoltPretension" load, activated according to the bolt tightening sequence).

[0103] Running a genetic algorithm with a population size of P=50 and 100 generations, the optimal permutation variables are obtained after optimization. The given numbers are [1, 3, 5, 2, 4].

[0104] The offset of the resultant force center corresponding to this sequence The peak value of the curve is approximately 0.08 mm, much smaller than the peak value of the random sequence (average 0.35 mm), resulting in a final flange warping deformation. =4.2μm, O-ring contact stress standard deviation =0.12 MPa. Comparing this arrangement variable with a manual sequence (e.g., [1, 2, 3, 4, 5]), warpage is reduced by about 65%, and O-ring contact uniformity is improved by 55%.

[0105] Example 5:

[0106] The basic content is the same as Example 3, except that:

[0107] Each bolt is tightened in three stages, including:

[0108] The target torque value of the bolt is calculated using the following formula:

[0109] ;

[0110] in, For the target torque value, For coefficients, For preload, The nominal diameter of the bolt thread;

[0111] In the first stage, a tightening torque of 30% of the target torque value is applied to the bolts, while the initial compression rate of the O-ring is 15%-20%.

[0112] In the second stage, a tightening torque is applied to the bolts, with the tightening torque value being 60% of the target torque value;

[0113] In the third stage, a tightening torque is applied to the bolt, which is 100% of the target torque value.

[0114] In this embodiment, the O-ring material is selected as fluororubber (FKM). The compressive stress relaxation curves of fluororubber (FKM), ethylene acrylate rubber (AEM), and nitrile rubber (NBR) are compared, as follows: Figure 2As shown, FKM (fluororubber) demonstrates superior performance in high-temperature stability, compression stress relaxation retention, oil resistance, and aging resistance, making it the optimal choice for high-temperature, oil-resistant, and long-life sealing applications. Furthermore, the viscoelastic properties of fluororubber are well-matched with the three-stage loading process; 30% torque is the physical threshold for generating effective sealing stress, while the holding time (60-120 seconds) corresponding to 60% torque represents the most severe stage of stress relaxation in fluororubber. Utilizing this time to allow the material to complete molecular chain rearrangement effectively prevents subsequent torque decay due to stress relaxation.

[0115] Example 6:

[0116] The basic content is the same as Example 5, except that:

[0117] Before the first stage, place the reference plane of the depth micrometer on the flange sealing surface, contact the top of the O-ring with the measuring rod and read the value. Measure at 4-6 points evenly distributed around the circumference and take the average value to obtain the initial height of the O-ring above the flange sealing surface. ;

[0118] After the first stage is completed, use a feeler gauge or laser displacement sensor to measure the gap between the upper and lower flanges. The initial compression rate of the O-ring for:

[0119] ;

[0120] Determine the initial compression ratio If the initial compression rate is outside the 15%-20% range, it indicates an error in the tightening torque. Adjust the tightening torque to achieve the correct initial compression rate. It falls within the range of 15%-20%.

[0121] In this embodiment, since the O-ring compression is achieved stepwise through three-stage torque loading of the bolt, the compression rate has three progressive stages: the compression rate range corresponding to the first stage is 15%-20%; the compression rate range corresponding to the second stage is 20%-23%; and the compression rate range corresponding to the third stage is 23%-25%. Therefore, the compression amount of the O-ring each time is related to the target torque and the compression amount is proportional to the torque applied by the bolt. In the second and third stages, the tightening torque value of the current stage can be determined by detecting the compression rate of the O-ring.

[0122] Example 7:

[0123] The basic content is the same as in Example 1, except that:

[0124] The leak detection and repair process includes:

[0125] After the welding cools down, all load-bearing welds are subjected to ultrasonic phased array testing. When the ultrasonic phased array testing finds that the weld defect level is greater than II or that there are cracks, lack of fusion, or incomplete penetration, the defect location is determined and preheating, repair welding, and post-weld treatment are carried out.

[0126] The weld repair area was subjected to ultrasonic phased array flaw detection again. At the same time, the system was roughly evacuated to ≤10Pa, and helium was sprayed point by point on the weld and flange joint for rough inspection. If the leak rate was greater than the rough inspection leak rate threshold, the leak point was repaired by welding and the rough inspection was repeated until the entire area was qualified.

[0127] Then the system is evacuated to ≤0.01Pa and fine inspection is performed by scanning with a suction gun. If the leak rate is greater than the fine inspection leak rate threshold, the leak points are repaired by welding and then fine inspection is performed again until the entire area is qualified.

[0128] In this embodiment, the weld defect level can be referenced from the weld defect level classification in GB / T 19418-2003. When ultrasonic phased array flaw detection finds a weld defect level greater than II, or there are harmful defects such as cracks, lack of fusion, or incomplete penetration, the 3D imaging function of the flaw detector is used to accurately determine the defect location, mark it on the workpiece surface, and take photos for archiving. According to the defect type and depth, the defect is removed by grinding, carbon arc gouging, or machining. The removal depth should exceed the bottom of the defect by at least 2mm to form a U-shaped or V-shaped bevel. Preheating, repair welding, and post-weld treatment are carried out according to the welding process specification. The interpass temperature is strictly controlled. Repair welding at the same location should not be performed more than twice. If two repair weldings are still unqualified, the weld section should be cut off and the joint should be remade.

[0129] The formula for calculating the leakage rate threshold is:

[0130] ;

[0131] ;

[0132] in, Leakage rate For the volume of the molecular pump, For the rate of increase in pressure, For the flywheel to operate normally The maximum allowable pressure value after a certain time. Initial vacuum level;

[0133] Taking into account the contribution of material outgassing (approximately 50%), the engineering safety factor (approximately 3 times), and the economic sensitivity of the helium mass spectrometer leak detector, the precise leak detection threshold is determined by rounding up. The threshold for the crude detection leak rate is ;

[0134] The residual gas composition was analyzed using a quadrupole mass spectrometer to determine whether the leak was a genuine leak or a material release.

[0135] If venting is required for the material:

[0136] When releasing water vapor, bake the cavity to remove gas (150-200℃, 24-48 hours), or purge with dry nitrogen; when releasing hydrocarbons, clean the inner wall of the cavity with acetone / ethanol, ultrasonically clean the removable parts, and replace with a low-release-rate sealing ring if necessary; when releasing hydrogen, perform high-temperature vacuum baking (300-400℃) or multiple evacuation-replacement cycles.

[0137] After the aforementioned treatment, if the residual gas composition analysis still shows the presence of adsorbed gases that are difficult to remove (e.g., water vapor partial pressure is still higher than 1 μPa, or inert gases such as argon and nitrogen are present), or if there are atomic-level surface contaminants on the inner wall of the cavity, glow discharge plasma cleaning is used to remove the stubborn adsorbed gases.

[0138] After treatment, the vacuum was re-evacuated to ≤0.01 Pa, and quadrupole mass spectrometry and helium mass spectrometry leak detection were performed again until the residual gas composition met the requirements (water vapor partial pressure ≤1 μPa, hydrocarbon partial pressure ≤0.1 μPa) and the total leak rate ≤ ;

[0139] If it is a genuine leak, repair the leak points that exceed the acceptable rate by welding, and then re-perform a fine inspection until the entire area is qualified.

[0140] Example 8:

[0141] The basic content is the same as in Example 1, except that:

[0142] Adjust the layout of the connecting pipes between the molecular pump and the vacuum chamber, including:

[0143] ;

[0144] ;

[0145] ;

[0146] in, For flow guidance, The diameter of the pipeline. The length of the pipeline, For the effective pumping speed of the molecular pump, This is the nominal pumping speed of the molecular pump.

[0147] In this embodiment, the pipe diameter and length can be obtained through the above formula. Since the effective pumping speed of the molecular pump is ≥70% of the nominal pumping speed, the high vacuum requirements of flywheel energy storage can be met while ensuring that the flow conductance is large enough.

[0148] Example 9:

[0149] The basic content is the same as in Example 1, except that:

[0150] Leakage rate values ​​are collected in real time, and a function of leakage rate versus time is obtained. ;

[0151] Linear regression was performed on the leak rate-time series within the window to obtain the slope. Simultaneously, the root mean square error between the actual leakage rate and the fitted straight line is calculated to determine if there is a sudden change point. If a sudden change point exists, it is determined to be a sudden leakage; if there is no sudden change point, the function expression is used. middle A number of consecutive sampling points were obtained Calculate a series of consecutive leakage rate values. Average leakage rate of consecutive leakage rate values ;

[0152] when >0.8 × the precision detection miss rate threshold and If the value is >0.05 and persists for 3 consecutive windows, it is determined to be a seal degradation. Replace the O-ring and retighten the bolts.

[0153] The vibration frequency of the flywheel is acquired in real time to obtain the time-domain signal. The time-domain signal is transformed by FFT. Convert to frequency domain signal The natural frequency of the bolted connection structure is set as , the natural frequency Substitute into the frequency domain signal The amplitude is obtained from Calculate the change in leakage rate With amplitude correlation coefficient ,like If the value is greater than 0.7, the bolt is considered loose, and it should be tightened again using a digital torque wrench.

[0154] In this embodiment, an ionization vacuum meter (or quadrupole mass spectrometer) can be used to collect the leak rate value in real time. The sampling frequency of the leak rate is ≥1Hz, and the sampling frequency of the flywheel is ≥2kHz. (In the time domain signal...) Convert to frequency domain signal Then, characteristic frequencies are extracted and identified in relation to rotor speed. Related characteristic peaks, while monitoring amplitude changes at key frequency points. Critical frequency points refer to multiple characteristic frequencies that are sensitive to faults; fundamental frequency. : Corresponds to rotor imbalance; frequency multiplication , : Corresponding to rotor misalignment or looseness; sideband Modulation: corresponds to bearing failure;

[0155] when >1.5× If the fundamental frequency amplitude exceeds the limit, the rotor imbalance is aggravated, and the adaptive notch filter of the electromagnetic bearing is adjusted to suppress the same-frequency vibration. If this is ineffective, a "dynamic balancing correction recommended" warning is issued, and on-site dynamic balancing is performed during the planned downtime. If the problem recurs, an installation deviation tracing inspection is triggered. The rotational speed frequency of the flywheel rotor under healthy conditions The vibration amplitude reference value at the location is automatically established by continuously acquiring vibration signals after the system's initial commissioning is successful, and is used to correlate with the real-time amplitude. Compare them to determine the degree of imbalance;

[0156] when >0.5× If the frequency harmonics are significant, it is determined that the rotor is misaligned or loose. The source of looseness is located through cross-verification using multiple sensors. The looseness is then graded according to its severity: slight looseness is automatically tightened; moderate looseness is planned for a shutdown for re-tightening; severe looseness is an emergency shutdown for repair. If loosening treatment is ineffective, a laser alignment instrument is activated to correct the shaft alignment. All processing results are recorded in the BIM model for optimizing subsequent installation process parameters and predicting deviation thresholds.

[0157] Example 10:

[0158] The basic content is the same as Example 9, except that:

[0159] Correlation coefficient The calculation formula is:

[0160] ;

[0161] in, No. The change in leak rate at each sampling point for The sample mean, For the first The vibration amplitude at a specific frequency at each sampling point for The sample mean;

[0162] when When the threshold is >0.7, the increase in leakage rate is determined to be strongly correlated with the vibration of the bolted connection structure, which is inferred to be due to bolt loosening. This threshold is based on test data of the flywheel energy storage prototype (loosening condition). =0.75~0.92, under normal operating conditions <0.3) and statistical significance requirements ( <0.01) Determined comprehensively; when the correlation coefficient between the leakage rate change and the vibration amplitude at a specific frequency is 0.3< When the correlation is less than 0.7, the system determines it to be of moderate correlation and does not directly trigger a bolt loosening alarm. Instead, it initiates an enhanced diagnostic process to check sensor and data quality, increasing the sampling rate or extending the sampling time if necessary. It fuses data from temperature, torque, and other frequency components to determine if the issue is thermal deformation, misalignment, or background noise. The system then analyzes the data. If the trend continues to rise over time, a "pre-loosening warning" will be issued; a light tightening (torque increased by 5%) will be performed on the critical flange. A decrease confirms a slight loosening; if it cannot be ruled out, it is recorded as "unknown coupling" and the next monitoring cycle begins.

[0163] Example 11:

[0164] The basic content is the same as in Example 1, except that:

[0165] After tightening the bolts, use a marker to draw a continuous vertical line between the hexagonal head of the bolt and the surface of the flange base.

[0166] Wait for the marking paint to dry completely, forming the zero-position baseline;

[0167] During the later inspection, observe the zero-position baseline. When the displacement arc length of the zero-position baseline on the bolt head relative to the flange in the circumferential direction exceeds 1mm, it is judged as "suspected loosening". Then, use a digital torque wrench to conduct spot checks and tighten it again.

[0168] When the zero-point reference line is completely broken into two independent lines with obvious misalignment, it is directly judged as "significantly loose" without the need for instrument measurement. The operator should stop the machine and isolate it to ensure that the system is in a safe state. Use a digital torque wrench to tighten the bolts according to the original process and assess whether plastic deformation has occurred. Open the flange, check the sealing surface, O-ring and bolt condition, and replace any damaged parts.

[0169] In this embodiment, the physical causes of "significant loosening" are mainly threefold: the lateral vibration generated by the high-speed rotation of the flywheel exceeds the static friction limit of the threaded pair, causing the bolt to rotate; the thermal cycling effect during operation causes periodic changes and attenuation of the preload due to the difference in the thermal expansion coefficient of the materials; and the compression creep of the sealing O-ring during long-term service leads to a reduction in the flange clearance and a decrease in the bolt preload.

[0170] Although embodiments of the present invention have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for controlling the sealing performance of a flywheel energy storage vacuum system, characterized in that, include: Place the O-ring in the sealing groove of the flange sealing surface of the vacuum chamber, and place the bolts on the flange sealing surface. Use a digital torque wrench to tighten the bolts in sequence, and tighten each bolt in three stages. Welding is performed to connect the vacuum chamber and pipelines. After the welding cools, leak detection is performed and leak repair is carried out. The current leak rate of the vacuum chamber is recorded as the initial leak rate. Install a molecular pump and adjust the layout of the connecting pipeline between the molecular pump and the vacuum chamber. Then install a vacuum gauge to measure the real-time vacuum level at the center of the vacuum chamber and calculate the real-time leakage rate. The flywheel undergoes trial operation, and the real-time leakage rate is monitored and recorded. At the same time, it is compared with the initial leakage rate. When the real-time leakage rate is greater than the initial leakage rate, the corresponding structure is adjusted synchronously to make the real-time leakage rate less than the initial leakage rate until the flywheel completes the trial operation.

2. The method for controlling the sealing performance of a flywheel energy storage vacuum system according to claim 1, characterized in that: The step of aligning the O-ring and placing it in the sealing groove of the flange sealing surface of the vacuum cavity includes: Before placement, use a lint-free cloth dampened with anhydrous ethanol to clean the sealing groove of the flange sealing surface, and apply a layer of vacuum grease evenly inside the sealing groove, and then place the O-ring into the sealing groove. Start the laser alignment instrument, fix the laser emitter to one side of the flange, align the receiver or vision probe with the circumference of the O-ring, and measure the radial and angular deviations of the center of the O-ring relative to the center of the sealing groove. According to the radial deviation, use a nylon tool to move the O-ring until the distance and angle deviation between the center of the O-ring and the center of the sealing groove reach the set value. At this time, the center of the O-ring coincides with the center of the sealing groove, and the gap between the O-ring and the wall of the sealing groove is uniform. Inspect the installation status of the O-rings using an industrial endoscope or reflector to ensure that the O-rings are not flipped or twisted.

3. The method for controlling the sealing performance of a flywheel energy storage vacuum system according to claim 1, characterized in that: The tightening sequence of the bolts is as follows: Number all bolts 1, 2, 3, ..., n in a clockwise direction starting from the 12 o'clock position; When n=4k, k∈N, the bolts are divided into several groups of four after numbering. The four bolts in each group are distributed in a cross-shaped symmetrical pattern on the circumference. The bolts in the first group are tightened in order of increasing number, and then the bolts in the second group are tightened in order, until all groups of bolts are tightened. When n=4k-2, k∈N, the bolts after the number are divided into several groups of two. The two bolts in each group are symmetrically distributed at 180 degrees on the circumference. The first group of bolts is tightened in order of increasing number, and then the second group of bolts is tightened in order, until all groups of bolts are tightened. When n≠4k and n≠4k-2, k∈N, the optimal loading sequence of the bolts is determined by the finite element simulation optimization method, and each bolt is tightened sequentially according to the optimal loading sequence.

4. The method for controlling the sealing performance of a flywheel energy storage vacuum system according to claim 3, characterized in that: The determination of the optimal loading sequence for the bolts using the finite element simulation optimization method includes: Establish a parametric finite element model; Encode the bolt tightening sequence as an arrangement variable. ,in, This is the number of the first bolt that was tightened. This is the number for the second tightening bolt. For the first The number of the tightening bolt; Determine the minimum weighting : ； in, For the first The offset of the center of the resultant force corresponding to each bolt This represents the maximum offset of the center of force during the process of tightening the first bolt to the last bolt. This represents the final flange warpage deformation. The standard deviation of O-ring contact stress. , , These are the weighting coefficients. This is the reference maximum value for the offset of the resultant force center. This represents the maximum allowable value for flange warping deformation. This is the reference maximum value for the standard deviation of O-ring contact stress; Based on minimizing weighted Genetic algorithms are used to search for the optimal permutation of variables. ; Output the optimal permutation of variables. The optimal arrangement of variables was evaluated through physical experiments and simulation comparisons. Reliability.

5. The method for controlling the sealing performance of a flywheel energy storage vacuum system according to claim 4, characterized in that: Each bolt is tightened in three stages, including: The target torque value of the bolt is calculated using the following formula: ； in, For the target torque value, For coefficients, For preload, The nominal diameter of the bolt thread; In the first stage, a tightening torque of 30% of the target torque value is applied to the bolts, while the initial compression rate of the O-ring is 15%-20%. In the second stage, a tightening torque is applied to the bolts, with the tightening torque value being 60% of the target torque value; In the third stage, a tightening torque is applied to the bolt, which is 100% of the target torque value.

6. The method for controlling the sealing performance of a flywheel energy storage vacuum system according to claim 5, characterized in that: Before the first stage, place the reference plane of the depth micrometer on the flange sealing surface, contact the top of the O-ring with the measuring rod and read the value. Measure at 4-6 points evenly distributed around the circumference and take the average value to obtain the initial height of the O-ring above the flange sealing surface. ; After the first stage is completed, use a feeler gauge or laser displacement sensor to measure the gap between the upper and lower flanges. The initial compression rate of the O-ring for: ； Determine the initial compression ratio If the initial compression rate is within 15%-20%, and exceeds this range, it indicates an error in the tightening torque. Adjust the tightening torque to achieve the correct initial compression rate. It falls within the range of 15%-20%.

7. The method for controlling the sealing performance of a flywheel energy storage vacuum system according to claim 1, characterized in that: The leak detection and repair process includes: After the welding cools down, all load-bearing welds are subjected to ultrasonic phased array testing. When the ultrasonic phased array testing finds that the weld defect level is greater than II or that there are cracks, lack of fusion, or incomplete penetration, the defect location is determined and preheating, repair welding, and post-weld treatment are carried out. The weld repair area was subjected to ultrasonic phased array flaw detection again. At the same time, the system was roughly evacuated to ≤10Pa, and helium was sprayed point by point on the weld and flange joint for rough inspection. If the leak rate was greater than the rough inspection leak rate threshold, the leak point was repaired by welding and the rough inspection was repeated until the entire area was qualified. Then the system is evacuated to ≤0.01Pa and fine inspection is performed by scanning with a suction gun. If the leak rate is greater than the fine inspection leak rate threshold, the leak points are repaired by welding and then fine inspection is performed again until the entire area is qualified.

8. The method for controlling the sealing performance of a flywheel energy storage vacuum system according to claim 7, characterized in that: The simultaneous comparison with the initial leak rate, and when the real-time leak rate is greater than the initial leak rate, synchronously adjusting the corresponding structure to make the real-time leak rate less than the initial leak rate, includes: Leakage rate values ​​are collected in real time, and a function of leakage rate versus time is obtained. ; Linear regression was performed on the leak rate-time series within the window to obtain the slope. Simultaneously, the root mean square error between the actual leakage rate and the fitted straight line is calculated to determine if there is a sudden change point. If a sudden change point exists, it is determined to be a sudden leakage; if there is no sudden change point, the function expression is used. middle 1 consecutive sampling points were obtained Calculate a series of consecutive leakage rate values. Average leakage rate of consecutive leakage rate values ; when >0.8 × the precision detection miss rate threshold and If the value is >0.05 and persists for 3 consecutive windows, it is determined to be a seal degradation. Replace the O-ring and retighten the bolts. The vibration frequency of the flywheel is acquired in real time to obtain the time-domain signal. The time-domain signal is transformed by FFT. Convert to frequency domain signal The natural frequency of the bolted connection structure is set as , will the natural frequency Substitute into the frequency domain signal The amplitude is obtained from Calculate the change in leakage rate With amplitude correlation coefficient ,like If the value is greater than 0.7, the bolt is considered loose, and it should be tightened again using a digital torque wrench.

9. The method for controlling the sealing performance of a flywheel energy storage vacuum system according to claim 8, characterized in that: The correlation coefficient The calculation formula is: ； in, No. The change in leak rate at each sampling point for The sample mean, For the first The vibration amplitude at a specific frequency at each sampling point for The sample mean.

10. The method for controlling the sealing performance of a flywheel energy storage vacuum system according to claim 6, characterized in that: After tightening the bolts, use a marker to draw a continuous vertical line between the hexagonal head of the bolt and the surface of the flange base. Wait for the marking paint to dry completely, forming the zero-position baseline; During the later inspection, observe the zero-position baseline. When the displacement arc length of the zero-position baseline on the bolt head relative to the flange in the circumferential direction exceeds 1mm, it is judged as "suspected loosening". Then, use a digital torque wrench to conduct spot checks and tighten it again. When the zero-position baseline is completely broken into two independent lines with obvious misalignment, it is not necessary to use instruments to measure it. It is directly judged as "significantly loose". Tighten the bolts with a digital torque wrench and assess whether plastic deformation has occurred. At the same time, open the flange, check the sealing surface, O-ring and bolt condition, and replace the damaged parts.

Images