Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
ANALYZING A PUMP TO DETERMINE VALVE WEAR AND WASHOUT USING A SELF-ORGANIZING MAP
Document Type and Number:
WIPO Patent Application WO/2018/160174
Kind Code:
A1
Abstract:
Examples of techniques monitoring pump health are disclosed. In one example implementation, a method may include: collecting, by a processing device, pump data about the pump, wherein the pump data comprises vibration data, timing data, and pressure data collected from vibration sensors, a timing sensor, and pressure sensors respectively; conditioning, by the processing device, the pump data by dividing the pump data into a plurality of segments, wherein each of the plurality of segments represents one completion cycle of the pump; analyzing, by the processing device, the plurality of segments of pump data to determine a valve and valve seat wear state of the pump using a self-organizing map generated from historical pump data; and changing, by the processing device, an operational parameter of the pump responsive to determining that the valve and valve seat wear state of the pump represents a risk level that is above a threshold.

Inventors:
SINGH, Gulshan (17021 Aldine Westfield, Houston, TX, 77073, US)
JAEGER, Thomas (17021 Aldine Westfield, Houston, TX, 77073, US)
GAUGHAN, Kevin (17021 Aldine Westfield, Houston, TX, 77073, US)
Application Number:
US2017/020075
Publication Date:
September 07, 2018
Filing Date:
March 01, 2017
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
BAKER HUGHES INCORPORATED (17021 Aldine Westfield, HOUSTON, Texas, 77073, US)
International Classes:
F04B51/00; F04B15/02; F04B49/06
Domestic Patent References:
WO2008118775A12008-10-02
Foreign References:
US20100300683A12010-12-02
US20130211811A12013-08-15
US20120234539A12012-09-20
US20110125332A12011-05-26
Download PDF:
Claims:
CLAIMS

What is claimed is:

1. A method for monitoring health of a pump, the method comprising:

collecting, by a processing device, pump data about the pump, wherein the pump data comprises vibration data, timing data, and pressure data collected from a vibration sensor, a timing sensor, and a pressure sensor respectively;

conditioning, by the processing device, the pump data by dividing the pump data into a plurality of segments, wherein each of the plurality of segments represents one completion cycle of the pump;

analyzing, by the processing device, the plurality of segments of pump data to determine a valve and valve seat wear state of the pump using a self-organizing map generated from historical pump data; and

changing, by the processing device, an operational parameter of the pump responsive to determining that the valve wear state of the pump represents a risk level that is above a threshold.

2. The method of claim 1, further comprising generating, by the processing device, an alert that the valve wear state of the pump represents a risk level that is above the threshold responsive to determining that the valve wear state of the pump represents a risk level that is above the threshold.

3. The method of claim 1, wherein conditioning the pump data further comprises filtering the vibration data using a high-pass filter to isolate a valve impact signature from a low frequency content that represents a pump motion.

4. The method of claim 1, wherein the self-organizing map is trained by:

conditioning historical pump data over m cycles, wherein the historical pump data comprises historical vibration data, historical timing data, and historical pressure data;

dividing each of the m cycles into n bins of binned historical pump data;

creating a representation of the binned historical pump data;

selecting a plurality of features and a number of groups for the binned historical pump data;

generating a feature plot for each of the groups of the binned historical pump data; and

characterizing the group as one of healthy, worn out, or bad valves.

5. The method of claim 4, wherein the plurality of features and the number of groups is selected based on at least one of a sampling rate, cutoff frequencies, number of frequency bins, a number of bins, and a pressure range.

6. The method of claim 1, wherein the operation parameter comprises one of a pressure, a speed, timing.

7. The method of claim 1, wherein changing the operational parameter of the pump comprises disabling the pump.

8. The method of claim 1, wherein the pump is a reciprocating positive displacement pump.

9. A method for monitoring health of a pump, the method comprising:

collecting, by a processing device, pump data about the pump, wherein the pump data comprises vibration data, timing data, and pressure data collected from vibration sensors, a timing sensor, and pressure sensors respectively;

conditioning, by the processing device, the pump data by dividing the pump data into a plurality of segments, wherein each of the plurality of segments represents one completion cycle of the pump;

analyzing, by the processing device, the plurality of segments of pump data to determine a washout potential of the pump using a self-organizing map generated from historical pump data; and

disabling, by the processing device, the pump responsive to determining that the washout potential of the pump represents a risk level that is above a threshold.

10. The method of claim 9, wherein the self-organizing map is trained by:

conditioning historical pump data over m cycles, wherein the historical pump data comprises historical vibration data, historical timing data, and historical pressure data;

dividing each of the m cycles into n bins of binned historical pump data;

creating a representation of the binned historical pump data;

selecting a plurality of features and a number of groups for the binned historical pump data;

generating a feature plot for each of the groups of the binned historical pump data; and

characterizing the group as one of healthy or impending washout.

11. The method of claim 10, wherein the plurality of features and the number of groups is selected based on at least one of a sampling rate, cutoff frequencies, number of frequency bins, a number of bins, and a pressure range.

12. A system for training a self-organizing map for monitoring pump health, the system comprising:

a memory having computer readable instructions; and

a processing device for executing the computer readable instructions, the computer readable instructions comprising:

collecting pump data about the pump, wherein the pump data comprises vibration data, timing data, and pressure data collected from vibration sensors, a timing sensor, and pressure sensors respectively;

training a self-organizing map by:

conditioning historical pump data over m cycles, wherein the historical pump data comprises historical vibration data, historical timing data, and historical pressure data,

dividing each of the m cycles into n bins of binned historical pump data,

creating a representation of the binned historical pump data,

selecting a plurality of features and a number of groups for the binned historical pump data,

generating a feature plot for each of the groups of the binned historical pump data, and

characterizing the group as one of healthy or impending washout;

analyzing the collected pump data to determine a valve wear state of the pump using the self-organizing map; and

changing an operational parameter of the pump responsive to determining that the valve wear state of the pump represents a risk level that is above a threshold.

13. The system of claim 12, wherein the plurality of features and the number of groups is selected based on at least one of a sampling rate, cutoff frequencies, number of frequency bins, a number of bins, and a pressure range.

14. The system of claim 12, wherein the operation parameter comprises one of a pressure, a speed, and a timing.

15. The method of claim 12, wherein changing the operational parameter of the pump comprises disabling the pump.

Description:
ANALYZING A PUMP TO DETERMINE VALVE WEAR AND WASHOUT USING A

SELF-ORGANIZING MAP

BACKGROUND

[0001] The present disclosure relates to well operations and, more particularly, to analyzing a pump to determine valve wear and washout using a self-organizing map.

[0002] Boreholes are drilled into earth formations having reservoirs of hydrocarbons in order to extract the hydrocarbons through the boreholes to the surface. Various

components (e.g., pipe segments, pipe couplings, pipe valves, manifolds, etc.) connect equipment trucks (e.g., blending trucks, pumping trucks, etc.) at the earth's surface to the bore holes. The components that connect the equipment trucks to the boreholes carry fluid, such as hydraulic fracturing fluid, to the boreholes to be used to extract the hydrocarbons through the boreholes. In particular, the pumping trucks use pumps to pressurize the fluid to fracture earth formations to extract the hydrocarbons.

BRIEF SUMMARY

[0003] According to aspects of the present disclosure, techniques including methods, systems, and/or computer program products for monitoring health of a pump are provided. An example method may include: collecting, by a processing device, pump data about the pump, wherein the pump data comprises vibration data, timing data, and pressure data collected from a vibration sensor, a timing sensor, and a pressure sensor respectively;

conditioning, by the processing device, the pump data by dividing the pump data into a plurality of segments, wherein each of the plurality of segments represents one completion cycle of the pump; analyzing, by the processing device, the plurality of segments of pump data to determine a valve wear state of the pump using a self-organizing map generated from historical pump data; and changing, by the processing device, an operational parameter of the pump responsive to determining that the valve wear state of the pump represents a risk level that is above a threshold.

[0004] In another example, a method may include: collecting, by a processing device, pump data about the pump, wherein the pump data comprises vibration data, timing data, and pressure data collected from vibration sensors, a timing sensor, and pressure sensors respectively; conditioning, by the processing device, the pump data by dividing the pump data into a plurality of segments, wherein each of the plurality of segments represents one completion cycle of the pump; analyzing, by the processing device, the plurality of segments of pump data to determine a washout potential of the pump using a self-organizing map generated from historical pump data; and disabling, by the processing device, the pump responsive to determining that the washout potential of the pump represents a risk level that is above a threshold.

[0005] According to additional aspects of the present disclosure, an example system may include: a memory having computer readable instructions; and a processing device for executing the computer readable instructions. The computer readable instructions may include: collecting pump data about the pump, wherein the pump data comprises vibration data, timing data, and pressure data collected from vibration sensors, a timing sensor, and pressure sensors respectively; training a self-organizing map by conditioning historical pump data over m cycles, wherein the historical pump data comprises historical vibration data, historical timing data, and historical pressure data, dividing each of the m cycles into n bins of binned historical pump data, creating a representation of the binned historical pump data, selecting a plurality of features and a number of groups for the binned historical pump data, generating a feature plot for each of the groups of the binned historical pump data, and characterizing the group as one of healthy or impending washout; analyzing the collected pump data to determine a valve wear state of the pump using the self-organizing map; and changing an operational parameter of the pump responsive to determining that the valve wear state of the pump represents a risk level that is above a threshold.

[0006] Additional features and advantages are realized through the techniques of the present disclosure. Other aspects are described in detail herein and are considered a part of the disclosure. For a better understanding of the present disclosure with the advantages and the features, refer to the following description and to the drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

[0007] The subject matter which is regarded as the invention is particularly pointed out and distinctly claimed in the claims at the conclusion of the specification. The foregoing and other features, and advantages thereof, are apparent from the following detailed description taken in conjunction with the accompanying drawings in which:

[0008] FIG. 1 illustrates a block diagram of a pumping environment using a system for monitoring health of a pump according to examples of the present disclosure;

[0009] FIG. 2 illustrates a flow diagram of a method for monitoring health of a pump according to examples of the present disclosure; [0010] FIG. 3 illustrates a flow diagram of a method for monitoring health of a pump according to examples of the present disclosure;

[0011] FIG. 4 illustrates a flow diagram of a method for training a self-organizing map used for monitoring health of a pump according to aspects of the present disclosure; and

[0012] FIG. 5 illustrates a block diagram of a processing system for implementing the techniques described herein according to examples of the present disclosure.

DETAILED DESCRIPTION

[0013] Positive displacement reciprocating pumps are used for multiple applications including offshore drilling, drilling, directional drilling, hydraulic fracturing, and the like. In drilling, the pump is used to circulate fluid (i.e., drilling mud) in a drilling string. Although the present techniques are described using the example of hydraulic fracturing, the same techniques can be applied to any industry where positive displacement pumps are used.

[0014] In hydraulic fracturing, the equipment used to facilitate the fracturing may include an engine or drive system, a cooling system, a transmission, and a pump that includes a power end and a fluid end. The engine or drive system may be a diesel engine, a gasoline engine, an electric drive, or other suitable engine or drive systems. The transmission connects the engine to the power end and provides speed control and dampening and allows for proper power distribution for the torque and horsepower used for a particular fracturing application.

[0015] The fluid end of the pump is directly connected to the power end. The power end transforms the supplied rotational motion and energy from the transmission into the reciprocating motion that drives the pump's plungers in the fluid end. The pump may include a power end casing, a connecting rod, a pony rod, a plunger, a suction cover assembly, a suction & discharge valve assembly, and stay rods. A pressure chamber in the pump is used to create high pressure for well servicing. High pressure is used for stimulation operations processes to create a path in the earth formation (i.e., rock) that enhances the hydrocarbon recovery. Development of tight formation and natural gas resources has seen a significant increase in the required operating pressure of pumps. In addition, different districts maintain pumps a certain way that was interpreted from a procedure or learned over the years. These differences affect the repair and maintenance (R&M) expense per pump and provide an opportunity for cost savings and reliability improvements.

[0016] R&M expenses can decrease profitability but provide for the safe operation of fracturing equipment. Generally, R&M expenses can be categorized as major failures or minor/regular maintenance. Major failures include, for example, repair or replacement of truck frame, engine, radiator, transmission, power end, and fluid end. Minor failures and regular maintenance include, for example, change or repair of valves, seats, springs, plungers, packing, seals, filters, and engine oils and lubricants. In some cases, fluid end cost is the highest percentage of all R&M expenses of a unit over its life. Valve & seat cost is also a major portion of the overall R&M cost. The present techniques reduce the fluid end expenses, improve equipment reliability and availability, and improve personnel safety.

[0017] Mainly, fluid ends are manufactured from two materials: carbon steel and stainless steel. A significant portion (e.g., greater than 75%) of the carbon steel fluid end failures are fatigue, corrosion fatigue, or washout. A majority of the stainless steel failures are washouts. Fatigue and corrosion fatigue of carbon steel fluid ends is a material limitation for a given environment or operational parameters. There are techniques such as surface treatment, cavitation avoidance, and proactive maintenance that can help mitigate fatigue or corrosion failures of carbon steel fluid ends.

[0018] A majority of the washouts for either stainless steel or carbon steel are preventable. Repairing a washed out fluid end takes time, increases labor and parts costs, and increases non-production time. In addition, it takes additional R&M expenses to change a failed fluid end on a truck.

[0019] To solve these problems, a neural network based model is provided herein to predict valve and valve seat health (i.e., good, worn out, bad, etc.) and provide warnings about an impending washout. The present techniques collect data (e.g., pressure data, vibration data, timing data, etc.) and divide (e.g., bin) the data into cycle-by-cycle segments. The segmented data is then filtered and analyzed to predict valve and valve seat condition and washout possibility. In particular, a type of neural network, self-organizing map (SOM), is used to develop a predictive model that uses high-speed vibration, pressure, and timing signals to predict valve and valve seat wear state and impending washout of a fluid end mounted on a positive displacement reciprocating pump. The same techniques can be extended to monitor the health of a pump sub-system, a pump, a fracturing fleet, a directional drilling pump, and the like.

[0020] A predictive capability of potential failure events provides an opportunity to reduce maintenance cost, improve equipment reliability and equipment availability. These reliability and availability improvements reduce the number of assets required to complete a job and achieve higher customer satisfaction. The additional assets (e.g., horsepower) can be valuable for additional revenue generation. Another advantage of this real-time model is that it does not need to know the history of the parts or assembly. The present techniques can predict bad valves and valve seat or impending washout irrespective of when the parts were deployed.

[0021] The teachings of the present disclosure can be applied in a variety of well operations. These operations may involve using one or more treatment agents to treat a formation, the fluids resident in a formation, a wellbore, and/or equipment in the wellbore, such as production tubing. The treatment agents may be in the form of liquids, gases, solids, semi-solids, and mixtures thereof. Illustrative treatment agents include, but are not limited to, fracturing fluids, acids, steam, water, brine, anti-corrosion agents, cement, permeability modifiers, drilling muds, emulsifiers, demulsifiers, tracers, flow improvers etc. Illustrative well operations include, but are not limited to, hydraulic fracturing, stimulation, tracer injection, cleaning, acidizing, steam injection, water flooding, cementing, etc.

[0022] FIG. 1 illustrates a block diagram of a pumping environment 100 using a system 110 for monitoring health of a pump 120 according to examples of the present disclosure. The example of FIG. 1 is described as being used in a fracturing operation but other implementations are also possible and within the scope of the present disclosure. In this example, the pump 120 pumps a fluid, such as a drilling fluid, along a fluid path 130 into a well 140. The pump 120 may be any suitable pump including a reciprocating positive displacement pump.

[0023] A pump health monitoring processing system 110 performs data collection, data conditioning, and data analysis using a pump data collection module 112, a pump data conditioning module 114, and a pump data analysis module 116 respectively. The pump data collection module 112 collects pump data (e.g., vibration data, timing data, pressure data, etc.) from sensors mounted on or close to the fluid end of the pump 120. In the present example, three sensors 122, 124, 126 are used: the sensor 122 is vibration sensors to sense pump vibration, the sensor 124 is a timing sensor to sense pump speed, and the sensor 126 is pressure sensors to sense operating pressure. The pump data may be collected at a high sampling rate (e.g., 7.5 KHz, 15 KHz, 30 KHz, 45 KHz, 51.2 KHz, 72 KHz, etc.) by the sensors 122, 124, 126 in conjunction with the pump data collection module 112.

[0024] Once the pump data are collected, the pump data conditioning module 114 conditions the pump data by dividing the pump data into segments. Each of the segments represents one completion cycle of the pump 120. Dividing the pump data into segments may be based on the speed (in revolutions-per-minute (RPM)) of the pump 120. Data from the timing sensor (e.g., the sensor 124) is used to determine a beginning or completion of a cycle. It is expected that the pump 120 is consistent from cycle to cycle because of the repetitive motion of various components and forces on and in the pump 120. Each segment represents a pump cycle.

[0025] The vibration data collected by the vibration sensors (e.g., the sensor 126) contains information including plungers, fluid, valves, and pump. The vibration data are digitally filtered using a high-pass filter to isolate valve impact signatures from low- frequency content that represents pump motions. After segmentation and filtering, the data are ready for valve wear and washout analysis.

[0026] The pump data analysis module 116 performs valve and valve seat wear and/or washout analysis, which may be calculated separately, consecutively, or in parallel. To perform the valve and valve seat wear and/or washout analysis, the pump data analysis module 116 utilizes a self-organizing map (SOM), which is a type of neural network.

[0027] A neural network, commonly referred to as an artificial neural network (ANN), is a biological-inspired collection of elements/nodes/neurons that process information in a coordinated manner. An ANN is typically organized by an input layer, hidden layers, and an output layer. The hidden layers include interconnected neurons and connect the input and output layers. An ANN uses a form of learning rules to store knowledge (stored

information/model) and utilize the knowledge to approximate a response. Numerical techniques can be used to train an ANN or to determine ANN parameters such as the number of neuron, weights, constants, and number of layers. An ANN is a powerful computational model with an ability to represent both linear and nonlinear relationships and learn these relationships from a given set of data. There are many types of neural networks including self-organizing maps.

[0028] A SOM, also known as a Kohonen self-organizing feature map, compresses high dimensional data into geometric relationships onto a low-dimensional representation. An input data set and an output data set are used to train a typical neural network. This is referred to as a supervised learning process. In contrast, the output data is not needed to train a SOM. This is referred to as an unsupervised learning process. A SOM includes an input layer, competition layers, and an output layer. As described in more detail below, FIG. 4 illustrates a flow diagram of a method for training, by the SOM generation module 118, a self- organizing map used for monitoring health of a pump according to aspects of the present disclosure.

[0029] Once the SOM is trained, the pump data analysis module 116 analyzes the plurality of segments of pump data to determine a valve wear state and/or a washout potential of the pump using the SOM generated from historical pump data. [0030] Regarding valve and valve seat wear of the pump 120, the main function of pump valves and seats (i.e., inlet and discharge) is to control the flow of the fluid by opening and closing during an operating cycle. The inlet valve opens (discharge closes) to feed the pressure chamber of the pump 120 with fluid from bottom of the fluid end when plunger is moving out. The discharge valve opens (inlet closes) to let the high-pressure fluid flow out from the top of the fluid end as plunger in moving in. Springs (e.g., helical springs) and differential pressure across valve provide the energy to open and close the valves.

Polyurethane annulus on valves is damaged over time from repeated impacts of opening and closing actions of the valves. If valves are not changed at optimal time, a lack of sealing due to damaged polyurethane can cause washout in the valve and seat area and ultimately loss of the fluid end.

[0031] Valve wear phenomenon in pumps tends to be slow as it develops over days or weeks (depending upon the usage). The pump data analysis module 116 can process a large number of pump cycles by being precise and detecting gradual changes in signals.

[0032] Valve and valve seat wear can be classified in three stages healthy, partially worn out, and bad. A healthy condition indicates that the valve and valve seat is operating as designed. A worn out condition indicates that more than half of the valve and valve seat life has been used but it is still functioning as designed and will not harm the fluid end. In some examples, it may be desirable to send an alert to an operator if the pump 120 is in a worn out state. Alternatively, it may be desirable to reduce the speed, pressure, etc., of the pump 120 if it is in a worn out state. A bad condition indicates that the valve should be changed as soon as possible in order to avoid a fluid end failure. In some examples, it may be desirable to disable the pump 120 or otherwise shut it down if it is determined that the pump is in a bad health.

[0033] Regarding washout potential, pump plungers move in and out of the pressure chamber. This movement opens and closes the inlet and discharge valves and creates high pressure in the pump chambers. Pump packing is stationary, held in place by stuffing boxes and around the plungers. These packings provide seals around the plungers to maintain pressure inside fluid end chambers. If seals are not mounted or maintained properly, these can provide a leaking area for high-pressure abrasive. This leaking can cause a fluid end washout and loss of the fluid end.

[0034] Washouts can develop in a short time (e.g., in less than 50 pump cycles) and can lead to fluid end failure if the washout is not addressed immediately. To solve this problem, the present techniques predict washouts conditions and take action to prevent a washout event. The proposed techniques for washout detection and prevention classify the pump in one of two states: healthy and impending washout. SOM operates in the frequency domain to detect and alert (or take corrective action) for impending washouts. For example, an energy level between 10 KHz and 16 KHz is consistently higher for impending washouts as compared to normal operation. SOM is above to detect this difference.

[0035] The various components, modules, engines, etc. described regarding FIG. 1 may be implemented as instructions stored on a computer-readable storage medium, as hardware modules, as special-purpose hardware (e.g., application specific hardware, application specific integrated circuits (ASICs), as embedded controllers, hardwired circuitry, etc.), or as some combination or combinations of these. In examples, the engine(s) described herein may be a combination of hardware and programming. The programming may be processor executable instructions stored on a tangible memory, and the hardware may include a processing device for executing those instructions. Thus, a system memory can store program instructions that when executed by a processing device implement the modules described herein. Other modules may also be utilized to include other features and

functionality described in other examples herein.

[0036] Alternatively or additionally, the processing system 100 may include dedicated hardware, such as one or more integrated circuits, Application Specific Integrated Circuits (ASICs), Application Specific Special Processors (ASSPs), Field Programmable Gate Arrays (FPGAs), or any combination of the foregoing examples of dedicated hardware, for performing the techniques described herein.

[0037] FIG. 2 illustrates a flow diagram of a method 200 for monitoring health of a pump according to examples of the present disclosure. The method 200 may be performed by a processing system, such as the pump health monitoring processing system 110 of FIG. 1, the processing system 20 of FIG. 5, and/or by another suitable processing system. In describing the method 200, the modules of the processing system 110 of FIG. 1 are referenced; however, such reference is not intended to be limiting.

[0038] At block 202 of the method 200, the pump data collection module 112 of the processing system 110 collects pump data about the pump 120. The pump data includes vibration data, timing data, and pressure data collected from vibration sensors, a timing sensor, and pressure sensors (e.g., sensors 122, 124, 126) respectively.

[0039] At block 204 of the method 200, the pump data conditioning module 114 of the processing system 110 conditions the pump data by dividing the pump data into a plurality of segments. Each of the plurality of segments represents one completion cycle of the pump. In some examples, conditioning the pump data further includes filtering the vibration data using a high-pass filter to isolate a valve impact signature from a low- frequency content that represents a pump motion.

[0040] At block 206 of the method 200, the pump data analysis module 116 of the processing system 110 analyzes the plurality of segments of pump data to determine a valve wear state of the pump using a self-organizing map generated from historical pump data.

[0041] At block 208 of the method 200, the processing system 110 then changes an operational parameter of the pump responsive to determining that the valve wear state of the pump represents a risk level that is above a threshold. For example, changing the operational parameter of the pump may include changing a pressure, speed, and/or timing of the pump 120. Additionally, changing the operational parameter of the pump 120 may include disabling the pump 120. The threshold may be preset and/or adjustable and defines a threshold above which a valve wear indicates that the pump may be partially worn out and/or bad. It should be appreciated that multiple thresholds may be used to differentiate various stages of pump health. If the risk level is above the threshold, the processing system 110 disables the pump. Otherwise, the pump remains active.

[0042] In some examples, the method 200 may further include generating, by the processing system 110, an alert that the valve wear state of the pump represents a risk level that is above the threshold responsive to determining that the valve wear state of the pump represents a risk level that is above the threshold. In yet other examples, the SOM generation module 118 trains the self-organizing map as described in FIG. 4 below.

[0043] It should be understood that the processes depicted in FIG. 2 represent illustrations and that other processes may be added or existing processes may be removed, modified, or rearranged without departing from the scope and spirit of the present disclosure.

[0044] FIG. 3 illustrates a flow diagram of a method 300 for monitoring health of a pump according to examples of the present disclosure. The method 300 may be performed by a processing system, such as the pump health monitoring processing system 110 of FIG. 1, the processing system 20 of FIG. 5, and/or by another suitable processing system. In describing the method 300, the modules of the processing system 110 of FIG. 1 are referenced; however, such reference is not intended to be limiting.

[0045] At block 302 of the method 300, the pump data collection module 112 of the processing system 110 collects pump data about the pump 120. The pump data includes vibration data, timing data, and pressure data collected from vibration sensors, a timing sensor, and pressure sensors (e.g., sensors 122, 124, 126) respectively. [0046] At block 304 of the method 300, the pump data conditioning module 114 of the processing system 110 conditions the pump data by dividing the pump data into a plurality of segments. Each of the plurality of segments represents one completion cycle of the pump. In some examples, conditioning the pump data further includes filtering the vibration data using a high-pass filter to isolate a valve impact signature from a low- frequency content that represents a pump motion.

[0047] At block 306 of the method 300, the pump data analysis module 116 of the processing system 110 analyzes the plurality of segments of pump data to determine a washout potential of the pump using a self-organizing map generated from historical pump data.

[0048] At block 308 of the method 300, the processing system 110 then disables the pump responsive to determining that the washout potential of the pump represents a risk level that is above a threshold. The threshold may be preset and/or adjustable and defines a threshold above which a washout may occur. If the risk level is above the threshold, the processing system 110 disables the pump. Otherwise, the pump remains active.

[0049] It should be understood that the processes depicted in FIG. 3 represent illustrations and that other processes may be added or existing processes may be removed, modified, or rearranged without departing from the scope and spirit of the present disclosure.

[0050] FIG. 4 illustrates a flow diagram of a method 400 for training a self-organizing map used for monitoring health of a pump according to aspects of the present disclosure. The method 400 may be performed by a processing system, such as the pump health monitoring processing system 110 of FIG. 1, the processing system 20 of FIG. 5, and/or by another suitable processing system. In describing the method 300, the modules of the processing system 110 of FIG. 1 are referenced; however, such reference is not intended to be limiting.

[0051] For example, at block 402, the SOM generation module 118 conditions historical pump data over m cycles. The historical pump data includes historical vibration data, historical timing data, and historical pressure data. At block 404, the SOM generation module 118 divides each of the m cycles into n bins of binned historical pump data. At block 406, the SOM generation module 118 creates a representation of the binned historical pump data. At block 408, the SOM generation module 118 selects a plurality of features and a number of groups for the binned historical pump data and generates a feature plot for each of the groups of the binned historical pump data. The features and the number of groups is selected based on at least one of a sampling rate, cutoff frequencies, number of frequency bins, a number of bins, and a pressure range. Finally, at block 410, the SOM generation module 118 characterizes the group. In the case of valve wear, the group is characterized as one of healthy, worn out, or bad valves. In the case of impending wash out, group is characterized as one of healthy or impending washout.

[0052] It should be understood that the processes depicted in FIG. 4 represent illustrations and that other processes may be added or existing processes may be removed, modified, or rearranged without departing from the scope and spirit of the present disclosure.

[0053] Being able to evaluate and trend the state of wear or condition of components in each pump provides benefits for a service provider and an operator. For example, the present techniques provide for pro-active, pre-planning and scheduling of maintenance breaks, provide for alignment of several operational steps, and provide for streamlining and optimizing the entire operation.

[0054] The present techniques described herein can be expanded to provide almost real-time wear and tear progression of pumps to a central location on site. This provides regular state of wear and progression updates to the equipment operators, which can be achieve through either wireless and/or wired integration of the data acquisition and processing units to a central control or monitoring unit, providing regular updates.

[0055] The present techniques also provide real-time alarms for unit emergency shutdown, in case of catastrophic events, which can be a warning to the operators to react and adjust or possibly expanded upon for automated intervention and shut down through relevant software. Additionally, the processing system 110 can be expanded upon to perform a remaining life to next maintenance interval required real-time calculation through techniques considering the historical use and wear rate provided on regular basis. The relevant data can be refreshed and displayed to the operators and crew on regular basis, allowing for pro-active pre-planning and scheduling of maintenance. Data can be displayed in the form of data or via a graphic interface. The data may be displayed in a format that includes remaining operating time left to next maintenance, percentage life consumed, or percentage life remaining, and the like.

[0056] It is understood that the present disclosure is capable of being implemented in conjunction with any other type of computing environment now known or later developed. For example, FIG. 5 illustrates a block diagram of a processing system 20 for implementing the techniques described herein. In examples, processing system 20 has one or more central processing units (processors) 21a, 21b, 21c, etc. (collectively or generically referred to as processor(s) 21 and/or as processing device(s)). In aspects of the present disclosure, each processor 21 may include a reduced instruction set computer (RISC) microprocessor. Processors 21 are coupled to system memory (e.g., random access memory (RAM) 24) and various other components via a system bus 33. Read only memory (ROM) 22 is coupled to system bus 33 and may include a basic input/output system (BIOS), which controls certain basic functions of processing system 20.

[0057] Further illustrated are an input/output (I/O) adapter 27 and a communications adapter 26 coupled to system bus 33. I/O adapter 27 may be a small computer system interface (SCSI) adapter that communicates with a hard disk 23 and/or a tape storage drive 25 or any other similar component. I/O adapter 27, hard disk 23, and tape storage device 25 are collectively referred to herein as mass storage 34. Operating system 40 for execution on processing system 20 may be stored in mass storage 34. A network adapter 26 interconnects system bus 33 with an outside network 36 enabling processing system 20 to communicate with other such systems.

[0058] A display (e.g., a display monitor) 35 is connected to system bus 33 by display adaptor 32, which may include a graphics adapter to improve the performance of graphics intensive applications and a video controller. In one aspect of the present disclosure, adapters 26, 27, and/or 32 may be connected to one or more I/O busses that are connected to system bus 33 via an intermediate bus bridge (not shown). Suitable I/O buses for connecting peripheral devices such as hard disk controllers, network adapters, and graphics adapters typically include common protocols, such as the Peripheral Component Interconnect (PCI). Additional input/output devices are shown as connected to system bus 33 via user interface adapter 28 and display adapter 32. A keyboard 29, mouse 30, and speaker 31 may be interconnected to system bus 33 via user interface adapter 28, which may include, for example, a Super I/O chip integrating multiple device adapters into a single integrated circuit.

[0059] In some aspects of the present disclosure, processing system 20 includes a graphics processing unit 37. Graphics processing unit 37 is a specialized electronic circuit designed to manipulate and alter memory to accelerate the creation of images in a frame buffer intended for output to a display. In general, graphics processing unit 37 is very efficient at manipulating computer graphics and image processing, and has a highly parallel structure that makes it more effective than general-purpose CPUs for algorithms where processing of large blocks of data is done in parallel.

[0060] Thus, as configured herein, processing system 20 includes processing capability in the form of processors 21, storage capability including system memory (e.g., RAM 24), and mass storage 34, input means such as keyboard 29 and mouse 30, and output capability including speaker 31 and display 35. In some aspects of the present disclosure, a portion of system memory (e.g., RAM 24) and mass storage 34 collectively store an operating system such as the AIX® operating system from IBM Corporation to coordinate the functions of the various components shown in processing system 20.

[0061] The present techniques may be implemented as a system, a method, and/or a computer program product. The computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present disclosure.

[0062] The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. A non- exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.

[0063] Computer readable program instructions described herein can be downloaded to respective computing/processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and/or a wireless network. The network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. A network adapter card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing/processing device. [0064] Computer readable program instructions for carrying out operations of the present disclosure may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, or either source code or object code written in any

combination of one or more programming languages, including an object oriented

programming language such as Smalltalk, C++ or the like, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider). In some examples, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present disclosure.

[0065] Aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to aspects of the present disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer readable program instructions.

[0066] These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function/act specified in the flowchart and/or block diagram block or blocks.

[0067] The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions/acts specified in the flowchart and/or block diagram block or blocks.

[0068] The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various aspects of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.

[0069] Set forth below are some embodiments of the foregoing disclosure:

[0070] Embodiment 1 : A method for monitoring health of a pump, the method comprising: collecting, by a processing device, pump data about the pump, wherein the pump data comprises vibration data, timing data, and pressure data collected from a vibration sensor, a timing sensor, and a pressure sensor respectively; conditioning, by the processing device, the pump data by dividing the pump data into a plurality of segments, wherein each of the plurality of segments represents one completion cycle of the pump; analyzing, by the processing device, the plurality of segments of pump data to determine a valve and valve seat wear state of the pump using a self-organizing map generated from historical pump data; and changing, by the processing device, an operational parameter of the pump responsive to determining that the valve wear state of the pump represents a risk level that is above a threshold. [0071] Embodiment 2: The method of any prior embodiment, further comprising generating, by the processing device, an alert that the valve wear state of the pump represents a risk level that is above the threshold responsive to determining that the valve wear state of the pump represents a risk level that is above the threshold.

[0072] Embodiment 3 : The method of any prior embodiment, wherein conditioning the pump data further comprises filtering the vibration data using a high-pass filter to isolate a valve impact signature from a low frequency content that represents a pump motion.

[0073] Embodiment 4: The method of any prior embodiment, wherein the self- organizing map is trained by: conditioning historical pump data over m cycles, wherein the historical pump data comprises historical vibration data, historical timing data, and historical pressure data; dividing each of the m cycles into n bins of binned historical pump data;

creating a representation of the binned historical pump data; selecting a plurality of features and a number of groups for the binned historical pump data; generating a feature plot for each of the groups of the binned historical pump data; and characterizing the group as one of healthy, worn out, or bad valves.

[0074] Embodiment 5 : The method of any prior embodiment, wherein the plurality of features and the number of groups is selected based on at least one of a sampling rate, cutoff frequencies, number of frequency bins, a number of bins, and a pressure range.

[0075] Embodiment 6: The method of any prior embodiment, wherein the operation parameter comprises one of a pressure, a speed, timing.

[0076] Embodiment 7: The method of any prior embodiment, wherein changing the operational parameter of the pump comprises disabling the pump.

[0077] Embodiment 8: The method of any prior embodiment, wherein the pump is a reciprocating positive displacement pump.

[0078] Embodiment 9: A method for monitoring health of a pump, the method comprising: collecting, by a processing device, pump data about the pump, wherein the pump data comprises vibration data, timing data, and pressure data collected from vibration sensors, a timing sensor, and pressure sensors respectively; conditioning, by the processing device, the pump data by dividing the pump data into a plurality of segments, wherein each of the plurality of segments represents one completion cycle of the pump; analyzing, by the processing device, the plurality of segments of pump data to determine a washout potential of the pump using a self-organizing map generated from historical pump data; and disabling, by the processing device, the pump responsive to determining that the washout potential of the pump represents a risk level that is above a threshold. [0079] Embodiment 10: The method of any prior embodiment, wherein the self- organizing map is trained by: conditioning historical pump data over m cycles, wherein the historical pump data comprises historical vibration data, historical timing data, and historical pressure data; dividing each of the m cycles into n bins of binned historical pump data;

creating a representation of the binned historical pump data; selecting a plurality of features and a number of groups for the binned historical pump data; generating a feature plot for each of the groups of the binned historical pump data; and characterizing the group as one of healthy or impending washout.

[0080] Embodiment 11 : The method of any prior embodiment, wherein the plurality of features and the number of groups is selected based on at least one of a sampling rate, cutoff frequencies, number of frequency bins, a number of bins, and a pressure range.

[0081] Embodiment 12: A system for training a self-organizing map for monitoring pump health, the system comprising: a memory having computer readable instructions; and a processing device for executing the computer readable instructions, the computer readable instructions comprising: collecting pump data about the pump, wherein the pump data comprises vibration data, timing data, and pressure data collected from vibration sensors, a timing sensor, and pressure sensors respectively; training a self-organizing map by:

conditioning historical pump data over m cycles, wherein the historical pump data comprises historical vibration data, historical timing data, and historical pressure data, dividing each of the m cycles into n bins of binned historical pump data, creating a representation of the binned historical pump data, selecting a plurality of features and a number of groups for the binned historical pump data, generating a feature plot for each of the groups of the binned historical pump data, and characterizing the group as one of healthy or impending washout; analyzing the collected pump data to determine a valve wear state of the pump using the self- organizing map; and changing an operational parameter of the pump responsive to determining that the valve wear state of the pump represents a risk level that is above a threshold.

[0082] Embodiment 13 : The system of any prior embodiment, wherein the plurality of features and the number of groups is selected based on at least one of a sampling rate, cutoff frequencies, number of frequency bins, a number of bins, and a pressure range.

[0083] Embodiment 14: The system of any prior embodiment, wherein the operation parameter comprises one of a pressure, a speed, and a timing.

[0084] Embodiment 15: The system of any prior embodiment, wherein changing the operational parameter of the pump comprises disabling the pump. [0085] The descriptions of the various examples of the present disclosure have been presented for purposes of illustration, but are not intended to be exhaustive or limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described techniques. The terminology used herein was chosen to best explain the principles of the present techniques, the practical application or technical improvement over technologies found in the marketplace, or to enable others of ordinary skill in the art to understand the techniques disclosed herein.

[0086] Additionally, the term "about" is intended to include the degree of error associated with measurement of the particular quantity based upon the equipment available at the time of filing the application. For example, "about" can include a range of ± 8% or 5%, or 2% of a given value.

[0087] 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.