Patent Application
20080204270
This invention is directed toward measurements made within a borehole, and more particularly directed toward a system for minimizing adverse effects of pulse reflections in mud pulse telemetry systems used in measurement-while-drilling (MWD) or logging-while-drilling (LWD) systems.
Measurements of various properties of earth formation penetrated by a well borehole are widely used in hydrocarbon and mineral exploration and production. In addition, measurements of borehole parameters are typically used in defining a path of the borehole, optimizing production from the borehole, and maintaining borehole integrity during and after drilling.
Many types of systems have been used to measure properties and parameters of earth formation penetrated by a borehole, as well as to measure properties and parameters related to the borehole itself. The first systems used downhole instruments or “tools” comprising one or more sensors. The tools were conveyed along the borehole by means of a “wireline” cable. In addition to a conveyance means, the wireline served as a communication conduit between the one or more sensors disposed in the downhole tool and equipment at the surface of the earth. The “surface equipment” typically processed measured sensor data to obtain parameters of interest as a function of depth within the borehole. These measurements as a function of depth, using wireline conveyance, are commonly referred to as “wireline well logs” or simply “wireline logs”. Logs relating to formation properties include measures of formation natural gamma radiation, thermal neutron flux, epithermal neutron flux, elastic and inelastically scattered neutron flux, capture gamma radiation, scattered gamma radiation, and the like. A variety of formation parameters are obtained from these measurements, or combinations of these measurements, such as shale content, porosity, density, lithology and hydrocarbon saturation. Logs relating to borehole properties include temperature, pressure, caliper, orientation and the like.
Wireline logging is applicable only after the borehole has been drilled. It was recognized in the 1960's that certain operational and economic advantages could be realized if formation and borehole properties measurements could be made while the borehole is being drilled. One or more sensors responsive to formation and borehole parameters are typically disposed near the lower end of a drill string and preferably within a drill collar. Systems for obtaining real time measures of borehole and drilling related parameters, as a function of depth, are generally referred to as measurement-while-drilling (MWD) systems. MWD systems measure properties and parameters such as weight on the drill bit, borehole orientation, and the like. Systems for obtaining real time measures of formation properties, as a function of depth, are generally referred to as logging-while-drilling (LWD) systems. LWD systems measure formation properties such as density, porosity, hydrocarbon saturation, permeability, and the like. The LWD measurements should conceptually be more accurate than their wireline counterparts. This is because the formation is less perturbed in the immediate vicinity of the borehole by the invasion of drilling fluids into the formation. This invasion alters the virgin state of the formation. This effect is particular detrimental to the more shallow depth of investigation measurements such as nuclear logging measurements.
As mentioned previously, the cable of a wireline system serves as a means for telemetering data between one or more sensors disposed in the tool and the surface of the earth. Wireline data transmission rates can be quite high. MWD and LWD systems can not rely on a cable as a means for telemetering sensor data to the surface. This is, of course, prevented by the rotation of the drill string used to convey LWD and MWD tools while drilling. One technique for transporting MWD and LWD to the surface is to record sensor response in the downhole tool, and to subsequently retrieve these data when the borehole assembly is returned to the surface of the earth or “tripped”. This method eliminates the observation of data at the surface in real time. Other types of MWD and LWD telemetry have been used to telemeter tool sensor data, in real time, to the surface of the earth. One type of telemetry is based upon electromagnetic transmission between the downhole tool and the surface. An example of an electromagnetic telemetry system is disclosed in U.S. Pat. No. 7,145,473. A second type telemetry is based upon acoustic transmission between the downhole tool and the surface using the drill string. An example of an acoustic telemetry system is disclosed in U.S. Pat. No. 6,847,585. A third type telemetry uses modulated pressure variations induced in the drilling fluid as a means for transmitting tool sensor data to the surface. Mud pulse telemetry will be discussed in detain in the following sections.
A borehole drilling operation typically uses drilling fluid commonly referred to as drilling “mud”. Drilling mud is pumped down the drill string, exits through ports in the drill bit, and returns to the surface via the drill string-borehole annulus. The mud is cleaned at the surface and recirculated throughout the drilling operation. During the drilling operation, the mud serves as a lubricant for the drill bit, as a means for returning drill bit cuttings to the surface, and as a means for provide pressure balance within the borehole.
As mentioned above, the drilling mud can function as a telemetry link between one or more sensors in the borehole tool and the surface of the earth. Mud telemetry systems are broadly classified as either mud pulse systems or continuous wave systems.
Mud pulse telemetry systems use valving to momentarily perturb mud flow through the downhole tool. This perturbation generates positive or negative pulses in the mud column within the borehole. The pulses are modulated to represent the response of a tool sensor. The modulated pulse sequence travels via the mud column to the surface where it is sensed by a pressure transducer. The output of the pressure transducer is demodulated thereby yielding a signal representative of the response of the downhole tool sensor. An example of a mud pulse telemetry system is disclosed in U.S. Pat. No. 5,586,084, which is incorporated into this disclosure by reference.
Continuous wave telemetry systems use a rotary valve or “mud siren” which restricts mud flow through the downhole tool to generate a positive pressure wave. As in the mud pulse system, the pressure wave is modulated to represent the response of a tool sensor, travels to the surface via the mud column, is sensed by a pressure transducer at the surface, and the output of the transducer is demodulated again yielding a signal representative of the response of the tool sensor. An example of a continuous wave telemetry system is disclosed in U.S. Pat. No. 3,309,656.
Using a mud pulse system as an example, pulse amplitude and frequency can be modulated to represent sensor response data. The pulse sequence degrades as it travels up the mud column to the surface. Attenuation, distortion and reflection cause this degradation. It is desirable to maximize data rates in order to maximize accuracy and precision of the data being telemetered. As data transmission frequency increases, the period for each individual pulse proportionally decreases. Data transmission rates are, therefore, limited by resolution requirements to separate and demodulate pulses at the surface. Pulse sequence degradation imposes further limitations on the period for each individual pulse thereby further reducing practical data transmission rates. As a result of all of these factors, typical mud pulse data transmission rates are limited to several pulses per second.
The present invention is directed toward optimizing mud pulse telemetry data rate by processing a sequence of pulses, measured at the surface, in order to remove or at least minimize adverse effects of pressure pulse reflections. It is also important to note that these reflections can, and usually do, undergo attenuations and distortions. The process of minimizing adverse effects of pulse reflections, which also incorporates attenuation and distortion of these reflections, will be referred to as a pulse sequence filter. A downhole telemetry unit disposed in a MWD or LWD tool generates a sequence of pulse within the mud column. The sequence is encoder to represent the response of one or more sensors disposed within the logging tool. The encoded pulse sequence is sensed at the surface of the earth using preferably a pressure transducer disposed in the standpipe or swivel of a drilling rig. The output of the transducer is an electrical signal representative of the pulse sequence that is typically attenuated and can contain reflections and distortions. This measured signal output is processed to minimize the effects of pulse reflections, considering any attenuation and distortion of the pulse reflections, thereby yielding a “primary” pulse sequence. In the context of this disclosure, the terms “primary” pulse or “primary” pulse sequence refer to pulses that are measured preferably at the surface of the earth and that have been mathematically processed by pulse sequence filtering so that they closely represent pulses or pulse sequences generated by the downhole telemetry unit. Stated another way, “primary” pulses are pulses that are measured at the surface and have been mathematically processed to minimize perturbing factors associated with pulse reflections. With these perturbing factors minimized, the pulse period can be reduced while still maintaining the desired pulse resolution needed for pulse sequence demodulation. This, in turn, allows the data transmission rate to be increased.
So that the manner in which the above recited features, advantages and objects the present invention are obtained and can be understood in detail, more particular description of the invention, briefly summarized above, may be had by reference to the embodiments thereof which are illustrated in the appended drawings.
a is an actual mud pulse sequence that exhibits reflections; and
b shows the same pulse sequence after pulse sequence filtering illustrating the removal of the reflections.
The present invention is a system for optimizing data rate from LWD or MWD logging systems using pulse sequence filtering of mud pulse telemetry. Although LWD and MWD logging systems have different connotations in the art, the invention will de disclosed in the context of a MWD logging system. Its should be understood that the invention is equally applicable to LWD logging systems.
Again referring to
Still referring to
Again referring to
As mentioned previously, mud pulse reflections can severely inhibit pulse detection, or at least force the telemetry data rate to decrease below desirable limits. There are two primary problems associated with reflection interference. The first problem is caused by an opposite sign reflection. An example would be a negative reflection in a sequence of positive primary mud pulses. The second problem is false pulse detection caused by a same sign reflection. An example would be a positive reflection in a sequence of positive mud pulses, where the amplitude of the reflected pulse is comparable to the amplitude of unreflected “primary” pulses. Reflection removal can help to obtain faster data rates. If primary pulses can be placed closer together (i.e. the pulse period T reduced), without interference from reflections, then data rate can be substantially increased. Stated another way, minimization of the pulse period T tends to maximize the data rate as long as resolution of individual pulses in the measured pulse sequence can be maintained. Reflected pulses can also be attenuated and distorted. Attenuation and distortion of reflected pulses must be considered in correcting for pulse reflections.
The adverse effects of pulse reflection are minimized using pulse sequence filtering. The following section sets forth mathematical algorithms used in pulse sequence filtering. It should be understood that the mathematical formalism can be varied while still performing the same mathematical functions and maintaining the desired results of the pulse sequence filtering concept.
Referring again to
s(t)=g(t)+cg(t−a). (1)
where:
t=time
s(t)=the measured signal from the response of the surface transducer;
g(t)=the primary pulse corrected for reflection and incorporating reflection attenuation;
g(t−a)=a reflection of the primary pulse delayed by a pulse sequence delay time parameter a; and
c=a pulse sequence parameter that is an amplitude multiplier to account for attenuation and/or sign reversal in the pulse reflection.
Equation (1) is more easily solved in the frequency domain. Equation (1) transformed into the frequency domain is
S(w)=G(w)+cG(w)eiaw. (2)
where:
w=frequency
S(w)=the measured signal in the frequency domain;
G(w)=the primary pulse in the frequency domain; and
a and c are corresponding delay and amplitude terms defined above.
Solving equation (2) for G(w) yields
G(w)=S(w)/(1+ce−iaw). (3)
The term G(w) can be determined since everything on the right side of the equation is measured, known or can be calculated. More specifically, term S(w) is the Fast Fourier Transform of the measured signal, c is a measured or known reflected pulse attenuation factor and a is a measured or known delay of the reflection. An example of s(t) as a measured pulse sequence, from which measured pulse sequence parameters a and c can be determined, is illustrated in
Equation (1) is applicable to a single reflector. In practice, the sequence of telemetered pulses can be adversely affected by a plurality of reflections. This situation is represented mathematically by expanding equation (1) as
s(t)=g(t)+c1g(t−a1)+c2g(t−a2)+ . . . +cng(t−an) (4)
where:
t=time
s(t)=the measured signal from the response of the surface transducer;
g(t)=the primary pulse corrected for reflection and attenuation;
g(t−aj)=a reflection j of the primary pulse delayed by time aj; and
cj=an amplitude multiplier to account for attenuation and/or sign reversal in the reflection j; where
j=1, 2, . . . , n; and
n=the number of reflections.
It is also important to expand equation (1) to include the possibility of distortion in a reflection. A distorted reflection is expressed mathematically in the time domain as
s(t)=g(t)+d(t)g(t−a) (5)
where:
t=time
s(t)=the measured signal from the response of the surface transducer;
g(t)=the primary pulse corrected for reflection and attenuation;
g(t−a)=a reflection of the primary pulse delayed by time a; and
d(t)=a measured pulse sequence parameter indicative of the distortion of the pulse reflection.
Expressing equation (5) in the frequency domain and again solving for G(w) yields
G(w)=S(w)/(1+D(w)e−iaw). (6)
As in equation (3), everything on the right side of the equation (6) is measured, known or can be calculated. More specifically, term S(w) is again the Fast Fourier Transform of the measured signal, D(w) is a measured or known distortion term, and a is a measured or known delay of the reflection. As an example, the distortion term d(t) in the time domain, and thus the distortion term D(w) in the frequency domain, can be determined from the measure of a sequence s(t) in the same manner as the previously discussed terms a and c. The primary pulse sequence g(t) of equation (1), which is the pulse parameter of interest, is obtained by performing a Reverse Fast Fourier Transform of G(w) obtained from equation (6). Physically, g(t) is the measured mud pulse sequence corrected for the adverse effects of a distortion and a reflection delay a. Once again, the sequence of corrected mud pulses g(t) is then demodulated in the processor disposed in the surface equipment 64 to obtain a signal representative of the response of a sensor 14 disposed in a MWD logging tool 10. As discussed above, the demodulated signal is again converted to a parameter of interest, preferably in the processor, and output to a recorder 66 and preferably tabulated as a function of depth thereby forming a log 68.
Pulse sequence filtering for cancellation of pulse reflection is presented as a functional flow diagram in
a is a plot of a measured signal s(t) from a pressure transducer as a function of time t. The curve 94 illustrates pairs of primary and reflected pulses denoted by the excursions 97. As an example, a primary pulse g(t) occurs at a time t denoted as 93, and a reflected pulse occurs at a time t denoted as 99. The primary and reflected pulses are separated by the previously defined delay time interval a. As discussed previously, the delay time interval a can be obtained from the measured signal sequence shown in
b is a plot of a primary pulse sequence g(t) as a function of time t obtained by pulse sequence filtering the measured pulse sequence s(t) shown in
While the foregoing disclosure is directed toward the preferred embodiments of the invention, the scope of the invention is defined by the claims, which follow.
