Method for identifying deposits in a process plant using a Coriolis mass flow sensor

Abstract

Method (100) for classifying coatings (230) in a measuring tube (200), comprising: determining (110) a damping value D for at least one vibration mode of an oscillator with at least one measuring tube (200) of a Coriolis mass flow sensor for guiding a medium (220); determining (120) a modal stiffness value k for at least one vibration mode; and classifying (150) the coating (230) as a function of the damping value D and the stiffness value k.

Description

The present invention relates to a method for identifying deposits in a process installation having a Coriolis mass flow measurement sensor.Generic methods are described in WO 2007 / 045539 A2, DE 10 2011 080 415 A1, EP 02 513 612 B1 and DE 10 2018 101 923 A1. The publication WO 2007 / 045539 A2 teaches to interpret a change in the torsional vibration frequency of a straight measuring tube as an indicator of a mass covering. The method according to the laid-open specification DE 10 2011 080 415 A1 evaluates changes in thermal properties of a mass flow sensor in order to conclude that there is formation of deposits or corrosion. EP 02 513 612 B1 and DE 10 2018 101 923 A1 teach interpreting an increased damping of the flexural vibration mode as an indicator of formation of a coating, wherein a control test is provided in each case in order to exclude a polyphase medium as the cause of damping. EP 02 513 612 B1 teaches to evaluate the characteristic of harmonics, i.e. whether the amplitudes of the harmonics correspond to an expected value for formation of deposits. DE 10 2018 101 923 A1, on the other hand, checks whether the so-called resonator effect can be excluded as the cause of the attenuation. Although the above-mentioned methods may indicate formation of deposits as a result, they let the operator of a plant know in the unknown what type of deposit is. It is the object of the invention to remedy this situation.The object is achieved by the method according to independent claim 1.The method according to the invention for classifying coatings in a measuring tube comprises:determining an attenuation value for at least one oscillation mode of an oscillator having at least one measuring tube of a Coriolis mass flow measuring sensor for conducting a medium;determining a modal stiffness value for at least one vibration mode; andclassifying a lining as a function of the damping value and the stiffness value.In a further development of the invention, the method further comprises: determining, on the basis of a comparison of the damping value with a reference value, whether an indication of formation of a coating is given.In a further development of the invention, the method further comprises: checking whether compressibility of the medium guided in the measuring tube can be excluded as a cause of the attenuation value.In a further development of the invention, the method further comprises: checking whether a fluctuation of a resonant frequency of the measuring tube can be excluded as a cause of the attenuation value.In a further development of the invention, classifying a lining comprises assigning the lining to a lining class at least on the basis of the stiffness value.In a further development of the invention, the assignment of the lining to a lining class takes place as a function of a relationship between the stiffness value and the damping value.In a further development of the invention, the method further comprises the determination of a mass value of the lining, wherein the mass value enters into the classification of the lining in relation to the stiffness value and / or damping value.In a further development of the invention, the mass value of the coating is determined on the basis of the change of a resonant frequency of the vibrating measuring tube.In a further development of the invention, the attenuation value is determined on the basis of a ratio of excitation current for driving the vibration mode in resonance and an amplitude of the vibration mode achieved therewith.In a further development of the invention, the attenuation value is determined on the basis of the decay of the oscillation of an oscillation mode when the exciter is switched off.In a further development of the invention, the determination of the stiffness value is effected by exciting an oscillation mode out of resonance and determining a relationship between the oscillation amplitude out of resonance and the excitation current out of resonance. Here, the excitation can take place intermittently with respect to the resonance excitation, except for resonance, as is described, for example, in WO 2012 / 062551 A1, or continuously according to the teaching of the post-published patent application DE 10 2019 124 709 A1.The invention is explained in more detail below with reference to the exemplary embodiments shown in the drawings. It shows: FIG. 1 is a diagram illustrating different classes of plots in a two-dimensional parameter space; FIG. 2 shows a flow diagram of an exemplary embodiment of the method according to the invention; FIG. 3 ashows a schematic cross section through a straight measuring tube in a flexural oscillation mode for carrying out an exemplary embodiment of the method according to the invention; and FIG. 3 bshows a schematic cross section through the straight measuring tube in a torsional vibration mode for carrying out an exemplary embodiment of the method according to the invention.The diagram in FIGS. 1 and 1 shows typical positions of data tuple of attenuation D and modal stiffness k of a measuring tube, which were obtained in a process plant in which monomers in a solvent flow through a pipeline in which a Coriolis mass flow measuring sensor is arranged. Depending on the process conditions, different coating shapes are produced in the pipeline or the Coriolis mass flow sensor. The ellipses mark regions in which combinations of a damping D and a modal stiffness k of an oscillation mode of a measuring tube of a Coriolis mass flow sensor have been observed in a clustered manner. A rigidity reference value k 0 denotes the modal rigidity of the measuring tube in the cleaned, coating-free state. A damping reference value D 0 denotes the damping of the vibration mode in the cleaned, coating-free state, wherein the measuring tube is filled with a medium, the flow of which is to be measured by the Coriolis mass flow measuring sensor. The horizontal spacing of the grid lines of the diagram corresponds to a change in modal attenuation by the amount of the attenuation reference value D 0. The dotted ellipse identifies a lining class W for soft linings which merely bring about an increased damping D of the vibration mode, but have hardly any effect on the modal stiffness k of the measuring tube. The broken ellipse, on the other hand, identifies a lining class H for hard linings which not only increase the damping D but also the modal stiffness k. Around the reference values for damping and stiffness D 0, k 0 a zero range 0 is defined, which is marked with the solid circular arc and in which the deviations from the reference values are still so small that a reliable classification is not yet possible. In addition, classification at this stage is also not absolutely necessary in most cases, since the impairment of the flow measurement or of the process control is still negligible. In principle, it is helpful for the operator of a process plant to obtain knowledge about the type of coating, since this can result on the one hand in approaches for process control and on the other hand in the cleaning of the pipeline in which the Coriolis mass flow measuring sensor is installed being able to be carried out more efficiently. This knowledge is provided by the method according to the invention, for which an exemplary embodiment is now explained with reference to FIG. 2.Method 100 begins with ascertaining 110 attenuation D of an oscillation mode, in particular the bending oscillation usable mode of a measuring tube of a Coriolis mass flow sensor. The attenuation D can be characterized, for example, on the basis of the ratio of an excitation current for maintaining the oscillation and the oscillation amplitude at resonance achieved therewith. Likewise, the attenuation can be characterized by the reciprocal of a time constant with which the oscillation decays after the excitation current has been switched off.In a next step (120), a modal stiffness k of the vibration mode is determined. In this case, the modal stiffness is proportional to a quotient of a quality-independent oscillation amplitude and the excitation current used for this purpose. For this purpose, the oscillation amplitude is to be measured out of resonance, so that the influence of the damping or quality on the oscillation amplitude is negligible. The excitation of this oscillation out of resonance and measurement of the associated amplitude can take place alternately to normal measurement operation periodically or if necessary, wherein the excitation takes place in particular with the entire available excitation power. Details thereof are taught, for example, in international publication WO 2012 062551 A1, which is referred to here for details of the amplitude measurement. Instead of the alternating excitation, the oscillation except for resonance can also take place simultaneously with the resonance excitation, wherein in this case the available power for the excitation except for resonance is significantly lower than in the alternating case, since here a sufficient amplitude is required for the flow measurement. As a result, the amplitude of the vibration other than resonance is very small. With suitable filter algorithms, the amplitude value can nevertheless be determined, as is described in the post-published patent application DE 10 2019 124 709 A1, which is referred to here for details of the amplitude measurement. According to this method, updated values for the amplitude out of resonance or the modal stiffness k are available at a lower cycle rate, but this is not problematic in that a coating usually forms over a period of days or weeks, so that updates for the values for the modal stiffness k in the period of minutes or hours are completely sufficient.If there is a tuple of values for the damping D and the modal stiffness k, an optional series of test steps is first carried out. In this case, a check is made in a first test step 132 as to whether the magnitude of a vector (D,k) reaches a minimum value, which can be, for example, the radius of a circle which bounds the zero range in FIG. 1. In principle, however, only the attenuation value D can also be compared with a corresponding minimum value, as long as the minimum value is not reached, the method is terminated and begins again from the beginning. However, when the minimum value is determined to be reached, other causes of an increase in attenuation D are to be excluded. For this purpose, a check is made in a second test step 134 as to whether density fluctuations which are reflected in frequency fluctuations remain below a critical value. Such density fluctuations can be caused in particular by free gas bubbles in a liquid and bring about a considerable damping of the measuring tube oscillations. If such critical fluctuations are detected, the method is terminated; otherwise, in a third test step 136, a check is made as to whether the so-called resonator effect as a cause of the observed attenuation D can be excluded. In the resonator effect, suspended microbubbles in a liquid have the result that the liquid which is compressible due to the microbubbles oscillates against the oscillating measuring tube containing it, and the latter thus extracts oscillation energy. Details of the procedure for how the presence of the resonator effect can be detected by density measurements at two different frequencies are already described in the publication DE 10 2018 101 923 A1, which is referred to here for these details. If the resonator effect cannot be excluded as the cause of the monitored attenuation, the method is terminated. Otherwise, the classification of the covering 150 can now take place. The optional determination of a mass value 140 of the covering is initially skipped.In classification step 150, it is now checked whether the coating belongs to coating class H of the hard coatings or to coating class W of the soft coatings. In the simplest case, for this purpose, a quotient is formed from the difference between the current modal stiffness k and the stiffness reference value k 0 and from the difference between the current damping D and the damping reference value D 0 and is compared with the gradient of the straight line L in FIG. 1. If the quotient is greater than the gradient, a coating from the coating class of the hard coatings is given, otherwise a coating from the coating class of the soft coatings.The result of the classification is output to a process control system in a signaling step 160.The aforementioned method steps are carried out in particular by an electronic operating and measuring circuit with a computing unit of a Coriolis mass flow measuring sensor.In addition to evaluating damping D and modal stiffness k, in Coriolis mass flow measuring sensors having a straight measuring tube, the coating mass can also be determined and used for classifying the coating, for example by comparing a quotient of the increase in stiffness and the coating mass which is the cause thereof with a reference value.The determination of the coating mass will now be explained with reference to FIGS. 3 aand 3 b, which both show a cross section through the same straight measuring tube 200 in different operating modes. The measuring tube 200 comprises a cylindrical tube wall 210 which encloses a lumen in which a liquid medium 220 is guided. A coating 230 has formed on the tube wall 210. In the operating mode in FIG. 3 a, the measuring tube 200 carries out a flexural oscillation, for example in the flexural oscillation usable mode. The flexural vibration mode of use is excited to resonance in order to determine damping in the manner discussed above. Similarly, the modal stiffness k for the bending vibration mode of use is determined on the basis of excitation other than resonance. In FIG. 3 b, the measuring tube executes a torsional oscillation, wherein the resonant frequency of the torsional oscillation is a function of modal stiffness kmfor the torsional oscillation mode and the distribution of the twisting masses. These are the mass of the tube wall 210 and the mass of the covering 230 adhering to the tube wall. The liquid 220 remains essentially at rest during the torsion. Assuming that the modal stiffness k t of the torsion mode under the influence of the lining develops in proportion to the modal stiffness k of the flexural vibration mode, the modal stiffness k t of the torsion mode can be calculated. The coating mass can thus be slid off from the resonant frequency of the torsional vibration, whereby a further parameter for classifying coatings is obtained.How to define the classes in detail depends, if appropriate, on the peculiarities of the media processed in a plant and can be specified by the operator by setting suitable checking criteria.

Claims

Method (100) for classifying coatings (230) in a measuring tube (200), comprising: determining (110) an attenuation value D for at least one oscillation mode of an oscillator having at least one measuring tube (200) of a Coriolis mass flow measurement pickup for guiding a medium (220); determining (120) a modal stiffness value k for at least one oscillation mode; and classifying (150) the coating (230) as a function of the attenuation value D and the stiffness value k.Method according to claim 1, further comprising: determining (132), on the basis of a comparison of the damping value D and / or the stiffness value k with a reference value, whether a minimum value is reached.Method according to one of the preceding claims, further comprising: checking (134) whether a fluctuation of a resonant frequency of the measuring tube (200) can be excluded as a cause of the attenuation value D.Method according to Claim 1 or 2, further comprising: checking (136) whether compressibility of the medium (220) guided in the measuring tube (200) can be ruled out as a cause of the attenuation value D.Method according to one of the preceding claims, wherein the classification (150) of the lining (230) comprises the assignment of the lining (230) to a lining class at least on the basis of the stiffness value k.Method according to Claim 4, wherein the assignment of the lining (230) to a lining class takes place as a function of a relationship between the rigidity value k and the damping value D.Method according to one of the preceding claims, wherein the method further comprises the determination (140) of a mass value of the lining (230), wherein the mass value enters into the classification of the lining (230) in relation to the stiffness value k and / or damping value D.Method according to Claim 7, wherein the mass value (140) of the covering (230) is determined on the basis of the change in a resonant frequency of the oscillating measuring tube (200).Method according to one of the preceding claims, wherein the attenuation value D is determined (110) on the basis of a ratio of excitation current for driving the vibration mode in resonance and an amplitude of the vibration mode achieved therewith.Method according to one of Claims 1 to 8, wherein the attenuation value D is ascertained (110) on the basis of the decay of the oscillation of the oscillation mode when the exciter is switched off.Method according to one of the preceding claims, wherein the ascertainment (120) of the stiffness value k is carried out by exciting the vibration mode out of resonance and ascertaining a relationship between the vibration amplitude out of resonance and the excitation current out of resonance.

Images