Patent Application
20140279772
Earth formations may be used for various purposes such as hydrocarbon or geothermal production or carbon dioxide sequestration. Boreholes drilled into the earth provide access to the formations. In order to drill a borehole or survey a formation, one or more downhole tools may be conveyed through a borehole penetrating the formation while the borehole is being drilled or through a previously drilled borehole. A downhole tool may contain one or more actuators that need to be controlled by a controller based on inputs received from one or more sensors. Unfortunately, controllers disposed downhole face some challenges. The space available in a downhole tool for the controller and, thus, the complexity of controlling may be limited by the size of the borehole. In addition, the downhole conditions may be extreme both environmentally and from noise and, consequentially, pose reliability concerns. Hence, it would be well received in the drilling and geophysical exploration industries if a downhole controller could be developed that would increase the controlling power and robustness available.
Disclosed is an apparatus for processing signals downhole. The apparatus includes: a carrier configured to be conveyed through a borehole penetrating an earth formation; a container disposed at the carrier and configured to carry biological material; a cultured biological neural network disposed at the container, the neural network being capable of processing a network input signal and providing a processed network output signal; and one or more electrodes in electrical communication with the neural network, the one or more electrodes being configured to communicate the network input signal into the neural network and to communicate the network output signal out of the neural network.
Also disclosed is a method for processing signals downhole. The method includes: conveying a carrier through a borehole; receiving a network input signal using one or more electrodes coupled to a cultured biological neural network; processing the network input signal using the neural network to provide a processed network output signal; and outputting the network output signal using one or more electrodes coupled to the neural network.
The following descriptions should not be considered limiting in any way. With reference to the accompanying drawings, like elements are numbered alike:
A detailed description of one or more embodiments of the disclosed apparatus and method presented herein by way of exemplification and not limitation with reference to the figures.
Disclosed are apparatus and method for processing data downhole. The apparatus includes a cultured biological neural (or neuronal) network that can process one or more inputs and provide one or more outputs based on the processing of the one or more inputs. Electrical stimuli are applied to the neural network via one or more electrodes in order to train the network to respond in a desired manner. The neurons in the network form neural connections from the training that result in processing inputs in the desired manner, which may be viewed as a processing algorithm. Electrical inputs, such as from a sensor, are input into the neural network via the one or more electrodes and the network processes the inputs according to the training received by the network. One or more outputs resulting from the processing are provided via the one or more electrodes or other electrodes and may be used to control a device, such as an actuator, may be recorded for future use, or may be transmitted for use by another device.
The downhole tool 10 is configured to sense a parameter of interest using a sensor 13. A controller 14 is configured to receive a sensor signal 15 that conveys parameter measurements from the sensor 13. Non-limiting embodiments of the sensor 13 are a pressure sensor, temperature sensor, orientation sensor, direction sensor, pH sensor, photodetector, chemical sensor, radiation detector (alpha, beta, gamma, or, neutron), spectrometer, acoustic sensor, seismic sensor, magnetic field sensor, electric field sensor, and antenna for receiving electromagnetic signals. The controller 14 is configured to implement an algorithm or procedure of interest that provides an output signal 16 based on the received parameter measurements. The output signal 16 is transmitted to an actuator 17 that is configured to perform a function based on the received output signal 16. The function may be related to a downhole activity such as steering the drill string 6 or performing other measurements with other sensors or analysis devices. Other functions that may be performed include recording an output signal, transmitting an output signal such as to another device or uphole to the processing system 12 using telemetry.
The algorithm or procedure implemented by the controller 14 is performed by a cultured biological neural network 18. In one or more non-limiting embodiments, the cultured biological neural network 18 includes rat cortical neurons, neurons of a lamprey, or other cultured biological neurons. The neural network 18 is a series of interconnected neurons, which when activated define a recognizable pathway. If the sum of input signals into one neuron exceeds a certain threshold, then that neuron sends an action potential to a neighboring interconnected neuron. An action potential (AP) of a neuron is a short-lasting event in which the electrical potential of the neuron rapidly rises and falls generally following a consistent trajectory. A temporal sequence of action potentials may be referred to as a spike train. Electrical signals are used to communicate with the neuron network 18. Non-limiting embodiments of the electrical signals include a voltage level and/or a frequency (such as a pulse-train frequency) that corresponds to a parameter value. In response to receiving an electrical signal, neurons in the neural network 18 will fire sending APs through the network. By sensing the APs, an output signal is provided by the neural network 18.
In order for the neural network to implement a selected algorithm or procedure, the neural network may be trained or taught using selected stimulation signals that result in the neural network providing the desired response. In one or more embodiments, the neural network may be mapped by applying various stimulus signals or combinations of stimulus signals to a multi-electrode array (MEA) that is coupled to the neural network and monitoring responses in the MEA to learn how the network operates and what neural connections result from the stimulation signals. Once the neural network is mapped, stimulus signals are applied to program or drive the network towards a desired response. Once, the neural network is programmed and operating as desired, further training stimulus signal are no longer necessary. Alternatively, literature describes other methods and procedures for training a biological neural network to achieve a desired result.
In one or more embodiments, a first one or more electrodes are used to input a signal into the neural network and a second one or more electrodes are used to receive a processed output signal. Electrodes may be common to the first one or more electrodes and the second one or more electrodes. A desired network input signal is input into the neural network and electrodes are monitored to detect and identify which electrodes output a corresponding desired output signal that corresponds to the selected algorithm. The above steps may be repeated using another (i.e., different) input signal. In this manner, several different input signals may be used to obtain desired responses at locations that correspond to the selected algorithm. It can be appreciated that using a greater number of electrodes increases the likelihood of achieving the desired response.
Reference may now be had to
An input interface 24 is coupled to one or more electrodes and is configured to convert received signals 25 (such as the sensor signal 15) into the network input signals 22 that are suitable for stimulating the neural network. Because the sensor signal 15 may supply an output signal as a voltage level that is not compatible with stimulating neurons, the input interface 24 converts the sensor signal 15 to a signal that is compatible to stimulating the neurons in the neuron network. Alternatively, if the sensor signal 15 is compatible with stimulating neurons directly in the neural network, then the input interface 24 may not be required.
An output interface 26 is coupled to one or more electrodes in the plurality and is configured to convert network output signals 27 into compatible output signals 28 (e.g., the output signals 16) that are compatible with performing desired functions downhole such as being recorded or transmitted or activating the actuator 17. In one or more embodiments, the output interface 26 includes an amplifier configured to amplify the network output signals 27 to a level that is compatible or suitable for being transmitted to other devices. Alternatively, if the network output signals 27 are compatible with a receiving device such as the actuator 17, then the network output signals 27 may be used directly as the output signal 16 or 28 and the output interface 26 may not be required.
Reference may now be had to
Reference may now be had to
It can be appreciated that the cultured biological neural network provides several advantages in a downhole environment. One advantage is that the biological neural network is robust both in computational power and against electrical noise, dysfunction or disruption. In addition, in that the biological neural network is pliable and has some elasticity, the neural network is also robust against vibration. Theoretically, the velocity of a “conventional” (i.e., electronic) computer (CC) should be superior to a cultured neuronal network by generally about 6 orders of magnitude because the response time/circuit time of the elements of a CC (e.g., transistors) is faster. However, the brain/connected neurons are computing massively in parallel (massive parallel computing). Most of the neurons are computing in parallel and are operating while in a CC normally most of the elements are passive during operation. While just a few transistors in a CC may be computing in an instant, in a brain or cultured neural network all or most neurons may be active at any instant to provide greater computational power.
Because the biological neural network has an ability to grow, the neural network has a self-healing or repair property, which can be useful when the network is disposed in a borehole as it can take a significant amount of time to remove a conventional controller/processor from the borehole. The “knowledge” in a biological neural network is distributed throughout the network and, thus, has fault tolerance or an ability to continue to operate with the loss or blackout of some neurons or a region of the neural network. Further, other new neurons may be replacing the lost neurons due to self-healing. In contrast, in a CC the blackout of some elements or algorithms can cause inoperability of the whole CC. Another advantage of the biological neural network is its ability to keep learning or being retrained downhole as conditions and requirements change.
In support of the teachings herein, various analysis components may be used, including a digital and/or an analog system. For example, the downhole electronics 11, the computer processing system 12, the sensor 13, the actuator 17, the input interface 24, or the output interface 26 may include digital and/or analog systems. The system may have components such as a processor, storage media, memory, input, output, communications link (wired, wireless, pulsed mud, optical or other), user interfaces, software programs, signal processors (digital or analog) and other such components (such as resistors, capacitors, inductors and others) to provide for operation and analyses of the apparatus and methods disclosed herein in any of several manners well-appreciated in the art. It is considered that these teachings may be, but need not be, implemented in conjunction with a set of computer executable instructions stored on a non-transitory computer readable medium, including memory (ROMs, RAMs), optical (CD-ROMs), or magnetic (disks, hard drives), or any other type that when executed causes a computer to implement the method of the present invention. These instructions may provide for equipment operation, control, data collection and analysis and other functions deemed relevant by a system designer, owner, user or other such personnel, in addition to the functions described in this disclosure.
The term “carrier” as used herein means any device, device component, combination of devices, media and/or member that may be used to convey, house, support or otherwise facilitate the use of another device, device component, combination of devices, media and/or member. Other exemplary non-limiting carriers include drill strings of the coiled tube type, of the jointed pipe type and any combination or portion thereof. Other carrier examples include casing pipes, wirelines, wireline sondes, slickline sondes, drop shots, bottom-hole-assemblies, drill string inserts, modules, internal housings and substrate portions thereof.
Elements of the embodiments have been introduced with either the articles “a” or “an.” The articles are intended to mean that there are one or more of the elements. The terms “including” and “having” are intended to be inclusive such that there may be additional elements other than the elements listed. The conjunction “or” when used with a list of at least two terms is intended to mean any term or combination of terms. The terms “first,” “second” and the like do not denote a particular order, but are used to distinguish different elements.
While one or more embodiments have been shown and described, modifications and substitutions may be made thereto without departing from the spirit and scope of the invention. Accordingly, it is to be understood that the present invention has been described by way of illustrations and not limitation.
It will be recognized that the various components or technologies may provide certain necessary or beneficial functionality or features. Accordingly, these functions and features as may be needed in support of the appended claims and variations thereof, are recognized as being inherently included as a part of the teachings herein and a part of the invention disclosed.
While the invention has been described with reference to exemplary embodiments, it will be understood that various changes may be made and equivalents may be substituted for elements thereof without departing from the scope of the invention. In addition, many modifications will be appreciated to adapt a particular instrument, situation or material to the teachings of the invention without departing from the essential scope thereof. Therefore, it is intended that the invention not be limited to the particular embodiment disclosed as the best mode contemplated for carrying out this invention, but that the invention will include all embodiments falling within the scope of the appended claims.
