FIELD

[001] The invention relates to methods, systems and sensors for detecting acceleration, including seismic motion and vibrations. BACKGROUND

[002] In harsh and remote environments, including in connection with offshore oil and gas exploration, seismic data collection or surveys are performed to gain a better understanding of subterranean geological formations. In a seismic survey, energy sources create seismic pulses or waves which travel into the earth and are then reflected back from various subsurface features. The reflected signal is detected by sensors deployed generally along the surface for analysis and interpretation. Power supply and placement are often determining factors in seismic surveys, so while sensors may be deployed along a land surface, streamers are dragged behind vessels in maritime seismic survey.

[003] Historically, 2-axis and 3-axis accelerometers used for seismic measurements feature pre-tensioned strings supporting a central proof mass in orthogonal directions within a housing, and use electromagnetic drivers to detect tensile stresses in each string, as shown in prior art Figure 1. This particular measurement utilized the concept of modulating the transverse resonant frequency by the tensile stresses, a familiar concept used in tuning musical instruments such as guitar strings. Power constraints on this method reduce its ability to be deployed remotely.

[004] The majority of current seismometers that utilize an inertial reference, i.e. seismic mass or proof mass, use position feedback to simulate an unrealistic condition of the mass being suspended on extremely long and soft springs that would result in a very low natural frequency of oscillations. This is needed to detect the sub-hertz spectra in many applications. Position feedback using often electromagnetic actuators tend to draw constant power which is a costly commodity in power limited autonomous systems.

[005] The ability and theoretical basis for using fibre Bragg grating to detect acceleration in seismic detection is discussed in the work of Jack Wu, graduate student Dalhousie University, and Dr. Vlastimil Masek, assistant professor Memorial University of Newfoundland, in April 2009, and also in United States Patent Application publication 201 1/0096624, published April 28, 201 1. In each case, the practical application is not suited for two-dimensional or three-dimensional measurement. Asymmetries in these older designs lead to resonant and underlying harmonic frequencies of the frames themselves, which may corrupt output data from the devices, or fail to detect acceleration in directions orthogonal to the sensor.

[006] Other multi-dimensional systems use multiple masses which, as a result of small difference in location, can lead to measurement inaccuracy or difficulties on correlating the data.

[007] There is a need for improved, low-cost, low-power, sensors designed for deployment in remote environments. [008] There is a need for a practical acceleration sensor design, capable of three dimensional seismic readings in a low power operating environment.

SUMMARY

[009] The invention is a fibre optic based accelerometer sensor that is capable of measuring seismic activity in three directions from a single inertial mass. Dual fibre Bragg grating (FGB) fibres suspend a seismic mass within a framing and tensioning structure, in at least two directions. The optical fibre features Bragg grating in appropriate wavelengths, and changes in tension resulting from acceleration of the seismic mass are measured on each fibre. As the fibre uses low power light rather than high power electromagnetic fields, the device demonstrates both high reliability and the ability to be deployed remotely. Rather than using the string accelerometer known in the art and shown in Figure 1 , which compares tensile stresses in the vibrating strings 63 suspending inertial/proof/seismic mass 62 using electromagnetic drivers 61 , the proposed accelerometers replace the strings with dual fibre Bragg grating (FBG), in different modes as compact, low-cost, low-power, fibre-optic inertial multi-dimensional accelerometers that feature broad dynamic range. In a first mode, the fully orthogonal implementation features low cross-sensitivity. In a second mode, tensioning in the third direction allows prolonged use before tuning, and common mode operation allows acceleration readings in the direction perpendicular to the fibres.

[010] Two modes of the accelerometer are shown. In a first mode, the instrument features a proof/seismic/inertial mass being suspended by three optical fibres strings passing through the mass in three substantially orthogonal directions, and measures the acceleration through the measurement of light interference in each fibre.

[011] Using a single mass measured on multiple axes requires control software capable of interpreting the data from the multiple 6 FBGs so as to take cross coupling into account. Although the IM-DD method naturally attenuates some cross components (if both wires are being stretched the overlap region will not go below nil, for instance), each pair of wires will have affects, both dampening and resonant, on the other orthogonal wires and vice versa.

[012] In another mode, 2 optical fibres pass through the inertial mass in two first orthogonal directions, and are tensioned in a third substantially orthogonal direction by a tensioning system. In this mode, the acceleration in the two first directions is determined in the normal fashion, and cross correlation in the 2 optical fibres can be used to resolve acceleration in the third substantially orthogonal direction.

[013] At its most basic, the accelerometer comprises a 2-Dimensional acceleromoter, having at least two taut optical fibres in a substantially orthogonal arrangement affixed through a mass and suspending the mass from a rigid frame. Each optical fibre has two fibre Bragg gratings (FGBs) inscribed thereon, one on each side of the mass, and the FGBs on each optical fibre are apodized, so they define a first centre wavelength and a second centre wavelength distinct from the first such that a partially overlapping interference pattern results from reflection of an input laser at the dual FGBs. Acceleration of the mass in the direction of each optical fibre can be measured as changes in intensity of the partially overlapping interference pattern using a photo detection circuitry according to the Intensity Modulation-Direct Detection (IM-DD) technique. Optional elements can be added, including: reclosable clamps on the rigid frame to receive the optical fibres through the rigid frame; a tensioning attachment system for affixing and simultaneously tensioning the optical cables to the rigid frame; a third optical fibre to provide 3-D measuring; the ability to gain additional information by analysing additional information in the light signal beyond the IM-DD technique in the so-called common mode of operation.

[014] A sensor of the type disclosed herein may find use in: land based and subsea petrochemical exploration & production, for seismic reflection surveys and 4-D seismic surveys; geological exploration & research, for seismic reflection surveys, seismic monitoring and gravimetric studies; construction, for vibration monitoring; ocean technology, for ocean bottom seismometers and inertial navigation and guidance systems for AUVS; and defense and security applications, for inertial navigation and guidance systems for aircraft, UAVs, etc. [015] Certain implementations of the accelerometer will now be described by example only, in greater detail with reference to the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

[016] Certain embodiments will be described in relation to the drawings in which: [017] Figure 1 is a partially exposed perspective view of an example of the three-axis suspended proof mass, tensioned string accelerometer of the prior art.

[018] Figure 2 is a diagram showing the detection principles of suspending a mass using FGB fibres. [019] Figure 3 is a set of diagrams and charts showing how relative motion of the inertial/seismic mass of Figure 2 affects reflectivity of light in the FBG on the upper and lower optical fibres.

[020] Figure 4 is a perspective view of a first embodiment of a 3-Dimensional dual FGB accelerometer implemented using 6 suspending optical fibres in 3 dimensions affixed to a rigid frame by a connecting and torsioning mechanical attachment in 6 positions on the frame.

[021] Figure 5 is a perspective view drawing of the pre-tensioning system used to tune the FBGs in the dual FBG accelerometer of Figure 4.

[022] Figure 6 is a block diagram side view of the pre-tensioning system is shown in

Figure 5 in use. [023] Figure 7 is a perspective view of a test device used to test the performance of the

3-Dimensional accelerometer, by causing known accelerations in one of the directions.

[024] Figure 8 are graphs showing signal output in volts over acceleration and non- linearity in volts over acceleration, for a static test.

[025] Figure 9 are graphs showing simulated results of reflected power for an idealized 3-Dimensional FGB accelerometer of Figure 4.

[026] Figure 10 is a graphical comparison of simulated acceleration curves derived in the manner of Figure 9 compared to actual data from the 3D accelerometer of the type shown in Figure 4, using the test platform of Figure 7. [027] Figure 1 1 is a side view of a diagram of a 1 -Dimensional accelerometer in which the optical fibre is tensioned in a substantially orthogonal direction.

[028] Figure 12 is a perspective view of a diagram of 2-Dimensional FBG accelerometer using 2 optical fibres in an orthogonal configuration pre-tensioned in a third substantially orthogonal direction by a tensioning element.

[029] Figure 13 is a perspective view of a 2-Dimensional FBG accelerometer using 3 optical fibres connected by a coupler in a symmetrical configuration pre-tensioned in a substantially orthogonal direction by a tensioning element.

[030] Figure 14 is a diagram of how the 3D dual FBG accelerometer is connected to the laser light source and control software (possibly on a PC).

DETAILED DESCRIPTION

[031] To illustrate the principle of operation of the accelerometer, consider a single axis system of Figure 2. The basic 1-D system consists of a seismic/test/inertial 44 mass, an optical fibre 41 passing through the mass 44 and a rigid frame & tensioning structure 45. The mass 44 is fixed exactly in the centre of two FBGs 42 and 43, possibly with an adhesive, clamp or other suitable means. The optical fibre is pre-tensioned and attached to the rigid frame with a clamping system. Fibre Bragg grating 42 and 43 on the optical fibre on both sides of the inertial mass 44, create an interference pattern in the input light which is disturbed by tensioning of the fibre under an acceleration of the inertial mass. The disturbance on both sides of the inertial mass can be resolved as acceleration in the direction of the fibre.

[032] Figure 3 shows the dual FBG accelerometer of Figure 2 with the two apodized

FBGs imprinted on each single-mode fibre with center wavelengths of 1550nm and 1550.4nm (for instance). The opposing strain changes of relative compression (relaxation from a pretension point) on one fibre and corresponding extension of the opposing fibre on the other side of the inertial mass, cause a perceived interference pattern of overlapping regions of light intensity, which grow or shrink depending on the direction of mass displacement. This, in turn, causes the intensity of the reflected light power to be modulated and directly detected using a simple photo detection circuitry. This is called the Intensity Modulation-Direct Detection (IM-DD) technique.

[033] Once attached, the optical fibre is tensioned within the housing to remove any slack from the optical fibre and firmly couple the seismic mass/fibre system. Any acceleration induced displacements along the axis of the fibre will cause opposing strain changes in the two FBGs. When a mechanical strain is placed on an FBG the reflected (and corresponding transmitted) Bragg wavelength in the optical domain will shift. Positive and negative strains cause opposite shifts in the reflected/transmitted Bragg wavelengths from the zero strain wavelengths. This, in turn, implies that in the optical domain the reflected Bragg wavelengths of the two FBGs will shift in opposite directions as shown in Figure 3.

[034] The wavelengths of the two FBGs can be designed and manufactured in such a way that they partially overlap in zero acceleration. In the example discussed and tested herein, two apodized FBGs were imprinted on a single-mode fibre with center wavelengths of 1550nm and 1550.4nm. The opposing strain changes cause the overlapped region to grow or shrink depending on the direction of mass displacement. This, in turn, causes the intensity of the reflected light power to be modulated and directly detected using a simple photo detection circuitry. This is called the Intensity Modulation-Direct Detection (IM-DD) technique.

[035] In order to create a first practical embodiment, the sensor is connected in 3- dimensions as shown in Figure 4. The mass 54 is supported along the three orthogonal axes by the optical fibres 53 with gratings (not shown) on either side of the mass 53 in each of the six directions. The mass 54 itself has holes in 3 orthogonal directions through its centre, which may be bored in, or created by assembly. Each of three fibres, having been appropriately scribed with FBGs, are passed through separate holes in the interial mass, and through the applicable holes on the rigid frame, and then tensioned (possibly as shown in Figure 5 and Figure 6). Once tensioned, the clamps 51 on the rigid frame 52 are locked, and then the fibres 53 bonded or clamped to the inertial mass 54. In this fashion, symmetrical positioning of the mass 54 within the frame 52 is preserved. The rigid frame comprises the necessary optical sensors and adjustable torsion connections 51 where the FBG inscribed fibres 53 connect to the frame 52. In Figure 4, optional openings 55 in the frame 52, allow the pre-tensioning and assembly process to be observed, but are not essential, and as shown in the view of Figure 5, may be covered over to protect the mass 52. [036] The proposed accelerometer uses the novel integration of a two-grating / dual- grating design and optical signal interrogation (IM-DD) technique, in three dimensions. Each grating in the sensor not only acts as a sensing element, but also automatically interrogates the induced optical signal changes in a simple but effective way. The two-grating design combined with IM-DD method provides a solution to compensate for the inherent temperature-sensitive nature of FBGs and eliminates additional signal interrogation devices. This reduces the size and cost of the entire sensing system and prevents unforeseen complications during system assembly and packaging.

[037] Another feature is the optional addition of an integrated fibre clamping mechanism. By integrating the fibre clamping mechanism to the mechanical transducer, the fibre pre-tensioning can be performed after the fibre has been bonded or affixed to the seismic mass. This makes the fibre bonding process much simpler and eliminates any variation in the fibre tension from unit to unit. Such variations could have a significant impact on sensor performance. As well, pre-tension on the fibre can be re-adjusted at any time. This clamping system also improves the mechanical coupling between the frame and the optical fibre. [038] Figure 5 and Figure 6 show a pre-tensioning system which may be employed with the 3D dual FBG accelerometer. In Figure 5, the 3D dual FBG accelerometer has mechanical clamps 91 to readjustably attach the optical fibres 92 to the frame 90. The tensioning device features a rigid base 80 and orthogonal pulleys 81. As shown in the side view block diagram of Figure 6, the pre-tensioning system, shown by the pulleys 81 , clamps 91 and 82 weights 83 for the fibres 92, is used to apply a small (~200C^strain) tension to the optical fibre once the mass has been attached and is suspended in the 3-D frame 90. The system illustrated comprises 4 main components: (1) a low friction, large diameter pulley 81 mounted to the stable mounting plate 80 so that the pulley groove is at the height of the fibres 72 exiting the clamps 91 of the 3-D frame 90; (2) a fibre clamp 82 at the end of the fibre 92 to be tensioned; (3) a known mass 83 to attach to the clamp 82 so as to exert the correct pre-tension force on the fibre 92; and (4) a stable mounting plate 80 for the 3-D frame 90 and pulleys 81. The accelerometer may be rotated and placed into the mounting plate 80 to pre-tension each of the fibres 92. The pre-tension is applied by suspending the known mass 83 from the fibre clamp 82 while the fiber clamp 91 integral to the 3-D seismometer frame 90 has been loosened (nearest the pulley 81). Once the mass 83 is suspended the clamp 91 on the 3-D frame 90 is tightened to 'lock-in' the applied pre-tension. This system was developed so that all three axes could be pre- tensioned simultaneously to eliminate pre-tensioning errors. The set up in Figure 6 shows a single pulley representing the two pulleys of Figure 5 in the 2 horizontal directions, and the vertical axis would be tensioned via a mass suspended through a hole directly below the mounting plate 80. This system would also allow for re-tensioning if necessary.

[039] By expanding the single axis design into a compact multi-axes accelerometer with a common inertial mass, the proposed accelerometer overcomes the need for additional support structures for the inertial/seismic mass required in other one or two dimensional designs. The three orthogonal pairs of FBGs to give more robust 3-axis measurements and simultaneously provide the required support for the inertial mass.

[040] Dynamic and static testing was performed. A dynamic testing platform is shown in Figure 7, in which a motor 71 drives rotor to linear conversion stages 72 and a linear motion test platform 73.

[041] For testing purposes, the elements are connected as shown in Figure 14, and according to the following parameters, which were used in testing but not meant to limit the scope of invention. The central waves of the two gratings on each optical fibre are set at 1549.936 nm and 1550.368 nm, respectively. The half width of the overlapped spectrum is chosen equal to 0.615 nm. The grating length of each Bragg grating is chosen to be 10 nm. The space between the two gratings is chosen to be 40 mm. There is no recoating surrounding the bare fibre in order to reduce the creeping effect of the polymeric recoating in the grating region. The 3D dual FBG accelerometer is connected to an Amplified Spontaneous Emission (ASE) broadband laser source with gain flatten filter (GFF) in the wavelength range from 1535 nm to 1570 nm, in order to maintain the reflected laser intensity of individual gratings during the wavelength shift. In this example, a 2x2 50:50 optical fibre coupler is used to split the laser which goes into and is reflected back from the FBG. An optical isolator is inserted between the laser source and the optical coupler. In this example, a germanium photo-detector of 0.85 A/W output current at 1550nm with a tunable gain trans-impedance amplifier converts the incident light power into voltage changes. In this example, the housing for the accelerometer is a 75 mm x 75 mm x 75 mm aluminum frame cube, and the proof mass is an 11 mm diameter hollow alloy sphere, with 3 mm orthogonal bore holes for the insertion of holders; the holders having through holes matched to the diameter of the fibre.

[042] Static calibration was performed on a vertically mounted turn (indexing) table that allowed for accurate angular positioning. A cosine(cp) component of the gravity vector, g, is the actual acceleration the measured/primary axis is facing. By turning the primary axis from +Z to - Z direction, i.e. by 180 degrees, we can change the acceleration from +g to -g. The data is recorded per 3 degrees with the sampling rate of 10 milliseconds and the sampling time of 30 seconds (61 sample points, 3,000 samples per sample point). When the data is collected from +g to -g, the rotor comes back to start a new test. The test is repeated 10 times for a large samples pool (30,000 samples per test point) to do the statistic analysis of the static test. Figure 8 illustrates the static test result and the linear fitting curve. The data is processed and plotted by MATLAB™ software. For this case, the reflected laser power at zero acceleration is compensated as an offset. The candle bar is used here to describe the standard deviation (STD) and maximum/minimum deviation of the static test. The residual from the fitting curve is also plotted in lower graph of Figure 8.

[043] It can be seen from Figure 8 that the non-offset output voltage of the photo- detector is linear to the acceleration from -g to +g. At the same time, a good agreement between the experimental result and the theoretical result in Figure 9 is achieved. [044] Then, a series of simple dynamic tests using the dynamic test platform of Figure

7 were performed. A flywheel 72 with a tunable rotation radius acts as the key role of the dynamic test platform of Figure 7 because it can not only generate a regular (sinusoidal) acceleration, but also stabilize the rotation. A motor 71 is used to drive the flywheel and the whole test platform 73. The acceleration of the test target is equal to the second derivative of its displacement.

[045] Figure 10 compares the simulated results (solid lines) with the corresponding measured results (symbols) for different magnitudes of horizontal displacement from 5.5 mm to 210 mm, in each case over accelerating input frequencies from about 0.05 Hz to 1.2 Hz. Again there is good agreement between the output of the 3D dual FBG accelerometer and the expected acceleration of the platform based on the theoretical calculations. [046] Figure 1 1 , Figure 12 and Figure 13 each show alternative configurations of the accelerometer disclosed herein, in which one dimension is given over to a tensioning element to automatically tension the optical fibres, which tend to elongate over time.

[047] The X-Z planar arrangement shown in Figure 1 1 demonstrates the concept of a proof mass 113 being tensioned in Z-axis using a mechanical spring 114 and a set screw 115, and in the X-axis by optical fibre 116 attached, to the rigid frame 117. Here the differential signal provided by the IM-DD method is proportional to the tension in FBG2 112 minus the tension in FBG1 111 , and thus the force due acceleration along the positive X axis is F(x)=(ma) I cos(ct), where (m) is the weight of the proof mass, (a) is the acceleration, and (a) is the angle between the fibre suspending the proof mass and the X-axis. If an extensive creep (plastic deformation) stretches the optical fibre (due to perhaps to elevated temperatures or long term operation) the spring will maintain the system pre-tensioned for proper operation. In cases where it is also desirable to measure the Z-axis component of the acceleration, a common mode signal will need to be employed, i.e. the tension in FBG1 111 plus the tension in FBG2 112 corresponds to the force due to acceleration along the negative Z-axis as F(z) = (ma) with the constant force of spring (114) subtracted and the result scaled by 1/sin(ct). This particular arrangement will require the signal from both FBG's (111 , 112) to be demodulated separately/individually; and is more computationally intensive than the IM-DD technique. The open end of the optical fibre is shown on the right, and on the left, an absorber 110 absorbs light which is not reflected back for the interference pattern of the FBGs 111 and 112.

[048] Figure 12 extends the principals of Figure 1 1 to three dimensions X-Y-Z, in which acceleration in the X-Y plane can be resolved using the differential principle (IM-DD). Optical fibre 123 connected to the frame 120 has a first FGB 124, passes through the mass 122, has a second FGB 125 and connects again to the rigid frame at the absorber 126. Similarly, optical fibre 127 connected to the frame 120 has a first FGB 128, passes through the mass 122, has a second FGB 129 and connects again to the rigid frame at the absorber 130. The tensioning spring 121 maintains keeps the optical fibres 123 and 127 taut.

[049] Figure 13 is yet another example of a possible configuration of a system implementing multiple dual FBG accelerometers, and is a variation on is a variant of Figure 12 using only three arms of the optical fibre to suspend the proof-mass. In this example, a 3dB 2x1 coupler 135 is used inside of the proof mass 132 to couple input optical fibre 133 having FGB 134 with each of the dead end optical fibres 137 (having FGB 136) and 139 (having FGB 138). The FBG pairs 134-136 and 134-138 are set apart from each other in the frequency domain to discriminate the two differential signals in the reflected spectra. As in Figure 11 and Figure 12, a tensioning element 131 pulls the mass 132 in the vertical direction to draw the optical fibres taut.

[050] Figure 14 shows the generalized interconnection of parts to create the system employing the present accelerometer element. A broadband laser source 151 is connected by optical fibre 152 to an optical isolator 153 (to reject reflected light from the FGBs and isolate the laser source), and then to an optical coupler 154. In one path, the optical coupler is connected to an input optical fibre 155 of an accelerometer 158 of the type disclosed herein, in which first FBG 156 is before the mass 157, and second FBG 159 is after the mass 157. Along the other path, the reflected light from the dual FGBs is detected by the photodector 160, amplified by the trans-impedance amplifier 161 and interpreted by control software implementing, at least, the IM-DD technique so as to measure the acceleration in the direction of fibre 155 within the accelerometer 158.

[051] It would be readily apparent to a person of skill in the art that the above procedures may be configured with minor adjustments to either the frequency of the grating, the use of a plurality of grating frequencies, or the use of an alternative means of affixing the fibre to either the seismic mass or the frame.

[052] The foregoing embodiments and advantages are merely exemplary and are not to be construed as limiting the present invention. The present teaching can be readily applied to other types of apparatuses. Also, the description of the embodiments of the present invention is intended to be illustrative, and not to limit the scope of the claims, and many alternatives, modifications, and variations will be apparent to those skilled in the art.