Machine learning-based time-to-event models to optimize asset health and failure prediction.

Abstract

A system is provided for determining an asset cleaning schedule. In some embodiments, the system may include a plurality of sensors tuned to monitor the asset, and a computing system, the computing system including at least one data processor and memory for storing instructions, the instructions causing the at least one data processor to perform an action when executed by the at least one data processor. In some embodiments, the actions performed by the processor may include receiving data from the plurality of sensors characterizing the asset's operational efficiency, determining the asset's operational efficiency, determining an operational efficiency threshold characterizing undesirable operational efficiency, determining an asset cleaning schedule based on the asset's operational efficiency and the operational efficiency threshold using an optimization algorithm, and providing the cleaning schedule.

Description

Technical Field

[0001] (Related Application) This application claims priority to U.S. Provisional Patent Application No. 63 / 459,684, filed on April 16, 2023, under 35 U.S.C. § 119(e), the entire content of which is hereby expressly incorporated by reference herein.

Background Art

[0002] In the process of operating industrial assets such as heat exchangers and compressors, the accumulation of unwanted deposits or contaminants may occur on their surfaces. This accumulation is called fouling. Fouling within industrial assets can affect the operating efficiency of the assets, increase operating costs, and potentially damage the assets. For example, fouling can potentially impede heat transfer efficiency in a heat exchanger, restrict flow paths within heat exchangers and compressors, leading to potential operating problems such as reduced flow rate, increased pressure loss, and cavitation.

Summary of the Invention

[0003] In one aspect, a method for determining a cleaning schedule for an asset is provided. In some aspects, the method includes obtaining data characterizing the operating efficiency of the asset over time from a plurality of sensors adjusted to monitor the asset, receiving the data characterizing the operating efficiency of the asset via a computing system including at least one data processor and a memory storing instructions, determining, by the at least one data processor, the operating efficiency of the asset and an operating efficiency threshold characterizing an undesirable operating efficiency, determining, by the at least one data processor, a cleaning schedule for the asset based on the operating efficiency of the asset and the operating efficiency threshold, and providing the cleaning schedule.

[0004] In some embodiments, the cleaning schedule includes one or more light cleanings for non-invasive cleaning of the asset. In some embodiments, the cleaning schedule includes one or more heavy cleanings for mechanical cleaning of the asset.

[0005] In some embodiments, the method may further include, by at least one data processor, determining the time to a predicted event corresponding to the time to reach an operating efficiency threshold, and providing the time to the event.

[0006] In some embodiments, the method may further include receiving data from an asset via a computing system that characterizes one or more operating states of the asset, and using an optimization algorithm, by at least one data processor, determining a cleaning schedule based on the asset's operating efficiency, an operating efficiency threshold, and the data characterizing one or more operating states of the asset.

[0007] In some embodiments, the method may further include, using an optimization algorithm, determining an optimized operating schedule for the operation of an asset under one or more operating conditions, by at least one data processor. In some embodiments, the time to a predicted event may be determined based on the cleaning schedule and the optimized operating schedule.

[0008] In some embodiments, the method may further include determining the amount of fouling that occurred within the asset by at least one data processor.

[0009] In some embodiments, the asset may be a heat exchanger or a compressor, and the multiple sensors may include at least one of a temperature sensor and a pressure sensor.

[0010] In some embodiments, the operating efficiency threshold can be determined based on at least one of the asset's operating history data, client preferences, predetermined efficiency requirements, and predetermined energy consumption thresholds.

[0011] In another embodiment, a system is provided for determining an asset cleaning schedule. In some embodiments, the system may include a plurality of sensors tuned to monitor an asset, at least one data processor, and a computing system including memory for storing instructions, wherein when an instruction is executed by the at least one data processor, the computing system causes the at least one data processor to perform an action. In some embodiments, the actions performed by the processor may include receiving data from the plurality of sensors characterizing the operational efficiency of the asset, determining the operational efficiency of the asset, determining an operational efficiency threshold characterizing undesirable operational efficiency, determining an asset cleaning schedule based on the operational efficiency of the asset and the operational efficiency threshold using an optimization algorithm, and providing the cleaning schedule.

[0012] In some embodiments, the cleaning schedule includes one or more light cleanings for non-invasive cleaning of the asset. In some embodiments, the cleaning schedule includes one or more heavy cleanings for mechanical cleaning of the asset.

[0013] In some embodiments, the operations performed by the processor may further include determining the time to a predicted event, which corresponds to the time it takes to reach an operating efficiency threshold, and providing the time to the event.

[0014] In some embodiments, the operations performed by the processor may further include receiving data from the asset that characterizes one or more operating states of the asset, and using an optimization algorithm to determine a cleaning schedule based on the asset's operating efficiency, an operating efficiency threshold, and the data characterizing one or more operating states of the asset.

[0015] In some embodiments, the operations performed by the processor may further include, by at least one data processor, determining an optimized operating schedule for the operation of an asset under one or more operating conditions using an optimization algorithm. In some embodiments, the time to a predicted event may be determined based on the cleaning schedule and the optimized operating schedule.

[0016] In some embodiments, the actions performed by the processor may further include determining the amount of fouling that has occurred within the asset.

[0017] In some embodiments, the asset may be a heat exchanger or a compressor, and the multiple sensors may include at least one of a temperature sensor and a pressure sensor.

[0018] In some embodiments, the operating efficiency threshold can be determined based on at least one of the asset's operating history data, client preferences, predetermined efficiency requirements, and predetermined energy consumption thresholds. [Brief explanation of the drawing]

[0019] [Figure 1] This is a block diagram of an exemplary asset monitoring system described herein, configured to monitor the operational efficiency of assets and determine various non-invasive and invasive cleaning schedules for those assets. [Figure 2] Figure 1 shows multiple graphs of exemplary non-invasive and invasive cleaning schedules for assets, as determined by the asset monitoring system. [Figure 3] This is an exemplary graph showing the operating efficiency of an asset over time, using an asset operating in multiple operating modes and predictive analysis performed by the asset monitoring system described herein to predict the future operating efficiency of the asset. [Figure 4]Figure 3 is an exemplary graph illustrating the directive analysis performed by the asset monitoring system described herein on the operating efficiency data, in order to optimize asset performance by defining operating schedules for multiple operating modes of the asset. [Figure 5] This flowchart illustrates an exemplary method for determining directive non-invasive and invasive cleaning schedules for assets using the systems and methods described herein.

[0020] Please note that the drawings are not necessarily to scale. The drawings are intended to depict only typical embodiments of the subject matter disclosed herein and should not be considered to limit the scope of this disclosure. [Modes for carrying out the invention]

[0021] During the operation of industrial assets such as heat exchangers and compressors, operational degradation known as fouling can occur, resulting in the accumulation of unwanted deposits or contaminants on their surfaces. Fouling within industrial assets can affect the operational efficiency of the asset, increase operating costs, and potentially damage the asset. For example, fouling can impair heat transfer efficiency in heat exchangers, restrict flow paths within heat exchangers and compressors, leading to potential operational problems such as reduced flow rates, increased pressure losses, and cavitation. Fouling can be caused by impurities in the heated material, biological impurities if water is present, or impurities in other areas. If appropriate measures are not taken to remove fouling from the asset, such as through non-invasive cleaning operations (e.g., decoking operations) or thorough mechanical cleaning of the asset, fouling can cause operational inefficiencies and operational failures. Traditionally, fouling in assets has been addressed retrospectively, depending on whether the asset is operating at undesirable efficiency or causing operational failures requiring attention. Therefore, in order to extend asset lifespan, improve efficiency, and reduce asset downtime, it is desirable to predict fouling within assets and provide information on when to proactively clean up assets.

[0022] The systems and methods described herein combine the use of sensors configured to monitor an asset with a computing system tuned to receive data from the sensors characterizing the asset's operational efficiency. Using the data characterizing the asset's operational efficiency and operational efficiency thresholds that characterize undesirable operational efficiency, the computing system can determine various non-invasive and invasive cleaning schedules for the asset, as will be described in more detail below.

[0023] The systems and methods described herein advantageously combine the use of machine learning algorithms with descriptive data characterizing asset efficiency / soundness to provide various non-invasive and invasive cleaning schedules for assets, extending asset life, improving efficiency, and reducing asset downtime. The systems and methods described herein can be applied to any industrial asset whose operating efficiency decreases by experiencing performance degradation over time and / or an asset that is susceptible to the effects of abnormal behavior. The described asset monitoring systems and methods utilize a proactive / predictive model to provide an intervention schedule aimed at managing fouling, enhancing efficiency, and extending asset life.

[0024] Figure 1 shows an exemplary asset monitoring system 100 for determining various non-invasive and invasive cleaning schedules for asset 110 based on the asset's operating efficiency over time. As shown in Figure 1, the system 100 may include a plurality of sensors 120a, 120b configured to monitor the asset's condition indicators (e.g., indicators of operating efficiency) over time. In some embodiments, asset 110 may be an industrial asset (e.g., a heat exchanger, a compressor). As shown in Figure 1, the plurality of sensors 120a, 120b may be configured to detect a plurality of condition indicators of asset 110 while asset 110 is in operation. The condition indicators monitored by sensors 120a, 120b can be used by the system to determine the change in the asset's operating efficiency over time as the asset deteriorates. For example, in some embodiments, asset 100 may be a heat exchanger or a compressor, and a plurality of sensors 120a, 120b may include temperature sensors and / or pressure sensors configured to measure various temperatures and / or pressures of asset 110 (e.g., inlet temperature / pressure and outlet temperature / pressure). Based on the temperature and / or pressure measurements, system 100 can determine the operating efficiency of the asset and / or the degradation of asset 110 over time, as will be described in more detail below. In some embodiments, condition indicators monitored by sensors 120a, 120b may inform system 100 of the amount of fouling occurring in asset 110 that could cause a decrease in the operating efficiency and degradation of the system.

[0025] System 100 can also include a computing system 130 that includes at least one data processor 140 and a memory 150 that stores instructions. In some embodiments, the computing device 130 can further include a user interface display 160. The computing system can receive data characterizing the operating efficiency of an asset acquired by a plurality of sensors 120a, 120b. Using the data received from the plurality of sensors 120a, 120b, the processor 140 can be configured to determine the operating efficiency of the asset 110. For example, if the plurality of sensors 120a, 120b are temperature sensors, the operating efficiency can be determined based on the temperature ratio from the plurality of sensors 120a, 120b (e.g., using the logarithmic mean temperature difference (LMTD) method, etc.). In some embodiments, the processor 140 can also determine an operating efficiency threshold characterizing an undesirable operating efficiency of the asset 110. In some embodiments, the operating efficiency threshold can be determined based on the operating history data of the asset, the client's preferences, predetermined efficiency requirements, and predetermined energy consumption thresholds. For example, in some cases, the client may desire that the asset 110 operate at a certain level of efficiency to comply with industry efficiency standards or to reduce operating / energy costs. In some embodiments, the operating efficiency threshold can be determined by the processor 140 using the measurements received from the plurality of sensors 120a, 120b. For example, in some embodiments, the system 100 can be configured to determine the operating efficiency threshold based on a predetermined LMTD threshold corresponding to a predetermined undesirable efficiency set by the system 100. The operating efficiency threshold is an undesirable operating efficiency when the asset 110 reaches it and can represent an operating efficiency at which maintenance is required to improve the performance of the asset 110.

[0026] Therefore, in some embodiments, based on the asset's operating efficiency and operating efficiency threshold, the processor 140 can determine a cleaning schedule for the asset 110 based on the asset's operating efficiency and operating efficiency threshold over time, as will be described in more detail below, and provide the cleaning schedule to the user.

[0027] Figure 2 shows two graphs 200 and 230 of the operational efficiency of assets over time. As shown in Figure 2, graphs 200 and 230 may each include cleaning schedules 215 and 245, respectively, determined using the systems and methods described herein (e.g., the asset monitoring system 100 in Figure 1). Exemplary graphs 200 and 230 shown in Figure 2 are described below with reference to Figure 1.

[0028] As described above, in some embodiments, the system described herein may include a plurality of sensors 120a, 120b configured to monitor asset 110 for a status indicator (e.g., an indicator of operating efficiency). For example, if asset 110 is a heat exchanger and the plurality of sensors 120a, 120b are temperature sensors, temperature readings from the sensors can be received by the computing system 130. Using the temperature readings from the sensors, as shown in Graph 200, the processor 140 can determine the operating efficiency 205 of the asset over time. As described above, in some embodiments, the operating efficiency 205 can be determined based on the temperature ratio from the plurality of sensors 120a, 120b (e.g., using the log-mean temperature difference (LMTD) method). The processor 140 can also determine an operating efficiency threshold 210 that characterizes undesirable operating efficiency of the asset. In some embodiments, the operating efficiency threshold 210 can be determined based on a predetermined LMTD threshold. In some embodiments, the operating efficiency threshold can be determined based on asset operating history data, client preferences, predetermined efficiency requirements, and / or a predetermined energy consumption threshold. For example, when asset 110 was operating above an operating efficiency threshold 210, system 100 can determine that maintenance was needed or that asset 110 had failed, based on the operating history data of asset 110 (stored in the memory 150 of computing system 130 and / or in a remote database). In another example, if client and / or industry standards require asset 110 to operate at a certain level of efficiency (e.g., to ensure that energy consumption costs are minimized), system 100 can determine the operating efficiency threshold 210 based on user input indicating client preferences, predetermined efficiency requirements, and / or predetermined energy consumption thresholds, as described above. In some embodiments, the operating efficiency threshold 210 may further include a time T1 to a predicted event, corresponding to the time it takes to reach the operating efficiency threshold 210.

[0029] As shown in Graph 200, the system 100 can also determine an asset cleaning schedule 215 based on the asset's operating efficiency 205 and operating efficiency threshold 210. The cleaning schedule 215 can represent a directive schedule for cleaning the asset over its entire operating period until the asset's operating efficiency reaches the operating efficiency threshold 210. In some embodiments, the cleaning schedule 215 may include one or more light cleanings 220a-220d and / or one or more heavy cleanings 220e. In some embodiments, the one or more light cleanings 220a-220d and / or one or more heavy cleanings 220e in the cleaning schedule 215 can be determined by the processor 140 at regular intervals. However, in some embodiments, the cleaning schedule can be determined by the processor 140 using an optimization algorithm, as will be described in more detail below. The one or more light cleanings 220a-220d in the cleaning schedule 215 may indicate to the user that non-invasive cleaning of the asset 110 is required. In some embodiments, non-invasive cleaning operations to be performed between one or more specified mild cleanings 220a–220d may include, but are not limited to, chemical decoking, steam-air decoking, or inspoling decoking. In some embodiments, the systems and methods described herein may be configured to instruct asset 110 to perform one or more specified mild cleanings 220a–220d at scheduled times defined by the cleaning schedule 215 when a human operator is unable to clean asset 110. In some embodiments, one or more heavy cleanings 220e within the cleaning schedule 215 may indicate to the user that asset 110 requires more thorough / invasive cleaning (e.g., due to excessive material fouling within asset 110). In some embodiments, invasive cleaning operations to be performed between one or more specified heavy cleanings 220e may include, but are not limited to, mechanical and / or manual cleaning of the asset by an operator.In some embodiments, the graph 200 and / or the cleaning schedule 215 may be provided on the user interface display 160 of the system 100 for the client / user to view. For example, in some cases, the cleaning schedule 215 may be provided as one or more future times corresponding to times when one or more light / heavy cleanings are scheduled to be performed.

[0030] As described above, in some embodiments, the cleaning schedule can be determined by the processor 140 using an optimization algorithm. Graph 230 illustrates this optimization process. As shown in Graph 230, the processor 140 can determine the asset's operating efficiency 235 over time and an operating efficiency threshold 240 that characterizes undesirable operating efficiency of the asset. In some embodiments, the operating efficiency 235 and the operating efficiency threshold 240 can be determined as described above, with reference to Graph 200. In the case of Graph 230, the system 100 can also determine an optimized cleaning schedule 245 for the asset based on the asset's operating efficiency 205 and the operating efficiency threshold 210. Similar to the cleaning schedule 215, the optimized cleaning schedule 245 may include one or more light cleanings 250a-250d and / or one or more heavy cleanings 250e. In some embodiments, the operating efficiency threshold 210 may further include a time T2 to a predicted event, which corresponds to the time it would take to reach the operating efficiency threshold 210 once the optimized cleaning schedule 245 is defined. The optimized cleaning schedule 245 can be determined by the processor 140 using one or more optimization algorithms. In some embodiments, the optimization algorithms can be trained using operational efficiency data stored in memory 150. For example, in some embodiments, the systems and methods described herein can perform simulations using a machine learning algorithm to optimize the total operating duration with intervention intervals (e.g., one or more light cleanings 250a-250d and / or one or more heavy cleanings 250e) as the optimization variable. In some embodiments, the machine learning / optimization algorithm may include a Monte Carlo algorithm that indicates the optimal time to intervene using either a light cleaning run or a heavy cleaning run to extend the operating duration of the asset. By using a machine learning / optimization algorithm as described herein, the system 100 can favorably increase the operating duration of the asset 110.For example, as shown in Figure 2, the time to the predicted event when cleaning schedule 215 is specified is T1, and the time to the predicted event when the optimized cleaning schedule 225 is specified is T2. Therefore, the time to the predicted event can be increased by ΔT(T2-T1) when the optimized cleaning schedule 245 is specified.

[0031] In some embodiments, the monitored asset (e.g., asset 110 in Figure 1) can be configured to operate in multiple operating modes. For example, in some cases, the asset may be a heat exchanger that can operate in a first mode and a second mode. Thus, in some embodiments, the systems and methods described herein (e.g., computing system 130) can also be configured to receive data characterizing one or more operating states of the asset, which can be used to generate an optimized cleaning schedule, as will be described in more detail below. In some embodiments, the data characterizing one or more operating states of the asset can be received directly from the asset.

[0032] Figure 3 shows another exemplary graph 300 illustrating the operating efficiency of an asset over time when the asset is operating in multiple operating modes. The exemplary graph 300 shown in Figure 3 is described below with reference to Figures 1 and 2. As shown in Figure 3, graph 300 includes a first operating efficiency 305a of the asset when it is operating in a first mode and a second operating efficiency 305b of the asset when it is operating in a second mode. In some embodiments, the operating efficiency (shown on the Y axis) can be measured in units of temperature or can be a unitless measurement of the operating efficiency. For example, if the sensor monitoring the asset is a temperature sensor and the operating efficiency is determined using LMTD, then, as described above, the operating efficiency can be measured in units of temperature. In some embodiments, if the temperature sensor and the operating efficiency are determined using the LMTD ratio, then the operating efficiency can be a unitless measurement of efficiency. In another example, if the asset is a compressor or pump, and the sensor monitoring the asset is a pressure sensor, then the operating efficiency can be a unitless measurement of efficiency determined, for example, by comparing the actual pressure output of the asset with a reference value or a desired output. In some embodiments, when the sensor is a pressure sensor, the operating efficiency can be a pressure difference measured in units of pressure. In some embodiments, the processor 140 can receive operating indicators from multiple sensors 120a, 120b over a period of time (e.g., from day 0 to the current date T1), as described above with reference to Figures 1-2, and can determine the operating efficiency 305a, 305b of the asset over time and an operating efficiency threshold 310 that characterizes the asset's undesirable operating efficiency. In some embodiments, as shown in Figure 3, the data received from the sensors of the system 100 can also indicate the time the asset has been operating in a standby state (or has stopped operating). As shown in Graph 300, the asset may have a lower efficiency 305b while operating in a second mode and a higher efficiency 305a while operating in a first mode. The system 100 can also determine an optimized cleaning schedule 315 for the asset based on the asset's operating efficiency 305a, 305b and the operating efficiency threshold 310.As described above, the optimized cleaning schedule 315 may include one or more cleaning cycles 320a-320b. In some embodiments, the one or more cleaning cycles 320a-320b may include both light and heavy cleaning cycles, as described above. In some embodiments, the operating efficiency threshold 310 may further include a time to a predicted event T2, which corresponds to the time it takes to reach the operating efficiency threshold 310 once the optimized cleaning schedule 315 is defined. In some embodiments, the optimized cleaning schedule 315 may be provided to the user interface for the user to view, as described above. In some embodiments, the user may interact with the display 160 to request information corresponding to the time to a predicted event (at T2) for the asset during its operating window. In some embodiments, the user may also request a prediction of the asset's degradation profile via the display 160, with and / or without intervention, as defined in the cleaning schedule 315. Based on a request provided by the user, the processor 140 can be configured to retrieve data stored in the computing system's memory 150 and provide predictive analytics regarding the time to predicted events (in T2), as well as the development of asset operational efficiency with and / or without intervention.

[0033] If the user wishes to operate the asset in the first and second modes as they have done in the past, the optimized cleaning schedule 315 can easily predict future operating modes, corresponding asset operating efficiencies 305a', 305b', and the optimized cleaning schedule 315 using the optimization algorithm described herein. However, in some embodiments, the user may interact with the display 160 to request that the system 100 propose an instructive operating schedule and a cleaning schedule for operating and cleaning the asset in a manner that maximizes the asset's operating life. In this case, the system and method described herein can be configured to further determine an optimized operating schedule for the asset in addition to the optimized operating schedule, using operating indicators received from a plurality of sensors 120a, 120b over a period of time (e.g., from day 0 to the current date T1) and the optimization algorithm described herein, as will be described in more detail below.

[0034] Figure 4 shows another exemplary graph 400, similar to that shown in Figure 3, illustrating the operating efficiency of an asset over time when the asset is operating in multiple operating modes. The exemplary graph 400 shown in Figure 4 is described below with reference to Figures 1 to 3. As shown in Figure 4, graph 400 includes a first operating efficiency 405a of the asset when it is operating in a first mode and a second operating efficiency 405b of the asset when it is operating in a second mode. In some embodiments, the processor 140 can receive operating indicators from multiple sensors 120a, 120b over a period of time (e.g., from day 0 to the current date T1) and determine the operating efficiency 405a, 405b of the asset over time and an operating efficiency threshold 410 that characterizes undesirable operating efficiency of the asset, similar to those described above. The data received from the sensors of the system 100 may also indicate the time the asset has been operating in a standby state (or has been shut down). As described above, in some cases, a user of the system and method may interact with the display 160 to request that the system 100 propose an instructive schedule that includes both an optimized operating schedule and an optimized cleaning schedule for operating and cleaning the asset in a manner that maximizes the asset's operating life. In this case, the system and method described herein may be configured to further determine an optimized schedule 415 for the asset using operating indicators received from a plurality of sensors 120a, 120b over a period of time (e.g., from day 0 to the current date T1) and the optimization algorithm described herein. The optimized schedule 415 may include an optimized operating schedule that informs the user of the best way to combine the operation of the asset in a first operating mode and a second operating mode in order to extend the asset's life. Furthermore, the optimized schedule 415 may include one or more cleanings 420a-420b, including a mixture of light and / or heavy cleaning, as described above. For example, the optimized schedule may include a number of instruction times 425 for operating the asset in a first operating mode and a number of instruction times 430 for operating the asset in a second operating mode.

[0035] Furthermore, in some embodiments, the user may be given the option to specify one of the optimized operating schedules 425, 430 and one or more washes 420a to 420b. Thus, if the user provides input to the user interface indicating that they wish to specify one or more washes 420a to 420b but do not wish to specify the optimized operating schedules 425, 430, the system can be configured to determine a time T2 to a first predicted event, corresponding to the time it takes to reach the operating efficiency threshold 410, assuming that one or more washes 420a to 420b are specified, but the user wishes to operate the asset in the first and second modes as they have done in the past. In this case, the system can easily predict future operating modes, the corresponding operating efficiencies 405a', 405b' of the asset, and one or more washes 420a to 420b using the optimization algorithm described above. Alternatively, if the user provides an input to the user interface indicating that they wish to define a complete optimized schedule 415 including one or more washes 420a-420b and optimized operating schedules 425, 430, the system can be configured to determine a time T3 to a second predicted event, corresponding to the time it takes to reach the operating efficiency threshold 410 once the complete optimized schedule 415 is defined. In this case, the system can use state indicator data received from sensors and operating history data received from assets, along with the optimization algorithm described above, to predict the optimized operating schedules 425, 430, the operating efficiencies 405a”, 405b”, and one or more washes 420a-420b for the corresponding assets. In some embodiments, the optimization algorithm can be trained using the operating efficiency data as described above. Furthermore, in some embodiments, the optimization algorithm can be trained using the operating history schedules of the asset or other similar assets. For example, in some embodiments, the systems and methods described herein can be simulated using machine learning algorithms to optimize the operating schedule and intervention intervals (e.g., one or more mild / severe washes 420a-420b) as optimization variables.By defining a fully optimized schedule 415, the system 100 can favorably increase the operating duration of asset 110. For example, as shown in Figure 4, if one or more washes 420a to 420b are defined but the optimized operating schedules 425 and 430 are not defined, the predicted time to event is T2, and if the fully optimized schedule 415 is defined, the predicted time to event is T3. Thus, the predicted time to event can be increased by ΔT(T3-T2) when the fully optimized schedule 415 is defined. In some embodiments, the prediction results based on defining only a portion of the optimized schedule 415 and the prediction results based on defining the fully optimized schedule 415 (including both the predicted times to event T2 and T3, and ΔT) can be provided to the user interface for the user to view.

[0036] As described above, the cleaning events described herein may include mechanical cleaning required by the user, and alternative interventions such as light, non-invasive cleaning in the absence of an operator. For example, the user may request, via the user interface display of the system described herein, that the asset monitoring system include provisions for light cleaning of assets in the cleaning schedule described herein, such as decoking (e.g., inspoling, steam-air decoking, chemical decoking, etc.). Both the light and heavy cleaning events defined by the systems and methods described herein advantageously allow the user to extend the operating window of those assets while maintaining operating efficiency below a determined operating efficiency threshold.

[0037] In some embodiments, the systems and methods described herein can be further configured to monitor a fleet of assets, where each asset may have different operating efficiencies depending on a different degree of fouling. Conventionally, in this case, it can be difficult for the user to determine the best cleaning schedule for each asset in the fleet. Therefore, by leveraging the machine learning and optimization capabilities of the systems and methods described herein, it is possible to train the systems described herein using operating data from a fleet of assets and operating efficiency data from sensors monitoring the fleet to generate a robust network of systems capable of providing descriptive, predictive, and directive analytical models for optimizing asset health and performance.

[0038] Figure 5 is a flowchart illustrating an exemplary method 500 for determining directive light and severe (e.g., non-invasive and invasive) cleaning schedules for an asset using the systems and methods described herein. In some embodiments, method 500 may include a step 510 of acquiring data characterizing the operational efficiency of the asset over time from a plurality of sensors configured to monitor the asset. The plurality of sensors may include temperature sensors and / or pressure sensors, but the use of other sensor types capable of monitoring operational efficiency is also realized.

[0039] Method 500 may also include a step 520 of receiving data characterizing the operational efficiency of an asset via a computing system which includes at least one data processor and memory for storing instructions. In some embodiments, the asset monitoring system may store data corresponding to the operational efficiency of the asset over the period in which the asset is in operation.

[0040] Method 500 may also include a method step 530 in which at least one data processor determines the operating efficiency of an asset and an operating efficiency threshold that characterizes undesirable operating efficiency, as described above with reference to Figures 1 to 4.

[0041] Method 500 may also include a step 540 of determining a cleaning schedule for an asset based on operating efficiency and an operating efficiency threshold. In some embodiments, the cleaning schedule may be determined using an optimization algorithm such as those described herein. In some embodiments, the method may also include a step of determining a time to a predicted event corresponding to the time it takes to reach the operating efficiency threshold. Furthermore, in some embodiments, step 540 may further include determining a plurality of cleaning schedules having different levels of intervention and / or different predicted times to failure, which may be provided to a user interface for the user to view and select. In some embodiments, an asset may include a plurality of operating modes which may require a particular cleaning method or optimized approach for cleaning the asset in order to extend its operating life. In this case, the method may also include a step of receiving from the asset data characterizing one or more operating states of the asset, and a step of determining a cleaning schedule using an optimization algorithm based on the asset's operating efficiency, an operating efficiency threshold, and the data characterizing one or more operating states of the asset.

[0042] Method 500 may also include step 550, which provides the user with one or more cleaning schedules via a user interface display.

[0043] To provide an overall understanding of the structure, function, manufacturing and use principles of the systems, devices, and methods disclosed herein, certain exemplary embodiments have been described. One or more examples of these embodiments are illustrated in the accompanying drawings. Those skilled in the art will understand that the systems, devices, and methods specifically described herein and illustrated in the accompanying drawings are non-limiting exemplary embodiments, and that the scope of the invention is defined solely by the claims. Features illustrated or described in relation to one exemplary embodiment may be combined with features of other embodiments. Such modifications and variations are intended to fall within the scope of the invention. Furthermore, in this disclosure, components with similar names in embodiments generally have similar features, and therefore, each feature of each component with a similar name within a particular embodiment is not necessarily fully detailed.

[0044] The subject matter described herein may be implemented in analog electronic circuits, digital electronic circuits, and / or computer software, firmware, or hardware, or in combination thereof, including the structural means and structural equivalents thereof disclosed herein. The subject matter described herein may be implemented as one or more computer program products, such as one or more computer programs, which are explicitly embodied in information carriers (e.g., in machine-readable storage devices) or embodied in propagated signals, for execution by or control of the operation of a data processing device (e.g., a programmable processor, a computer, or a number of computers). A computer program (also known as a program, software, software application, or code) can be written in any form of programming language, including compiled or interpreted languages, and can be deployed in any form, such as as a standalone program or as modules, components, subroutines, or other units suitable for use in a computing environment. A computer program does not necessarily correspond to a file. A program may be stored in a single file dedicated to the program, in part of a file that holds other programs or data, or in a number of coordinated files (e.g., a file that stores one or more modules, subprograms, or parts of code). Computer programs can be deployed to run on a single computer, on multiple computers at a single site, or distributed across multiple sites and interconnected by a communication network.

[0045] The processes and logic flows described herein, including the subject matter method steps described herein, may be performed by one or more programmable processors that execute one or more computer programs to perform the functions of the subject matter described herein by operating on input data and generating outputs. The processes and logic flows may also be performed by dedicated logic circuits, such as FPGAs (field programmable gate arrays) or ASICs (application-specific integrated circuits), and the devices of the subject matter described herein may be implemented as such dedicated logic circuits.

[0046] Suitable processors for executing computer programs include, for example, both general-purpose and dedicated microprocessors, as well as any one or more processors in any type of digital computer. Generally, a processor receives instructions and data from read-only memory, random-access memory, or both. Essential elements of a computer are a processor for executing instructions, and one or more memory devices for storing instructions and data. Generally, a computer also includes one or more mass storage devices for storing data, such as magnetic disks, magneto-optical disks, or optical disks, or is operablely coupled to receive data from there, transfer data to there, or both. Suitable information carriers for embodying computer program instructions and data include, for example, all forms of non-volatile memory, including semiconductor memory devices (e.g., EPROM, EEPROM, and flash memory devices), magnetic disks (e.g., internal hard disks or removable disks), magneto-optical disks, and optical disks (e.g., CDs and DVDs). Processors and memory may be complemented by or incorporated into dedicated logic circuits.

[0047] To provide user interaction, the subjects described herein may be implemented on a computer having a display device for displaying information to the user, such as a CRT (cathode ray tube) or LCD (liquid crystal display) monitor, and a keyboard and pointing device (e.g., mouse or trackball) on which the user can provide input to the computer. User interaction may also be provided using other types of devices. For example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback), and input from the user may be received in any form, including acoustic, voice, or tactile input.

[0048] The technologies described herein may be implemented using one or more modules. As used herein, the term “module” refers to computing software, firmware, hardware, and / or various combinations thereof. However, at a minimum, a module should not be interpreted as software not implemented on hardware, firmware, or on a readable and recordable storage medium of a non-temporary processor (i.e., a module is not software itself). In practice, a “module” should always be interpreted as including at least some physical non-temporary hardware, such as a processor or part of a computer. Two different modules may share the same physical hardware (for example, two different modules may use the same processor and network interface). The modules described herein may be combined, integrated, separated, and / or duplicated to support various applications. Furthermore, functions described herein as performed by a particular module may be performed by one or more other modules and / or one or more other devices instead of, or in addition to, functions performed by a particular module. Moreover, modules may be implemented across numerous devices and / or other components, local or remote to each other. Additionally, modules can be moved from one device to another, and / or integrated into both devices.

[0049] The subject matter described herein may be implemented in a computing system including backend components (e.g., data servers), middleware components (e.g., application servers), or frontend components (e.g., client computers having a graphical user interface or a web browser through which users can interact with implementations of the subject matter described herein), or any combination of such backend, middleware, and frontend components. The components of the system may be interconnected by any form or medium of digital data communication, such as a communication network. Examples of communication networks include local area networks ("LANs") and wide area networks ("WANs"), such as the Internet.

[0050] Where used herein and throughout the claims, the approximate term can be applied to modify any quantitative expression that may change acceptablely without altering the underlying function of the expression. Thus, values ​​modified by one or more terms, such as “about,” “approximately,” and “substantially,” are not limited to the specified exact value. In at least some examples, the approximate term may correspond to the precision of an instrument used to measure a value. Herein, throughout this specification and the claims, range limitations may be combined and / or interchangeable, but such ranges include all subranges identified and contained therein unless the context or wording indicates otherwise.

[0051] Those skilled in the art will understand further features and advantages of the present invention based on the embodiments described above. Therefore, this application is not limited to what has been specifically illustrated and described, except as indicated by the appended claims. All publications and references cited herein are expressly incorporated by reference in their entirety.

Claims

1. It is a method, To acquire data characterizing the operational efficiency of the asset over time from multiple sensors configured to monitor the asset, Receiving the data characterizing the operational efficiency of the asset via a computing system including at least one data processor and memory for storing instructions, The at least one data processor determines the operating efficiency of the asset and an operating efficiency threshold that characterizes undesirable operating efficiency. The at least one data processor determines the cleaning schedule for the asset based on the operating efficiency and operating efficiency threshold of the asset, A method comprising providing the aforementioned cleaning schedule.

2. The method according to claim 1, wherein the cleaning schedule includes one or more light cleanings for non-invasive cleaning of the asset.

3. The method according to claim 2, wherein the cleaning schedule includes one or more heavy cleanings for mechanical cleaning of the asset.

4. The at least one data processor determines the time to a predicted event corresponding to the time it takes to reach the operating efficiency threshold, The method according to claim 1, further comprising providing time until the aforementioned event.

5. The computing system receives data from the asset that characterizes one or more operating states of the asset, The method according to claim 4, further comprising using an optimization algorithm to determine the cleaning schedule based on the operating efficiency of the asset, the operating efficiency threshold, and the data characterizing one or more operating states of the asset, by the at least one data processor.

6. The method according to claim 5, further comprising using the optimization algorithm to determine an optimized operating schedule for the operation of the asset in one or more operating states, using the at least one data processor.

7. The method according to claim 6, wherein the time to the predicted event is determined based on the cleaning schedule and the optimized operation schedule.

8. The method according to claim 1, further comprising determining the amount of fouling that occurred within the asset using the at least one data processor.

9. The method according to claim 1, wherein the asset is a heat exchanger or a compressor, and the plurality of sensors include at least one of a temperature sensor and a pressure sensor.

10. The method according to claim 1, wherein the operating efficiency threshold is determined based on at least one of the operating history data of the asset, client preferences, predetermined efficiency requirements, and predetermined energy consumption thresholds.

11. It is a system, Multiple sensors configured to monitor assets, A computing system comprising, the computing system includes at least one data processor and memory for storing instructions, and when an instruction is executed by the at least one data processor, the at least one data processor receives Receiving data characterizing the operational efficiency of the asset from the aforementioned multiple sensors, To determine the operating efficiency of the aforementioned assets, Determining the driving efficiency threshold that characterizes undesirable driving efficiency, Using an optimization algorithm, the cleaning schedule for the asset is determined based on the asset's operating efficiency and the operating efficiency threshold. A system that causes the system to perform operations including providing the aforementioned cleaning schedule.

12. The system according to claim 11, wherein the cleaning schedule includes one or more light cleanings for non-invasive cleaning of the asset.

13. The system according to claim 12, wherein the cleaning schedule includes one or more heavy cleanings for mechanical cleaning of the asset.

14. The operation performed by the at least one data processor is: Determining the time until a predicted event that corresponds to the time required to achieve the aforementioned driving efficiency threshold, The system according to claim 11, further comprising providing time until the aforementioned event.

15. The operation performed by the at least one data processor is: Receiving data from the said asset that characterizes one or more operating states of said asset, The system according to claim 14, further comprising using an optimization algorithm to determine the cleaning schedule based on the operating efficiency of the asset, the operating efficiency threshold, and the data characterizing one or more operating states of the asset.

16. The system according to claim 15, wherein the operation performed by the at least one data processor further comprises the at least one data processor using the optimization algorithm to determine an optimized operating schedule for the operation of the asset in one or more operating states.

17. The system according to claim 16, wherein the time until the predicted event is determined based on the cleaning schedule and the optimized operating schedule.

18. The system according to claim 11, wherein the operation performed by the at least one data processor further comprises determining the amount of fouling that occurred within the asset.

19. The system according to claim 11, wherein the asset is a heat exchanger or a compressor, and the plurality of sensors include at least one of a temperature sensor and a pressure sensor.

20. The system according to claim 11, wherein the operating efficiency threshold is determined based on at least one of the operating history data of the asset, client preferences, predetermined efficiency requirements, and predetermined energy consumption thresholds.

Images