A method and system for intelligently detecting liquid density of a storage tank

Abstract

The application discloses a kind of storage tank liquid density intelligent detection method and system, it is related to liquid physical property detection technical field, comprising: drive vibrating element formula density probe moves along the height of storage tank and conversion initial density, determine upper layer reference area, lower layer reference area and interface candidate area between them;In the same target depth of interface candidate area, respectively by upper layer reference area and lower layer reference area reach, collect two frequency relaxation sequences and obtain two starting densities;When two starting densities are respectively inclined to corresponding layer, and the difference between the end densities of two sequences is not greater than convergence threshold, the end frequencies of two sequences are combined into common frequency, and the liquid density of target depth is converted according to frequency-density corresponding relationship.The application solves the problem that moving vibrating type density probe carries the last liquid phase when crossing the interface, which causes the apparent density distortion at the same depth, can distinguish the real spatial density from the probe motion history interference and keep the density value traceable.

Description

Technical Field

[0001] This invention relates to the field of liquid property testing technology, and in particular to an intelligent method and system for detecting the density of liquids in storage tanks. Background Technology

[0002] The density of liquid in a storage tank is a crucial parameter for material metering, stratification identification, and storage process monitoring. Contact vibration density testing typically involves bringing a resonant cylinder, tuning fork, or other vibrating element into contact with the liquid being measured. It utilizes the added mass and dissipation of the vibrating element by the adjacent liquid, determining the liquid density based on the resonant frequency and a calibration-derived correspondence. Patent CN112945796A discloses a tuning fork-type density meter and its coefficient calibration method, reflecting the general technical approach of characterizing liquid density using the resonant response of a vibrating element and obtaining conversion coefficients through calibration. To obtain the liquid density at different depths in the storage tank, existing detection systems can also use a lifting mechanism to move the probe with an integrated density sensor along the tank's height, allowing the probe to stop and sample at multiple depths, and performing temperature compensation on the resonant response. All of these approaches rely on the premise that the medium in the vibrating element's vibrating element, after the probe reaches the target depth, can represent the original liquid at that depth.

[0003] However, large storage tanks often present local operating conditions that do not meet the above prerequisites. Different batches of products, oil-water two-phase or emulsion transition layers are distributed along the height of the tank. When the probe crosses the liquid-liquid interface, the previous liquid phase may enter the next liquid phase as a wetting film, stagnant liquid in the groove, or residual liquid in the protective cover. This makes the initial resonance response at the same depth depend not only on the local liquid but also on which liquid layer the probe arrives from. Similar historical effects also occur at the gas-liquid surface and the tank bottom: when a dry probe first crosses the gas-liquid surface, the gaps in the vibrating element or the dead corners of the protective cover may trap gas in the head space; when the probe approaches the bottom of the tank and turns back, its drainage and turning movements may stir up sediment. As a result, the apparent density formed at the same depth is easily recorded as the true surface, transition layer, or bottom layer density.

[0004] Extending the probe's dwell time at the target depth only allows for passive waiting and makes it difficult to determine when the observed value has detached from the incoming liquid film or trapped gas. Increasing vibration intensity or adding flushing actions may accelerate replacement, but they may disturb the actual interface or further stir up sediment at the bottom of the tank. When using a fixed combination of difference, stability, and recovery thresholds for judgment, each threshold will vary with liquid viscosity, wettability, temperature, probe structure, movement speed, and dwell time, making it difficult for a single set of artificial parameters to cover different tank conditions. Simply adding measurement points or post-processing the resonant frequency at a single moment cannot provide a basis for distinguishing whether the reading at a certain depth belongs to the actual spatial density change or the residual data from the probe's movement history.

[0005] Therefore, the technical problem that needs to be solved in this field is how to obtain observational information that can characterize the formation and dissipation process of interference in the probe's motion history while retaining the physical main chain of directly measuring liquid density at the resonant frequency, and how to make a verifiable distinction as to whether the interference has dissipated, so that only the verified true liquid density is output at the same depth. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide an intelligent method and system for detecting the density of liquid in storage tanks, which is used to distinguish between the actual spatial density inside the storage tank and the motion history interference introduced by the moving vibration element density probe, and to perform traceable detection of the liquid density after the interference has subsided.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: A method for intelligent detection of liquid density in a storage tank includes: driving a vibration element density probe to move along the height of the storage tank, collecting the resonant frequency of the vibration element and converting it into an initial density, and determining an upper reference zone, a lower reference zone, and a candidate interface zone between the two. Select the target depth within the interface candidate region, reach the target depth from the upper reference region, and acquire a first frequency relaxation sequence; The target depth is crossed into the lower reference region, and then the lower reference region is returned to the target depth to acquire the second frequency relaxation sequence; The frequencies at the starting points of the two frequency relaxation sequences are respectively converted into a first starting density and a second starting density; When the difference between the first starting density and the upper reference density is less than the difference between the first starting density and the lower reference density, the difference between the second starting density and the lower reference density is less than the difference between the second starting density and the upper reference density, and the difference between the terminal densities of the two frequency relaxation sequences is not greater than the convergence threshold, the terminal frequencies of the two frequency relaxation sequences are merged into a common frequency. The common frequency is converted into the liquid density at the target depth according to the frequency-density correspondence.

[0008] Optionally, determining the upper reference region, the lower reference region, and the interface candidate region between them includes: determining the segment whose initial density variation range of continuous detection depth is not greater than a reference stability threshold as a reference region; when the difference in reference density between two adjacent reference regions is greater than an interface detection threshold, determining the depth segment located between the two reference regions as the interface candidate region.

[0009] Optionally, obtaining the terminal density includes: selecting a continuous measurement window forward from the end of each frequency relaxation sequence; determining the continuous measurement window whose difference between the maximum and minimum density values ​​within the window is not greater than a stability threshold as the terminal stable window; determining the average density within the terminal stable window as the terminal density; and determining the average frequency within two terminal stable windows as the common frequency.

[0010] Optionally, when the difference between the two end densities is greater than the convergence threshold, the vibration element is driven with a release drive amplitude that is greater than the measured drive amplitude and within the rated drive range of the vibration element, and the first frequency relaxation sequence and the second frequency relaxation sequence are repeatedly acquired; if the difference between the two end densities is still greater than the convergence threshold after repeated acquisition, a continuous membrane or dynamic interface marker is output.

[0011] Optionally, multiple target depths are selected within the interface candidate region, and the liquid density is determined for each. The segment among multiple adjacent target depths that simultaneously satisfies the condition that the difference between the two terminal densities is not greater than the convergence threshold is determined as an effective transition region. A density profile of the effective transition region is generated according to each target depth and the corresponding liquid density.

[0012] Optionally, after the density probe first crosses the gas-liquid surface from the storage tank space, it acquires the first liquid entry frequency sequence and the first driving energy sequence at the target surface depth; keeping the density probe fully submerged, it moves the density probe down one venting stroke and returns from below to the target surface depth, acquiring the full immersion re-entry frequency sequence and the re-entry driving energy sequence; when the initial end density is lower than the re-entry end density and the difference between the two is greater than the cavitation density threshold, and the initial end driving energy is higher than the re-entry end driving energy and the difference between the two is greater than the cavitation energy threshold, the initial liquid entry cavitation marker is output.

[0013] Optionally, the exhaust stroke and the acquisition of the full immersion re-entry frequency sequence are repeated; when the difference between the densities of two adjacent re-entry ends is not greater than the density repeatability threshold, and the difference between the driving energies of two adjacent re-entry ends is not greater than the energy repeatability threshold, the frequency of the last re-entry end is converted into the liquid density of the target surface depth according to the frequency-density correspondence.

[0014] Optionally, when the density probe descends to the bottom clearance depth, the pre-return frequency and pre-return damping are acquired; the density probe is further descended to the turning depth, then returned to the bottom clearance depth, and the post-return frequency relaxation sequence and post-return damping relaxation sequence are acquired; when the initial density after reversal is higher than the pre-return density and the difference between the two is greater than the resuspension density threshold, the initial damping after reversal is higher than the pre-return damping and the difference between the two is greater than the resuspension damping threshold, and the post-reversal density and post-reversal damping decrease towards the pre-reversal density and the pre-return damping respectively, the probe-induced resuspension marker is output.

[0015] Optionally, after outputting the probe-induced resuspension mark, the pre-reversal frequency is converted into the liquid density at the bottom clearance depth according to the frequency-density correspondence; the adjacent detection range above the turning depth is determined as the lowest effective detection boundary; and the post-reversal frequency relaxation sequence located below the lowest effective detection boundary is excluded from the tank density profile.

[0016] The present invention also discloses an intelligent detection system for liquid density in a storage tank for implementing the method described in the present invention, comprising: a lifting acquisition module for driving a vibration element density probe to move along the height of the storage tank, acquiring the resonant frequency of the vibration element and converting it into an initial density, and determining an upper reference area, a lower reference area and a candidate interface area between the two. The bidirectional sequence module is used to select the target depth within the interface candidate area, reach the target depth from the upper reference area and collect a first frequency relaxation sequence, cross the target depth and enter the lower reference area, and then return to the target depth from the lower reference area and collect a second frequency relaxation sequence. The frequencies at the starting points of the two frequency relaxation sequences are converted into a first starting density and a second starting density, respectively. The direction determination module is used to compare the differences between the first starting density and the second starting density and the upper reference density and the lower reference density, respectively, and to compare the difference between the terminal densities of the two frequency relaxation sequences and the convergence threshold. The density determination module is used to merge the terminal frequencies of the two frequency relaxation sequences into a common frequency when the difference between the first initial density and the upper reference density is less than the difference between the first initial density and the lower reference density, the difference between the second initial density and the lower reference density is less than the difference between the second initial density and the upper reference density, and the difference between the terminal densities of the two frequency relaxation sequences is not greater than the convergence threshold, and convert the common frequency into the liquid density at the target depth according to the frequency-density correspondence.

[0017] The beneficial effects of this invention are as follows: This invention uses the resonant frequency of a vibrating element as a direct physical measurement quantity to determine the density of the liquid in a storage tank. By constructing two frequency relaxation sequences with opposite directional histories at the same target depth, motion history disambiguation is added to this physical measurement quantity. Thus, without changing the main chain of density measurement at the resonant frequency, the transient deviation caused by the probe carrying the previous liquid phase is distinguished from the true liquid state at the target depth.

[0018] When arriving from the upper reference region, the probe carries the upper liquid film, and the initial density is biased towards the upper reference density. When returning from the lower reference region, the probe carries the lower liquid film, and the initial density is biased towards the lower reference density. The true local density at the target depth is independent of the probe's direction of arrival. As the incoming liquid film is replaced by the liquid at the target depth, the two sequences converge towards the same local density. Accordingly, this invention uses the condition that the two initial densities are biased towards the corresponding incoming layers, and the difference in the end densities of the two sequences is not greater than the convergence threshold as the acceptance condition. Only when this condition is met is the liquid density at the target depth calculated using the common frequency within the two end stability windows. Compared to directly writing unidirectional instantaneous values ​​into the density profile, this acceptance condition utilizes both the opposite starting point bias and the direction-independent common endpoint, which can exclude the false transition band created by the probe itself from the density profile without deleting the true emulsified region or intermediate density layer that can stably converge in both opposite directions.

[0019] Because each effective density point must pass the verifiable acceptance condition of initial direction splitting and terminal convergence, each value in the density profile carries a clear basis for acceptance. When two sequences fail to converge after an allowable number of retests, the system outputs a continuous film-carrying or dynamic interface marker, rather than forcibly providing a density by extending the waiting period or extrapolating. Therefore, probe raising and lowering, temperature compensation, and threshold determination all serve the purpose of resonant density measurement, and the final density result can still be traced back to the actual coupling state between the vibrating element and the local liquid.

[0020] At the gas-liquid surface, the initial liquid entry and subsequent re-entry into the liquid after full immersion constitute a motion history comparison for the same surface target depth. The lower apparent density and higher initial driving energy of the initial liquid entry are used as dual criteria to identify trapped gas. Surface density is calculated based on the re-entry frequency only after adjacent re-entry results enter a repeatable range. This allows for different treatments for continuous gas evolution and one-time cavitation, reducing the impact of occasional headspace gas on surface density measurements.

[0021] In the tank bottom region, the event of the probe passing a lower turning point is used as a time marker for sediment disturbance. Using the reading of the same bottom clearance depth before the turnback as a baseline, when density and damping increase simultaneously after the turnback and then decrease as particles resettle, it can be confirmed that the high value originates from probe-induced resuspension. Based on this, the system retains the density before the turnback, discards the disturbed sequence, and sets the minimum effective detection boundary to avoid interpreting the silt, rust, or heavy phase particles swirled up by the probe itself as the original dense liquid layer in the tank. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the intelligent liquid density detection system for storage tanks provided in an embodiment of the present invention; Figure 2 A depth-density profile diagram illustrating the division of the reference region and interface candidate region in an embodiment of the present invention; Figure 3 A schematic diagram illustrating the mechanism of bidirectional arrival and film carrying of the interface candidate region target depth provided in an embodiment of the present invention; Figure 4 A coordinate diagram illustrating the convergence of a bidirectional frequency relaxation sequence provided in an embodiment of the present invention; Figure 5 A schematic flowchart of the intelligent detection method for liquid density in storage tanks provided in an embodiment of the present invention; Figure 6 A schematic diagram illustrating the mechanism of the cavitation criterion for the first liquid entry at the gas-liquid surface provided in an embodiment of the present invention; Figure 7 This is a schematic diagram illustrating the mechanism of resuspension induced by the bottom probe in an embodiment of the present invention. Figure 8 This is a schematic diagram of the modules of the intelligent liquid density detection system for storage tanks provided in an embodiment of the present invention.

[0023] Figure reference numerals: 10-Storage tank; 11-Lifting mechanism; 12-Density probe; 13-Vibration element; 14-Protective cover; 15-Position detection component; 16-Drive and frequency acquisition circuit; 17-Processor; 18-Upper liquid film; 19-Lower liquid film; 20-Upper reference area; 21-Lower reference area; 22-Interface candidate area; 23-Target depth; 24-Gas-liquid surface; 25-Storage tank head space; 26-Sediment; 27-Bottom clearance depth; 28-Turning depth; 29-Head space gas; 30-First frequency relaxation sequence; 31-Second frequency relaxation sequence; 32-Minimum effective detection boundary; 34-Density profile; 35-Effective transition zone; 40-Intelligent detection system for storage tank liquid density; 41-Lifting acquisition module; 42-Bidirectional sequence module; 43-Direction of arrival determination module; 44-Density determination module. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions in the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0025] Figure 1 This is a schematic diagram of the intelligent liquid density detection system for storage tanks provided in an embodiment of the present invention. The system is installed in storage tank 10 and is used to detect the liquid density at different heights in storage tank 10. Storage tank 10 contains stratified liquid from top to bottom, with the upper liquid surface being the gas-liquid surface 24. The liquid is divided along the height direction into an upper reference area 20, a lower reference area 21, and an interface candidate area 22 located between the two.

[0026] A lifting mechanism 11 is mounted on top of the storage tank 10, and its output end is connected to a density probe 12. This mechanism moves the density probe 12 up and down along the height of the storage tank 10 and stops at a selected depth. The sensitive element of the density probe 12 is a vibrating element 13. A protective cover 14 is fitted around the vibrating element 13, forming an open accommodating space around it. This allows the liquid being measured to submerge the vibrating element 13 while reducing direct scouring of the vibrating element 13 by the liquid flow during the lifting process. In some embodiments, the vibrating element 13 is a resonant cylinder or tuning fork, and its resonant frequency changes with the density of the adjacent liquid. A position detection component 15 works in conjunction with the lifting mechanism 11 to provide the current depth of the density probe 12 in real time. Its position repeatability accuracy can be 0.5 mm to 2 mm in some embodiments. The drive and frequency acquisition circuit 16 is connected to the vibration element 13. On the one hand, it excites the vibration element 13 to vibrate with a set drive amplitude, and on the other hand, it acquires the resonant frequency of the vibration element 13. The processor 17 receives the resonant frequency output by the drive and frequency acquisition circuit 16 and the depth output by the position detection component 15, calculates the liquid density at each depth according to the frequency-density correspondence, and executes the various judgments described below. In the above hardware, the resonant frequency is a measurement quantity that already exists in contact vibration density detection, so there is no need to add a separate sensor for this invention. Information such as depth, drive status, and direction of arrival are also recorded by the lifting mechanism 11, the position detection component 15, and the drive and frequency acquisition circuit 16 during normal operation for subsequent judgment.

[0027] Before formal testing, a frequency-density correspondence is established. In one embodiment, at least two standard liquids with known and traceable densities are used. At multiple temperature points covering the operating temperature range of the storage tank 10, the vibrating element 13 is immersed in the standard liquids, and its resonant frequency or vibration period is collected. Based on this, a correspondence between the resonant frequency and the liquid density is fitted, and temperature is incorporated into this relationship as a compensation variable. Once calibrated, this correspondence is fixed for use. All subsequent conversions from resonant frequency to density, including initial density, starting density, and final density, are based on the same frequency-density correspondence. During calibration, the fitting residual and the permissible error of the densitometer are recorded as references for the values ​​of the stability criteria in the following text. The calibration results have an expiration date and are recalibrated after the expiration date or after replacing the vibrating element 13. The consideration for this approach is that the value of liquid density is ultimately determined by the actual coupling relationship between the vibrating element 13 and the local liquid. Fixing the conversion relationship to the physical relationship obtained from a single calibration allows for the traceability of the density at each depth. The subsequent criteria only address the reliability of a particular reading and do not change the physical source of the density value itself.

[0028] Figure 5 This is a schematic flowchart of the intelligent liquid density detection method for storage tanks provided in an embodiment of the present invention. (Refer to...) Figure 5 The method first drives the density probe 12 to move along the height of the tank 10 and calculates the initial density to determine the upper reference region 20, the lower reference region 21, and the interface candidate region 22. Then, at the target depth 23 of the interface candidate region 22, the first frequency relaxation sequence 30 is acquired from the upper reference region 20 and then the second frequency relaxation sequence 31 is acquired from the lower reference region 21. The first and second initial densities biased towards their respective incoming layers are calculated respectively. Then, it is determined whether the two initial densities are biased towards their respective incoming layers and whether the difference between the end densities of the two sequences is not greater than the convergence threshold. If both the incoming and convergence aspects are met, the end frequencies of the two sequences are merged into a common frequency and the liquid density at the target depth 23 is calculated. If either aspect is not met, the two sequences are reacquired. Finally, the above process is repeated at multiple target depths 23 of the interface candidate region 22. Adjacent depths that both meet the convergence conditions are included in the effective transition region 35 and a density profile 34 is generated. Figure 5 The text within each box describes the processing content for each step, while the arrows and decision branches between boxes indicate the order and transfer relationships of the processing. Each step will be explained in detail below.

[0029] At the start of a complete inspection, the density probe 12 first performs a full tank scan. The lifting mechanism 11 drives the density probe 12 from top to bottom through each inspection depth of the storage tank 10. At each depth, the drive and frequency acquisition circuit 16 acquires the resonant frequency of the vibration element 13. The processor 17 converts it into the initial density at that depth according to the frequency-density correspondence, thereby obtaining the distribution of the initial density along the height of the storage tank 10. Figure 2 This is a schematic diagram of the depth-density profile for dividing the reference area and interface candidate area according to an embodiment of the present invention. Figure 2 The left side shows the stratification of storage tank 10 along its height, and the right side shows the corresponding depth-density profile. The density profile shown by label 34 is the solid line in the figure, with density on the horizontal axis and depth on the vertical axis, reflecting the change of initial density with depth.

[0030] When determining the stratification, the processor 17 examines the continuous change of initial density along the depth. For continuous detection depths, if the range of initial density variation is not greater than a reference stability threshold, this depth interval is defined as a reference zone; the reference zone located at the top of the tank 10 with lower density is the upper reference zone 20, and the reference zone located at the bottom with higher density is the lower reference zone 21. The reference density of each of the two adjacent reference zones is taken as the representative value of the initial density within that reference zone. In some embodiments, the average value of the initial density within the reference zone is taken, and in other embodiments, the median value is taken. When the difference in reference density between two adjacent reference zones is greater than the interface detection threshold, it indicates that there is a significant density transition between the two. The processor 17 defines the depth segment between these two reference zones as the interface candidate zone 22. The reason for first defining stable reference zones according to the reference stability threshold and then determining the transition segment between them according to the interface detection threshold is that the reference zone represents a liquid layer with a definite probe direction and stable readings, while the interface candidate zone 22 is a segment where the reading at the same depth may change with the probe direction. Only by separately identifying the latter is it necessary to perform subsequent bidirectional acquisition within it.

[0031] The reference stability threshold reflects the permissible fluctuations in initial density within the same liquid layer. In some embodiments, the reference stability threshold can be taken as 1 to 3 times the calibrated permissible error of the densitometer, typically 2 times. If the reference stability threshold is too small, minor fluctuations within a normal liquid layer will be mistakenly divided into multiple reference regions, causing the interface candidate region 22 to be overly fragmented; if it is too large, genuine stratification with slow transitions will be incorporated into the same reference region, leading to the missed detection of the interface candidate region 22. The interface detection threshold reflects the density difference between adjacent liquid layers that is sufficient to be considered as stratification. Its value is greater than the reference stability threshold, thereby distinguishing genuine transitions between liquid layers from normal fluctuations within the liquid layer; in some embodiments, the interface detection threshold can be taken as 3 to 10 times the calibrated permissible error of the densitometer, typically 5 times. If the interface detection threshold is too large, genuine but small-difference emulsion transition layers will be missed; if it is too small, normal fluctuations within the same liquid layer will be misjudged as interfaces. In the above-mentioned whole-can scanning, the resonant frequency acquisition uses a frequency sampling rate of 20 Hz to 100 Hz in some embodiments. If the sampling rate is too low, it is insufficient to characterize the relaxation process, and if it is too high, it increases the amount of data without providing substantial help for the determination.

[0032] After selecting the target depth 23 within the interface candidate area 22, the density probe 12 performs two acquisitions at the same target depth 23 with opposite directions. Figure 3 This is a schematic diagram illustrating the mechanism of bidirectional arrival at the target depth of the interface candidate region and film carrying provided in an embodiment of the present invention. Figure 3 The two data acquisitions at the same target depth 23 are shown in a two-column comparison. During the first acquisition, the density probe 12 descended from the upper reference zone 20 to the target depth 23, as shown... Figure 3 As shown in the left column, at this time, the upper liquid film 18 brought in by the probe from the upper layer remains around the vibrating element 13. The drive and frequency acquisition circuit 16 continuously acquires the resonant frequency of the vibrating element 13 after the density probe 12 reaches the target depth 23, obtaining the first frequency relaxation sequence 30. During the second acquisition, the density probe 12 first crosses the target depth 23 and continues downward into the lower reference region 21, then ascends from the lower reference region 21 back to the target depth 23. (The text repeats itself here.) Figure 3 As shown in the right column, the liquid film 19 brought in from the lower layer by the probe remains around the vibrating element 13 at this time; the drive and frequency acquisition circuit 16 continuously acquires the resonant frequency after returning to the target depth 23, and obtains the second frequency relaxation sequence 31. The processor 17 takes the frequencies of the starting points of the first frequency relaxation sequence 30 and the second frequency relaxation sequence 31 respectively, and converts them into the first starting density and the second starting density according to the frequency-density correspondence.

[0033] The reason for arranging two samplings at the same target depth 23 with opposite directions is that the true local liquid density at the target depth 23 depends only on the liquid itself and is independent of the direction from which the probe arrives. The liquid film carried by the vibrating element 13 has a direction memory: when arriving from the upper layer, it carries the upper liquid film 18, and the initial reading is biased towards the reference density of the upper reference area 20; when returning from the lower layer, it carries the lower liquid film 19, and the initial reading is biased towards the reference density of the lower reference area 21. The two frequency relaxation sequences converge towards the same true density from different starting points as their respective incoming liquid films are gradually replaced by the liquid at the target depth 23. In some embodiments, the positional deviation of the target depth 23 in the two samplings is controlled within the repeatability accuracy of the position detection component 15, and the ascending and descending speeds and dwell times of the two descents and ascents are consistent with the sampling clock. Temperature, depth, drive state, and arrival direction are recorded simultaneously for subsequent determination. In some embodiments, the starting point is the first sampling point of the sequence; in other embodiments, it is a representative value within a small sampling window at the beginning of the sequence to reduce the impact of single-point noise on the initial density.

[0034] After being fully replaced by the liquid at the target depth of 23, the two frequency relaxation sequences tended to stabilize. Figure 4 This is a coordinate diagram illustrating the convergence of a bidirectional frequency relaxation sequence provided in an embodiment of the present invention. Figure 4 The horizontal axis represents the sampling point, and the vertical axis represents the density. The first frequency relaxation sequence, indicated by label 30, is the solid line in the figure, which moves upward from the starting point biased towards the upper reference density. The second frequency relaxation sequence, indicated by label 31, is the dashed line in the figure, which moves downward from the starting point biased towards the lower reference density. As the acquisition progresses, the two sequences move towards the common end located between the upper and lower reference densities.

[0035] Processor 17 obtains the end densities of the two sequences as follows: For each frequency relaxation sequence, a continuous measurement window is selected forward from its end, and the density within the window, calculated from the resonant frequency, is examined; if the difference between the maximum and minimum density within a certain continuous measurement window is not greater than a stability threshold, then that measurement window is determined as the end-stabilization window, such as... Figure 4The section near the end is shown. The average density within the end stabilization window is used as the end density of the sequence, while the average resonant frequency within the end stabilization windows of the two sequences is used for subsequent merging of common frequencies. The stabilization threshold reflects the allowable residual fluctuations after the sequence enters a stable state. In some embodiments, it can be taken as 0.5 to 2 times the allowable error of the calibrated densitometer, typically 1 times. If the stabilization threshold is too small, normal sampling noise will prevent the end stabilization window from forming in a long time. If it is too large, the transition section that has not yet completed the replacement will be misjudged as stable, causing the end density to deviate from the true value. The number of sampling points for the end stabilization window is taken as 20 to 60 sampling points in some embodiments. If the number of sampling points is too small, the average within the window is difficult to suppress noise. If the number of sampling points is too large, the samples that have not yet stabilized will be included in the window. The single relaxation acquisition time for each sequence is taken as 5 to 30 seconds in some embodiments. If the acquisition time is too short, it is not enough to form an effective end stabilization window. If it is too long, it will reduce the efficiency of the whole tank detection. Using the average value within the window, rather than a single end sampling point, as the source of the end density and common frequency is to prevent the representative value from being affected by fluctuations in individual sampling points. In other embodiments, the end density and common frequency may also be the median value within the end stabilization window, and this embodiment of the invention does not impose any restrictions on this.

[0036] After obtaining the two initial densities and the terminal densities of the two sequences, the processor 17 determines whether the bidirectional acquisition is acceptable, considering both the direction of arrival and convergence. Regarding the direction of arrival, the difference between the first initial density and the reference density of the upper reference region 20 is required to be less than the difference between it and the reference density of the lower reference region 21, and simultaneously, the difference between the second initial density and the reference density of the lower reference region 21 is less than the difference between it and the reference density of the upper reference region 20; that is, the two initial densities are indeed biased towards their respective incoming layers. Regarding convergence, the difference between the terminal densities of the two sequences is required to be no greater than the convergence threshold; that is, the two acquisitions with opposite directions of arrival converge to a consistent density at the end. Only when both the direction of arrival and convergence are met simultaneously does the processor 17 merge the resonant frequencies within the stabilization window at the end of the two frequency relaxation sequences into a common frequency, and convert the common frequency into the liquid density at the target depth 23 according to the frequency-density correspondence.

[0037] The convergence threshold is a key quantity in the acceptance criteria, reflecting the permissible residual difference at the end of two sequences traveling in opposite directions. In some embodiments, the convergence threshold can be set to 1 to 3 times the permissible error of the calibrated densitometer, typically 2 times. If the convergence threshold is set too low, even if the two sequences have actually converged to the same liquid state, they will be judged as non-converged due to residual noise, resulting in excessive rejection of valid readings; if it is set too high, situations where the probe has not yet broken free of the incoming liquid film and the two sequences still have systematic deviations will be misjudged as convergence, causing the deviation caused by the film to be mixed into the density results. The convergence threshold, stability threshold, and the aforementioned reference stability threshold all vary with liquid viscosity, wettability, and temperature. In some embodiments, they are calibrated separately according to different operating conditions during the calibration stage and stored as lookup tables. During detection, the corresponding value is selected according to the current temperature and liquid type. The aforementioned joint judgment of direction and convergence ensures that each accepted density point has two verifiable criteria: a starting point bias with opposite directions and a common endpoint regardless of direction. This excludes the false transition zone created by the probe itself carrying the liquid phase from the results, while retaining the true emulsion region or intermediate density layer that can converge stably in both directions.

[0038] When the difference in density at the ends of the two sequences exceeds the convergence threshold, it indicates that the bidirectional acquisition has not yet converged. The processor 17 does not immediately provide the density but instead performs a release retest. During the release retest, the drive and frequency acquisition circuit 16 drives the vibration element 13 with a release drive amplitude greater than the measurement drive amplitude used in conventional acquisition but not exceeding the rated drive range of the vibration element 13. In some embodiments, the release drive amplitude is 1.2 to 2 times the measurement drive amplitude. The reason for increasing the drive amplitude during the retest is that a larger vibration helps accelerate the shedding and replacement of the liquid film around the vibration element 13, causing the two sequences to converge faster. Limiting it to the rated drive range avoids damage to the vibration element 13 or disturbance to the real interface. After increasing the drive amplitude, the density probe 12 repeatedly acquires the first frequency relaxation sequence 30 and the second frequency relaxation sequence 31 and determines convergence again. If the difference in end densities between the two sequences after release and retest is still greater than the convergence threshold, it is determined that there is a persistent film-carrying structure at the target depth 23 that the probe cannot get rid of within the allowable operation, or a dynamic interface that changes over time. The processor 17 outputs a persistent film-carrying structure or dynamic interface marker accordingly, instead of forcibly giving a density value by extending the wait or extrapolating.

[0039] To obtain the complete density distribution within the interface candidate region 22, the density probe 12 selects multiple target depths 23 within the interface candidate region 22 and performs the aforementioned bidirectional acquisition and acceptance judgment one by one to obtain the accepted liquid density at each target depth 23. Based on this, the processor 17 examines the convergence of these target depths 23, determining the depth segments among multiple adjacent target depths 23 that simultaneously satisfy the condition that the difference between the densities at the ends of two sequences is not greater than the convergence threshold as effective transition regions 35. Then, a density profile 34 of the effective transition region 35 is generated according to each target depth 23 and its corresponding liquid density. (Refer to...) Figure 2 The effective transition zone 35 corresponds to the depth band in the figure where the density changes abruptly. Within this depth band, the density profile 34 smoothly transitions from the upper reference density to the lower reference density. In the density profile 34 obtained in this way, the density at each depth comes from a single accepted bidirectional acquisition. Depths that do not pass acceptance are retained as markers and are not written into the density profile 34. This ensures that the density profile 34 reflects the true layer transition without being contaminated by false transition bands caused by the probe's movement history.

[0040] In addition to the liquid-liquid interface, the density probe 12 will also encounter two other types of interference caused by the probe's movement history near the gas-liquid surface 24 and at the bottom of the storage tank 10. The embodiments of the present invention identify these interferences in a similar bidirectional comparison manner. Figure 6 This is a schematic diagram illustrating the mechanism of the first liquid cavitation criterion at the gas-liquid surface provided in an embodiment of the present invention. When the dry density probe 12 first descends from the storage tank space 25, passing over the gas-liquid surface 24, and enters the liquid, as... Figure 6 As shown in the left column, the gaps in the vibrating element 13 or the dead corners of the protective cover 14 may trap a portion of the gas 29 in the head space, causing air bubbles to mix into the medium adjacent to the vibrating element 13; after the density probe 12 reaches the target depth 23 of the surface, it acquires the initial liquid entry frequency sequence and simultaneously the drive and frequency acquisition circuit 16 records the initial drive energy sequence required to maintain this vibration amplitude. Subsequently, as Figure 6 As shown in the right column, the density probe 12 remains fully submerged, moves downwards for one venting stroke, and then returns to the same surface target depth 23 from below, collecting the full immersion re-entry frequency sequence and re-entry drive energy sequence. Since this re-entry is inside the liquid, the gas 29 in the head space is no longer trapped around the vibrating element 13. In some embodiments, the venting stroke is 2 to 5 times the axial length of the vibrating element 13. Its function is to completely remove the vibrating element 13 from the gas-trapping position during the initial liquid immersion and allow the trapped gas to be discharged during the downward movement.

[0041] Processor 17 identifies cavitation by comparing the initial liquid immersion with the full immersion re-immersion. When the end density obtained from the initial liquid immersion is lower than that obtained from the full immersion re-immersion, and the difference between the two is greater than the cavitation density threshold, and simultaneously the end driving energy obtained from the initial liquid immersion is higher than that obtained from the full immersion re-immersion, and the difference between the two is greater than the cavitation energy threshold, processor 17 determines that gas 29 in the head space was trapped during the initial liquid immersion and outputs an initial liquid immersion cavitation marker. The reason for using both low density and high driving energy as dual criteria is that when air bubbles are mixed into the medium adjacent to the vibrating element 13, on the one hand, the added mass decreases, making the apparent density appear lower; on the other hand, the added dissipation of the gas and liquid phases increases, increasing the driving energy required to maintain vibration. Both characteristics must appear simultaneously to distinguish it from a true low-density surface. In some embodiments, the cavitation density threshold is taken as 2 to 5 times the allowable error of the calibrated densitometer, and in some embodiments, the cavitation energy threshold is taken as 0.1 to 0.3 times the end driving energy of the full immersion re-immersion. In some embodiments, the driving energy is accumulated by the driving and frequency acquisition circuit 16 over a sampling period as the product of the excitation current and excitation voltage required to maintain a set vibration amplitude. In some embodiments, a cavitation criterion is calibrated in a transparent calibration cavity to synchronously image and confirm the presence of gas in the vibrating element 13 neighborhood, covering different gas volumes, surface tensions, protective cover 14 orientations, and liquid ingress velocities.

[0042] For occasional one-time cavitation and continuous gas evolution at 24 locations on the gas-liquid surface, this embodiment of the invention distinguishes between them based on the repeatability of the re-entry results. The density probe 12 repeatedly performs the aforementioned exhaust stroke and full immersion re-entry acquisition to obtain the terminal density and terminal driving energy of multiple re-entries. When the difference between the terminal densities obtained from two adjacent re-entries is not greater than the density repeatability threshold, and the difference between the terminal driving energies obtained from two adjacent re-entries is not greater than the energy repeatability threshold, it indicates that the re-entry results have entered a stable repeatability range. Based on this, the processor 17 calculates the liquid density at the target depth 23 of the surface layer according to the frequency-density correspondence using the terminal frequency of the last re-entry. If the re-entry results consistently do not enter the repeatability range, it indicates that there is continuous gas evolution rather than one-time gas interception. The marker is retained without providing a density, following the aforementioned handling method for membrane carrying and dynamic interfaces. In some embodiments, the density repeatability threshold and energy repeatability threshold are taken as 0.3 to 0.6 times the cavitation density threshold and cavitation energy threshold, respectively. These values ​​are less than the corresponding cavitation thresholds to ensure that the determination of entering the repeatability range is stricter than the detection of cavitation.

[0043] Figure 7 This is a schematic diagram illustrating the mechanism of resuspension induced by the bottom probe in an embodiment of the present invention. When the density probe 12 descends to the bottom of the storage tank 10, the frequency and damping before the reversal are first collected at a bottom clearance depth of 27. Figure 7As shown in the left column, the bottom sediment 26 remains stationary at this point. Subsequently, the density probe 12 continues to descend to the turning depth 28, then returns to the bottom clearance depth 27, acquiring the frequency relaxation sequence and the damping relaxation sequence after the turnaround; as shown... Figure 7 As shown in the right column, the probe's movement to descend to the turning depth 28 and then return disturbs the sediment 26, causing the particles to resuspend in the liquid near the bottom clearance depth 27. The turning depth 28 is lower than the bottom clearance depth 27, and the difference between the two is, in some embodiments, 3 to 8 times the axial length of the vibrating element 13. In some embodiments, the damping is obtained by the drive and frequency acquisition circuit 16 based on the logarithmic decay rate of the free decay of the vibrating element 13 or the drive energy required to maintain constant amplitude vibration, reflecting the dissipation of vibration by the liquid.

[0044] Processor 17 identifies probe-induced resuspension based on a comparison before and after the reversal. When the initial density after reversal is higher than the density before reversal, and the difference between the two is greater than the resuspension density threshold, and the initial damping after reversal is higher than the damping before reversal, and the difference between the two is greater than the resuspension damping threshold, and the density and damping after reversal decrease over time relative to the density and damping before reversal, respectively, processor 17 determines that the high density and high damping at the bottom clearance depth 27 originate from the resuspension of the sediment 26 rolled up by the probe during reversal, rather than the original dense liquid layer of tank 10, and outputs a probe-induced resuspension marker. The reason for using density increase, damping increase, and subsequent decay as criteria simultaneously is that suspended particles increase both apparent density and the dissipation of vibration by the liquid, while the resettling of particles will cause these two values ​​to decrease over time. Only when all three occur simultaneously can the transient high value induced by the probe be distinguished from the true dense bottom layer. In some embodiments, the resuspension density threshold is taken as 3 to 8 times the permissible error of the calibrated density meter, and the resuspension damping threshold is taken as 0.1 to 0.3 times the damping before the re-entry. In some embodiments, the resuspension criterion is calibrated in a calibration tank with a controllable sedimentation bed, and the increase in particles after the re-entry is confirmed by independent particle concentration detection, covering different particle sizes, sedimentation bed thicknesses, liquid viscosities, and turning velocities. For high-viscosity liquids with slow particle decay, if the density and damping after the re-entry fail to decay back to the pre-re-entry level within the permissible observation time, the processor 17 outputs probes to induce a resuspension marker, without forcibly providing the static density at that point.

[0045] After the output probe induces the resuspension marker, the processor 17 processes the density results of the bottom region accordingly. For the bottom clearance depth 27 itself, since the reading before the turnaround is taken from a time when the sediment 26 is undisturbed, the processor 17 calculates the liquid density of the bottom clearance depth 27 according to the frequency-density correspondence based on the frequency before the turnaround, as a reliable result for this depth. For the area near the turning depth 28 that is most disturbed, the processor 17 determines the adjacent detection level above the turning depth 28 as the lowest effective detection boundary 32, such as... Figure 7 As shown, the minimum effective detection boundary 32 is located between the bottom clearance depth 27 and the turning depth 28. The frequency relaxation sequence after the reversal, located below the minimum effective detection boundary 32, is excluded from the density profile 34 of the tank 10 because it is within the range of direct disturbance caused by the probe reversal, and does not participate in the generation of the density profile 34. The reason for this is that the reading before the reversal and the reading after the reversal correspond to two states of the sediment 26 before and after the disturbance, respectively. By retaining the former and discarding the latter, the reliable density of the bottom clearance depth 27 is maintained, and the mud and heavy phase particles rolled up by the probe itself are avoided from being written into the density profile 34 as the original dense liquid layer of the tank 10.

[0046] In some embodiments, the processor 17 also constrains the changes in the state of the tank 10 itself during a bidirectional acquisition. There is a time interval between two acquisitions from opposite directions at the same target depth 23. If, during this period, feeding, discharging, or strong agitation causes a significant drift in the reference density between the upper reference zone 20 and the lower reference zone 21, then the two acquisitions no longer represent the same liquid state. Therefore, the processor 17 considers the drift in reference density between the upper reference zone 20 and the lower reference zone 21 between the two acquisitions to be a valid round trip as the amount of drift not exceeding a set drift threshold. When the drift exceeds this threshold, the bidirectional acquisition is deemed invalid, the processor 17 refuses to output the density accordingly, and re-acquires the density after the state of the tank 10 stabilizes again. Accordingly, this embodiment of the invention is applicable to situations where the stratified state of the tank 10 is basically stable during two round trips. For strong disturbances where a real density change occurs during the round trip, the processor rejects the output rather than providing a questionable density.

[0047] After the above processing, the processor 17 provides structured detection results for each depth of the storage tank 10. In some embodiments, the results output by the processor 17 include the target depth 23 and its liquid density, the combined common frequency and the range of the end-stability window, and several decision markers. The aforementioned decision markers include continuous film-carrying or dynamic interface markers, first-time liquid cavitation markers, and probe-induced resuspension markers, as well as the lowest effective detection boundary 32. Based on this, accepted effective values, rejected values, and locations requiring retesting are distinguished in the density profile 34. The consideration for organizing the output in this way is that each density value is accompanied by the basis for its acceptance or the reason for its rejection, so that the stratified density results of the storage tank 10 can be used for measurement and condition identification, and can be traced back to the specific acquisition and decision process.

[0048] This invention also provides an intelligent liquid density detection system 40 for storage tanks, used to implement the aforementioned detection method. Figure 8 This is a schematic diagram of the modules of the intelligent liquid density detection system for storage tanks provided in an embodiment of the present invention. (Refer to...) Figure 8The intelligent liquid density detection system 40 for storage tanks includes a lifting acquisition module 41, a bidirectional sequence module 42, an incoming direction determination module 43, and a density determination module 44. The lifting acquisition module 41 drives a vibrating element-type density probe 12 to move along the height of the storage tank 10, acquiring the resonant frequency of the vibrating element 13 and converting it into an initial density. Based on this, it determines the upper reference area 20, the lower reference area 21, and the interface candidate area 22, corresponding to the aforementioned full-tank scanning and zoning processing. The bidirectional sequence module 42 acquires a first frequency relaxation sequence 30 from the upper reference area 20 and a second frequency relaxation sequence 31 from the lower reference area 21 at the target depth 23 of the interface candidate area 22. It then calculates the first and second initial densities from the frequencies at the starting points of the two sequences, corresponding to the aforementioned bidirectional acquisition processing. The incoming direction determination module 43 compares the differences between the first and second initial densities and the upper and lower reference densities, respectively, and compares the difference between the densities at the ends of the two sequences with a convergence threshold, corresponding to the aforementioned joint judgment of incoming direction and convergence. The density determination module 44 is used to merge the end frequencies of the two sequences into a common frequency when both the direction of arrival and the convergence condition are met, and convert the common frequency into the liquid density at the target depth 23 according to the frequency-density correspondence, corresponding to the density conversion after acceptance mentioned above. Each module corresponds to the steps of the aforementioned method, and its specific processing is the same as before, so it will not be repeated. In some embodiments, the lifting acquisition module 41, the bidirectional sequence module 42, the direction of arrival determination module 43, and the density determination module 44 are carried by the processor 17, and the resonant frequency and depth are obtained through the drive and frequency acquisition circuit 16 and the position detection component 15.

[0049] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit them. Those skilled in the art should understand that modifications can be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features, and such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for intelligent detection of liquid density in storage tanks, characterized in that, include: The vibration element density probe is moved along the height of the storage tank to collect the resonant frequency of the vibration element and convert it into the initial density, thereby determining the upper reference area, the lower reference area, and the candidate interface area between them. Select the target depth within the interface candidate region, reach the target depth from the upper reference region, and acquire a first frequency relaxation sequence; The target depth is crossed into the lower reference region, and then the lower reference region is returned to the target depth to acquire the second frequency relaxation sequence; The frequencies at the starting points of the two frequency relaxation sequences are respectively converted into a first starting density and a second starting density; When the difference between the first starting density and the upper reference density is less than the difference between the first starting density and the lower reference density, the difference between the second starting density and the lower reference density is less than the difference between the second starting density and the upper reference density, and the difference between the terminal densities of the two frequency relaxation sequences is not greater than the convergence threshold, the terminal frequencies of the two frequency relaxation sequences are merged into a common frequency. The common frequency is converted into the liquid density at the target depth according to the frequency-density correspondence.

2. The intelligent detection method for liquid density in storage tanks according to claim 1, characterized in that, The process of determining the upper reference region, the lower reference region, and the interface candidate region between them includes: determining the segment whose initial density variation range at continuous detection depth is no greater than the reference stability threshold as the reference region; When the difference in reference density between two adjacent reference regions is greater than the interface detection threshold, the depth segment located between the two reference regions is determined as the interface candidate region.

3. The intelligent detection method for liquid density in storage tanks according to claim 1, characterized in that, The acquisition of the terminal density includes: selecting a continuous measurement window forward from the end of each frequency relaxation sequence; A continuous measurement window in which the difference between the maximum and minimum density values ​​within the window is not greater than the stability threshold is defined as the end-stable window. The average density within the end stabilization window is determined as the end density; The average frequency within the two end-stabilization windows is determined as the common frequency.

4. The intelligent detection method for liquid density in storage tanks according to claim 1, characterized in that, When the difference between the two end densities is greater than the convergence threshold, the vibration element is driven with a release drive amplitude that is greater than the measured drive amplitude and within the rated drive range of the vibration element, and the first frequency relaxation sequence and the second frequency relaxation sequence are repeatedly acquired. If the difference between the two end densities after repeated sampling is still greater than the convergence threshold, a continuous membrane or dynamic interface marker is output.

5. The intelligent detection method for liquid density in storage tanks according to claim 1, characterized in that, Multiple target depths are selected within the interface candidate region, and the liquid density is determined for each. The effective transition zone is defined as the segment in which the difference between the two terminal densities of multiple adjacent target depths is not greater than the convergence threshold. The density profile of the effective transition zone is generated according to each target depth and the corresponding liquid density.

6. The intelligent detection method for liquid density in storage tanks according to claim 1, characterized in that, After the density probe passes the gas-liquid surface for the first time from the storage tank space, it collects the first liquid entry frequency sequence and the first driving energy sequence at the target depth on the surface. Keeping the density probe fully submerged, the density probe is moved down by one exhaust stroke and returned from below to the target depth of the surface layer, and the full immersion reentry frequency sequence and reentry drive energy sequence are collected; When the initial terminal density is lower than the re-entry terminal density and the difference between the two is greater than the cavitation density threshold, and the initial terminal driving energy is higher than the re-entry terminal driving energy and the difference between the two is greater than the cavitation energy threshold, the initial liquid entry cavitation marker is output.

7. The intelligent detection method for liquid density in storage tanks according to claim 6, characterized in that, Repeat the exhaust stroke and the acquisition of the full immersion reentry frequency sequence; When the difference between the densities of two adjacent reentry ends is not greater than the density repeatability threshold, and the difference between the driving energies of two adjacent reentry ends is not greater than the energy repeatability threshold, the frequency of the last reentry end is converted into the liquid density at the target depth of the surface layer according to the frequency-density correspondence.

8. The intelligent detection method for liquid density in storage tanks according to claim 1, characterized in that, The frequency and damping before the turnback were collected when the density probe descended to the bottom clearance depth. The density probe is lowered further to the turning depth, then returned to the bottom clearance depth and the frequency relaxation sequence and damping relaxation sequence after the turnaround are collected; When the initial density after reversal is higher than the density before reversal and the difference between the two is greater than the resuspension density threshold, the initial damping after reversal is higher than the damping before reversal and the difference between the two is greater than the resuspension damping threshold, and the density after reversal and the damping after reversal decrease towards the density before reversal and the damping before reversal, respectively, the output probe induces a resuspension marker.

9. The intelligent detection method for liquid density in storage tanks according to claim 8, characterized in that, After the probe induces a resuspension marker, the frequency before the reversal is converted into the liquid density at the bottom clearance depth according to the frequency-density correspondence. The adjacent detection gears above the steering depth are defined as the lowest effective detection boundary; The frequency relaxation sequence after the return, located below the minimum effective detection boundary, is excluded from the tank density profile.

10. A smart detection system for liquid density in a storage tank for implementing the method of claim 1, characterized in that, include: The lifting acquisition module is used to drive the vibration element density probe to move along the height of the storage tank, acquire the resonant frequency of the vibration element and convert it into the initial density, and determine the upper reference area, the lower reference area and the interface candidate area between the two. The bidirectional sequence module is used to select the target depth within the interface candidate area, reach the target depth from the upper reference area and collect a first frequency relaxation sequence, cross the target depth and enter the lower reference area, and then return to the target depth from the lower reference area and collect a second frequency relaxation sequence. The frequencies at the starting points of the two frequency relaxation sequences are converted into a first starting density and a second starting density, respectively. The direction determination module is used to compare the differences between the first starting density and the second starting density and the upper reference density and the lower reference density, respectively, and to compare the difference between the terminal densities of the two frequency relaxation sequences and the convergence threshold. The density determination module is used to merge the terminal frequencies of the two frequency relaxation sequences into a common frequency when the difference between the first initial density and the upper reference density is less than the difference between the first initial density and the lower reference density, the difference between the second initial density and the lower reference density is less than the difference between the second initial density and the upper reference density, and the difference between the terminal densities of the two frequency relaxation sequences is not greater than the convergence threshold, and convert the common frequency into the liquid density at the target depth according to the frequency-density correspondence.

Images