Measurement method for aviation-specific proximity sensor

Abstract

A measurement method for an aviation-specific inductive proximity sensor (IPS for short) includes steps of: 1) building a measurement circuit, wherein an IPS comprises an internal resistance r and an inductance L; 2) building a look-up table, wherein the step 2) specifically comprises steps of: sampling a first voltage measured value U.sub.1 corresponding to a first constant delay time T.sub.1 with the ADC; sampling a second voltage measured value U.sub.2 corresponding to a second constant delay time T.sub.2 with the ADC; then obtaining voltage discharge formulas U.sub.1(T.sub.1, R, r, L) and U.sub.2(T.sub.2, R, r, L) of an r-L circuit; and 3) compressing the look-up table, utilizing the compressed look-up table for calculation during measurement. The present invention ensures that the system works within a standard temperature range, and improves measurement stability, reliability, and real-time performance. Furthermore, there is no floating point calculation, which saves CPU or MCU hardware resources.

Claims

1. A measurement method for measuring an induction of an aviation-specific inductive proximity sensor (IPS), comprising steps of: 1) building a measurement circuit, wherein the IPS comprises an internal resistance r and an inductance L, a value of the internal resistance r increases when environmental temperature increases, a value of the inductance L relates to a distance of the IPS; the measurement circuit of the IPS comprises a current-limiting resistance R, the IPS and a controlled switch connected in series: an analog digital converter (ADC) is placed at a voltage measurement node between the current-limiting resistance R and the internal resistance r; at a time T.sub.0, when the controlled switch is off, the inductance L is charged, a field programmable gate array (FPGA) drives the measurement circuit, when the controlled switch is on, the inductance L slowly discharges through the internal resistance r and the current-limiting resistance, a first constant delay time T.sub.1 and a second constant delay time T.sub.2 control the ADC to sample; and 2) building a look-up table comprising steps of: sampling a first voltage measured value U(T.sub.1) which is a sampling voltage at the voltage measurement node corresponding to the first constant delay time T.sub.1 with the ADC; sampling a second voltage measured value U(T.sub.2) which is a sampling voltage at the voltage measurement node corresponding to the second constant delay time T.sub.2 with the ADC; using voltage discharge formulas U(T.sub.1) and U(T.sub.2) of an r-L circuit for solving L and r: $\begin{array}{cc}\{\begin{array}{c}{U}_1=\frac{U_{\max }}{R+r}\left[r+R\times e^{-\frac{R+r}{L}\times T_1}\right]\\ U_2=\frac{U_{\max }}{R+r}\left[r+R\times e^{-\frac{R+r}{L}\times T_2}\right]\end{array}& \left(4\right)\end{array}$ wherein T.sub.1, T.sub.2, U.sub.max, and R are constants; L and r are independent variables, in the two formulas for solving [L, r] corresponding to sample values [U(T.sub.1), U(T.sub.2)], wherein U.sub.max is equal to 2.sup.n−1, n is a resolution of the ADC, then building the look-up table of the internal resistance r and the inductance L corresponding to the sample values [U(T.sub.1), U(T.sub.2)].

2. The measurement method, as recited in claim 1, wherein in the step 2), between the first constant delay time T.sub.1 and the second constant delay time T.sub.2, the controlled switch is off which deactivates the measurement circuit, the inductance L slowly discharges; at an initial time T.sub.0, the controlled switch is on which activates the measurement circuit, then the inductance L is charged, and a current thereof is: $\begin{array}{cc}i=\frac{U_{\max }}{R+r}\left[1-e^{-\frac{R+r}{L}\times T}\right]& \left(1\right)\end{array}$ a voltage at the voltage measurement node is:
U(T)=U.sub.max−i×R  (2) therefore: $\begin{array}{cc}U\left(T\right)=\frac{U_{\max }}{R+r}\left[r+R\times e^{-\frac{R+r}{L}\times T}\right]& \left(3\right)\end{array}$ the first constant delay time T.sub.1 and the second constant delay T.sub.2 control the ADC to sample voltage at the voltage measurement node, U(T) is a monotone decreasing function, if T.sub.1<T.sub.2, then U(T.sub.1)>U(T.sub.2), the values of the internal resistance r and the inductance L are calculated.

3. The measurement method, as recited in claim 2, wherein based on a fact that the formula (4) has only one solution [L, r] with given sample values [U(T.sub.1), U(T.sub.2)], the solution is obtained by a least square method comprising steps of: building an object function:
min:(U(T.sub.1)−U(L,r,T.sub.1)).sup.2+(U(T.sub.2)−U(L,r,T.sub.2)).sup.2
s.t.1:r>0
s.t.2:L>0  (5) further obtaining: $\begin{array}{cc}\min \text{:}{\left(U_1\left(R+r\right)-U_{\max }\left(r+R\times e^{-\frac{R+r}{L}\times T_1}\right)\right)}^2+{\left(U_2\left(R+r\right)-U_{\max }\left(r+R\times e^{-\frac{R+r}{L}\times T_2}\right)\right)}^{2}s.t\mathrm{.1}\text{:}r>0s.t\mathrm{.2}\text{:}L>0& \left(6\right)\end{array}$ and applying each of the sample values in the [U(T.sub.1), U(T.sub.2)] for respectively calculating and obtaining a numerical solution of the [L, r] corresponding to the [U(T.sub.1), U(T.sub.2)], in such a manner that the look-up table is obtained.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows a schematic view of an equivalent circuit model of an IPS according to a preferred embodiment of the present invention.

(2) FIG. 2 shows a schematic view of a driving and detection module of the IPS according to the preferred embodiment of the present invention.

(3) FIG. 3 shows a waveform of a circuit output signal of the IPS according to the preferred embodiment of the present invention.

(4) FIG. 4 shows a response waveform of the IPS which is close to a target according to the preferred embodiment of the present invention.

(5) FIG. 5 shows a response waveform of the IPS which is far from the target according to the preferred embodiment of the present invention.

(6) FIG. 6 shows a response waveform of the IPS which is 1 mm˜2 mm far from the target according to the preferred embodiment of the present invention.

(7) FIG. 7 shows a discharge curve according to the preferred embodiment of the present invention.

(8) FIG. 8 shows a cluster of curves passing (U.sub.1, T.sub.1) according to the preferred embodiment of the present invention.

(9) FIG. 9 shows a cluster of curves passing (U.sub.2, T.sub.2) according to the preferred embodiment of the present invention.

(10) FIG. 10 shows a restriction condition of r-L according to the preferred embodiment of the present invention.

(11) FIG. 11 shows a restriction condition of r-U.sub.2 according to the preferred embodiment of the present invention.

(12) FIG. 12 shows a cluster of curves of U.sub.2 which have positive real crossover points with a given curve of U.sub.1 according to the preferred embodiment of the present invention.

(13) FIG. 13 shows coordinate mapping according to the preferred embodiment of the present invention.

(14) FIG. 14 shows coordinate mapping according to the preferred embodiment of the present invention.

(15) FIG. 15 shows an effective table area according to the preferred embodiment of the present invention.

(16) FIG. 16 shows a solution of a function L(U.sub.1, U.sub.2) within an effective definition domain according to the preferred embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

(17) Referring to the drawings, the present invention is further illustrated.

(18) According to the present invention, an internal circuit structure of an IPS is simple, which is formed by a group of metal winding wires. Referring to the FIG. 1, an equivalent circuit model thereof is formed by an internal resistance r in an induction coil and an inductance L connected in series (a parasitic capacitance is ignored). The internal resistance r increases when environmental temperature increases, and a value of the inductance L relates to a distance between the IPS and a metal target. If an external metal target is moving towards the IPS, electromagnetic field distribution around the IPS is greatly changed, and an equivalent induction of the IPS is increased. By driving and detecting an induction value, whether the external metal target is moving towards the IPS is able to be judged.

(19) For example, an induction value of a product of Crouzet changes from 4.5 mH (far) to 5.5 mH (close), which rarely changes with a temperature. A resistance thereof changes with a temperature, and a reference range is 10Ω to 15Ω, which rarely changes with a proximity

(20) By driving the IPS and detecting an inductive reactance of the induction, the proximity is able to be detected.

(21) Accordingly, driving and detecting methods require:

(22) 1) a peak of a driving current is no more than 15 mA, wherein a higher current is not able to pass an electromagnetic compatibility test; the peak is no less than 10 mA, a lower current is liable to be interfered by an external electromagnetic environment; and

(23) 2) a inductance change rate is 2%; a quantified measurement index of 5% is required; inductance detection accuracy of the system should be 0.1%; and

(24) 3) as the change rate of the internal resistance within a standard temperature range is over 60%, an impact of temperature drift on inductance detection must be considered; and

(25) 4) for improving system MTBF, hardware resources such as CPU, MCU and DSP are not utilized.

(26) Referring to the FIG. 2, FPGA controlling and driving circuit outputs a 0˜5V pulsing signal. A drive circuit outputs a driving signal with an amplitude of 0˜5V, a positive bandwidth of 2 ms and a period of 200 Hz. Referring to the FIG. 3, a waveform of the drive circuit output signal is illustrated, wherein the output signal passes through a divider circuit, a filter and a sensor, and grounds through the filter.

(27) If the IPS is close to a target, a waveform detected at a measurement node is as illustrated in the FIG. 4. If the IPS is far from a target, a waveform detected at the measurement node is as illustrated in the FIG. 5. If the IPS is 1 mm˜2 mm far from a target, a waveform detected at the measurement node is as illustrated in the FIG. 6.

(28) Two constant delay times are obtained by an ADC for sampling voltage measured values U.sub.1 and U.sub.2 corresponding to the constant delay times T.sub.1 and T.sub.2, and further obtaining two voltage discharge formulas U.sub.1(T.sub.1, R, r, L) and U.sub.2(T.sub.2, R, r, L) of an r-L circuit, wherein T.sub.1, T.sub.2, and R (limiting resistance) are constants; the two formulas are united for solving the internal resistance r and the inductance L.

(29) For building the formulas:

(30) Referring to the FIG. 1, between two measurements, a controlled switch is on, the inductance L slowly discharges through internal and external resistances; at a time T.sub.0, the controlled switch is off, then the inductance L is charged, and a current thereof is:

(31) 0$\begin{array}{cc}i=\frac{U_{\max }}{R+r}\left[1-e^{-\frac{R+r}{L}\times T}\right]& \left(1\right)\end{array}$

(32) a voltage at the voltage measurement node is:
U=U.sub.max−i×R  (2)

(33) therefore:

(34) $\begin{array}{cc}U=\frac{U_{\max }}{R+r}\left[r+R\times e^{-\frac{R+r}{L}\times T}\right]& \left(3\right)\end{array}$

(35) the first constant delay time T.sub.1 and the second constant delay time T.sub.2 control the ADC to sample, and the corresponding U.sub.1 and U.sub.2 are:

(36) $\begin{array}{cc}\{\begin{array}{c}{U}_1=\frac{U_{\max }}{R+r}\left[r+R\times e^{-\frac{R+r}{L}\times T_1}\right]\\ U_2=\frac{U_{\max }}{R+r}\left[r+R\times e^{-\frac{R+r}{L}\times T_2}\right]\end{array}.& \left(4\right)\end{array}$

(37) Parameters:

(38) A range of the inductance L is [4.5, 5.5] mH; a range of the internal resistance r is [10, 15]Ω.

(39) U.sub.1 and U.sub.2 are represented by values sampled by a 12-bit ADC, wherein U.sub.max is 4095.

(40) According to a current-limiting condition, the limiting resistance R is 230Ω.

(41) Referring to the FIG. 7, medians of the ranges of the internal resistance r and the inductance L are selected for forming a discharge curve. At a time T.sub.1, the inductance L discharges to 30%. At a time T.sub.2, the inductance L discharges to 60%. T.sub.1 is 10.284 μs, T.sub.2 is 24.287 μs.

(42) It is provable that the formula (4) has only on solution:

(43) after quantifying by the ADC:

(44) $\begin{array}{cc}U\left(R,T,r,L\right)=\frac{4095}{R+r}\left[r+R\times e^{-1\frac{R+r}{L}\times T}\right]& \left(5\right)\end{array}$

(45) U(R, T, r, L) is a monotone decreasing function about R and T, and also a monotone increasing function about r and L.

(46) Because U(R, T, r, L) is monotonic continuous function, an inverse function thereof exists, and the formula (4) has solutions.

(47) With given (U.sub.1, T.sub.1), the formula (4) is restricted by r and L. A space p is defined as a set of (r.sub.p, L.sub.p) which enables the function U-T to pass (U.sub.1, T.sub.1), as illustrated in the FIG. 8.

(48) A space q is defined as a set of (r.sub.q, L.sub.q) which enables the function U-T to pass (U.sub.2, T.sub.2), as illustrated in the FIG. 9.

(49) The spaces p and q are drawn in an r-L coordinate, as illustrated in the FIG. 10. The U-T curve corresponding to a crossover point of two curves in the FIG. 10 passes (U.sub.1, T.sub.1) and (U.sub.2, T.sub.2) at the same time, is the solution of the function. Because L(r) is monotone, there is at most one crossover point, which means that the formula (4) has at most one solution.

(50) Therefore, the formula (4) has only one solution.

(51) The formula (4) is a transcendental equation, and it is difficult to obtain a analytical solution. However, since the formula (4) has only one solution, a numerical solution satisfying an engineering accuracy requirement is obtained by iteration.

(52) The formula (4) is calculated by a least square method comprising steps of:
min:(U.sub.1−U.sub.1(L,r)).sup.2+(U.sub.2−U.sub.2(L,r)).sup.2
s.t.1:r>0
s.t.2:L>0  (6)

(53) further obtaining:

(54) $\begin{array}{cc}\min \text{:}{\left(U_1\left(R+r\right)-U_{\max }\left(r+R\times e^{-\frac{R+r}{K}\times T_1}\right)\right)}^2+{\left(U_2\left(R+r\right)-U_{\max }\left(r+R\times e^{-\frac{R+r}{L}\times T_2}\right)\right)}^{2}s.t\mathrm{.1}\text{:}r>0s.t\mathrm{.2}\text{:}L>0& \left(7\right)\end{array}$

(55) and applying each of the sample values in the [U.sub.1, U.sub.2] for respectively calculating and obtaining the numerical solution of the [L, r] corresponding to the [U.sub.1, U.sub.2], in such a manner that the look-up table is obtained. In practice, the value of L is obtained with U.sub.1 and U.sub.2, so as to define a state of the IPS.

(56) Look-up Table Compression:

(57) For example, in a 12-bit ADC, a size of a complete 2-dimensional look-up table is 2.sup.12×2.sup.12, and a storage volume thereof is 16M units, which means poor practicability. An effective method for compressing the look-up table must be found for ensuring practicability of the present invention. By taking full advantage of a distribution characteristic of sample values [U.sub.1, U.sub.2] in a real circuit, the look-up table is able to be effectively compressed.

(58) With given (rx, Lx), the only (U.sub.1x, U.sub.2x) is calculated and obtained by the formula (7). Theoretically, there are and only one solution, but the solution is restricted by conditions such as physical models and T.sub.1<T.sub.2.

(59) U(T) is a monotone decreasing function. If T.sub.1<T.sub.2, then U.sub.1>U.sub.2. Supposing that U.sub.1<U.sub.2, the solution of the function should at least comprise a negative number for changing monotone, which is physically unreasonable.

(60) After the formula (5) is converted to an inverse function of U about L:

(61) $\begin{array}{cc}\frac{1}{L}=-\frac{\log \left[\frac{U}{4096R}\left(R+r\right)-\frac{r}{R}\right]}{\left(R+r\right)\times T}& \left(8\right)\end{array}$

(62) Therefore, L may be a complex number, unless:

(63) $\begin{array}{cc}\left[\frac{U}{4096R}\left(R+r\right)-\frac{r}{R}\right]>0& \left(9\right)\end{array}$

(64) which means:

(65) $\begin{array}{cc}U>\frac{4096r}{R+r}& \left(10\right)\end{array}$

(66) If (U.sub.1x, U.sub.2x) is improperly given, (rx, Lx) obtained may comprise a negative or even complex number. In fact, physically sampled (U.sub.1, U.sub.2) is certainly reasonable. A restriction relationship exists between U.sub.1 and U.sub.2, in such a manner that (r, L) belongs to a positive real domain. That is to say, a crossover point of the two curves in the FIG. 10 is at a first quadrant of a positive real domain coordinate. (U.sub.1, U.sub.2) out of the restriction will not be obtained in the practical sampling, and do not need to be recorded.

(67) With the sample values (U.sub.1, T.sub.1) and (U.sub.2, T.sub.2), it is obtained that:

(68) $\begin{array}{cc}\{\begin{array}{c}L=-\frac{\left(R+r\right)\times T_1}{\log \left[\frac{U_1}{4096R}\left(R+r\right)-\frac{r}{R}\right]}\\ L=-\frac{\left(R+r\right)\times T_2}{\log \left[\frac{U_2}{4096R}\left(R+r\right)-\frac{r}{R}\right]}\end{array}& \left(11\right)\end{array}$

(69) then obtaining

(70) $\begin{array}{cc}{\left[\frac{U_1}{4096R}\left(R+r\right)-\frac{r}{R}\right]}^{\left(\frac{T_2}{T_1}\right)}=\left[\frac{U_2}{4096R}\left(R+r\right)-\frac{r}{R}\right]& \left(12\right)\end{array}$

(71) If a set of (U.sub.1, U.sub.2) satisfies both the formulas (10) and (12), the function has a positive real number solution.

(72) According to the formula (12), the given U.sub.1 and r should satisfy a condition of:

(73) 0$\begin{array}{cc}r\ge \frac{U_{1}R}{U_1-4096}& \left(13\right)\end{array}$

(74) wherein r.sub.min is able to be determined.

(75) After determining the U.sub.1, a restriction relationship of U.sub.2(r) is obtained, as illustrated in the FIG. 11.

(76) U.sub.2(r) is a monotone function, a range thereof is determined by the definition domain of the r. U.sub.2 within the range and the given U.sub.1 both enable a positive real number solution of the function, which is illustrated in the FIG. 12.

(77) Referring to the FIG. 13, essence of look-up calculation is mapping the U.sub.1-U.sub.2 coordinate to the r-L coordinate.

(78) Process A is a physical sampling process. Every point in the r-L coordinate is able to be mapped to the U.sub.1-U.sub.2 coordinate.

(79) Process B is a look-up calculation process. Some points in the U.sub.1-U.sub.2 coordinate are able to be mapped back to the r-L coordinate. Some points in the U.sub.1-U.sub.2 coordinate will fall in other quadrants (negative number) or a 4-dimensional complex space when being mapped to the r-L coordinate.

(80) The r is [10, 15]Ω, the L is [4.5, 5.5] mH; mapping relationship thereof is illustrated in the FIG. 14.

(81) A range of U.sub.1 is determined by the formula (13). With each given U.sub.1, a range of U.sub.2 corresponding to the r within a physical changing range is able to be obtained by the restriction relationship of the formula (12). The range of U.sub.2 is obtained in sequence, and a shadow area in the U.sub.1-U.sub.2 coordinate in the FIG. 14 is determined for obtaining the FIG. 15.

(82) A quantity of units of the look-up table is decreased from 4096*4096 to 9360, which means the look-up table is compressed by 1792 times.

(83) Beneficial Effects:

(84) According to the definition domain of (U.sub.1, U.sub.2) in the FIG. 15, the corresponding L is obtained in sequence by the formula (4), and the look-up table is built. Regarding L as a Z axis, an image of the function L(U.sub.1, U.sub.2) within the definition domain is obtained, as shown in the FIG. 16.

(85) After ergodic, a max quantization error of L during looking up is 6.30153 uH, at U.sub.1=2754 and U.sub.2=1644.

(86) It is illustrated that with a 12-bit ADC, measurement accuracy is up to 1%. L is [4.5, 5.5] mH, wherein at least 33 levels of effective resolutions are able to be obtained.

(87) According to the theory of the present invention, driving and detection systems for non-contact IPSs are developed. According to tests, a total measurement error of the system is lower than 1% at a temperature between −55° C. and 125° C.; all detection and calculation are directly processed by a logic device, wherein hardware such as CPU and DSP is not utilized. And the detection and calculation are realized by looking up a 10K look-up table. At the same time, the system has been in accordance with standards of MIL-STD-461F for providing CS114, RE102, and RS103 tests. Results thereof satisfy MIL-STD-461F requirements.

(88) One skilled in the art will understand that the embodiment of the present invention as shown in the drawings and described above is exemplary only and not intended to be limiting.

(89) It will thus be seen that the objects of the present invention have been fully and effectively accomplished. Its embodiments have been shown and described for the purposes of illustrating the functional and structural principles of the present invention and is subject to change without departure from such principles. Therefore, this invention includes all modifications encompassed within the spirit and scope of the following claims.