Method for normalizing acoustic signal

Abstract

A method for characterizing a liquid metal, the method employing a probe (1) comprising: a measurement transmitter (11) configured to transmit an incident ultrasonic wave in the liquid metal; a receiver (21) configured to detect ultrasonic waves reflected or diffracted by the liquid metal after the emission of the incident ultrasonic waves; the method includes normalizing an ultrasonic wave detected by the receiver by a normalization amplitude. The normalized magnitude is experimentally determined from ultrasonic waves detected when the metal is considered to be free of size-detectable inclusions.

Description

Technical Field

[0001] The technical field of this invention is to characterize liquid metal castings using ultrasound, such as aluminum alloy castings. Background Technology

[0002] In the field of metal casting, it is necessary to inspect the quality of molten metal to detect inclusions. This is especially true for aluminum. Some aluminum alloy products, such as can stock, have very high requirements for inclusion content. These inclusions are generated during the melting, processing, and casting processes when these alloys are in the liquid phase. The quality of molten metal is determined by the quantity and size of the inclusions present in the metal. These inclusions can consist of oxides, nitrides, carbides, fluorides, or borides. Common inclusions encountered in molten aluminum include Al₂O₃, MgAl₂O₄, Al₄C₃, MgO, CaO, CaF₂, TiB₂, TiVB₂, and SiO₂ particles. Their sizes range from a few micrometers to several hundred micrometers.

[0003] The quality of inclusions in castings can be monitored online using ultrasonic measurement technology. In this type of measurement, ultrasonic waves propagate through the molten metal. If inclusions are present, some of the emitted sound waves will be reflected or diffracted, thus becoming detectable. The amplitude of the reflected waves can be used to determine the size of the inclusions in the metal. Ultrasonic technology allows for the characterization of a portion of the molten metal, typically 2% when inspected in a tank as the molten metal is transferred from the furnace to the casting pit.

[0004] Probes that enable ultrasonic measurement of castings have been described in patents EP3204764B1 and EP3204763B1.

[0005] Document EP1194772B1 describes a method for calibrating a probe in which a reflector is disposed at the end of a rod, which is then placed in liquid metal. The reflectors can have different sizes. Therefore, a correlation can be established between the size of each reflector and the amplitude of the detected sound wave. This method needs to be applied to each probe individually. Furthermore, it implies setting the calibrated reflector in liquid metal, which can prove complex.

[0006] Another alternative for estimating inclusion size is to use a correlation curve between measurements obtained using an ultrasonic probe and measurements obtained using a method considered as a reference method. For example, the reference method could be a "LIMCA" (Liquid Metal Cleanliness Analysis) type analytical method, based on the analysis of the electrical resistance of a liquid metal sample. It could also consist of a method described in EP1194772B1, which involves calibrating a reflector. Therefore, a correlation function can be established between the inclusion size measured by the reference method and the amplitude of the ultrasonic signal.

[0007] In “In Situ Detection of Non-metallic Inclusions in Aluminum Melt (1xxx) - Comparison between a newly developed ultrasonic technique and LiMCA and PoDFA method” (Light Metals 2019, January 1, 2019, pp. 1623-1629), F. Feikus et al. describe the development of a method for detecting non-metallic inclusions in cast or machined aluminum products based on ultrasonic technology. The document acknowledges the calibration issue, noting in paragraph 2 on page 1629: “The greatest challenge to date has been the calibration of the ultrasonic system, which still requires further investigation.”

[0008] One problem associated with using ultrasound probes is that their responses may differ from one another. However, it is difficult to calibrate each probe using a reference method (such as the LIMCA-type reference method or the reference method described in EP1194772B1). Such calibration is time-consuming.

[0009] Ono Y et al. described an ultrasonic technique for evaluating the cleanliness of molten aluminum and magnesium using a coated steel buffer rod in “An online ultrasonic cleanliness analyzer for molten metals” (Journal of Metals, Vol. 56, No. 2, pp. 59-64, 2004). The authors’ method involved mounting an ultrasonic probe with a coated buffer rod having a high signal-to-noise ratio at the opening of a tube through which the molten metal was guided.

[0010] Ihara I et al. in "Ultrasonic imaging, particle detection, and V(z) The article "Measurements in molten zinc using focused clad buffer rods" (Review of Scientif Instruments, Vol. 71, No. 9, pp. 3579-3586, September 2000) describes the use of focused clad buffer rods for ultrasonic imaging, particle detection, and V(z) measurement in molten zinc at temperatures above 600°C.

[0011] Patent application WO2007 / 003058 A1 relates to a method and system for determining the properties of an object by measuring ultrasonic wave attenuation.

[0012] The variability in response between two probes with identical structures is particularly attributable to differences that may affect the electronic circuitry or the materials used. When encountering the same inclusion, two probes with identical but different designs may produce detection signals of different amplitudes due to the variability in probe response.

[0013] It is preferable, or even necessary, to consider this variability, for example, if one seeks to apply the same correlation curve to measurements produced by different ultrasound probes.

[0014] The inventors have proposed a method that allows for consideration of the variability in the responses of two different probes. The aim is to ensure that, in the presence of the same inclusions, two identically designed probes produce the same signal, or can be considered to produce the same signal. Summary of the Invention

[0015] The first objective of this invention is a method for characterizing liquid metals, the method employing a probe comprising: - A measuring transmitter configured to emit incident ultrasonic waves in the liquid metal; - A receiver configured to detect ultrasonic waves reflected or diffracted by inclusions in the liquid metal after the incident ultrasonic waves are emitted; The method includes the following steps: a) The probe is configured such that the measuring transmitter and receiver are positioned in or facing the liquid metal; b) Powering the measuring transmitter so that it emits incident ultrasonic waves propagating in the liquid metal; c) The receiver detects ultrasonic waves reflected or diffracted by inclusions in the liquid metal under the action of the incident wave; d) Repeat steps b) to c) during different measurement times, the measurement times forming a measurement time range, and measure the amplitude of the ultrasonic wave detected by the receiver at each measurement time; e) Within the measurement time range, select a normalized time period, which consists of continuous measurement times; f) Calculate the normalized amplitude based on the amplitude measured during the normalized period; g) Normalize the amplitude measured during each step c) within the measurement time range using the normalized amplitude calculated in step f).

[0016] According to one alternative approach, step f) includes: - Assign a normalized value to the normalized amplitude; - Multiply the normalized amplitude by the normalized value.

[0017] The normalized time period can correspond to a predefined portion of the measurement time range.

[0018] The normalized time period can correspond to the last time of the measurement time range.

[0019] The normalization period can be extended to the last few minutes of the measurement time range.

[0020] Step e) may include: - ei) Analyze the amplitudes measured within the specified measurement time range; - eii) Identify a minimum amplitude period that comprises different continuous measurement times during which the measured amplitude is minimum relative to the intensity measured outside the minimum intensity period; such that the normalized period includes all or part of the minimum amplitude period.

[0021] According to one alternative scheme, - Sub-step ei) includes calculating the moving average or moving median of the amplitude measured respectively within the measurement time range, the average or median being calculated based on different continuous time windows; - During sub-step eii), the minimum amplitude period corresponds to the period during which the moving average or median is at its minimum.

[0022] According to one alternative scheme, - The liquid metal flows along the tank; - The transmitter and the receiver are positioned facing the slot.

[0023] According to one alternative scheme, - The measurement transmitter is coupled to a first waveguide immersed in the liquid metal; - The receiver is coupled to a second waveguide immersed in the liquid metal.

[0024] A second objective of the present invention is an ultrasonic probe comprising: - A measuring transmitter (11) configured to emit incident ultrasonic waves in the liquid metal; - Receiver (21), the receiver being configured to detect ultrasonic waves reflected or diffracted by inclusions in the liquid metal after the incident ultrasonic waves are emitted; - Processing unit (32), the processing unit being configured to: - The amplitude of the ultrasonic waves detected by the receiver is measured at different measurement times, which form a measurement time range; - Using the amplitude measured within the stated measurement time range, implement steps e) to g) of the method according to the first objective of the invention.

[0025] The invention will be better understood by reading the disclosure of the embodiments presented in the following description in conjunction with the accompanying drawings. Attached Figure Description

[0026] Figure 1A A first embodiment of the probe that allows the implementation of the present invention is shown.

[0027] Figure 1B Another embodiment of the probe that allows the present invention to be implemented is shown.

[0028] Figure 2 The diagram schematically illustrates the sound waves detected at one measurement time.

[0029] Figure 3A The main steps of the method according to the invention are illustrated schematically.

[0030] Figure 3B A configuration is schematically shown in which the normalization period is located at the end of the measurement time range.

[0031] Figure 3C A configuration is schematically shown in which the normalized period is located in the middle of the measurement time range.

[0032] Figure 4 The correlation curve between the detected amplitude value and the inclusion size is shown.

[0033] Figure 5A The amplitudes measured at different measurement times are shown.

[0034] Figure 5B This shows the effect after applying the normalization function. Figure 5A The amplitude is shown in the figure. Detailed Implementation

[0035] The first embodiment of the ultrasonic probe that allows implementation of the present invention has been combined Figure 1A As shown. The probe includes a measuring transmitter 11 and a receiver 21. In this embodiment, the measuring transmitter 11 and receiver 21 are formed by the same acoustic transducer. This acoustic transducer is, for example, a piezoelectric transducer known to those skilled in the art.

[0036] For example, the transducer is a piezoelectric material wafer that emits sound waves at a frequency of 5 MHz. Typically, the frequency of the emitted sound waves is preferably between 1 MHz and 20 MHz.

[0037] This probe is designed to detect inclusions in a liquid medium 2 to be analyzed, more specifically, to detect inclusions in liquid metal, such as an aluminum alloy. Figure 1A and Figure 1B In the diagram, the horizontal dashed line represents the interface between liquid metal 2 and the ambient medium 3, such as air.

[0038] The probe 1 includes a waveguide 13 designed to be immersed in liquid metal 2. This waveguide is configured to facilitate the propagation of acoustic waves between the transducer, which acts as a transceiver, and the liquid metal 2. For example, the waveguide 13 is formed of a silicon nitride or silicon oxynitride (SiAlON) type material, or more generally of a metallic or refractory ceramic type material, which are considered inert to the liquid metal 2. The use of such a waveguide has been described in EP3204763B1 or EP1194772B1.

[0039] exist Figure 1A and Figure 1B The ultrasonic field 5 is schematically shown, corresponding to a portion of the liquid metal being examined by probe 1. As inclusion 4 passes through the ultrasonic field, a portion of the sound waves emitted by measuring transmitter 11 is reflected or diffracted toward receiver 21. The characteristics of the reflected waves, particularly their amplitude, allow for the detection and characterization of inclusions, such as estimation of their size.

[0040] Preferably, the probe may include a power transmitter 12 configured to emit high-power acoustic waves, i.e., greater than 10W, preferably greater than 100W, such as 120W, as described in EP3204764. In practice, applying high-power acoustic waves allows for the wetting of the first waveguide 13 by the liquid metal 2. The power transmitter 12 is activated at regular time intervals to allow for maintaining good wetting of the waveguide 13 by the liquid metal 2.

[0041] Figure 1BA preferred configuration is shown, in which the probe includes a measuring transmitter 11 and a receiver 21. The measuring transmitter 11 is connected to a first waveguide 13. The receiver 21 is connected to a second waveguide 23, which is preferably identical to the first waveguide 13 (same material, same dimensions), but separate from the first waveguide 13. For example, the length of each waveguide is between 10 cm and 50 cm. The first and second waveguides are preferably straight and extend around a first central axis Δ1 and a second central axis Δ2, respectively. The central axes Δ1 and Δ2 intersect and are inclined relative to each other according to an angle α. This angle is, for example, between 25° and 35°. Angle α is, for example, equal to 28°. This configuration has been described in EP3204764B1.

[0042] In this embodiment, the probe may include a first power transmitter 12 and a second power transmitter 22, as described in conjunction with the first embodiment. The first and second power transmitters are periodically activated to improve the wetting of the liquid metal 2 on the first waveguide 13 and the second waveguide 23, respectively.

[0043] The probe includes a control unit 31 configured to allow control of the measuring transmitter 11 to emit sound waves, and when the probe includes such a transmitter, to allow control of the power transmitter 12 to emit power sound waves.

[0044] The sound waves emitted by the measuring transmitter 11 are typically emitted in series. In practice, each series of sound waves is emitted by the measuring transmitter 11 at a regular frequency (e.g., 10 Hz). A series includes at least one sound wave, preferably several consecutive sound waves. Within the same series, the emission frequency of the sound waves can be between 100 Hz and 300 Hz, corresponding to the excitation frequency of the piezoelectric transducer of the measuring transmitter. Figure 2 The diagram schematically illustrates three waves detected by receiver 21 after transmitter 11 emits a series of three waves. Parentheses indicate that the three waves belong to the same series. Figure 2 In the diagram, the y-axis corresponds to the signal detected by receiver 21. The x-axis corresponds to time t.

[0045] The probe includes a processing unit 32 configured to calculate the amplitude of waves detected by the receiver 21 in response to a series of sound waves emitted by the measuring transmitter 11. The calculated amplitude, or measured amplitude, can be the average of the maximum amplitudes of each detected sound wave, or the maximum amplitude of all sound waves detected after a series of wave emissions. The calculated amplitude is then labeled as follows: , The measurement time corresponds to a series of sound waves detected by receiver 21.

[0046] The processing unit 32 may include one or more microprocessors or electronic circuits configured to determine the measured amplitude and perform the steps described below.

[0047] The presence of large inclusions in liquid metal can affect the measured amplitude. The measured amplitude has changed. It allows for the estimation of inclusion size based on correlation curves plotted experimentally using reference methods (such as the LIMCA method or methods based on the use of calibrated reflectors, which are described in conjunction with existing techniques).

[0048] The same correlation curve can be used to design different ultrasound probes, provided that the amplitudes produced by the probes are comparable. However, as mentioned earlier, it is necessary to consider the variability of probes, even if they are similar to each other: the same size and using the same materials.

[0049] Processing unit 32 includes an amplifier designed to amplify the signal from receiver 21, and the amplification gain G0 of the amplifier is typically set at the factory after the probe is manufactured. This gain is generally not modified during the process. It can be adjusted according to the conditions of probe use.

[0050] Probe 1 is designed to be immersed in molten metal 2 flowing from a furnace. For example, the molten metal may flow from a furnace or be present in a ladle or crucible. Alternatively, the molten metal may flow into a tank. One or more waveguides of the probe are immersed in the tank. The probe can be positioned directly at the furnace outlet or further downstream, at the outlet of a filter or deaerator. This filter can consist of a deep-bed filter, a ceramic foam filter (CFF), or any other type of filter. When the probe is immersed in the casting downstream of a CFF-type filter, the presence of air bubbles in the molten metal 2 may require adjustment of the amplification gain G0 to account for the attenuation effect caused by the air bubbles. The probe can also be positioned downstream of a deaerator and in a crucible.

[0051] Due to cost and time constraints, it is impractical to generate correlation curves for each probe using a reference method. However, it is necessary to consider the variability in the response of each probe. The inventors propose a method for obtaining comparable responses for each probe, enabling the use of the following combination... Figure 4 The correlation curve described.

[0052] The main steps of this method are as follows: Figure 3A Example in.

[0053] Step 100 Measurements were taken at different times.

[0054] During the casting process, measurements are performed periodically, for example, based on repeated measurement times at frequencies, such as between 0.1 Hz and 20 Hz. Perform the measurement. Measurement time. The set forms the measurement time range The measured amplitude Associated with each measurement time. (For example, combining...) Figure 2 As described, the measured amplitude is calculated by the processing unit 32 based on a wave or a series of waves detected by the receiver 21 after a sound wave or a series of sound waves is emitted by the measuring transmitter 11.

[0055] Step 110 Select the normalized time period.

[0056] An important aspect of this invention is that, within the measurement time range Within, you can select the normalized time period. The normalization period includes different continuous measurement times. Normalized time period Corresponding to the measured amplitude Measurement time point at the lowest point In practice, during the casting process, it is assumed that along a given position in the casting process (occupied by probe 1), a time period can be defined as the minimum amplitude period. The amplitude measured during the minimum amplitude period It is the smallest. The period of minimum amplitude. This corresponds to a period in casting considered "clean," meaning the metal has the lowest possible proportion of inclusions. Normalized period Corresponding to the period of minimum amplitude All or part of. During the normalization period. The measured amplitude during the period Measurement time relative to the time before and / or after the normalization period The amplitude measured during this period is the smallest. (During the normalization period) During this period, the average amplitude was less than that within the measurement time range. The average amplitude calculated during the period.

[0057] Normalized time period The position of probe 1 relative to the casting can be predefined, especially when probe 1 is positioned downstream of the filter. It is assumed that at this position, the liquid metal has the lowest proportion of inclusions compared to other positions in the casting. At this position, the normalized time... This can correspond to the final time of casting, such as the last minute or minutes of casting, as the duration of casting is typically between one and three hours or even longer. Normalized time period This corresponds to the measurement time range. The last measurement time. This situation occurs in Figure 3B Example in.

[0058] According to another alternative approach, the normalization time period is the measurement time... The measured amplitude during the period After analysis, the minimum amplitude period was identified through backtracking. It is defined based on measurement. The normalization period is during the period of minimum amplitude. Choose from. This corresponds to... Figure 3C The example shown. Minimum amplitude period. The minimum amplitude period can be identified by calculating the moving average or moving median of the amplitude measured separately within the measurement time range, based on different continuous time windows offset from each other by approximately one minute. This corresponds to the period when the moving average or median is at its minimum.

[0059] When the probe is placed at the outlet of the furnace, depending on the process implemented, it can be assumed that the liquid metal has a large number of inclusions at the beginning and end of casting.

[0060] Step 120 Determine the normalized amplitude.

[0061] After selecting the normalization period Then, the normalized amplitude representing the normalized period is determined. This can be determined by the normalized time period. The measured amplitude within It consists of the average or median.

[0062] Step 130 Normalization.

[0063] In this step, at each measurement time The amplitudes measured separately Normalized amplitude Normalization. Preferably, it is the normalized amplitude. Assign a normalized value (labeled as) Therefore, normalization is performed such that: in: - Corresponding normalization function; - It is the normalized amplitude; - For example, it could be 20. In this case, the normalized amplitude... Expressed as a percentage. A value of 20% corresponds to a casting considered clean, meaning that no inclusions will produce a significant ultrasonic signal. Therefore, for values ​​of 20% and below, this includes ultrasonic noise associated with inclusions or microbubbles, which may be numerous but insignificant in size. - It is the normalized amplitude defined above.

[0064] One advantage of this method is that it corresponds to the normalized value. (e.g., 20%) normalized amplitude The value represents clean casting and is independent of the probe used. Normalization allows us to avoid affecting probe variability: the normalized amplitude varies between different probes. They are comparable because, for each probe, the normalized amplitude... It was established under comparable conditions, in this case, in the presence of liquid metal that is considered clean.

[0065] The normalization assumption is that accepting liquid metal that is considered clean (i.e., free of detectable inclusions) will produce sound waves, the amplitude of which is assimilated as noise. This amplitude corresponds to the normalized amplitude. .

[0066] Normalization allows for the use of, for example Figure 4 The correlation curve is schematically shown in the diagram. This correlation curve was plotted experimentally using a probe after applying a normalization function determined for that probe. Therefore, the y-axis corresponds to... ,in The amplitude corresponds to the measured value associated with the known inclusion size. The x-axis corresponds to the size (e.g., diameter) measured by a reference method.

[0067] Figure 4 Multiple points are shown, each corresponding to one experimental measurement. A calibration curve, obtained by adjusting the experimental points, has also been plotted using dashed lines.

[0068] The correlation function plotted in this way can be applied to normalized amplitudes from different probes. The normalized values ​​can be used. The correlation curve is represented by the vertical axis.

[0069] A combination Figure 1B The described probe is implemented in an aluminum casting and positioned at the outlet of the CFF filter. The measurement time range lasts for 60.5 minutes. Figure 5A Corresponding to different measurement times Amplitude measured at (x-axis – sampling frequency 10 Hz) (y-axis). Figure 5BNormalization is shown back Figure 5A The amplitude represented in the figure has a normalized value N of 16%. Figure 5A and Figure 5B In this context, zero amplitude corresponds to the time when the power transmitter 12 is activated. During the activation of the power transmitter, the signal detected by the receiver is not taken into account.

[0070] This invention can be applied by implementing the same type of ultrasonic probe (i.e., having the same structure), and its variability is due to the differences in the components affecting the probe.

[0071] If the measurement is intended to characterize the size of inclusions, it is preferable, and even necessary, to establish a correlation between the measured amplitude and the measurement result of the inclusion size for at least one probe. These measurement results are obtained through a reference measurement method. Preferably, the different probes implemented are then positioned at the same location or considered equivalent locations relative to the casting.

Claims

1. A method for characterizing liquid metal, the method employing a probe (1), the probe comprising: - A measuring transmitter (11) configured to emit incident ultrasonic waves in the liquid metal; - Receiver (21), the receiver being configured to detect ultrasonic waves reflected or diffracted by inclusions in the liquid metal after the incident ultrasonic waves are emitted; The method includes the following steps: a) The probe (1) is configured such that the measuring transmitter and receiver are positioned in or facing the liquid metal; b) Powering the measuring transmitter so that it emits incident ultrasonic waves propagating in the liquid metal; c) The receiver detects ultrasonic waves reflected or diffracted by inclusions in the liquid metal under the action of the incident wave; d) At different measurement times ( During this period, steps b) to c) are repeated, and the measurement time forms a measurement time range. ), and at each measurement time, the amplitude of the ultrasonic wave detected by the receiver is measured ( ); e) within the measurement time range Within ) select the normalized time period ( The normalized time period consists of continuous measurement times; f) Calculate the normalized amplitude based on the amplitude measured at the time point of the normalized period. ); g) Normalize the amplitude measured during each step c) within the measurement time range using the normalized amplitude calculated in step f).

2. The method of claim 1, wherein the normalized time period includes the measurement time during which the measured amplitude is the smallest relative to the measurement time range.

3. The method according to any one of the preceding claims, wherein step f) comprises: - Assign a normalized value to the normalized amplitude ( ); - Multiply the normalized amplitude by the normalized value.

4. The method according to any one of the preceding claims, wherein the normalized time period corresponds to a predefined portion of the measurement time range.

5. The method of claim 4, wherein the normalized time period corresponds to the last time of the measurement time range.

6. The method of claim 4, wherein the normalization period extends to the last few minutes of the measurement time range.

7. The method according to any one of claims 1 to 3, wherein step e) comprises: - ei) Analyze the amplitudes measured within the specified measurement time range; - eii) Identify the minimum amplitude period, which includes different continuous measurement times during which the measured amplitude is the smallest relative to the intensity measured outside the minimum intensity period; This ensures that the normalized time period includes all or part of the minimum amplitude time period.

8. The method according to claim 6, wherein: - Sub-step ei) includes calculating the moving average or moving median of the amplitude measured respectively within the measurement time range, the average or median being calculated based on different continuous time windows; - During sub-step eii), the minimum amplitude period corresponds to the period during which the moving average or median is at its minimum.

9. The method according to any one of the preceding claims, wherein - The liquid metal flows along the tank; - The transmitter and the receiver are positioned facing the slot.

10. The method according to any one of the preceding claims, wherein - The measurement transmitter is coupled to a first waveguide (13) immersed in the liquid metal; - The receiver is coupled to a second waveguide (23) immersed in the liquid metal.

11. An ultrasonic probe for characterizing liquid metal, comprising: - A measuring transmitter (11) configured to emit incident ultrasonic waves in the liquid metal; - Receiver (21), the receiver being configured to detect ultrasonic waves reflected or diffracted by inclusions in the liquid metal after the incident ultrasonic waves are emitted; - Processing unit (32), the processing unit being configured to: - at different measurement times ( )Measure the amplitude of the ultrasonic wave detected by the receiver. The different measurement times form a measurement time range. ); - Use the amplitude measured within the measurement time range ( (e) to (g) of the method according to any one of the preceding claims.

Images