Forecasting industrial asset failure

Abstract

Forecasting industrial asset failures is described. A system determines a data start value associated with an industrial asset at a data start time. The system determines an end-of-data value associated with the industrial asset at an end-of-data time. The system estimates a fault time when a trend of estimating from a data start value to a data end value will reach a fault limit value. The system determines a distance to the fault based on the fault limit value and the data end value. The system outputs a fault forecast for the industrial asset associated with a fault time and a distance from the fault.

Description

[0001] Cross-references to related applications

[0002] This application claims priority to U.S. Provisional Patent Application No. 63 / 534,196, filed August 23, 2023, pursuant to 35 USC §119 or the Paris Convention, the entire contents of which are incorporated herein by reference as if set forth herein in their entirety. Background Technology

[0003] Analyzing operational trends in industrial facilities is time-consuming. Each industrial asset can have numerous sensors, and an industrial facility can comprise hundreds or thousands of industrial assets or pieces of equipment. This analytical challenge is exponentially amplified when a company has multiple industrial facilities scattered across a vast geographical area. The sheer volume of data generated by these sensors makes it impossible to maintain the very frequent physical monitoring and assessment of each trend. Further complicating matters is that only operators familiar with the equipment can identify risks from anomalous trends, as each trend has its own unit and failure limits, such as… Figure 1 As shown, Figure 1 The illustration depicts an example trend 100 of equipment data depicted by a system for predicting industrial asset failures, according to some embodiments.

[0004] Currently, conventional forecasting systems exist that use sample trend data to extrapolate the timing of failures. However, these existing systems are often found to be unreliable when forecasting equipment failures. Attached Figure Description

[0005] Figure 1 The illustration depicts example trends in equipment data depicted by a system for predicting industrial asset failures, according to some embodiments.

[0006] Figure 2 The illustration depicts an example display portion of a system, according to some embodiments, comprising software configured to look for anomalies in data trends to predict industrial asset failures.

[0007] Figure 3 The illustration depicts an example alarm tab section of a system for predicting industrial asset failures, according to some embodiments.

[0008] Figure 4 The illustration shows an example diagram highlighting the time fault domain by a system for predicting industrial asset failures, according to some embodiments.

[0009] Figure 5 The illustration shows an example diagram highlighting a unit fault domain by a system for predicting industrial asset failures, according to some embodiments.

[0010] Figure 6The illustration shows an example failure risk assessment formula used by a system for predicting industrial asset failures, according to some embodiments.

[0011] Figure 7 The illustration shows example risk designation categories used by systems for predicting industrial asset failures, according to some embodiments.

[0012] Figure 8 The illustration shows example graphs depicting time and distance risk assessments by a system for predicting industrial asset failures, according to some embodiments.

[0013] Figure 9 The diagram illustrates the predicted failure path as a linear trend in a graph depicted by a system for predicting industrial asset failures, according to some embodiments.

[0014] Figure 10 The illustration shows a real-world example of a current trend in a circular domain of a graph depicted by a system for predicting industrial asset failures, according to some embodiments.

[0015] Figure 11 The illustrations present examples of problems with conventional systems, according to some embodiments, in accurately predicting the changing trends of failures when forecasting industrial asset failures.

[0016] Figure 12 A comparison is shown between a unit-weighted left-hand risk representation (conventional system) defined by a system for predicting industrial asset failures, according to some embodiments, and a unit-weighted right-hand risk representation.

[0017] Figure 13 A modification of a risk assessment equation used by a system for predicting industrial asset failures, according to some embodiments, is shown, wherein the time to failure includes a minimum response time component.

[0018] Figure 14 The illustration depicts an example time-to-failure trend segment on an analysis display depicted by a system for predicting industrial asset failures, according to some embodiments.

[0019] Figure 15 The diagram illustrates a block diagram of an example system for predicting industrial asset failures, according to some embodiments.

[0020] Figure 16 This is a flowchart illustrating an example computer implementation of a method for predicting industrial asset failures, according to some embodiments.

[0021] Figure 17 This is a block diagram illustrating an example hardware device in which the subject matter can be implemented. Detailed Implementation

[0022] Existing systems are often found to be unreliable because they only provide estimates of the time to failure limits without considering how close the overall trend is to those limits. Conventional forecasting systems perform poorly in assessing the overall failure risk when dealing with issues involving high / low rates of change or quantities close to failure limits. For example, for the current trend, a conventional system might estimate the time to failure for a temperature limit as 10 days, without considering that even a mere degree Celsius increase in temperature as a change during the process could cause failure at any time. Considering only the time to failure often results in high-risk assets being inappropriately given low priority.

[0023] In some conventional systems, as the trend of one of the data points of an industrial asset moves toward the failure threshold, the slope of the data changes. In these systems, this change in slope will also alter the estimated time to failure. This time to failure can indicate that, based on the current trend, the risk is now minimal because the time to failure has significantly increased, such as from hours to days.

[0024] However, this conventional calculation ignores the unit (vertical) component of data trends and how close the data is to the edge of failure. Because the unit component is not considered, conventional systems often downgrade important issues to the point where they may no longer be considered a review priority. Therefore, there is a need for systems and methods that can perform concurrent analysis on a large number of different types of processes and accurately prioritize each failure risk.

[0025] This disclosure relates to a system and method for normalizing failure risk into a single metric used across different equipment types, so that system users have a way of viewing forecast results in a prioritized order that places the most important information first. The system allows users to assess risk without viewing trend graphs with specific labels and can communicate risk with parameters in different units in the same manner.

[0026] In some embodiments, the system determines a data start value associated with the industrial asset at the data start time. The system determines a data end value associated with the industrial asset at the data end time. The system estimates the failure time when the trend extrapolated from the data start value to the data end value will reach a failure limit value. The system determines the distance to failure based on the failure limit value and the data end value. The system outputs a failure prediction for the industrial asset associated with the failure time and the distance to failure.

[0027] For example, the system determines that the feedwater pump temperature was 144.9 degrees Fahrenheit (F) at noon on September 8th, and is currently 145.0 degrees Fahrenheit at noon on September 9th, representing an overall trend of a 0.1 degree Fahrenheit increase within one day. The system extrapolates the trend from the current temperature of 145.0 degrees Fahrenheit and estimates that the feedwater pump temperature will reach the equipment failure temperature of 146.0 degrees Fahrenheit at noon on September 19th, 10 days later. The remaining time to failure is calculated as 10 days from the end of the data period on September 9th to the estimated failure time on September 19th, which is 91% (10 days divided by 11 days) of the normalized remaining time to failure. The remaining distance to failure is the total distance to failure calculated from the expected start temperature of 124.0 degrees Fahrenheit to the calculated failure temperature of 146.0 degrees Fahrenheit, which is 1.0 degree Fahrenheit from the current temperature of 145.0 degrees Fahrenheit to the failure limit of 146.0 degrees Fahrenheit at noon on September 19th. This is 5% (1.0 degree Fahrenheit divided by 22.0 degrees Fahrenheit) of the normalized distance to failure. Due to the extreme urgency of the distance to failure assigned to the feedwater pump temperature, and despite the low urgency of the time to failure assigned to the feedwater pump, the system outputs the feedwater pump failure risk as the highest priority.

[0028] Various embodiments and aspects of this disclosure will be described with reference to the details discussed below, and the accompanying drawings will illustrate various embodiments. The following description and drawings illustrate this disclosure and should not be construed as limiting it. Numerous specific details are described to provide a thorough understanding of various embodiments of this disclosure. However, in some cases, well-known or conventional details have not been described in order to provide a concise discussion of embodiments of this disclosure.

[0029] While these embodiments have been described in sufficient detail to enable those skilled in the art to practice the disclosed embodiments, it should be understood that these examples are not limiting, allowing for the use of other embodiments and modifications that may be made without departing from their spirit and scope. For example, the operations of the methods shown and described herein do not necessarily have to be performed in the indicated order and may be performed in parallel. It should also be understood that a method may include more or fewer operations than indicated. Operations described herein as individual operations may be combined. Conversely, what is described herein as a single operation may be implemented in multiple operations.

[0030] References to "an embodiment," "an embodiment," or "some embodiments" in the specification mean that a particular feature, structure, or characteristic described in connection with that embodiment may include at least one embodiment of this disclosure. The appearance of the phrase "embodiment" or "this embodiment" in various places in the specification does not necessarily refer to the same embodiment.

[0031] Figure 2 The illustration shows an example display portion of a system 200 for predicting industrial asset failures according to some embodiments, and the system includes software configured to look for anomalies in data trends. System 200 allows a user to select training data 202, select one or more points representing anomalies in the trend, and select a range of trends. System 200 can compare the training data with other sample data and use the training data to train an artificial intelligence model.

[0032] After training on training data 202, system 200 can identify multiple anomalies and / or generate a list of these anomalies on a dashboard. The system can extrapolate trends based on these anomalies to predict the time it will take for anomalies to trigger system alarms and cause equipment failures. System 200 can display these alarms in conjunction with a prediction model, where each alarm corresponds to a discrete level of an urgency scale 204.

[0033] System 200 can perform time normalization to normalize the forecast model on the time to failure axis 206; perform unit normalization to normalize the forecast model in a graph 208 depicting the normalization of the measurement axis in units 210; and use time normalization and unit normalization to perform urgency calculations to assess risk across numerous considerations.

[0034] Figure 3 The illustration shows an example alarm tab section 302 of a system 300 for forecasting industrial asset failures according to some embodiments. Alarm tab section 302 includes a normalized forecast remaining time column 304 (such as the number of days or hours remaining until equipment failure) and a forecast urgency column 306 displaying urgency in discrete levels (such as low, medium, high, very high, and extreme). The assets represented in the forecast remaining time column 304 are associated with the forecast urgency column 306, and the resulting columns and rows categorize which assets have the highest urgency.

[0035] Figure 4 An example diagram 402, illustrating a time-domain failure, is provided by a system 400 for predicting industrial asset failures according to some embodiments. The system 400 can extrapolate trends toward failure limits and assess the time and distance to those limits. The system 400 can decompose the trend into axial components, such as x-axis and y-axis components. The system 400 can determine risk based on a separate analysis of each axial component, using the distance to the failure limit along the trend in each axial direction. Axial directions include horizontal and / or vertical directions.

[0036] System 400 can normalize the values ​​of axis components, enabling system users to compare different equipment. Normalization allows for risk analysis comparisons between different equipment types. Each axis component represents a fault domain. Normalization in both directions gives each axis component a value between 0 and 1 (or 0% and 100%).

[0037] In the example below, System 400 records the temperature of water pump 1B as 105 degrees Fahrenheit at 5:38:00 AM and 115 degrees Fahrenheit at 5:43:00 AM, then extrapolates that the linear trend between these two data points continues linearly until it reaches 128 degrees Fahrenheit at 5:46:20 AM. While this example uses a specific trend and a specific type of extrapolation to simplify example calculations of time to failure and distance to failure, System 400 can assess any type of trend and perform any number of extrapolations of any type.

[0038] Continuing this example, System 400 normalizes the time to failure by subtracting the data end date (which can be called the data end time) from the failure date (which is the date and time when the trend calculated by System 400 reaches the failure limit value, and is the expected time when the equipment failure begins) to determine the numerator. In this example, the failure time is 5:46:20 AM and the data end time is 5:43:00 AM, so the numerator is determined to be a difference of 3 minutes and 20 seconds or 3.33 minutes. Further, in this example, System 400 normalizes the time to failure by subtracting the failure start date (which can be called the data start time) from the failure date (which can be called the failure time, which is the date and time when the trend calculated by System 400 reaches the failure limit value, and is the expected time when the equipment failure begins) to determine the denominator. In this example, the failure time is 5:46:20 AM and the data start time is 5:38:00 AM, so the denominator is determined to be a difference of 8 minutes and 20 seconds or 8.33 minutes.

[0039] System 400 normalizes the time to failure by dividing a defined numerator by a defined denominator; in this example, 3.33 minutes divided by 8.33 minutes, which normalizes the remaining time to failure to 40%. This means that in the expected trend of 8 minutes and 20 seconds from the data start value to the failure limit, there are still 3 minutes and 20 seconds (40% of 8 minutes and 20 seconds) to elapse. Example chart 402 shows the time domain (vertical line) at 40% of the time to failure.

[0040] Figure 5The illustration shows an example graph highlighting a unit domain by a system 500 for predicting industrial asset failures, according to some embodiments. The unit domain is also referred to herein as the distance from the failure domain. Corresponding to the previous example, system 500 normalizes the distance from the failure by subtracting the data end value from the failure limit (which may be referred to as the failure limit value, which is the value of the assessed data type that begins to indicate equipment failure of the corresponding industrial asset (such as feedwater pump 1B)).

[0041] In this example, the fault limit is 128 degrees Fahrenheit and the data endpoint is 115 degrees Fahrenheit, so the numerator is determined to be a difference of 13 degrees Fahrenheit. Further in this example, system 500 normalizes the distance to the fault by subtracting the expected value (which can be called the expected data value) from the fault limit (which is the value of the assessed data type that begins to indicate the equipment fault of the corresponding industrial asset (such as feedwater pump 1B)). In this example, the fault limit is 128 degrees Fahrenheit and the expected data value is 105 degrees Fahrenheit, so the numerator is determined to be a difference of 23 degrees Fahrenheit.

[0042] System 500 normalizes the distance to fault by dividing the determined numerator by the determined denominator; in this example, 13 degrees Fahrenheit divided by 23 degrees Fahrenheit. This produces a normalized distance to fault of 57%. This means that of the expected trend experiencing 23 degrees Fahrenheit from the initial data value to the fault limit, there are still 13 degrees Fahrenheit (57% of 23 degrees Fahrenheit) to be experienced. Example chart 502 shows that the unit field (vertical line) is 75% away from the fault.

[0043] Figure 6The illustration shows an example failure risk assessment formula 602 used by a system 600 for predicting industrial asset failures, according to some embodiments. The system 600 analyzes the failure risk of industrial assets by evaluating the input parameters required by the example failure risk assessment formula 602. The Dtf parameter 604 can be a distance-to-failure value, measured as a percentage (%), defined between 0% and 100%, and can be calculated based on the failure limit (failure limit value) minus the numerator of the current value (data end value) divided by the failure limit (failure limit value) minus the denominator of the expected value (expected data value). The Ttf parameter 606 can be a time-to-failure value, measured as a percentage (%), defined between 0% and 100%, and can be calculated based on the failure date (failure time) minus the numerator of the current date (data end time) divided by the failure date (failure time) minus the denominator of the failure start date (data start time). The system can use Wdtf parameter 608 (which can be the user priority weight of distance to failure parameter Dtf 604), Wttf parameter 610 (which can be the user priority weight of time to failure parameter Ttf 606), and Cf parameter 612 (which can be the curvature factor and can be a value between 0 and 1) to evaluate formula 602, which produces a number bounded between 0 and 100.

[0044] Figure 7 The illustration depicts an example risk designation category 702 used by a system 700 for predicting industrial asset failures, according to some embodiments. After calculating a failure risk analysis value (which can be a number between 0 and 100), the system 700 can use the failure risk analysis value to support multiple “buckets” for risk designation or labeling. For example, if the risk analysis value ranges from 0 to less than 22, the risk is low; if the risk analysis value ranges from 22 to less than 45, the risk is medium; if the risk analysis value ranges from 45 to less than 65, the risk is high; if the risk analysis value ranges from 65 to less than 85, the risk is very high; and if the risk analysis value ranges from 85 to 100, the risk is extreme.

[0045] Figure 8An example graph 802 depicts a time and distance risk assessment drawn by a system 800 for predicting industrial asset failures, according to some embodiments. Since the risk analysis values ​​are numbers from 0 to 100, the system 800 can visually depict the risk assessment by indicating the position of the risk analysis values ​​on the graph 802. Graph 802 can be based on a horizontal time axis 804 depicting values ​​from 0 to 100, where position 0 is shown in green, position 100 in red, and intermediate positions are shown as a gradient from green to yellow to orange to red. Similarly, graph 802 can be based on a vertical distance axis 806 depicting values ​​from 0 to 100, where position 0 is shown in green, position 100 in red, and intermediate positions are shown as a gradient from green to yellow to orange to red.

[0046] In general, if the time to failure of water pump 1B has a risk value of 52 and the distance to failure of water pump 1B has a risk value of 59, then system 800 can position the icon representing water pump 1B at xy coordinates (52, 59) on graph 802. A human operator can easily and quickly interpret the position of this icon as indicating high risk because the icon is in the middle or middle of various shades of orange, as both the horizontal axis 804 (time to failure) and the vertical axis 806 (distance to failure) are transitioning from yellow to red. However, in Figure 8 Depicting a chart like 802 was a challenge because the numbering of 1000 positions (100 columns by 100 rows) could not be represented on a single sheet of paper using a font large enough to meet the plotting requirements. Similarly, the requirement to depict the chart in black and white presented a challenge in representing it using a color gradient from green to yellow to orange to red. Therefore, chart 802 is based on 25 horizontal positions (numbered 4, 8, 12...100) and 25 vertical positions (numbered 4, 8, 12...100), with positions 4 to 32 depicted in white, positions 36 to 64 depicted in white with black grid lines, and positions 68 to 100 depicted in black.

[0047] Figure 9 The diagram 902 illustrates a predicted failure path with a linear trend depicted by a system 900 for predicting industrial asset failures, according to some embodiments. Trend arrows 904 in diagram 902 indicate how the predictive trend will behave according to some embodiments. In this example, the failure trend is perfectly linear.

[0048] Figure 10The illustration depicts a real-world example of a current trend in a circular domain of a graph 1002 drawn by a system 1000 for predicting industrial asset failures, according to some embodiments. As is clear from the image including trend arrows 1004 and circular icons 1006 and 1008, system 1000 is configured to assign a risk as high if either the time domain or the unit domain is within predetermined limits. Conventional techniques cannot achieve proper risk assessment in the unit domain.

[0049] Figure 11 The illustration illustrates an example comparison of problems with conventional systems according to some embodiments in accurately predicting trends in failures compared to system 1100 used for forecasting industrial asset failures. In some conventional systems, as the trend of one data point of the industrial asset moves towards the failure threshold, the slope of the data trend changes. In conventional systems, this change in slope will cause the estimated time to failure to also change. This time to failure can indicate that, based on the current trend, the risk is now minimal because the time to failure has increased, for example, from several hours to several days.

[0050] However, this conventional calculation ignores the unit (vertical) component of the data trend and how close it is to the edge of failure. Because the unit component is not considered, conventional systems often downgrade important issues to the point where these issues may no longer be considered as review priorities. Therefore, system 1100 enables concurrent analysis of both the time to failure 1102 and distance to failure 1104 for a large number of different types of processes and accurately prioritizes their respective failure risks.

[0051] Without normalization of units and / or time components, assets cannot be compared because of unit mismatch. For example, risk assessments of a temperature 10 degrees Fahrenheit away from a fault and a pressure of 100 pounds per square inch (psi) away from a fault cannot be compared because the numerical values ​​give the illusion that a temperature of 10 degrees Fahrenheit automatically has higher priority, as the number 10 seems closer to any fault limit than the number 100. However, if a temperature of 10 degrees Fahrenheit is far from a fault limit, while a pressure of 100 psi is close to a fault limit, then the pressure trend carries greater risk. In some conventional systems, analyzing risk only in the time domain does not produce this information.

[0052] Figure 12 A comparison is shown between a unit-weighted left-hand risk representation 1202 (conventional system) defined by a system 1200 for predicting industrial asset failures, according to some embodiments, and a unit-weighted right-hand risk representation 1204. System 1200 allows a user to adjust the weight of each risk category, enabling the user to select which urgency (time or distance) should be given the highest priority.

[0053] Figure 13 A modification of risk assessment equation 1302 by a system 1300 for predicting industrial asset failures, according to some embodiments, is shown, where the time to failure includes a minimum response time component. The minimum response time component 1302 is an increase in the risk component based on how much time is needed to correct the issues associated with the risk. The modified Ttf can be a time to failure value measured as a percentage (%), defined between 0% and 100%, and can be calculated by subtracting the current date (data end time) and also the alarm time span (minimum response time) from the failure date (failure time) and dividing the numerator by subtracting the failure start date (data start time) and also the alarm time span (minimum response time). Including the alarm time span (minimum response time) adjusts the risk value by removing the required minimum response time from the time buffer of the time to failure risk. For example, if the time to failure for turbine 3 is 4 hours and the operator assigned to maintain the turbine is willing to use one hour as a minimum response time to carefully transition to another turbine that is in operation while the turbine is stopped, then the time to failure is effectively reduced from four hours to three hours.

[0054] Figure 14 The illustration depicts an example time-to-failure trend segment 1402 on an analysis display depicted by a system 1400 for predicting industrial asset failures, according to some embodiments. As each segment of the trend is analyzed, the system 1400 can display the corresponding trend in the time-to-failure trend segment 1402. The system 1400 leads the user to believe that assets in a non-alarm state are of low risk, while items in an alarm state require more attention due to their corresponding trends.

[0055] This reduces costs because businesses don't need as many human operators to monitor various trends in industrial assets. System 1400 allows users to generate forecasts for one or more sensors in the asset sensor list section 1404 of the analysis display. Then, once the forecasting model is deployed, System 1400 can automatically select data from the trends.

[0056] Figure 15 The illustration shows a block diagram of an example system 1500 for predicting industrial asset failures in an embodiment. Figure 1 As shown, System 1500 can illustrate a cloud computing environment in which data, applications, services, and other resources are stored and delivered through a shared data center and appear as a single access point for users. System 1500 can also represent any other type of distributed computer network environment in which servers control the storage and distribution of resources and services for different client users.

[0057] In this embodiment, system 1500 represents a cloud computing system, which includes a first client 1502, a second client 1504, a third client 1506, a fourth client 1508, a server 1510 which may be provided by a hosting company, and an optional cloud computing environment 1512. Clients 1502–1508, server 1510, and cloud computing environment 1512 communicate via network 1514. Although Figure 15 The first client 1502 is described as a laptop computer 1502, the second client 1504 as a desktop computer 1504, the third client 1506 as a smartphone 1506, and the fourth client 1508 as a server, but each of the system components 1502-1510 can be any type of computer system, and each can be substantially related to... Figure 17 The hardware device 1700 described in the text and below is similar.

[0058] Server 1510 can host and execute industrial asset failure prediction system 1516, which can be accessed via graphical user interface 1518 (e.g., Figure 15 (as depicted), and / or reside on any of clients 1502–1508. Although Figure 15 It is described that all industrial asset failure prediction systems 1516 reside entirely on server 1510, but any or all of the industrial asset failure prediction systems 1516 may reside entirely on clients 1502-1508, entirely on cloud computing environment 1512, or partially on clients 1502-1508, partially on server 1510, partially on cloud computing environment 1512, and / or partially on [other platforms / systems]. Figure 15 Any combination on another server not described in the text. Figure 15 A system 1500 is described, comprising four clients 1502-1508, a server 1510, a cloud computing environment 1512, a network 1514, an industrial asset failure prediction system 1516, and a graphical user interface 1518. However, the system 1500 may include any number of clients 1502-1508, any number of servers 1510, any number of cloud computing environments 1512, any number of networks 1514, any number of industrial asset failure prediction systems 1516, and any number of graphical user interfaces 1518.

[0059] Figure 16 This is a flowchart illustrating a computer-implemented method for predicting industrial asset failures in an embodiment. Flowchart 1600 depicts the method shown as... Figure 15 The flowchart blocks for the method actions of system components 1502-1518 and / or certain actions involved between them.

[0060] The data start time is determined by setting the data start value associated with the industrial asset, box 1602. The system begins collecting data for predicting equipment failure. For example, and not as a limitation, this may include the industrial asset failure prediction system 1516 determining that the water pump 1B has a water temperature of 144.9 degrees Fahrenheit at noon on September 8.

[0061] The starting value of data can be the initial number or mathematical object of an information sequence. An industrial asset can be equipment used to manufacture or produce products and / or services. The starting time of data can be a chronological measurement of the beginning of an information sequence.

[0062] After determining the data start value for the industrial asset at the data start time, the data end value associated with that industrial asset at the data end time is determined, box 1604. The system completes the data collection for predicting equipment failures. For example, and not as a limitation, this may include the industrial asset failure prediction system 1516 determining that the water pump 1B has a water temperature of 145.0 degrees Fahrenheit at noon on September 9th. The data end value can be the final numerical or mathematical object of the information sequence. The data end time can be the final time-sequential measurement for the information sequence.

[0063] After determining the start and end time data values, the failure time is estimated when the trend extrapolated from the start to the end of the data will reach the failure limit value, box 1606. The system predicts when the equipment will begin to fail. In an embodiment, this may include an industrial asset failure prediction system 1516 extrapolating the trend from the current water temperature of 145.0 degrees Fahrenheit to the future and estimating that the water temperature of feedwater pump 1B will reach the equipment failure temperature of 146.0 degrees Fahrenheit at noon on September 19, 10 days later.

[0064] Failure time can be a time-series measurement associated with functional loss. A trend can be the overall direction of development or change of something. A failure limit can be a threshold number or mathematical object associated with functional loss.

[0065] After estimating the failure time based on trends extrapolated from the start and end values ​​of the data, the distance to failure is determined based on the failure limit value and the end value of the data, box 1608. The system identifies the tolerance up to the point where the equipment fails. For example, and not as a limitation, this could include an industrial asset failure prediction system 1516 subtracting the current temperature of 145.0 degrees Fahrenheit from a projected failure temperature of 146.0 degrees Fahrenheit to arrive at a distance to failure of 1 degree Fahrenheit. The distance to failure can be an interval between numerical values ​​and associated with loss of function.

[0066] In addition to determining the distance to failure based on the values ​​of the failure limit and the end of the data period, the time-based failure risk can optionally be determined based on a normalized time-to-failure time, which is normalized based on the data start time, data end time, and failure time (box 1610). The system identifies the time-to-failure risk of the equipment. For example, and without limitation, this could include the Industrial Asset Failure Forecasting System 1516 determining that the remaining time-to-failure is 10 days from the data end time at noon on September 9th to the failure time at noon on September 19th, out of the 11 days from the data start time at noon on September 8th to the predicted failure time at noon on September 19th. This is 91% (10 days divided by 11 days) of the normalized remaining time-to-failure time. The Industrial Asset Failure Forecasting System 1516 uses the 91% normalized remaining time-to-failure time to assign the low urgency of the time-based failure risk to the maintenance request of the feedwater pump operator, because the estimated failure risk of feedwater pump 1B occurs after a relatively long period of time. Time-based failure risk can be a time-series measurement associated with the probability of functional loss.

[0067] Determining the risk of time-related failures can be based on adjustments to the required minimum response time. For example, the Industrial Asset Failure Forecasting System 1516 takes into account the following scenario: the operator assigned to maintain a feedwater pump is willing to use one hour as the minimum response time to cautiously transition to another feedwater pump that is in operation while the pump is stopped. The Industrial Asset Failure Forecasting System 1516 uses 10 days (240 hours) of the estimated remaining time to the failure and subtracts the one-hour minimum response time for switching the feedwater pump, resulting in a numerator of 239 hours.

[0068] Next, the Industrial Asset Failure Prediction System 1516 uses the estimated total time to failure, 11 days (264 hours), and subtracts the minimum one-hour response time for switching the feedwater pump, resulting in a denominator of 263 hours. Then, the Industrial Asset Failure Prediction System 1516 divides the numerator of 239 hours by the denominator of 263 hours to obtain a normalized percentage of the unchanging time to failure, 91%, which still represents a low urgency of the failure risk for assigning maintenance requests to feedwater pump operators. Adjustments can be minor changes made to achieve desired fit, appearance, or results. Minimum response time can be the smallest time-sequential measurement of the response.

[0069] Similar to determining time-based failure risk, unit failure risk can optionally be determined based on normalized distance to failure, where the distance to failure is normalized based on the expected data value, the data end value, and the failure limit value (block 1612). The system identifies the distance to failure risk of the equipment. In an embodiment, this may include the industrial asset failure prediction system 1516 using the remaining distance to failure (which is 1.0 degree Fahrenheit from the current temperature of 145.0 degrees Fahrenheit to the failure limit value of 146.0 degrees Fahrenheit at noon on September 19th) to calculate 5% (1.0 degree Fahrenheit divided by 22.0 degrees Fahrenheit). The industrial asset failure prediction system 1516 then uses the 5% normalized distance to failure to assign the extreme urgency of the unit failure risk to the maintenance request of the feedwater pump operator. The unit failure risk can be a criterion for quantity and associated with the probability of functional loss. The expected data value can be a possible number or mathematical object in the information sequence.

[0070] After determining the time-based failure risk and the unit failure risk, the failure risk of the industrial asset can optionally be determined based on the time-based failure risk and the unit failure risk, wherein the output failure prediction includes outputting the failure risk of the industrial asset, box 1614. The system combines the probability of equipment failure. For example, and without limitation, this can include the industrial asset failure prediction system 1516 combining the low urgency of the time-based failure risk assigned to feedwater pump 1B with the extreme urgency of the unit failure risk assigned to feedwater pump 1B, resulting in assigning high urgency to the failure risk of feedwater pump 1B. When outputting a failure prediction, the industrial asset failure prediction system 1516 outputs the high urgency of the failure risk assigned to feedwater pump 1B. The failure risk can be the probability of functional loss.

[0071] Determining failure risk can be based on the time weight assigned by the system user to the time failure risk and / or the unit weight assigned by the system user to the unit failure risk. For example, the industrial asset failure prediction system 1516 can combine the time failure risk and unit failure risk of feedwater pump 1B using the weights assigned by the system user, instead of combining the time failure risk and unit failure risk of feedwater pump 1B using the same default weights for both. If the system user assigns a time weight of 0.50 to the time failure risk and a unit weight of 2.0 to the unit failure risk, then the industrial asset failure prediction system 1516 can assign a very high urgency to the failure risk of feedwater pump 1B because the assigned weights downplay the factors contributing to the low urgency of the time failure risk while emphasizing the factors emphasizing the extreme urgency of the unit failure risk.

[0072] Time weight can be a measure of importance associated with measurements taken in chronological order. A system user can be a person operating a computer. Unit weight can be a measure of importance associated with a quantitative standard.

[0073] After determining the failure risk, optionally one of several discrete urgency levels can be identified based on that failure risk, wherein the output failure risk of the industrial asset includes outputting one of the identified discrete urgency levels, box 1616. The system identifies the urgency level of the probability of functional loss. For example, and not as a limitation, this could include the industrial asset failure prediction system 1516 assigning high urgency to the failure risk of feedwater pump 1B. When outputting a failure prediction, the industrial asset failure prediction system 1516 outputs high urgency for the failure risk of feedwater pump 1B. Discrete urgency levels can be distinctly different categories requiring different actions.

[0074] After identifying one of the urgency levels, a priority may be assigned to the failure risk of the industrial asset based on a corresponding one of the assigned discrete urgency levels, box 1618. The system assigns priority to the probability of functional failure. In an embodiment, this may include the industrial asset failure prediction system 1516 assigning the highest priority to the failure risk of the feedwater pump 1B, which is also assigned extreme urgency. Priorities may be based on the order in which processes are arranged and executed according to their assigned importance or urgency level.

[0075] After assigning priorities to the failure risks of industrial assets, alternatively, another priority can be assigned to another failure risk of another industrial asset, as shown in box 1620. The system prioritizes the probability of functional failure for all industrial assets. For example, and without limitation, this could include the industrial asset failure prediction system 1516 assigning a moderate overall priority to the failure risk of steam turbine 4. Although the failure risk of the turbine is based on pressure data and the failure risk of the pump is based on temperature data, the normalization of the time to failure and distance to failure for steam turbine 3 and feedwater pump 1B allows the operator to easily compare the overall priority, urgency level, and failure risk of steam turbine 3 and feedwater pump 1B.

[0076] After determining the failure time and distance from failure, the system outputs a failure prediction associated with the failure time and distance from failure for the industrial asset, box 1622. The system outputs a priority for the predicted probability of functional failure. For example, and not as a limitation, this could include the industrial asset failure prediction system 1516 outputting the highest priority for the failure risk of the feedwater pump 1B, given the extreme urgency of the distance from failure assigned to the feedwater pump 1B temperature, and despite the low urgency of the time from failure assigned to the feedwater pump 1B. The failure prediction can be a prediction associated with functional failure.

[0077] Although Figure 16 Blocks 1602-1622 are depicted occurring in a specific order, but blocks 1602-1622 may occur in another order. In other embodiments, each of blocks 1602-1622 may also be performed in combination with other blocks and / or some blocks may be divided into different sets of blocks.

[0078] Exemplary hardware devices in which this subject matter may be implemented will now be described. Those skilled in the art will recognize that... Figure 17 The components illustrated in the diagram may vary depending on the system implementation. (Reference) Figure 17 An exemplary system for implementing the subject matter disclosed herein includes a hardware device 1700, including a processing unit 1702, a memory 1704, a storage device 1706, a data entry module 1708, a display adapter 1710, a communication interface 1712, and a bus 1714 that couples the components 1704-1712 to the processing unit 1702.

[0079] Bus 1714 can include any type of bus architecture. Examples include memory bus, peripheral bus, local bus, etc. Processing unit 1702 is an instruction execution machine, device, or apparatus, and may include a microprocessor, digital signal processor, graphics processing unit, application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), etc. Processing unit 1702 can be configured to execute program instructions stored in memory 1704 and / or storage device 1706, and / or program instructions received via data entry module 1708.

[0080] Memory 1704 may include read-only memory (ROM) 1716 and random access memory (RAM) 1718. Memory 1704 may be configured to store program instructions and data during operation of hardware device 1700. In various embodiments, for example, memory 1704 may include any of a variety of memory technologies, such as static random access memory (SRAM) or dynamic RAM (DRAM), including variations such as double data rate synchronous DRAM (DDR SDRAM), error correction code synchronous DRAM (ECCSDRAM), or RAMBUS DRAM (RDRAM).

[0081] Memory 1704 may also include non-volatile memory technologies, such as non-volatile flash RAM (NVRAM) or ROM. It is anticipated that memory 1704 may include combinations of technologies such as those described above, as well as other technologies not specifically mentioned. When the subject matter is implemented in a computer system, a basic input / output system (BIOS) 1720 containing basic routines that facilitate the transfer of information between components within the computer system (such as during startup) is stored in ROM 1716.

[0082] Storage device 1706 may include a flash memory data storage device for reading and writing to flash memory, a hard disk drive for reading and writing to a hard disk, a disk drive for reading and writing to a removable disk, and / or an optical disc drive for reading and writing to a removable optical disc such as a CD-ROM, DVD, or other optical media. The drive and its associated computer-readable medium provide non-volatile storage of computer-readable instructions, data structures, program modules, and other data for hardware device 1700.

[0083] It should be noted that the methods described herein can be implemented in executable instructions stored in a computer-readable medium for use by or in connection with an instruction execution machine, apparatus, or device (such as a computer-based or processor-containing machine, apparatus, or device). Those skilled in the art will recognize that, for some embodiments, other types of computer-readable media capable of storing computer-accessible data, such as magnetic tape cassettes, flash memory cards, digital video discs, Bernoulli cartridges, RAM, ROM, etc., can also be used in the exemplary operating environment. As used herein, “computer-readable medium” can include one or more suitable media for storing executable instructions of a computer program in one or more of electronic, magnetic, optical, and electromagnetic formats, such that an instruction execution machine, system, apparatus, or device can read (or retrieve) the instructions from the computer-readable medium and execute the instructions to perform the described methods. A non-exhaustive list of common exemplary computer-readable media includes: portable computer floppy disks; RAM; ROM; erasable programmable read-only memory (EPROM or flash memory); optical storage devices, including portable optical discs (CDs), portable digital video discs (DVDs), high-definition DVDs (HD-DVD™), BLU-RAY discs; and so on.

[0084] Multiple program modules can be stored on storage device 1706, ROM 1716, or RAM 1718, including operating system 1722, one or more application programs 1726, program data 1726, and other program modules 1728. Users can input commands and information into hardware device 1700 via data input module 1708. Data input module 1708 may include mechanisms such as a keyboard, touchscreen, or pointing device.

[0085] Other external input devices (not shown) are connected to hardware device 1700 via external data input interface 1710. For example, and not limitingly, external input devices may include microphones, joysticks, game controllers, satellite antennas, scanners, etc. External input devices may include video or audio input devices, such as camcorders, still cameras, etc. Data input module 1708 may be configured to receive input from one or more users from hardware device 1700 and deliver such input to processing unit 1702 and / or memory 1704 via bus 1714.

[0086] Display 1712 is also connected to bus 1714 via display adapter 1710. Display 1712 can be configured to display the output of hardware device 1700 to one or more users. A given device (e.g., a touchscreen) can be used as both data entry module 1708 and display 1712. External display devices can also be connected to bus 1714 via external display interface 1734. Other peripheral output devices, such as speakers and printers (not shown), can also be connected to hardware device 1700.

[0087] Hardware device 1700 can operate in a networked environment using a logical connection to one or more remote nodes (not shown) via communication interface 1712. The remote node can be another computer, server, router, peer device, or other common network node, and typically includes many or all of the elements described above with respect to hardware device 1700. Communication interface 1712 can interface with wireless and / or wired networks. Examples of wireless networks include, for example, BLUETOOTH networks, wireless personal area networks, wireless 802.21 local area networks (LANs), and / or wireless telephone networks (e.g., cellular, PCS, or GSM networks).

[0088] Examples of wired networks include, for example, LANs, fiber optic networks, wired personal area networks, telephone networks, and / or wide area networks (WANs). Such networking environments are common in intranets, the Internet, office networks, enterprise-wide computer networks, etc. Communication interface 1712 may include logic configured to support direct memory access (DMA) transfers between memory 1704 and other devices.

[0089] In a networked environment, the program modules or portions thereof depicted relative to hardware device 1700 may be stored in a remote storage device, such as, for example, a server. It will be appreciated that other hardware and / or software may be used to establish communication links between hardware device 1700 and other devices.

[0090] It should be understood that, Figure 17The arrangement of the hardware device 1700 shown is merely one potential implementation, and other arrangements are possible. It should also be understood that the various system components (and parts) defined by the claims described below and illustrated in the block diagrams represent logical components configured to perform the functions described herein. For example, one or more of these system components (and parts) may be implemented wholly or partially by at least some of the components shown in the arrangement of the hardware device 1700.

[0091] Furthermore, while at least one of these components is implemented at least partially as an electronic hardware component and thus constitutes a machine, the other components may be implemented using software, hardware, or a combination of both. More specifically, at least one component as defined in the claims is implemented at least partially as an electronic hardware component, such as an instruction execution machine (e.g., a processor-based or processor-integrated machine) and / or as a dedicated circuit or circuit system (e.g., discrete logic gates interconnected to perform a dedicated function), such as... Figure 17 Those shown.

[0092] Other components can be implemented using software, hardware, or a combination of both. Furthermore, some or all of these other components can be combined, some can be omitted entirely, and additional components can be added, while still achieving the functionality described herein. Therefore, the subject matter described herein can be implemented in many different variations, and all such variations are within the scope of the claims.

[0093] In the above description, unless otherwise stated, the subject matter is described with reference to symbolic representations of actions and operations performed by one or more devices. Therefore, it is understood that such actions and operations (sometimes referred to as actions and operations performed by a computer) involve the manipulation of data in a structured form by a processing unit. This manipulation transforms the data or maintains it at a location in the computer's memory system, which reconfigures or otherwise alters the operation of the device in a manner well known to those skilled in the art. The data structure that maintains the data is a physical location in memory having specific properties defined by the format of the data. However, while the subject matter is described in context, this is not intended to be limiting, as those skilled in the art will recognize that the various actions and operations described below can also be implemented in hardware.

[0094] To facilitate understanding of the subject matter described above, many aspects are described in terms of sequences of actions. At least one of these aspects as defined by the claims is performed by an electronic hardware component. For example, it will be appreciated that various actions can be performed by a dedicated circuit or circuit system, by program instructions executed by one or more processors, or by a combination of both. The description of any sequence of actions herein is not intended to imply that the sequence must be performed in a specific order as described. Unless otherwise stated herein or explicitly contradicted by the context, all methods described herein can be performed in any suitable order.

[0095] While one or more embodiments have been described by way of example and specific examples, it should be understood that the one or more embodiments are not limited to the disclosed embodiments. Rather, it is intended to cover various modifications and similar arrangements that will be apparent to those skilled in the art. Therefore, the scope of the appended claims should be given the broadest interpretation to cover all such modifications and similar arrangements.

[0096] This disclosure describes in detail how a machine comprising one or more computers, including one or more processors and one or more non-transitory computer-readable media, can implement a system and its improvements over the prior art. Instructions executed by a machine cannot be executed in the human brain or derived by humans using pen and paper; rather, the machine needs to transform process input data into useful output data. Furthermore, the claims presented herein do not attempt to tie judicial exceptions to known conventional steps implemented by general-purpose computers; nor do they attempt to tie judicial exceptions by simply linking them to the technical field. In fact, the systems and methods described herein were not known and / or existed in the public domain at the time of filing, and they offer technical improvements and advantages unknown in the prior art. Moreover, the system includes unconventional steps that limit the claims to useful applications.

[0097] It should be understood that the system is not limited in its application to the details of the construction and arrangement of the components set forth in the foregoing description or shown in the accompanying drawings. The systems and methods disclosed herein fall within the scope of many embodiments. The foregoing discussion is presented to enable those skilled in the art to make and use embodiments of the system. Any part of the included structure and / or principles can be applied to any and / or all embodiments: it should be understood that features from some embodiments presented herein can be combined with other features according to some other embodiments. Therefore, some embodiments of the system are not intended to be limited to what is shown, but are to be given the broadest scope consistent with all the principles and features disclosed herein.

[0098] Any text in the accompanying drawings is part of the system disclosure and is to be readily incorporated into any description of the system boundaries. Any functional language in the accompanying drawings is a reference to the system configured to perform said functions, and the structures shown or described in the drawings should be considered as including the system containing said structures. Any graphics depicting content to be displayed on a graphical user interface are a disclosure of the system configured to generate and display the content of the graphical user interface. It should be understood that using descriptions of images in the accompanying drawings to define the boundaries of the system does not require a corresponding textual description in a written specification to fall within the scope of this disclosure.

[0099] Furthermore, as the applicant's own lexicographer, the applicant assigns explicit meaning to the following terms and / or waives the scope of the claims: the applicant defines any use of "and / or," such as, for example, "A and / or B" or "at least one of A and / or B," as referring only to element A, only to element B, or elements A and B together. Additionally, for example, the statements "at least one of A, B, and C," "at least one of A, B, or C," or "at least one of A, B, or C, or any combination thereof," are each defined as referring only to element A, only to element B, only to element C, or any combination of elements A, B, and C, such as AB, AC, BC, or ABC. As used herein, "can" or "may" or their derivatives (e.g., the system display may show X) are used for descriptive purposes only and should be understood as synonymous with and / or interchangeable with "configured to" (e.g., the computer is configured to execute instruction X) in defining the boundaries and scope of the system. The phrase "configured to" also indicates the steps of configuring a structure or computer to perform a function.

[0100] It should be understood that the wording and terminology used herein are descriptive and should not be considered limiting. The terms “comprising,” “including,” or “having,” and variations thereof, as used herein, are intended to cover the items listed thereafter and their equivalents, as well as additional items. Unless otherwise specified or limited, the terms “installation,” “connection,” “support,” and “coupling,” and variations thereof, are used extensively and cover both direct and indirect installation, connection, support, and coupling. Furthermore, “connection” and “coupling” are not limited to physical or mechanical connections or couplings.

[0101] Refer to the preceding detailed description in conjunction with the figures, where similar elements in different figures have similar reference numerals. The figures, which are not necessarily drawn to scale, depict some embodiments and are not intended to limit the scope of embodiments of the system.

[0102] Any operation that forms part of the system described herein is a useful machine operation. The system also relates to devices or apparatuses for performing these operations. All flowcharts presented herein represent computer-implemented steps and / or visual representations of algorithms implemented by the system.

[0103] Devices can be specifically configured for a particular purpose, such as a dedicated computer. When defined as a dedicated computer, the computer can also perform other processing, program execution, or routines that are not specific to that purpose, while still being able to operate for that specific purpose. Alternatively, operations can be handled by a general-purpose computer selectively activated or configured by one or more computer programs stored in computer memory, cache, or obtained via a network. When data is obtained via a network, the data can be processed by other computers on the network (e.g., a cloud of computing resources).

[0104] An embodiment of the system can also be defined as a machine that transforms data from one state to another. The data can represent an object, which can be represented as an electronic signal and the data manipulated electronically. In some cases, the transformed data can be visually depicted on a display, thereby representing the physical object resulting from the data transformation. The transformed data can be stored in a storage device, either generally or in a specific format that enables the construction or depiction of physical and tangible objects. This manipulation can be performed by a processor.

[0105] In such an example, the processor thus transforms data from one thing to another. Furthermore, some embodiments include methods that can be processed by one or more machines or processors that can be connected via a network. Each machine can transform data from one state or thing to another, and can also process the data, store the data in a storage device, transmit the data over a network, display the results, or transmit the results to another machine. As used herein, computer-readable storage medium means a physical or tangible storage device (as opposed to a signal), and includes, but is not limited to, volatile and non-volatile, removable and non-removable storage media implemented in any method or technology for the tangible storage of information such as computer-readable instructions, data structures, program modules, or other data.

[0106] While the method operations are presented in a specific order according to some embodiments, the execution of these steps does not necessarily occur in the listed order unless explicitly specified. Furthermore, other housekeeping operations may be performed between operations, operations may be adjusted to occur at slightly different times, and / or operations may be distributed across a system that allows processing operations to occur at various intervals associated with the processing, as long as the processing of the superimposed operations is performed in the desired manner and produces the desired system output. Those skilled in the art will recognize that while the system has been described above in conjunction with specific embodiments and examples, the system is not necessarily limited thereto, and many other embodiments, examples, uses, modifications and deviations from the described embodiments, examples, and uses are intended to be covered by the appended claims. The full disclosure of each patent and publication cited herein is incorporated by reference as if each such patent or publication were individually incorporated herein by reference. Various features and advantages of the system are set forth in the following claims.

Claims

1. A system for predicting industrial asset failures, the system comprising: One or more processors; as well as A non-transitory computer-readable medium storing multiple instructions, which, when executed, cause the one or more processors to: Determine the data start value associated with the industrial assets at the data start time; Determine the data end value associated with the industrial assets at the data end time; Estimate the failure time when the trend calculated from the start value to the end value of the data will reach the failure limit; The distance to the fault is determined based on the fault limit value and the data end value. as well as Outputs fault predictions for industrial assets, correlated with fault time and distance from fault.

2. The system of claim 1, wherein the output fault prediction includes outputting a time to fault, the time to fault being based on the fault time and the data end time, and the time to fault being normalized based on the data start time, the data end time and the fault time, and the distance to fault being normalized based on the expected data value, the data end value and the fault limit value.

3. The system of claim 2, wherein the plurality of instructions further cause the processor to: Determining time-of-failure risk based on normalized time-of-failure distance; Determining unit failure risk based on normalized distance to failure; as well as The failure risk of industrial assets is determined based on time failure risk and unit failure risk, wherein the output failure forecast includes the failure risk of the output industrial assets.

4. The system of claim 3, wherein determining the failure risk of an industrial asset based on time failure risk and unit failure risk is further based on at least one of a time weight assigned by the system user to time failure risk or a unit weight assigned by the system user to unit failure risk.

5. The system of claim 3, wherein determining the risk of time failure is based on an adjustment to the required minimum response time.

6. The system of claim 3, wherein the plurality of instructions further cause the processor to assign a level corresponding to a plurality of discrete urgency levels to the failure risk, wherein outputting the failure risk of the industrial asset includes outputting the level corresponding to the level assigned to the plurality of discrete urgency levels.

7. The system of claim 6, wherein the plurality of instructions further cause the processor to: Based on the assigned priority of failure risk for industrial assets within the multiple discrete urgency levels; and Assign another priority to another failure risk of another industrial asset based on another of the multiple discrete urgency levels.

8. A computer-implemented method for predicting industrial asset failures, the computer-implemented method comprising: Determine the data start value associated with the industrial assets at the data start time; Determine the data end value associated with the industrial assets at the data end time; Estimate the failure time when the trend calculated from the start value to the end value of the data will reach the failure limit; The distance to the fault is determined based on the fault limit value and the data end value. as well as Outputs fault predictions for industrial assets, correlated with fault time and distance from fault.

9. The computer-implemented method of claim 8, wherein the output fault prediction includes outputting a time to fault, the time to fault being based on a fault time and a data end time, and the time to fault being normalized based on a data start time, a data end time, and a fault time, and the distance to fault being normalized based on an expected data value, a data end value, and a fault limit value.

10. The computer-implemented method of claim 9, wherein the computer-implemented method further comprises: Determining time-of-failure risk based on normalized time-of-failure distance; Determining unit failure risk based on normalized distance to failure; as well as The failure risk of industrial assets is determined based on time failure risk and unit failure risk, wherein the output failure forecast includes the failure risk of the output industrial assets.

11. The computer-implemented method of claim 10, wherein determining the failure risk of an industrial asset based on time failure risk and unit failure risk is further based on at least one of a time weight assigned by the system user to time failure risk or a unit weight assigned by the system user to unit failure risk.

12. The computer-implemented method of claim 10, wherein determining the risk of time failure is based on an adjustment to the required minimum response time.

13. The computer-implemented method of claim 10, wherein the computer-implemented method further comprises assigning a corresponding level from a plurality of discrete urgency levels to the failure risk, wherein outputting the failure risk of the industrial asset includes outputting the corresponding level assigned from the plurality of discrete urgency levels.

14. The computer-implemented method of claim 13, wherein the computer-implemented method further comprises: Priority is assigned to the failure risk of industrial assets based on the corresponding level assigned among the multiple discrete urgency levels. as well as Assign another priority to another failure risk of another industrial asset based on another of the multiple discrete urgency levels.

15. A computer program product comprising a non-transitory computer-readable medium having computer-readable program code implemented therein, executable by one or more processors, the program code including instructions for performing the following operations: Determine the data start value associated with the industrial assets at the data start time; Determine the data end value associated with the industrial assets at the data end time; Estimate the failure time when the trend calculated from the start value to the end value of the data will reach the failure limit; The distance to the fault is determined based on the fault limit value and the data end value. as well as Outputs fault predictions for industrial assets, correlated with fault time and distance from fault.

16. The computer program product of claim 15, wherein the output fault prediction includes outputting a time to fault, the time to fault being based on a fault time and a data end time, and the time to fault being normalized based on a data start time, a data end time, and a fault time, and the distance to fault being normalized based on an expected data value, a data end value, and a fault limit value.

17. The computer program product of claim 16, wherein the program code includes further instructions for performing the following operations: Determining time-of-failure risk based on normalized time-of-failure distance; Determining unit failure risk based on normalized distance to failure; as well as The failure risk of industrial assets is determined based on time failure risk and unit failure risk, wherein the output failure forecast includes the failure risk of the output industrial assets.

18. The computer program product of claim 17, wherein determining the failure risk of an industrial asset based on time failure risk and unit failure risk is further based on at least one of a time weight assigned by the system user to time failure risk or a unit weight assigned by the system user to unit failure risk.

19. The computer program product of claim 17, wherein the determination of time failure risk is based on an adjustment to the required minimum response time.

20. The computer program product of claim 17, wherein the program code includes further instructions for performing the following operations: Assign a level from a plurality of discrete urgency levels to the failure risk, wherein the failure risk of the output industrial asset includes the output of the level assigned from the plurality of discrete urgency levels. Priority is assigned to the failure risk of industrial assets based on the corresponding level assigned among the multiple discrete urgency levels. as well as Assign another priority to another failure risk of another industrial asset based on another of the multiple discrete urgency levels.

Images