Patent Application
20080190604
The present invention relates generally to sensing characteristics of operating well pumping systems, and, more particularly, to a system and method for implementing coordinated monitoring of one or more oil well pump systems.
An oil pump system is the primary equipment used to extract oil from an oil well. In a conventional oil pump system, the mechanical operation of the pump is monitored by a single sensing device, such as a load cell that is used to measure the load on the pump throughout the cycle of the pump crank.
In a typical operation, a control unit excites and measures the load cell via a load cell cable. As the control unit receives load measurements from the load cell, it correlates this data with the vertical position of the crank (which the control unit calculates from the speed and angular position of the motor) to determine if the measured load is within normal operating parameters. If not, then the control unit will adjust the speed of the motor in order to bring the load back to normal or otherwise shut down the pump system altogether.
The load cell cable is a high-maintenance item that frequently breaks and, when it fails, the pump must be shut down for repair. Additionally, because of frequent failures using a cable to communicate with the load cell, the present approach cannot be readily extended to other sensor modalities that may also be useful. Further, pumps are treated as independent systems even though pumping from a given well may actually employ many pumps. Thus, to achieve optimum pumping performance, all the pumps associated with given well should be viewed as an aggregate system.
Another problem associated with an oil pump system is pounding, which refers to the vibration that occurs when the pump's traveling valve contacts the surface of the oil in the well on the down stroke of the pump's crank. This phenomenon occurs when the oil being sucked up during an upstroke cannot fill the bore of the well as fast as the traveling valve moves upward. The optimum pump cycle is when the oil flow rate matches the rate at which the traveling valve rises on an upstroke. Excessive pounding can damage the pump system and reduce its overall life.
In view of the above, it would be desirable to be able to improve the reliability monitoring capability of individual pumps for well pumping systems such as oil wells, and in a manner that manages a number of parameters of an aggregate pump system, including for example, well yield, fault mitigation, and power consumption.
The foregoing discussed drawbacks and deficiencies of the prior art are overcome or alleviated by, in an exemplary embodiment, a pump monitoring and control system, including a plurality of pumps associated with a well; each of the plurality of pumps including one or more sensors associated therewith; and one or more communication devices linking each of the sensors with a well system controller; wherein the well system controller is configured to process information received from the one or more sensors and control any of the plurality of pumps based on information received from one or more of the plurality of pumps.
In another embodiment, a method of monitoring and controlling a pump system includes initializing, through a well system controller, a plurality of pumps associated with a well, the initializing including one or more of: powering on each of the pumps, configuring each of a plurality of pump controllers associated with the pumps, configuring each of a plurality of sensors associated with the pumps, wherein one or more communication devices link each of the sensors with a well system controller, and initiating a pumping operation with one or more of the pumps; entering an event-driven loop by polling for fault status events and internal timeout events, the fault status events received from the plurality of sensors from one or more communication devices; sending a data request to each of the plurality of pumps and waiting for responses therefrom for specified amount of time; determining, from data returned by the plurality of the pumps, desired operating parameters for each of the pumps; and issuing an update command, with new parameters, to each of the pumps and returning to the beginning of the event-driven loop.
Referring to the exemplary drawings wherein like elements are numbered alike in the several Figures:
Disclosed herein is a system and method for implementing coordinated monitoring of one or more well pump systems, thereby addressing problems such as cable maintenance and pounding associated with conventional systems, as well as providing enhanced monitoring and management capabilities. By using arrays of sensors, reliability is improved, and the overall monitoring of oil well pumping systems for individual pumps is enabled. In addition, by using wireless communications to network the sensor arrays and to provide feedback to a well controller, a system can manage a number of pertinent goals of an aggregate pump system, including for example, well yield, fault mitigation, and power consumption.
Referring initially to
As further illustrated in
As indicated above, however, the load cell cable 122 is a high-maintenance component that frequently breaks. Upon failure of the cable, the pump 156 must be shut down for repair. Moreover, due to frequent failures using a hardwired cable to the load cell, such as shown in
Accordingly,
As an additional benefit over existing solutions, having co-located sensor modalities lends itself to distributing some of the control logic to the localized data capture sites. For example, the logic that correlates the load with crank position may be embedded in a controller at the load cell site. It should be appreciated that the system and method embodiments described herein need not be limited to oil pumps. Rather, they may be applied to any type of pumping system that uses automated operation, such as a water pump, for example.
As particularly shown in
In contrast, when the embedded controller 220 detects a fault, it communicates the appropriate information to a control unit 234 for the motor 236 over a wireless link 230. The information communicated may be as simple, for example, as the type of event detected, or as extensive as the complete data history for some number crank cycles. This approach also facilitates software updates as well as the configuration of the embedded device over a wireless link 232 in the reverse direction (i.e., from the control unit 234 to the embedded controller 220. Data may also be transferred by wireless link to a control system for the pump. Alternatively, the processor unit may be placed on the sensor site on the pump, which may be on the sucker rod. The processor may also be placed anywhere else on the pump and may communicate commands to a control unit for the pump, by wireless means or wired means.
In a conventional system, the sensor measures a single parameter (e.g., the load on the load cell), which is transferred over a wire. However, with the wireless system such as shown in
The addition of the multi-parameter capability adds new capabilities to the system 200. For instance, if the vertical position of the crank 226 is measured with respect to the ground by means of a sonar-ranging sensor, then a gradual decrease in the maximum and minimum distance could indicate the depth of snowfall on the ground. On the other hand, the sensed distance to the ground may also indicate flooding by water or by oil resulting from a leak or rupture in a nearby oil storage tank. Thus, the addition of temperature and atmospheric pressure sensors could further provide data that may be used to indicate local weather at the site of the pump. Such data may relayed by an additional wireless link 280, e.g., through a cell phone, satellite phone, radio etc., or wired link to a remote location where it could be used as feedback in determining the schedule for the performance of maintenance at the pump site.
Vibration sensing data may also be used to indicate an incipient failure of the apparatus before the failure occurs. Thus, the pump operation could be modified or the pump could be shut down before the position vs. load data indicates that there is cause for concern. When information is transferred by wireless link to a remote location, additional functionalities may be obtained. For instance, if the system of pumps is supplied from a common electric source, the loads on the source may be balanced by adjusting the phases of the individual pumps so that they are not all functioning in phase. This may apply to pumps that are in continuous operation or to a collection or plurality of pumps where one or more of them are idle.
Additionally, the pumps may be also used to gather weather information, e.g., temperature, atmospheric pressure, humidity, wind velocity (speed and direction) in areas where there are not sufficient whether monitoring stations. It is also contemplated that information from vibration sensors may be used to chart the distribution and severity of earthquakes. Still other sensors may be used to determine the distance from the pump to the ground in order to measure the depth of water flooding over a region.
Referring now to
The sensors 350-359 may include for example load, range, vibration, acceleration, magnetic field, temperature, humidity, liquid level, and atmospheric pressure sensors. Additionally, each sensor may have an integrated communication and/or a computing means so that it can perform local analytics on the data and communicate status and results instead of raw measured data. Alternatively, one or more of the sensors may be a discrete device that is polled by a computing device via a wired or wireless link. Similarly, a communication means may also have one or more integrated sensors or computing means.
A pump controller 370, 371, . . . , 379 included within a corresponding pump system 340-349 controls the mechanical operation of the pump. A controller may also have one or more integrated sensors and communication means. The well control system 301 is a computing system that supervises each of the pump systems 340-349 operating in the well network 302. In one embodiment, the communication devices, sensors, and pump controllers depicted in
Proceeding from start block 405, the method waits for internal or external driving events in block 410 such as, for example, a timeout, a data request or update command. A periodic cycle is driven by timer events. At decision block 415, if an update command event is received, the method proceeds to block 420 to update parameters, looping back to block 410. On the other hand, if a timeout occurs the method proceed through the NO branch of decision block 415 to block 425, where sensor measurements from the various sensor modalities are taken and analyzed. The operational status of the pump system is the result of the analysis.
Then, at decision block 435, if no catastrophic failure has occurred, and if no other fault (decision block 445) is detected, then the cycle ends and the method proceeds back to block 410 to wait for the next event. However, if there is a catastrophic failure determined at decision block 435 (e.g., such as a broken traveling valve), then the method issues a shutdown command to the pump controller at block 440. The method then sends the operation status and other data to the well system controller, as shown at block 455. The cycle ends and the method proceeds back to block 410 to wait for the next event.
If there is no catastrophic failure at decision block 435, but some other fault is detected at decision block 445 (e.g., such as excessive pounding), then the method sends the operation status and other data to the well system controller at block 455. Again, the cycle ends and the method proceeds back to block 410 to wait for the next event. In an exemplary embodiment, taking and analyzing sensor measurements (block 425) is a multi-step process that may involve the use of multiple sensors at multiple discrete sites. For example, a combination of load cell and ranging sensors attached to a polished rod may be used to detect a broken or faulty traveling valve. As another example, a liquid level sensor in the well, combined with a vibration sensor attached to the sucker rod may be used to detect pounding. In essence, such combinations of sensors comprise fault-detection subsystems within a pump system and a pump system may have a plurality of such fault-detection subsystems.
Whenever a fault is detected (at decision blocks 435 or 445) or the well system controller requests data 450, the operational status of the pump, sensor data, and other pertinent information is fed back to the well system controller over the well network in block 455. This feedback means allows all of the pump systems for a given well to be monitored and managed as an aggregate system.
Upon receiving a fault status event, the well system controller will calculate the optimum parameters for each pump system in the well network, and then send an update command to each pump. When a pump system receives an update command (block 410), it will update the appropriate parameters (block 420) of the pump system. For example, if pounding is detected for one or more pump systems, the well system controller may calculate the optimum motor speed for each pump system, and then send an update command to change the parameter for motor speed in each pump controller.
Beginning from start block 505, the method proceeds to block 510 where the well system controller initializes all the pumps in its well network 510. The initialization process includes, for example, powering on all pumps, configuring all pump controllers, configuring all sensors, and starting the pumping operation. Once all pumps are operating, the well system controller enters an event-driven loop where it polls for fault status events from the well network or internal timeout events, as reflected in block 515.
Proceeding to decision block 520, if there is a status event that indicates a catastrophic failure, then the method issues a maintenance request at block 525. The method then sends a data request to all the pumps in the well network at block 530 and waits for the responses or a for specified amount of time at block 535. Using the data returned by the pumps in the well network, the method calculates the optimum parameters for each pump system as reflected at block 540. If, at decision block 545, the calculations show that one or more pumps need to be shut down for maintenance, then the method issues a maintenance request as shown at block 550. The appropriate update command, with new parameters, are then sent to each pump at block 555. The cycle ends and the method waits for the next event 515.
On the other hand, if there is a status event that is not a catastrophic failure or there is a timeout event (i.e., a NO determination at decision block 520), then the method sends a data request to all the pumps in the well network a block 530 and the method continues as described above at block 535.
More specifically, the operation(s) described in block 540 is/are a process that involves analytics to determine the optimum operational parameters for each pump by considering all the pumps as an aggregate system. The operational parameters of a pump may be generally expressed as the tuple {enable, speed, phase}, where enable represents the power on/off state of a pump, speed is the rotational speed of the pump's motor, and phase is the phase offset of a pump's cycle from a reference cycle.
These three parameters enable the management of a number of pertinent goals of the aggregate system, including well yield, fault mitigation, and power consumption. For example, given a target for well yield, it may be desirable to minimize the overall power consumption. This scenario may occur if a system of pumps is supplied from a common electric source. For an individual pump, the power consumption is a function of the motor speed and the motor's load (i.e., position of the crank). Therefore, for multiple pumps, the minimum power consumption can be achieved by finding the optimum phase and speed for each pump that minimizes the aggregate load on the source. This may apply to pumps that are in continuous operation or to a collection of pumps where one or more of them are idle.
As another example, given a target for well yield, it may be desirable to minimize pounding across the entire well system to extend service lifetime. In the conventional approach, pounding is typically mitigated by adjusting the motor speed on the pump exhibiting pounding, without regard for how adjustments to one pump may affect well yield or the performance of other pumps in the well network. Thus, the conventional approach is unable to determine an optimum speed adjustment. To achieve the well yield target, the optimum speed adjustment may not even involve the specific pump exhibiting the problem. Using the above described approach, the right combination of sensors for pounding detection (e.g., vibration and liquid level) in conjunction with the feedback to a well system controller, allows for consideration of well yield and overall system health during fault mitigation.
While the invention has been described with reference to a preferred embodiment or embodiments, it will be understood by those skilled in the art 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 may be made to adapt a particular 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.
