METHOD FOR REAL-TIME MONITORING OF THE OPERATIONAL STATE OF A CAPACITIVE SENSOR

Abstract

A method is provided for real-time monitoring the operational state of a capacitive sensor capable of being mounted on a rotating machine, and connected to an electronic measuring module via a high frequency transmission line, the method comprising the steps of: generating within the electronic module a signal for compensating in capacitance parasitic effects from the transmission line and the sensor; generating within the electronic module a signal for compensating in conductance parasitic effects from the transmission line and the sensor; extracting a signal representative of the capacitance compensation and of a signal representative of the conductance compensation so as to determine an operating point of the sensor; and analyzing the operating point so as to check if it is located in a predetermined area.

Claims

1. A method for real-time monitoring of the operational state of a capacitive sensor capable of being mounted on a rotating machine, and connected to an electronic measuring module via a high frequency transmission line, this method comprising the steps of: generating within the electronic module a signal for compensating in capacitance parasitic effects from the transmission line and the sensor; generating within the electronic module a signal for compensating in conductance parasitic effects from the transmission line and the sensor; extracting a signal representative of the capacitance compensation and a signal representative of the conductance compensation so as to determine an operating point of the sensor; and analyzing the operating point so as to check if it is located in a predetermined area.

2. The method according to claim 1, characterized in that it further comprises a step of triggering an alarm signal when the operating point is outside the predetermined area.

3. The method according to claim 1, characterized in that it comprises a step of analyzing the evolution of the operating point so as to deduce a diagnosis therefrom.

4. The method according to claim 1, characterized in that the predetermined area is defined on the basis of temperature limit values of the sensor and/or of the transmission line and from capacitance and conductance limit values representative of at least one of the following parameters: short-circuit of electrodes of the sensor, breaking or short-circuit of the connection between the electronic module and the sensor, cracking of a ceramic contained in the sensor.

5. The method according to claim 4, characterized in that the predetermined area is further defined on the basis of capacitance and conductance limit values representative of at least one of the following parameters of the transmission line: breaking of the means of connection to the ground, breaking of the means of connection to a guard.

6. The method according to claim 4, characterized in that a risk factor related to short-circuit of the electrodes of the sensor when the operating point tends towards saturation conductance and capacitance values is determined.

7. The method according to claim 4, characterized in that a risk factor related to a ceramic cracking of the sensor when the operating point evolves towards higher and higher conductance values is determined.

8. The method according to claim 4, characterized in that a risk factor related to a high temperature of the sensor when the operating point evolves towards higher and higher conductance and capacitance values is determined.

9. The method according to claim 5, characterized in that a risk factor related to a high temperature of the transmission line when the operating point evolves towards higher and higher conductance values in absolute value, and towards higher and higher positive capacitance values is determined.

10. The method according to claim 5, characterized in that a risk factor related to a break in means of connection to the ground of the transmission line when the operating point evolves towards lower and lower conductance and capacitance values is determined.

11. The method according to claim 5, characterized in that a risk factor related to a break in means of connection to a guard of the transmission line when the operating point evolves towards higher and higher capacitance values is determined.

12. The method according to claim 1, characterized in that each measurement carried out by the capacitive sensor is accompanied by determination of said operating point; the measurement being validated only when the operating point is inside the predetermined area.

13. The method according to claim 1, characterized in that it comprises a step of transmitting a sound and/or visual signal when the operating point is outside the predetermined area.

14. A use of the method according to claim 1, for measuring the time taken for the tips of blades to pass in a rotating machine.

15. A capacitive measuring system comprising: a capacitive sensor capable of being mounted on a rotating machine, an electronic measuring module; and a high-frequency transmission line connecting the sensor to the electronic module; the electronic module is configured in order to carry out real time monitoring of the operational state of the sensor by the steps of: generating a signal for compensating in capacitance parasitic effects from the transmission line and the sensor; generating a signal for compensating in conductance parasitic effects from the transmission line and the sensor; extracting a signal representative of the capacitance compensation and of a signal representative of the conductance compensation so as to determine an operating point of the sensor; and analyzing the operating point in order to check if it is outside a predetermined area.

16. The capacitive measuring system according to claim 15, characterized in that the transmission line comprises a triaxial or coaxial cable.

17. The capacitive measuring system according to claim 15, characterized in that it comprises a capacitive sensor of the triaxial or coaxial type.

Description

[0050] Other advantages and characteristics will become apparent on examination of the detailed description of an embodiment which is in no way limitative, and the attached drawings, in which:

[0051] FIG. 1 is a diagrammatic view of an example of an active measuring system adapted for the implementation of the invention,

[0052] FIG. 2 is a simplified diagrammatic view of the incorporation of the diagnostic method according to the invention in an active measuring system,

[0053] FIG. 3 is a curve showing the predetermined area for a coaxial sensor, and

[0054] FIG. 4 is a curve showing the predetermined area for a triaxial sensor.

[0055] Although the invention is not limited thereto, a measuring system will now be described, comprising a capacitive sensor mounted on a housing of a turbomachine for measuring the time of passage of the blade tips. Aerodynamic, thermal and mechanical loads on the turbomachine when operating can alter the reliability of the sensor, thus distort the measurements. In the context of a general monitoring of the state of the blades, it is necessary to also take into account the evolution of the state of the sensor.

[0056] The implementation of the invention will be described in relation to an embodiment of a capacitive detection system implementing automatic compensation of the leakage capacitances and conductances as described in the document FR2784179.

[0057] Of course the invention can be implemented with other embodiments of a capacitive detection system implementing a compensation of the leakage capacitances and conductances.

[0058] For reasons of clarity and conciseness, FIG. 1 and the description relating thereto below are essentially taken from this document FR2784179.

[0059] With reference to FIG. 1, the capacitive measuring system comprises an input circuit CE and a capacitive measuring circuit CMC.

[0060] The input circuit CE essentially comprises a capacitive sensor 1, a connecting high-frequency transmission line 2 and a transformer 3 connected to a high-frequency voltage source 4.

[0061] In the embodiment presented, the capacitive sensor 1 is of the triaxial type and comprises first, second and third concentric electrodes 11, 12 and 13.

[0062] The first electrode is a central measuring electrode 11, of the order of a few millimetres in diameter, for example. The electrode 11 has a free end which is arranged facing a part A connected to a reference potential such as the ground M of a device including the part A, directly or by parasitic capacitances specific to the device. The distance J between the end of the measuring electrode 11 and the part A is to be measured. For example, the part A is successively constituted by the blades of a turbine rotating about a shaft perpendicular to the plane of FIG. 1, following the arrow F. The distance J is the variable clearance, of the order of a millimetre, between the ends of the blades A passing successively in front of the central electrode 11.

[0063] The variation in the clearance J generates a low-frequency signal whose amplitude variations vary little from one cycle to the next, each cycle corresponding to the passage of a respective blade. The low-frequency signal modulates in amplitude a carrier of frequency F.sub.0 into a modulated signal having an amplitude varying as a function of the clearance J. The second electrode 12 surrounds the first electrode 11 and constitutes a guard electrode.

[0064] The third electrode 13 is a shielding electrode connected to ground M and surrounding the electrode 12 and constitutes the cylindrical metallic body of the sensor 1. The front face of the body of the sensor is fixed in a hole in the, for example, cylindrical or conical revolution housing CT of the turbine.

[0065] The parasitic impedance values, both at the sensor 1 and at the connecting line 2 are variable with the temperature and the measuring system 1 must be very tolerant for these variations in capacitance and resistance.

[0066] The connecting line 2, according to the embodiment shown in FIG. 1, comprises a section of triaxial cable with three concentric conductors 21, 22 and 23 respectively connected to the electrodes 11, 12 and 13 of the sensor 1. In general, the connecting cable comprises a rigid triaxial cable LTR typically a few metres in length, one end of which is soldered directly to the sensor, and a flexible triaxial cable LTS, the length of which can be from a few meters to a few tens of meters. The connecting cable can also comprise a section of coaxial cable on the electronics side.

[0067] According to other embodiments, the capacitive sensor 1 can be of the coaxial type. In this case it comprises only one first measuring electrode 11 and one third shielding electrode 13 connected to ground M.

[0068] In this case the connecting line 2 is a coaxial cable which comprises two concentric conductors 21 and 23 respectively connected to the electrodes 11 and 13 of the coaxial sensor 1.

[0069] An important difference between the coaxial and triaxial sensors is that the loss impedances in the coaxial sensors are more highly dependent on the temperature. Depending on the applications, it can be useful to minimize this dependency, or on the contrary, to deliberately increase it.

[0070] The alternative voltage source 4 is a high-frequency HF oscillator, controlled by a quartz at a carrier frequency F.sub.0, and controlled in amplitude so as to improve the waveform of the generated carrier and to guarantee the constancy of the characteristics of the measuring system. The carrier, which is a sinusoidal polarization voltage, is applied by the oscillator 4 to the sensor 1 via the transformer 3.

[0071] In the embodiment presented, this carrier typically has an amplitude of the order of from a few volts to 10 volts RMS, and a frequency F.sub.0 of the order of MHz.

[0072] The oscillator also provides two reference voltages in phase and in quadrature.

VR.sub.P=V.sub.R sin(ωt)et VR.sub.Q=V.sub.R sin(ωt+π/2)

[0073] with ω=2πF.sub.0 the pulsation of the HF carrier. The voltages VR.sub.P and VR.sub.Q are used to control synchronous detectors and generate active compensation voltage signals that are necessary to the operation of the capacitive measuring circuit CMC.

[0074] In the input circuit CE, the transformer 3 has a primary connected to output terminals of the oscillator 4 producing the carrier V.sub.0 sin(ωt) of frequency F.sub.0 and a secondary, constituting a floating source, connected on the one hand to the sensor 1 via the connecting line 2 and on the other hand to the inputs of a charge amplifier 5 included in the measuring circuit CMC. As will be seen below, the connecting line can be of the shielded triaxial or two-wire type, or coaxial type; a conductor of the connecting line connects the measuring electrode 11 to the inverting input (−) of the amplifier 5 via the secondary of the transformer 3. A shield conductor of the connecting line connects at least the shielding electrode 13 of the sensor 1 to the ground terminal M.

[0075] The charge amplifier 5 is an operational amplifier 30, the output of which is connected in feedback to the inverting input (−) via a feedback capacitor 51 of capacitance C51 and a feedback resistor 52 in parallel, and via a control loop described below.

[0076] The control loop comprises a bandpass filter 61 and an amplifier 62 connected in cascade to the output of the charge amplifier 5, as well as two parallel paths between the output of the amplifier 62 and the inverting input (−) of the charge amplifier 5. The channels are assigned to the phase (P) and quadrature (Q) signal processings. Each path comprises in cascade a synchronous detector 7.sub.P, 7.sub.Q, an integrator 8.sub.P, 8.sub.Q, a multiplier 9.sub.P, 9.sub.Q et a reference capacitor C.sub.P, C.sub.Q of reference capacitance C.sub.R.

[0077] The path comprising the synchronous detector 7.sub.P, the integrator 8.sub.P, the multiplier 9.sub.P and the reference capacitor C.sub.P makes it possible to compensate the reactive part of the loss impedance of the sensor 1 and of the connecting line 2.

[0078] The path comprising the synchronous detector 7.sub.Q, the integrator 8.sub.Q, the multiplier 9.sub.Q and the reference capacitor C.sub.P makes it possible to compensate the resistive part of the loss impedance of the sensor 1 and of the connecting line 2.

[0079] This loss impedance can be globally modelled as a parasitic capacitance and resistance in parallel between the inverting input (−) of the charge amplifier 5 and the ground M.

[0080] The AF bandwidth of the filter 61 is centred on the frequency F.sub.0 of the oscillator 4 and a width typically fixed at approximately 300 kHz. The amplifier 62 is a unity-gain follower-amplifier and produces a filtered and amplified voltage signal SFA the amplitude of which varies inversely with the clearance J in the sensor 1. This signal is applied to the two synchronous detectors 7.sub.P and 7.sub.Q which are controlled by the reference phase and quadrature voltages VR.sub.P and VR.sub.Q provided by the oscillator 4. The synchronous detectors 7.sub.P and 7.sub.Q are phase detectors which amplitude demodulate the filtered and amplified voltage signal SFA into two phase and quadrature component signals S.sub.P and S.sub.Q one of which, S.sub.P, is used to measure the clearance J. The low-frequency component signals S.sub.P and S.sub.Q are respectively integrated in the integrators 8.sub.P and 8.sub.Q into a phase voltage V.sub.P and a quadrature voltage V.sub.Q in order to stabilize the control loop. The voltages V.sub.P and V.sub.Q leaving the integrators are applied to first inputs of the two multipliers 9.sub.P and 9.sub.Q with a multiplication factor K so as to multiply the voltages V.sub.P and V.sub.Q respectively by the reference voltages VR.sub.P=V.sub.R sin(2ωt) and VR.sub.Q=V.sub.R sin(2ωt+π/2) applied to second inputs of the multipliers.

[0081] The amplitude modulated signals [K V.sub.P V.sub.R sin(2ωt)] and [K V.sub.Q V.sub.R sin(2ωt+π/2)] produced by the multipliers 9.sub.P et 9.sub.Q are reinjected at the inverting input (−) of the charge amplifier 5 via the reference capacitors C.sub.P and C.sub.Q with suitable phases and added to the measuring signal transmitted by the sensor 1 via the transformer 3 in order to obtain the stability of the control.

[0082] Given the amplification factor ensured by the integrators which is very high at the low frequencies, the average value of the error signal ER leaving the amplifier 5 and thus the average values of the S.sub.P et S.sub.Q signals are maintained at zero.

[0083] The in-phase current passing through the sensor 1 and having (V.sub.0 C.sub.13ω) as amplitude is compensated by the current passing through the phase capacitor Cp, having the amplitude:

Ip=K Vp V.sub.R C.sub.Rω.

[0084] The quadrature current due to the sensor losses and having (V.sub.0 C.sub.13) as amplitude is compensated by a current passing through the quadrature capacitor CQ, having the amplitude:

I.sub.Q=K V.sub.QV.sub.R C.sub.Rω.

[0085] The capacitor C.sub.Q supplied with the quadrature current behaves as a resistor supplied with an in-phase current. It thus allows for a compensation of the resistive losses, while avoiding the thermal noise which would be introduced by the use of a resistor.

[0086] The nominal sensitivities at the outputs of the integrators I.sub.P and I.sub.Q are:

[00001]$S\left(V_P\right)=\frac{V_P}{C_{13}}=\frac{V_0}{{\mathrm{KV}}_R}.Math.\frac{1}{C_R}.Math.V.Math.\text{/}.Math.\mathrm{pF},\mathrm{and}$$S\left(V_Q\right)=\frac{V_Q}{1/R_{13}}=\frac{V_0}{{\mathrm{KV}}_R}.Math.\frac{1}{C_R.Math.\omega }.Math.V.Math.\text{/}.Math.\mathrm{Siemens}.$

[0087] At the output of the amplifier 5, the error signal ER is normally zero when the part A is immobile. In operation, when the turbine rotates, the signal ER comprises only the background noise of the capacitive sensor 1, as well as transient signals, for example the passage of a blade A, which are in the limited bandwidth AF of the control.

[0088] The signal S.sub.P (or S.sub.Q) of passage of a blade is present at the output of the clearance measuring system in the form of a signal with a zero average value.

[0089] The simplified diagram of FIG. 2 is a functional representation of a measuring and processing system implementing the method according to the invention. In the presented embodiment, it comprises an electronic module 26 incorporating the electronic elements of FIG. 1. According to the invention, the electronic module 26 also comprises a signal processing unit 24 capable of implementing the method according to the invention.

[0090] The sensor 20 is arranged at the end of a transmission line 21 constituted by a triaxial cable. In the presented embodiment, the sensor 20 is a coaxial sensor, the electrodes of which are connected respectively to the measuring conductor and the ground conductor of the cable. Other cables can of course be used, such as in particular a coaxial cable.

[0091] The input of the electronic module 26 is ensured by the preamplifier 22 comprising in particular the charge amplifier 5 of FIG. 1. Conductance and capacitance compensation signals are generated by the signal processing unit 24 and are injected into the preamplifier 22. An amplifier 23, incorporating in particular the amplifier 6 of FIG. 1, supplies, after amplification, the signal processing unit 24 from a signal coming from the preamplifier 22. The processing unit 24 advantageously comprises the synchronous detectors and the integrators of FIG. 1 as well as a microcontroller 27 configured in order to implement the method according to the invention. This microcontroller 27 receives the signals VQ and VP, then derives therefrom conductance and capacitance values, knowing that VQ is proportional to the conductance which can be named G.sub.L, and VP is proportional to the capacitance which can be named C.sub.L.

[0092] The signal processing unit 24 is capable of generating an output signal Vout which is proportional to the capacitance measured by the sensor 20, Vout=kC, k being a real number. This output signal can supply a RMS/DC converter 28.

[0093] The microcontroller is configured in order to determine in real time the values of G.sub.L and C.sub.L and store these values in memory so as to monitor their evolutions. A value G.sub.L associated with a value C.sub.L constitutes an operating point which can be represented in a two-dimensional space having the values of C.sub.L for the x-axis and the values of G.sub.L for the y-axis.

[0094] FIG. 3 shows curves representing the variations in the operating point in a G.sub.L/C.sub.L coordinate system for a coaxial sensor.

[0095] There can be seen a first curve T in the form of an oblique segment in the domain of the positive conductance and capacitance values.

[0096] In a so-called normal state of health, as a function of the variation in environmental parameters such as temperature, the line compensation, represented by the capacitance C.sub.L and the conductance G.sub.L, remains in or close to the predetermined curve T and the capacitance C.sub.L and the conductance G.sub.L follow a monotone evolution with respect to each other (for example the conductance increases when the capacitance increases).

[0097] In the case of a failure of the sensor or of the cable, the operating point materialized by the capacitance C.sub.L and the conductance G.sub.L moves away from the so-called normal operating curve T. Detection of the evolution of these parameters at a distance from the normal operating curve makes it possible to detect failures and/or to invalidate measurements, in the context for example of a blade state monitoring (BHM for “blade health monitoring”).

[0098] Under normal conditions, the current operating point PF can develop over the curve T between a lower point corresponding to a temperature Tmin and an upper point corresponding to a temperature Tmax.

[0099] According to the invention, in order to detect the failures and the risks, a curve defining a predetermined area ZP can be defined beforehand. This predetermined area includes the curve T. It is thus possible to implement the following method of detection: [0100] when the operating point is no longer on the curve T but is inside the predetermined area, it is considered that there is a risk to the operating state of the sensor and that it is necessary to monitor the evolution; [0101] Beyond the predetermined area, it is considered that the measurement is not reliable.

[0102] FIG. 3 shows arrows E1-E4 which start from the operating point and point in different directions. Each arrow represents a risky evolution due to one or more predefined characteristics.

[0103] For example, the evolution in the direction of the arrow E1 (the capacitance increases when the conductance decreases) is characteristic of a evolution towards a break in the transmission line.

[0104] The evolution in the direction of the arrow E2 (increase in the capacitance for a fixed value of the conductance) is characteristic of a evolution towards a break in a guard connection of the transmission line.

[0105] The evolution in the direction of the arrow E3 (reduction in the capacitance for a fixed value of the conductance) is characteristic of a evolution towards a break in the electrode of the sensor.

[0106] The evolution in the direction of the arrow E4 (increase in the conductance for a fixed value of the capacitance) is characteristic of a evolution towards a short-circuit at the sensor or the transmission line.

[0107] FIG. 4 shows a normal operating curve T1 in a predetermined area ZP1 for a triaxial sensor. In this case, the curve T1 is a curve passing through zero and having a minimum (corresponding to a temperature value Tmin) in the quadrant of the positive conductances and of the negative capacitances. The maximum (corresponding to a temperature value Tmax) is located in the quadrant of the negative conductances and of the positive capacitances.

[0108] As explained previously, the present invention makes it possible to carry out various diagnoses from the analysis of the operating point and of its evolution. Different scenarios are described below simply by way of non-limitative examples.

[0109] Concerning the capacitive sensor: [0110] a cracking in the ceramic constituting the insulator between the electrodes can result in an absorption of humidity or pollutants. It is characterized by abnormally high leakage conductance values G.sub.L. The associated risks are oxidation, definitive pollution, breakage and thus destruction of the sensor; [0111] a too high temperature causes an increase in the dielectric permittivity of the insulator. It is characterized by abnormally high leakage capacitance C.sub.L and leakage conductance G.sub.L values. The associated risks are irreversible damage to or destruction of the sensor; [0112] a short-circuit trigger between the measuring electrode and the ground is characterized by saturations of the compensation voltages and overconsumption of the oscillator 4 which excites the transformer. There is also a risk of destruction of the sensor; [0113] a break in the electrical connection to the measuring electrode, apart from the absence of measurement signals, is characterized by a reduction in the leakage capacitance C.sub.L due to the fact that the residual leakage capacitance relates only to the section of the measuring line up to the break.

[0114] Concerning the cable, and in particular a triaxial cable: [0115] a too high temperature causes an increase in the dielectric permittivity of the insulator (generally mineral). It is characterized by abnormally high leakage capacitance C.sub.L and leakage conductance G.sub.L values in absolute value, the leakage conductance G.sub.L (equivalent) having a negative sign because of the dephasing introduced by the resistance of the guard cable; [0116] a break in the connection of the ground conductors is characterized by abnormally low leakage capacitance C.sub.L and leakage conductance G.sub.L values since they correspond to a length of cable (up to the break) less than the normal length. [0117] a break in the connection of the guard conductors is characterized by abnormally high leakage capacitance C.sub.L since they correspond to increased leakage capacitance towards the ground.

[0118] Thus, it can be noted that the invention makes it possible to diagnose causes of failure, or groups of causes of failures, or an abnormal state, by detecting at least one of the following events: [0119] an abnormal evolution of a parameter among the leakage capacitance C.sub.L and leakage conductance G.sub.L, the other parameter retaining a normal value;—an abnormal evolution of the two parameters, the leakage capacitance C.sub.L and leakage conductance G.sub.L.

[0120] Of course, the range of possible failures or a more precise identification of the failure can be realized according to the invention by incorporating additional information into the analysis, such as: [0121] external temperature measurements; [0122] the quality of the measurements, or the absence of measurements; [0123] a monitoring of electricity consumption, in order to detect in particular saturated components.

[0124] Of course, the invention is not limited to the examples which have just been described and numerous adjustments can be made to these examples without exceeding the scope of the invention.