Method for determining angular position of signal generator, angle detection system and valve

Abstract

The invention relates to a method for determining the angular position of a magnetic signal generator in an angle detection system comprising a signal generator and a Hall sensor, the Hall sensor is designed to generate at least two measurement signals in at least two different measurement directions on the basis of a magnetic field which varies as a result of a rotation of the signal generator, the signal generator comprising a magnet which is polarized such that the magnetic field strength measured by the Hall sensor varies with the rotation of the signal generator. And wherein the signal generator is arranged within the detection range of the at least one Hall sensor. In a first method step, the signal generator is rotated about its axis of rotation. When the signal generator rotates, the at least one Hall sensor generates at least two measurement signals. The angular position of the signal generator is determined in the control unit on the basis of a measurement signal generated by the at least one Hall sensor. The invention further relates to an angle detection system and to a valve.

Description

Technical Field

[0001] The present invention relates to a method for confirming the angular position of a magnetic signal generator rotatably supported about a rotation axis, an angle detection system for confirming the angular position of the signal generator, and a valve having a corresponding angle detection system. Background Technology

[0002] It is known that the position of the valve tappet and the valve closing element connected thereto is detected in a valve to confirm the valve's open state. For this purpose, a stroke measurement system is typically used, comprising multiple Hall effect sensors arranged in series and a magnetic signal generator fixed at one end of the valve tappet and moving within the detection range of the Hall effect sensors. However, in this case, only the axial movement of the valve tappet is detected. Summary of the Invention

[0003] The purpose of this invention is to achieve the detection of rotational movement.

[0004] According to the invention, the objective is achieved by a method for confirming the angular position of a magnetic signal generator rotatably supported about a rotation axis in an angle detection system, particularly for determining the open state of a valve, the angle detection system comprising a signal generator and at least one Hall sensor with at least two axes, the Hall sensor being designed to generate at least two measurement signals in at least two different measurement directions based on a magnetic field that varies due to the rotation of the signal generator, wherein the signal generator comprises a magnet that is polarized such that the magnetic field strength measured by the Hall sensor varies with the rotation of the signal generator, and wherein the signal generator is arranged within the detection range of at least one Hall sensor.

[0005] In the first method step, the signal generator rotates about its rotation axis.

[0006] As the signal generator rotates, at least one Hall sensor generates at least two measurement signals in the form of magnetic field components in at least two different measurement directions. This means that a measurement signal is generated in each measurement direction. Due to the different orientations of at least two axes, the rotating magnetic field generates different measurement currents in the Hall sensor in its at least two measurement directions.

[0007] The angular position of the signal generator is determined in the control unit based on the measurement signal generated by at least one Hall sensor, particularly according to machine learning principles.

[0008] Therefore, by means of the method according to the invention, the rotation of a signal generator or a component connected thereto, such as the actuator of a valve, can be detected, and the angular position of the signal generator can be confirmed, thereby allowing the angular position of the actuator to be deduced.

[0009] The different measurement signals generated by the Hall sensor are particularly orthogonal to each other.

[0010] Preferably, at least one Hall sensor is a triaxial sensor, which is designed to generate three measurement signals in three different measurement directions based on a magnetic field that varies through the rotation of a signal generator.

[0011] Alternatively, at least one Hall sensor is a biaxial sensor designed to generate two measurement signals in two different measurement directions based on a magnetic field that varies through the rotation of a signal generator.

[0012] According to a preferred embodiment, the angle detection system includes only a single Hall sensor.

[0013] For example, a control unit is connected to at least one Hall sensor via signal transmission, wherein position determination software is stored in the control unit. This position determination software is designed to determine the angular position of a signal generator based on measurement signals generated by the at least one Hall sensor, and wherein the position determination software is designed to associate the measurement directions of the measurement signals generated by the Hall sensors with an axis of a coordinate system stored in the position determination software. It is feasible in this way to standardize measurement signals generated by Hall sensors from different manufacturers, so that the angular position can be correctly determined. The direction of the measurement signals generated by the Hall sensors is therefore associated with a coordinate system stored in the Hall sensors, which is not necessarily consistent with the coordinate system stored in the position determination software. More precisely, in the coordinate system of the Hall sensors, spatial directions can be exchanged with the coordinate system stored in the position measurement system, specifically the x and y directions.

[0014] The position recognition software identifies the type of Hall sensor used, for example, by a trigger, particularly an analog trigger, and associates the measurement signal with the axis of the coordinate system stored in the position determination software based on the identified Hall sensor.

[0015] In one method step, for example, a set of position-intensity data from at least one Hall sensor is provided for multiple preset angular positions of the signal generator. In another method step, a current measurement signal is generated by the Hall sensor for each measurement direction to determine the current angular position of the signal generator. Subsequently, the current angular position of the signal generator is determined from the position-intensity data and the current measurement signal. The position-intensity data is stored, in particular, in position determination software.

[0016] By comparing the current measurement signal with position-intensity data, the current angular position of the signal generator can be inferred, if necessary, without transforming the current measurement signal or performing calculations using it. In this way, angular position determination can be performed quickly and accurately. Specifically, the current measurement signal can be explicitly associated with the current angular position of the signal generator.

[0017] According to one implementation, the position-intensity data is immutable. This position-intensity data is provided once in advance, and only needs to be read for each current angular position determination, which also speeds up the method.

[0018] Preferably, each change in angular position of the signal generator within the detection range of at least one Hall sensor causes a change in the measurement signal for all measurement directions, and in particular, such change is unilaterally depicted by the characteristic curve of at least one Hall sensor for the corresponding measurement direction.

[0019] This characteristic curve can vary depending on the type and manufacturer of the Hall sensor itself, as well as other variables such as ambient temperature and aging condition. The magnetic field of the signal generator also depends on the type, temperature, and / or aging condition.

[0020] The characteristic curve can be pre-determined for each measurement direction for at least one Hall sensor, which can be determined through analysis, a partial empirical function, or by recording measurement curves. The characteristic curve is based on position-intensity data.

[0021] To provide position-intensity data, a signal generator is rotated, for example, within a predetermined angular range, particularly 360°, and the resulting measurement signal and its corresponding angular position are recorded. This allows for the generation of position-intensity data as realistically as possible.

[0022] According to another implementation, location-intensity data can be calculated. This can be done using analytical formulas or empirically derived formulas. For example, a sufficiently accurate empirically derived fitted curve from the measured characteristic curves would be considered.

[0023] Combining the two types of location-intensity data provision methods is also feasible.

[0024] The position-intensity dataset preferably includes characteristic curves from the entire family of curves, each coordinated to a separate angle detection system. Here, for each individual characteristic curve, not only the rotation of the signal generator can be considered, but also different, expected ambient temperatures and the aging conditions of the Hall sensor and signal generator during the expected lifespan of the angle detection system. Obviously, other additional parameters that affect the measurement signal detected by the Hall sensor or the magnetic field of the signal generator can also be included in the position-intensity data. These additional parameters specifically include temperature and / or aging data.

[0025] Here, taking into account the corresponding parameters, the position-intensity data should correspond as closely as possible to the actual characteristic curve of at least one Hall sensor.

[0026] To determine the location, for example, based on current external conditions such as the operating duration of the angle detection system or the ambient temperature, appropriate location-intensity data based on the corresponding characteristic curves are selected from the location-intensity dataset and used as the basis for location determination.

[0027] The appropriate control routines are used to select the characteristic curves and parameters to be used from the position-intensity dataset and to determine the position based on the position-intensity data itself. These control routines are typically integrated into the angle detection system itself and, for example, stored as software in the control unit.

[0028] The software used can include any suitable form of artificial intelligence or machine learning program.

[0029] The position-intensity dataset can be created outside the angle detection system and transmitted to and stored in the control unit of the angle detection system. The advantage of this is that a position-intensity dataset suitable for a large number of similar angle detection systems only needs to be created once. For example, this works for angle detection systems from the same batch from the manufacturer, equipped with Hall sensors and signal generators.

[0030] Here, for example, the position-intensity dataset can be based on the angle detection system and precise measurements of the characteristic curve. This can be achieved, for instance, by rotating a signal generator and measuring the characteristic curve of at least one Hall sensor for different environmental conditions, where the known angular position is recorded and correlated with the characteristic curve as position data. In particular, this can be performed within the scope of the teaching process of the angle detection system.

[0031] Another possibility is to create the position-intensity dataset in an appropriate manner within the angle detection system itself. A combination is also possible, where the majority of the position-intensity dataset is created externally and transmitted to the corresponding angle detection system, while the detection of additional parameters for the respective individual angle detection system is performed within the angle detection system itself. This could, for example, be done within the scope of a teaching process.

[0032] The current angular position of the signal generator can be determined using machine learning or deep learning, particularly based on nearest neighbor classification or random forest classification. Alternatively, nearest neighbor regression or random forest regression can be used. Clearly, any other suitable shallow learning or deep learning method (e.g., via neural networks) is also feasible, using a position-intensity dataset to generate the precise current angular position of the signal generator. A trainable model can also be used. It is also possible to consider retraining the model after a certain period of time.

[0033] Artificial intelligence methods can also be used to identify and hide or eliminate any inconsistencies and anomalies in a database of location-intensity data.

[0034] Preferably, downsampling reduces the amount of location-intensity data obtained, and weighting improves the accuracy of predicted locations, particularly by averaging the most probable predicted locations weighted by probability. More precisely, downsampling reduces measurement resolution, which can, of course, be compensated for by weighting in nearest neighbor regression or other suitable shallow or deep learning methods.

[0035] Besides averaging, the nearest neighbor can be approximated using a parabola, where the minimum value of the parabola is used to determine the location. This method is very robust to tolerances.

[0036] This variant is particularly advantageous if the method is to be executed on a microcontroller located on a circuit board, wherein at least one Hall sensor is mounted on the circuit board. This means that the method is executed within the angle detection system itself, rather than in a spatially separated control device connected to the angle detection system.

[0037] For example, a Hall sensor outputs a separate signal to a control unit corresponding to its configuration relative to the axis, and the control unit evaluates the measured signal based on the configuration. Therefore, Hall sensors with different configurations can be used without compromising the accuracy of determining the angular position of the signal generator.

[0038] The control unit is designed, for example, to determine whether the signal generator is rotating or moving axially, based on a measurement signal generated by at least one Hall sensor. This is particularly advantageous when the circuit board with at least one Hall sensor is used in both an angle detection system and an axial travel measurement system.

[0039] In addition, the control unit can be designed to: confirm the presence of axially polarized or diagonally polarized magnets. This serves as an additional check to ensure that the correct magnet has been used.

[0040] According to the invention, this objective is also achieved by an angle detection system specifically designed to perform the above-described method. The system comprises at least one Hall sensor with at least two axes and a magnetic signal generator rotatably supported about a rotation axis. The magnetic signal generator includes a magnet polarized such that the magnetic field strength measured by the Hall sensor varies with the rotation of the signal generator. The signal generator is arranged within the detection range of at least one Hall sensor, wherein the Hall sensor is designed to generate at least two measurement signals in at least two different measurement directions based on the magnetic field varying due to the rotation of the signal generator. As already described in conjunction with the method according to the invention, the current angular position of the signal generator can be inferred from the measurement signals generated by the Hall sensors. Therefore, the rotation of the signal generator can be detected by means of the measurement system according to the invention.

[0041] The diagonally polarized magnet can be toroidal or cylindrical and has at least two segments with different polarizations. Multiple segments with different polarizations arranged alternately along the circumference can also be considered. Alternatively, a bar magnet capable of pivoting about its transverse axis can also be considered. The diagonally polarized magnet ensures that, unlike axially polarized magnets where the magnetic field remains constant during rotation, the magnetic field within the Hall sensor detection region changes as the signal generator rotates.

[0042] Hall sensors are characterized by a central axis extending along the depth direction of the sensor, wherein the rotation axis of the signal generator is arranged concentrically, parallel to, or laterally, particularly perpendicular to the central axis of the Hall sensor. When arranged laterally to the central axis, the Hall sensor can be positioned such that the extension of the central axis intersects with or extends through the signal generator. Therefore, the signal generator can be flexibly positioned relative to the Hall sensor, allowing for flexible responses to different structural spatial conditions. Consequently, angle detection systems can be widely used.

[0043] According to the invention, this objective is also achieved by a valve having the angle detection system described above, wherein the valve includes an adjusting element for adjusting the flow cross-section through the valve, wherein the adjusting element is coupled to a signal generator such that the position of the adjusting element can be determined based on the angular position of the signal generator. In such a valve, therefore, the opening state of the valve, and in particular the set flow cross-section, can be determined by determining the angular position of the signal generator.

[0044] For example, a valve has a rotary actuator, wherein a signal generator is fixed at the actuator.

[0045] The adjusting element can be a flip cover that is pivotally supported by an actuator. Attached Figure Description

[0046] Further advantages and features of the invention will become apparent from the following description and the accompanying drawings. The drawings show:

[0047] - Figure 1 A valve according to the invention, having an angle detection system according to the invention, is schematically shown.

[0048] - Figure 2 The magnetic signal generator of the angle detection system according to the present invention is shown.

[0049] - Figure 3 A triaxial Hall sensor is shown in the angle detection system according to the present invention.

[0050] - Figure 4 This illustrates a signal generator with a Hall sensor arranged according to the first structural space.

[0051] - Figure 5 Shown in accordance with Figure 4 The measurement signal of the Hall sensor relative to the arrangement of the signal generator.

[0052] - Figure 6 This illustrates a signal generator with a Hall sensor arranged according to another structural space.

[0053] - Figure 7 Shown in accordance with Figure 6 The measurement signal of the Hall sensor relative to the arrangement of the signal generator.

[0054] - Figure 8 An angle detection system according to another embodiment of the present invention is schematically illustrated.

[0055] - Figure 9 Show Figure 8 The angle detection system in the middle uses a signal generator and a Hall sensor.

[0056] - Figure 10 Shown in accordance with Figure 8 and Figure 9 The measurement signal of the Hall sensor relative to the arrangement of the signal generator.

[0057] - Figure 11 An angle detection system according to another embodiment of the present invention is schematically illustrated.

[0058] - Figure 12 Show Figure 11 The angle detection system in the middle uses a signal generator and a Hall sensor.

[0059] - Figure 13 Shown in accordance with Figure 11 and Figure 12 The measurement signal of the Hall sensor relative to the arrangement of the signal generator.

[0060] - Figure 14 The steps for providing a location-intensity dataset are shown.

[0061] - Figure 15 Show other method steps, and

[0062] - Figure 16 The steps of a method for determining the angular position of a signal generator are shown. Detailed Implementation

[0063] Figure 1 A valve 10 having an angle detection system 12 according to the present invention is shown schematically.

[0064] Valve 10 includes a valve tap 14 that is rotatably supported and connected to an adjustment element 16, such as a valve disc, for adjusting the flow cross-section through valve 10.

[0065] The position of the adjusting element 16 is changed by rotating the valve stem 14, thereby adjusting the opening cross section of the valve 10.

[0066] An actuator (electric, hydraulic, pneumatic) is provided to rotate the valve tappet 14, but for simplicity, the actuator is not shown in the figure.

[0067] In this embodiment, according to Figure 1 The angle detection system 12 includes multiple (e.g., three) triaxial Hall sensors 18. However, in principle, a single Hall sensor 18 is sufficient.

[0068] Alternatively, the Hall sensor 18 can also be biaxial.

[0069] Hall sensor 18 is arranged on circuit board 20.

[0070] In addition, there is a control unit 22, which is connected to the Hall sensor 18 via signal transmission.

[0071] In this embodiment, the control unit 22 is configured as a microcontroller 30. The microcontroller is also arranged on the circuit board 20.

[0072] In addition, the angle detection system 12 includes a magnetic signal generator 24 that is rotatably supported about the rotation axis D.

[0073] The signal generator 24 is located at one end of the valve stem 14 and is coupled to the regulating element 16 so that the position of the regulating element 16 can be determined based on the angular position of the signal generator 24.

[0074] Signal generator 24 is arranged within the detection range of at least one of Hall sensors 18.

[0075] Here, the signal generator 24 generates a magnetic field around its rotation axis D.

[0076] The Hall sensor 18 is designed to generate three measurement signals in three different measurement directions based on the magnetic field that changes due to the rotation of the signal generator 24.

[0077] In the illustrated embodiment, the angle detection system 12 is housed in the control head 26 of the valve 10. However, it is also possible to arrange the control unit 22 outside the control head 26.

[0078] The signal generator 24 includes a diagonally polarized magnet 25. Figure 2 Such a magnet is illustrated in the example.

[0079] Figure 2 The magnet 25 shown is toroidal and has two segments 19 and 21 with different polarizations.

[0080] However, other shapes of magnets are also possible. For example, magnet 25 could be cylindrical. It is also possible that magnet 25 has two or more segments 19 and 21 with different polarizations.

[0081] Figure 3 An exemplary Hall sensor 18 is shown in three different views. The Hall sensor 18 has a central axis M extending along the depth direction of the Hall sensor 18.

[0082] The Hall sensor 18 generates signals in all spatial directions, including but not limited to directions a, b, and c, based on the magnetic field of the signal generator 24. Directions a, b, and c correspond to the axes of the coordinate system stored in the Hall sensor 18.

[0083] For example, direction a corresponds to the x-direction, direction b corresponds to the y-direction, and direction c corresponds to the z-direction.

[0084] However, depending on the configuration of the Hall sensor 18, the correlation within the Hall sensor 18 can also be performed differently. For example, the correlation described above applies to the Infineon sensor TLE493D. In the case of the Texas Instruments TMAG5173–Q1 sensor, the x-axis and y-axis are interchanged compared to the Infineon sensor, such that direction a corresponds to the y-direction and direction b corresponds to the x-direction.

[0085] Hall sensor 18 outputs a separate signal to control unit 22 corresponding to its configuration relative to the axis position.

[0086] Control unit 22 evaluates the measurement signal based on the configuration.

[0087] More precisely, the location determination software 28 is stored in the control unit 22, such as... Figure 1 It is shown schematically in the diagram.

[0088] Position determination software 28 is designed to determine the angular position of signal generator 24 based on the measurement signal generated by Hall sensor 18. Furthermore, position determination software 28 is designed to associate the measurement directions of the measurement signals generated by the Hall sensor with the axes of the coordinate system stored in position determination software 28. Standardization of the measurement signals of Hall sensor 18 in this manner makes it possible to use Hall sensors 18 with different configurations.

[0089] During the operation of valve 10, when signal generator 24 rotates about its rotation axis D, at least one of the Hall sensors 18 generates three measurement signals M in the form of magnetic field components in three different measurement directions a, b, and c as signal generator 24 rotates. a M b M c .

[0090] The angular position of the signal generator 24 is determined in the control unit 22 based on the measurement signal generated by at least one Hall sensor 18.

[0091] To confirm the angular position of the signal generator 24, typically a single Hall sensor 18 is sufficient. However, it is possible to consider using the circuit board 20 as a general-purpose component in the axial travel measurement system. In this case, multiple Hall sensors 18 can be arranged on the circuit board 20, such as... Figure 1 As shown in the diagram. In particular, except for the signal generator 24, the same hardware and software as those in the angle measurement system 12 can be used in the axial travel measurement system.

[0092] In this case, the control unit 22 is designed, for example, to determine whether the signal generator 24 is rotating or moving axially based on the measurement signal generated by at least one Hall sensor 18.

[0093] In addition, the control unit 22 can determine whether the magnet 25 is axially polarized or diagonally polarized.

[0094] The orientation of the signal generator 24 relative to at least one Hall sensor 18 can be different. Depending on the arrangement, the measurement signals of at least one Hall sensor 18 can also be different. Therefore, the arrangement of the Hall sensor 18 must be taken into account when evaluating the measurement signals.

[0095] exist Figure 4 The following arrangement is shown: the rotation axis D of the signal generator 24 is arranged transversely to the central axis M of the Hall sensor 18, and in particular, the Hall sensor 18 is arranged such that the extension of the central axis M intersects the rotation axis D of the signal generator 24.

[0096] exist Figure 5 The diagram shows the measurement signal M of the Hall sensor 18 corresponding to the structure when the signal generator 24 rotates. a M b M c .

[0097] In the case of the dual-axis Hall sensor 18, the Hall sensor 18 only generates the measurement signal M. a M c .

[0098] Figure 6 An alternative arrangement of the Hall sensor 18 relative to the signal generator 24 is shown, wherein the rotation axis D of the signal generator 24 is also arranged transversely to the central axis M of the Hall sensor 18, but the central axis M extends through the signal generator 24. In particular, the rotation axis D is offset relative to the central axis of the signal generator 24.

[0099] exist Figure 7 The diagram shows the signal generator 24 rotating according to... Figure 6 The corresponding measurement signal M of the Hall sensor 18 with the structure a M b M c .

[0100] In the case of the dual-axis Hall sensor 18, the Hall sensor 18 only generates the measurement signal M. a M b .

[0101] exist Figure 8 and Figure 9Another embodiment of the angle detection system 12 is shown, wherein the rotation axis D of the signal generator 24 is arranged parallel to the central axis of the Hall sensor 18.

[0102] exist Figure 10 The associated measurement signal M is shown in the figure. a M b M c .

[0103] In the case of the dual-axis Hall sensor 18, the Hall sensor 18 only generates the measurement signal M. a M b .

[0104] exist Figure 11 and 12 Another embodiment of the visible angle detection system 12 is that the rotation axis D of the signal generator 24 is arranged concentrically with the central axis of the Hall sensor 18.

[0105] exist Figure 13 The associated measurement signal M is shown in the figure. a M b M c .

[0106] In the case of the dual-axis Hall sensor 18, the Hall sensor 18 only generates the measurement signal M. a M b .

[0107] The method for confirming the angular position p of the signal generator 24 and thus the angular position p of the adjustment element 16 by means of the angle detection system 12 will be explained in more detail below.

[0108] Therefore, in one method step, multiple preset angular positions p of the signal generator 24 are defined. i Provide position-intensity data B from at least one Hall sensor 18 a B b B c At least one set. Location-intensity data B a B b B c It is stored in the location determination software 28.

[0109] As explained above, only a single Hall sensor 18 is required for the application of the angle detection system 12, but this method can also be applied when multiple Hall sensors 18 are used.

[0110] Position-intensity data B can be calculated or determined by rotating the signal generator 24 within a predetermined angular range (specifically 360°) and recording the resulting measurement signal and the corresponding angular position p. aB b B c .

[0111] Combining these two methods is also feasible. Thus, for example, for one type of angle detection system 12, the basic characteristic curve of the rotation of the signal generator 24 can be obtained, while other parameters such as ambient temperature and aging phenomena are incorporated into the characteristic curve by analytical or empirically determined correction factors.

[0112] Location-Intensity Data B a B b B c The creation of Figure 14 As shown in the image.

[0113] Location-Intensity Data B a B b B c Establish a preset angular position p throughout the entire measurement range i The measurement signal M generated by Hall sensor 18 a M b M c The relationship between them.

[0114] Location-Intensity Data B a B b B c This includes, for example, the characteristic curves of Hall sensor 18 for different parameters, such as Hall sensor 18 and signal generator 22 at different ambient temperatures or aging time points.

[0115] Location-Intensity Data B a B b B c The set of data can be created outside the angle detection system 12 and transmitted to the control unit 22 of the angle detection system 12 for storage. However, position-intensity data B can also be considered. a B b B c It is generated within the angle detection system 12 itself.

[0116] For example, location-intensity data B a B b B c Stored in matrix form, the matrix for each preset angular position p i For each Hall sensor 18, values ​​are included for each measurement direction a, b, c, and, if necessary, for other parameters.

[0117] For example, a large number of matrices can be stored, each created for a specific value of a particular parameter.

[0118] To determine the current angular position p of the signal generator 24 during valve operation, the Hall sensor 18 generates a current measurement signal M for each measurement direction based on the current angular position of the signal generator 24. a M b M c .

[0119] In another methodological step, from location-intensity data B a B b B c and the current measurement signal M a M b M c The current angular position p of the signal generator 24 is determined in the middle.

[0120] exist Figure 15 The process is illustrated in the figure.

[0121] This method is performed, in particular, on a microcontroller 30 located on a circuit board 20, on which at least one Hall sensor 18 is also arranged.

[0122] The current measurement signal M for the current angular position p of the signal generator 24 can be used. a M b M c Combined into measurement vector

[0123] From location-intensity data B a B b B c For the current measurement signal M a M b M c Determine the position p.

[0124] This only uses location-intensity data B a B b B c and the current measurement signal M a M b M c To perform this without relying on the current measurement signal M a M b M c No further computational operations can be performed, such as without using the current measurement signal M. a M b M c With location-intensity data B a B b B c Appropriate comparisons are made to perform further calculations.

[0125] To determine the location p, for example, using nearest neighbor classification or regression, such as its position in... Figure 16 As shown in the image.

[0126] In the method, for each preset position p i Find the corresponding current measurement signal M. a M b M c Distance-Intensity Data B a B b B c The spacing d of the characteristic curves of the set applicable to the detected or selected parameters i .

[0127] Spacing d i And the corresponding, belonging position p i Stored in a list. The list is based on spacing d. i The data is sorted in ascending order by size.

[0128] The minimum spacing d can be i Position p i The current position is identified as p, or it can be determined by k minimum spacings d. i Position p i A weighted average is applied, and then, for example, the weighted average is determined as the current position p. Here, for example, when there are a large number of positions p... i Choosing k=1 has proven appropriate when used to determine location-intensity data B. a B b B c Choosing k=3 has proven appropriate when the measurement data is noisy or weak, and in most other cases, choosing k=2 has proven appropriate.

[0129] The algorithm can be selectively optimized, for example, by reducing the range of data to be considered. Here, for example, a decision tree based on simple comparisons can be used, which utilizes the characteristic shape of a feature curve. The center of the region to be examined can be determined, for example, via one or two measurement signals M. a M b M c The number of levels or branches in a decision tree is defined by a threshold. The specific number of levels or branches a decision tree should have depends on the application.

[0130] Further optimization involves reducing the location-intensity data B using downsampling methods. a B b B c The resulting reduction in measurement resolution is compensated for by weighting, in particular by averaging, for example, for measurements with a spacing d.i Position p i Weighting, particularly averaging, is used to compensate (weighted kNN). That is, the accuracy of the predicted location is improved by averaging the most likely predicted locations in a probability-weighted manner.

[0131] This is particularly advantageous if the method is to be performed on a microcontroller 30, especially on a microcontroller with limited storage space, wherein the microcontroller is, for example, integrated into an angle detection system, and is a microcontroller without the capability of an external control device. The microcontroller is preferably located on the circuit board containing the one or more Hall sensors.

[0132] Alternatively, the measurement signal M can be used... a M b M c The angular position of the signal generator 24 is determined by approximating a parabola. In this method, some nearest neighbors are identified, such as the four nearest neighbors, and then a parabola containing these nearest neighbors is created. The minimum value of the parabola is used as the position prediction.

[0133] Clearly, the current measurement signal M can be taken in any suitable manner. a M b M c With location-intensity data B a B b B c Therefore, for example, in addition to nearest neighbor classification or regression described above, random forest classification or regression can also be used.

Claims

1. A method for confirming the angular position (p) of a magnetic signal generator (24) rotatably supported about a rotation axis in an angle detection system (12), particularly for determining the open state of a valve (10), said angle detection system comprising a signal generator (24) and at least one Hall sensor (18) with at least two axes, said Hall sensor being designed to generate at least two measurement signals (M) in at least two different measurement directions (a, b, c) based on a magnetic field that varies due to the rotation of said signal generator (24). a M b M c The method comprises the following steps: wherein the signal generator (24) includes a magnet (25) that is polarized such that the magnetic field strength measured by the Hall sensor (18) varies with rotation of the signal generator (24), and wherein the signal generator (24) is arranged within the detection range of the at least one Hall sensor (18). -The signal generator (24) rotates about its axis of rotation (D), - As the signal generator (24) rotates, the at least one Hall sensor (18) generates at least two measurement signals (M) in the form of magnetic field components in at least two different measurement directions (a, b, c). a M b M c ),and -Based on the measurement signal (M) generated by the at least one Hall sensor (18) a M b M c The angular position (p) of the signal generator (24) is determined in the control unit (22).

2. The method according to claim 1, characterized in that, Based on the measurement signal (M) generated by the at least one Hall sensor (18) a M b M c The angular position (p) of the signal generator (24) is determined in the control unit (22) according to the principle of machine learning.

3. The method according to claim 1 or 2, characterized in that, The control unit (22) is connected to the at least one Hall sensor (18) via signal transmission, wherein position determination software (28) is stored in the control unit (22), the position determination software being designed to: determine the angular position (p) of the signal generator (24) based on the measurement signal generated by the at least one Hall sensor (18), and wherein the position determination software (28) is designed to: convert the measurement signal (m) generated by the Hall sensor (18) into a signal generator (24). a M b M c The measurement direction (M) a M b M c Each of them is associated with an axis of the coordinate system stored in the location determination software (28).

4. The method according to claim 3, characterized in that, In one method step, position-intensity data (B) of at least one Hall sensor (18) is provided for a plurality of preset angular positions (p) of the signal generator (24). a B b B c In another method step, the Hall sensor (18) generates a current measurement signal (M) for each measurement direction (a, b, c) for the current angular position (p) of the signal generator (24). a M b M c ); and in another method step, from the location-intensity data (B a B b B c ) and the current measurement signal (M) a M b M c The current angular position (p) of the signal generator (24) is determined in the process.

5. The method according to claim 4, characterized in that, In order to provide the location-intensity data (B a B b B c The signal generator rotates within a predetermined angular range, specifically 360°, and records the resulting measurement signal (M). a M b M c ) and its corresponding angular position (p).

6. The method according to claim 4 or 5, characterized in that, Calculate the location-intensity data (B) a B b B c ).

7. The method according to any one of claims 4 to 6, characterized in that, Position-intensity data (B) is created outside the angle detection system (12). a B b B c The set of data is transmitted to the control unit (22) of the angle detection system (12) and stored there.

8. The method according to any one of claims 4 to 7, characterized in that, The current angular position (p) of the signal generator (24) is determined by means of machine learning or deep learning.

9. The method according to claim 8, characterized in that, The current angular position (p) of the signal generator (24) is determined by means of machine learning or deep learning based on nearest neighbor classification or nearest neighbor regression or random forest classification or random forest regression.

10. The method according to any one of claims 4 to 9, characterized in that, The obtained location-intensity data (B) is reduced by a downsampling method. a B b B c The quantity is weighted and the accuracy of the predicted location is improved by weighting.

11. The method according to any one of the preceding claims, characterized in that, The method is performed on a microcontroller (30) located on a circuit board (20), on which at least one Hall sensor (18) is also arranged.

12. The method according to any one of the preceding claims, characterized in that, The Hall sensor (18) outputs a separate signal corresponding to its configuration relative to the axis to the control unit (22), and the control unit (22) evaluates the measurement signal (M) according to the configuration. a M b M c ).

13. The method according to any one of the preceding claims, characterized in that, The control unit (22) is designed to: measure the signal (M) generated by the at least one Hall sensor (18). a M b M c To confirm whether the signal generator (24) is rotating or moving axially.

14. The method according to any one of the preceding claims, characterized in that, The control unit is designed to: confirm the presence of axially polarized magnets or diagonally polarized magnets.

15. An angle detection system (12) having at least one Hall sensor (18) with at least two axes and a magnetic signal generator (24) rotatably supported about a rotation axis (D), the magnetic signal generator comprising a magnet (25) polarized such that the magnetic field strength measured by the Hall sensor (18) varies with the rotation of the signal generator (24), and the signal generator is arranged within the detection range of the at least one Hall sensor (18), wherein the Hall sensor (18) is designed to generate at least two measurement signals (M) in at least two different measurement directions (a, b, c) based on the magnetic field varying due to the rotation of the signal generator (24). a M b M c ).

16. The angle detection system (12) according to claim 15, wherein the angle detection system is configured to perform the method according to any one of claims 1 to 14.

17. The angle detection system (12) according to claim 15 or 16, characterized in that, The diagonally polarized magnet (25) is toroidal or cylindrical and has at least two segments (19, 21) with different polarizations.

18. The angle detection system (12) according to any one of claims 15 to 17, characterized in that, The Hall sensor (18) has a central axis (M) extending along the depth direction of the Hall sensor (18), wherein the rotation axis (D) of the signal generator (24) is arranged concentrically, parallelly, or laterally, particularly perpendicularly to the central axis (M) of the Hall sensor (18).

19. An angle detection system (12) according to any one of claims 15 to 18, characterized in that, The Hall sensor (18) has a central axis (M) extending along the depth direction of the Hall sensor (18), wherein the rotation axis (D) of the signal generator (24) is arranged perpendicular to the central axis (M) of the Hall sensor (18).

20. A valve (10) having an angle detection system (12) according to any one of claims 15 to 19, wherein the valve (10) includes an adjustment element (16) for adjusting the flow cross section through the valve (10), wherein the adjustment element (16) is coupled to the signal generator (24) such that the position of the adjustment element (16) can be determined according to the angular position (p) of the signal generator (24).

Images