Continuous flow engine self-optimizing control method and system

Abstract

The invention relates to a continuous flow engine monitoring and control method, comprising improved procedures for adapting existing systems to detected deviations to provide an improved management and control system, thereby increasing the overall benefit provided by such a continuous flow engine. Still further, the invention relates to a system adapted to perform such a method. Additionally, the invention relates to a computer program product for implementing such a method. Still further, the invention relates to using such means to increase the utilization of such a continuous flow engine.

Description

[0001] This invention relates to a method for monitoring a continuous flow engine, which provides automatic adaptation of evaluation parameters. Furthermore, this invention relates to a system enabling such a method. Additionally, this invention relates to a computer program product suitable for performing such a method. Even further, this invention relates to the use of the method, system, or computer program product of this invention.

[0002] Continuous flow engines are well-established devices used in industry. Examples include compressors used in large-scale industrial production processes such as oil refineries, or turbines such as gas turbines or steam turbines used in energy production. In the past, these devices have offered significant increases in utilization. From experimental possibilities decades ago to highly reliable and durable equipment that is now difficult or even impossible to replace. Although general types of engines have been available for a long time, these units continue to undergo further development. In particular, the shift from past concepts of utilization, monitoring, and maintenance to modern, highly advanced and complex concepts has demanded significant changes in data acquisition and utilization in this context. However, the new possibilities for monitoring and controlling corresponding continuous flow engines have led to additional challenges. In this context, it is noted that at least some operators face the challenge of dealing with a large amount of data that requires manual monitoring, a problem that can only be partially addressed by support from automated systems based on established parameters and their limitations. In particular, modern continuous flow engines provide a significantly greater number of parameters to be determined to a significantly higher degree, and require monitoring systems based on automated limits. This presents the challenge of choosing between a wide range and limitations that easily overlook problems as initial indications or a narrower range and limitations that lead to numerous false alarms. In this context, it's important to note that even finding ideal constraints for a given timeframe doesn't guarantee those constraints will apply in the future. This presents further challenges beyond the ongoing challenges of continuously improving the corresponding continuous streaming engine, services, and tools. The goal is to provide a defined system capable of identifying problems at an early stage while still significantly reducing the number of false alarms in the long run.

[0003] These problems are addressed by the products and methods disclosed below and in the claims. Further advantageous embodiments are disclosed in the dependent claims and further description. These advantages can be used to adapt the corresponding solutions to specific needs or to solve further problems.

[0004] According to another aspect, the present invention relates to a method for monitoring an industrial plant comprising at least one continuous flow engine, wherein the method comprises the following steps:

[0005] - Obtain measurement values ​​from at least one sensor.

[0006] -Assess whether the measured value falls within the threshold range.

[0007] - If the measured value falls outside the threshold range, an alarm action is triggered, wherein the alarm action includes sending an evaluation request to the operator.

[0008] - Receive evaluation data from the operator, wherein the evaluation data is based on at least three options, wherein the options include:

[0009] i. Evaluate the alarm action as false; wherein such evaluation data triggers an increase in the threshold range,

[0010] ii. The assessment is that the alarm action was correct.

[0011] iii. Evaluating that the alarm action may be correct, wherein such evaluation data triggers the addition of a counter, wherein if the counter reaches a counter limit, an increase in the threshold range is triggered.

[0012] This is particularly important for continuous flow engines, as the corresponding engines offer high reliability and can be used for decades. However, the drawback of these engines is that their use provides for continuous changes to the engine itself, spontaneous changes based on maintenance and upgrade processes, and the like, which makes them highly reliable and durable on the one hand, but at the same time, constantly deviating from their initial state, requiring such monitoring adaptation. The method of the present invention allows for the provision of such an adaptation system, which significantly improves the possibility of monitoring and controlling such continuous flow engines. Typically, preferably, the at least one sensor monitors the continuous flow engine or the support structure connected to the continuous flow engine. For example, such a support structure can be a fuel supply for a gas turbine, a boiler for a steam turbine, or a device utilizing compressed gas supplied by a compressor. Although the overall design of an industrial plant utilizing such a continuous flow engine also provides a very continuous and predictable process, the characteristics of the continuous flow engine are not only best suited to such a method, but also benefit the most from it, which is naturally particularly evident for the continuous flow engine itself and the directly interconnected support structures (such as the connecting devices that enable the continuous flow engine to function).

[0013] As used herein, the term "operator" refers to a person responsible for, in particular, monitoring or controlling the continuous flow engine. Typically, preferably, such an operator is located on-site at the continuous flow engine. In this document, this person may also be temporarily present at that location, for example, during some inspection or maintenance. However, such an operator may also be located remotely, whereby such an operator may be substantially stationed on the other side of the world, closely monitoring the status of the flow engines from a distance.

[0014] As used in this article, the term "alarm action" refers to a signal indicating an unusual state that requires attention or even immediate action. For example, it could be an alarm indicating that an engine should be shut down as soon as possible to perform maintenance. Or it could be a warning requesting attention to keep the utilization of a continuous flow engine within certain limits to avoid damage or to prevent premature scheduling of maintenance tasks.

[0015] According to another aspect, the present invention relates to a system suitable for use in the method of the invention, the system comprising a processor and a non-transitory computer-readable medium, the non-transitory computer-readable medium comprising computer-executable instructions, which, when executed by the processor, cause the system to perform operations comprising the following steps:

[0016] - Obtain measurement values ​​from at least one sensor.

[0017] -Assess whether the measured value falls within the threshold range.

[0018] - If the measured value falls outside the threshold range, an alarm action is triggered, wherein the alarm action includes sending an evaluation request to the operator.

[0019] - Receive evaluation data from the operator, wherein the evaluation data is based on at least three options, wherein the options include:

[0020] i. Evaluate the alarm action as false; wherein such evaluation data triggers an increase in the threshold range,

[0021] ii. Assess the alarm action as correct.

[0022] iii. Evaluating that the alarm action may be correct, wherein such evaluation data triggers the addition of a counter, wherein if the counter reaches a counter limit, an increase in the threshold range is triggered.

[0023] According to another aspect, the present invention relates to a computer program product tangibly embodied in a machine-readable storage medium, the computer program product comprising instructions operable to cause a computing entity to perform the method of the present invention.

[0024] According to another aspect, the present invention relates to a storage device for providing a computer program product of the present invention, wherein the device stores the computer program product and / or provides the computer program product for further use.

[0025] According to another aspect, the present invention relates to using the methods, systems, or computer program products of the present invention to provide data indicating the benefits of servicing and / or placing filters in a continuous flow engine.

[0026] To simplify the understanding of the invention, reference is made to the following detailed description and the accompanying figures, as well as their description. In this document, the figures should be understood not to limit the scope of the invention, but rather to disclose preferred embodiments that further explain the invention.

[0027] Figure 1 A scheme for an industrial plant comprising a continuous flow engine suitable for utilizing the method of the present invention is shown.

[0028] Figure 2 The scheme of the method of the present invention is shown.

[0029] Preferably, unless otherwise stated, the embodiments described below include at least one processor and / or data storage unit to implement the method of the present invention.

[0030] According to one aspect, the present invention relates to the method as specified above.

[0031] For typical applications, it is advantageous that the method of the present invention also includes the possibility of automatically reducing the counter capacity. According to another embodiment, it is preferred that the evaluation data is based on option ii, where such evaluation data reduces the counter, wherein the minimum counter value is zero. Typically, it is preferred that the counter is reset. It is worth noting that such feedback from experts can be beneficially used to optimize the automated system, enabling monitoring to continuously adapt to long-term developments.

[0032] Typically, it is also advantageous to include a counter limit that is at least several times the maximum number added to such a counter value. According to another embodiment, preferably, when the alarm action is based on option iii, the counter limit is at least three times the maximum amount added to the counter value, more preferably at least four times, and even more preferably at least seven times. It should be noted that corresponding limits below such numbers are typically less advantageous for many applications. In particular, such automated systems for continuous monitoring of status, as described herein, provide that a large number of false alarms can be easily filtered out by such a counter limit.

[0033] Options for adapting threshold ranges, beneficial for many typical applications, are based on a combination of relative and absolute values. According to another embodiment, it is preferred that the adjustment of the threshold range be calculated according to formula (I).

[0034] T new =(A1·T·(1+x1))+(A2·(T+x2)) (I),

[0035] Where T new =Adjusted threshold range,

[0036] T = threshold range,

[0037] A1 = Relative factor,

[0038] A2 = absolute factor,

[0039] x1 is the specified increment factor

[0040] X2 is the specified increment value.

[0041] Where A1 + A2 = 1. In this document, either the specified increment factor or the specified increment value can be chosen to be zero, thus limiting the change to a relative or absolute change. For example, it is preferable to provide a simplified system that typically only provides relative or absolute adaptation. However, while the system could provide the possibility of completely disabling such dual adaptation by excluding relative or absolute adaptation, it is preferable that the method utilizes the possibility provided herein. Furthermore, by reducing A2, such adaptation can be advantageously used to switch from a relative-driven adaptation to an adaptation based primarily on the absolute factor. For example, relative adaptation can be utilized during the first phase of using a newly delivered continuous flow engine, and a switch to more or entirely absolute increment-based changes can be made upon completion of the introduction phase. Such a method can also be utilized after maintenance steps or, in particular, after an introduced upgrade.

[0042] It should be noted that such a seemingly simple approach offers the possibility of easily customizing the available system to meet specific needs. This makes it possible to leverage such a general-purpose system to apply to multiple available data sources and to easily adapt x1 and x2 based on the characteristics of the specific features being monitored. This significantly simplifies the entire system and, for example, allows for the easy introduction of improvements or adaptations, such as upgrades, into the entire continuous streaming engine with minimal effort.

[0043] Typically, a further preferred option is to include a modified upper limit. According to another embodiment, T new -T is at most B, where B is the adaptation constraint. This feature allows for narrowing down the possible modifications to the threshold range to a certain limit, preventing significant changes that were not initially intended from becoming possible, for example, by ignoring developments over longer periods of time. Surprisingly, it is noted that, when reviewing log files and the like, the likelihood of such topics arising, based on the observed interactions with different features and options providing comparable monitoring, represents a real problem for continuous streaming engines.

[0044] Surprisingly, it is noted that the adjustment of the threshold range can be specified by formula in many cases. According to another embodiment, preferably, the adjustment of the threshold range is calculated according to formula (II).

[0045] A = (ML)·y (II),

[0046] Where A = adjustment value,

[0047] M = measured value,

[0048] L = the limit value of the threshold range,

[0049] y is a predefined fit factor.

[0050] The threshold range is increased by 'a'. For typical applications, such a system that provides automatic adaptation of the threshold range is a significant benefit. Such a system is typically particularly useful for training several separate systems, usually decoupled from the distributed system, thereby providing continuous improvement in the threshold range. However, such a system can also be beneficially utilized in conjunction with several connected systems that receive data from the distributed system, where, for example, such local automatic direct adjustment is used in a first step for rapid response to the current situation, while general adaptation based on such a distributed system occurs periodically.

[0051] In this document, certain adaptation factors are typically efficient for many applications. According to another embodiment, preferably, the predefined adaptation factor is selected from the range of 0.01 to 0.1, more preferably from 0.2 to 0.08, and even more preferably from 0.03 to 0.07. It is noted that such predefined adaptation factors selected from such ranges are typically advantageous for continuous flow engines such as gas turbines and steam turbines. In this document, based on its broader applicability to such systems, such an adaptation factor can, for example, be used as a general starting point, wherein the adaptation factor undergoes further optimization based on a specific system.

[0052] Actions based on the formulas cited above, or the different automatic adaptations introduced, can be advantageously connected to a feedback loop. According to another embodiment, preferably, the method includes an automatically provided adaptation of a threshold range, which is forwarded to the operator for review. Typically, it is preferred herein that the operator must agree to the modification for its implementation. However, it is also possible to provide a system that first implements such a modification and then requests the operator to verify it. Such a system is typically preferred for highly relevant parameters and components, to combine automatic recommendations with their annual evaluations. It is noted that such an arrangement significantly reduces the time required by such operators while best utilizing the experience accumulated over a long period by such personnel.

[0053] Furthermore, it is noted that the number of adaptations within a certain time period is also a very useful parameter to consider. According to another embodiment, the number of adaptations within a time period is stored in a database. Typically, preferably, a further alert is triggered if the number of adaptations within a specified time period exceeds an adaptation limit. It is noted that the system provides a statistical count of adaptations that typically occur over time. Surprisingly, reviewing whether the required number of adaptations significantly exceeds the expected statistical amount provides valuable insight into whether the machine is deviating from normal behavior. Such information and alerts are particularly useful because not everything that needs to be done is viewed by a single operator who remembers everything; rather, even different operators switching shifts and responsible for multiple continuous streaming engines can detect anomalous behavior at a very early stage.

[0054] It is further noted that the method of the present invention is particularly well-suited for application to specific engines. According to another embodiment, the continuous flow engine is preferably a gas turbine, steam turbine, or compressor, more preferably a gas turbine, and even more preferably a gas turbine including a compressor unit. It is noted that the method of the present invention is particularly useful for monitoring and controlling such continuous flow engines. For typical applications, utilizing the other embodiments described herein for such continuous flow engines is particularly useful. For example, the large amount of data available for such engines and the experience collected on the manufacturer's side allows for highly sophisticated simulation models and a large amount of historical data that can be utilized as described herein.

[0055] According to another embodiment, it is preferred that the measured value is a measurement of the compressor component of a continuous flow engine. For example, such a compressor component may be the compressor itself of the continuous flow engine or a compressor component preceding the turbine section of a gas turbine.

[0056] Only by providing simulated data can you further support the evaluation of the measurements. According to another embodiment, preferably, the simulated values ​​are provided to the operator along with the measured values, wherein the simulated values ​​are based on a model of the continuous flow engine or a model of a portion of the continuous flow engine. While an operator, as an expert in such a continuous flow engine, can evaluate such data based on his / her experience, it is noted that providing such simulated values ​​provides significant benefits for evaluating, for example, anomalies or during shift changes, until the operator has gained a deep understanding of the continuous flow engine's recent use and behavior. This benefit surprisingly outweighs the additional effort required to provide such simulations.

[0057] However, it should be noted that historical data can also be advantageously utilized according to the invention. According to another embodiment, historical values ​​are preferably provided to the operator along with measured values, wherein historical values ​​are retrieved from a historical database based on the current state of the continuous streaming engine. Utilizing such historical values ​​is typically advantageous when a sufficient amount of historical data is available. The processing power required to identify comparable cases based on a historical database containing historical values ​​of available continuous streaming engines is typically far less than the processing power required to simulate such values. However, to increase the amount of available data, data collected from comparable continuous streaming engines, either locally stored or stored in a distributed database, can preferably be utilized. This typically requires automatic evaluation of whether the available data is transferable and suitable for utilization. While such automatic evaluation significantly increases the required processing power, the possibility of relying on a significantly larger amount of available data makes such a method highly beneficial for typical applications. Examples of historical data that can be forwarded include multiple potentially correct alerts within a specified time period and multiple verified alerts within such a time period.

[0058] Further improvements to this method could include communication with a remote database. According to another embodiment, it is preferred that the method includes the step of sending a dataset to a remote database (like a distributed database), wherein the dataset includes data containing threshold ranges and / or data related to the evaluation data. Utilizing such a distributed database is particularly useful. This allows for optimal use of advanced expertise at a central location, and especially at a third party (such as the manufacturer of a continuous streaming engine). In this paper, the possibility of significantly improved speed of problem identification, as well as improved quality of existing systems and corresponding checks, can be provided.

[0059] Another feature that can be advantageously incorporated into the method of the present invention is the introduction of smart contracts. According to another embodiment, preferably, a remote database, preferably a distributed database, contains smart contracts that are triggered when a threshold range falls outside a specified range and / or the operator's assessment differs from reference data regarding the operator's assessment. The term "smart contract" as used herein includes, in particular, data like program data or data that can be executed by a program to perform specific steps or actions, including control commands, specified values, requirements, related data like measurements, and corresponding actions in response to whether a predefined value is met or not compared to said measurement. For example, the execution of a smart contract can be accomplished by a correspondingly selected distributed database or a runtime environment like a virtual machine. Preferably, such means for executing smart contracts are becoming increasingly comprehensive. Typically, it is preferred to use the infrastructure of a distributed database to execute smart contracts. Such smart contracts can, for example, be advantageously used to automatically trigger maintenance actions or payment models in real time using more interactive and adaptable systems. In this document, for example, maintenance costs can be automatically adapted based on the use of the continuous streaming engine to provide the owner of the continuous streaming engine with as much operational freedom as possible. At the same time, maintenance providers can provide reasonable computations based on security enhancements that exceed the initially agreed level, which requires accountable increases in service costs.

[0060] While the present invention can also be used with very old continuous flow engines and with little insight into the corresponding processes, it is typically preferred to evaluate multiple measurements using this method. According to another embodiment, preferably, the method includes acquiring at least three, more preferably at least four, and even more preferably at least seven measurements for evaluation, wherein the measurements are measurements of the compressor components of the continuous flow engine. While it is possible to monitor and control such a device using a single measurement, it is noted that a combination of multiple measurements for evaluation is beneficial for typical applications. This results in a surprisingly enhanced level of insight, allowing for the review of multiple measurements. Comparing multiple measurements derived from, for example, different sensors allows for the identification of problems with significantly higher security.

[0061] Surprisingly beneficial in this paper is the combination of data from different types of data sources. According to another embodiment, preferably, the method utilizes at least two, or even more preferably at least three different measurements of a continuous flow engine. Utilizing such different measurements of the same continuous flow engine stream to monitor different parts of the continuous flow engine, where problems leading to deviations from expected behavior are typically reflected in changes in multiple measurements. Coupled, in such cases, multiple measurements that should all show deviations, allow for a significant reduction in the false positive detection rate. According to another embodiment, preferably, the measurements are based on sensor data from different sensors. Typically, preferably, the different sensors are based on different measurement methods. For example, such sensors may be based on pressure measurements, rotational speed measurements, density measurements, temperature measurements, and the like. According to another aspect, the present invention relates to a system suitable for use in the method of the invention, the system comprising a processor and a non-transitory computer-readable medium, the non-transitory computer-readable medium comprising computer-executable instructions, which, when executed by the processor, cause the system to perform operations comprising the following steps:

[0062] - Obtain measurement values ​​from at least one sensor.

[0063] -Assess whether the measured value falls within the threshold range.

[0064] - If the measured value falls outside the threshold range, an alarm action is triggered, wherein the alarm action includes sending an evaluation request to the operator.

[0065] - Receive evaluation data from the operator, wherein the evaluation data is based on at least three options, wherein the options include:

[0066] i. Assess the alarm action as false; wherein such assessment data triggers an increase in the threshold range.

[0067] ii. Assess the alarm action as correct.

[0068] iii. Evaluating that the alarm action may be correct, wherein such evaluation data triggers an addition to the counter, wherein if the counter reaches the counter limit, an increase in the threshold range is triggered.

[0069] According to another aspect, the present invention relates to an upgrade kit that includes at least the system of the present invention. Such an upgrade kit is particularly useful for upgrading existing industrial plants and continuous flow engines. Surprisingly, it is noted that the system of the present invention can be adapted to automatically establish corresponding connections and retrieve data, including what data is available in the data infrastructure of such industrial plants. Further, it is possible, with the granting of corresponding rights, to automatically retrieve updates from remote databases, for example, provided by manufacturers of continuous flow engines, in order to improve the initial setup of the system based on specific local requirements and unforeseen events.

[0070] Typically, and preferably, such upgrade kits are designed to communicate automatically with field personnel responsible for implementing such upgrades. Surprisingly, it is possible to automatically provide field personnel with a list of deficiencies, such as those related to sensor data and connectivity issues, and to forward explicit instructions on what is needed to optimize the benefits of the upgrade kit.

[0071] In most cases, it is preferable that such upgrade kits further include historical databases and simulation units, which are particularly well-suited for implementing the methods of the present invention without requiring further processing capacity from the local industrial plant. While future and state-of-the-art industrial plants utilizing continuous flow engines are expected to provide corresponding databases and processes that enable the simple reallocation of corresponding processing capacity and data storage, it is noted that introducing such elements into the upgrade kit offers the significant benefit of providing this possibility to most existing industrial plants.

[0072] According to another aspect, the present invention relates to a computer program product tangibly embodied in a machine-readable storage medium, the computer program product comprising instructions operable to cause a computing entity to perform the method of the present invention.

[0073] According to another aspect, the present invention relates to a storage device for providing a computer program product of the present invention, wherein the device stores the computer program product and / or provides the computer program product for further use.

[0074] According to another aspect, the present invention relates to upgrading an industrial plant containing at least one continuous flow engine using the method of the present invention, the system of the present invention, the upgrade kit of the present invention, or the computer program product of the present invention.

[0075] The invention has been described in further detail for illustrative purposes only. However, the invention should not be construed as limited to these embodiments, as they represent embodiments that provide benefits to solve a particular problem or meet a particular need. The scope of protection should be understood to be defined only by the appended claims.

[0076] Figure 1A scheme for an industrial plant comprising continuous flow engines 2, 2' is illustrated, wherein the continuous flow engines 2, 2' are gas turbines that provide compressor components within turbine components, the turbine components being adapted to utilize the method of the present invention. In this document, sensors 1, 1', 1″ are exemplarily shown located within the compressor components of the two continuous flow engines 2, 2' in an industrial plant 6. It should be noted that the corresponding continuous flow engines 2, 2' contain a significantly higher number of sensors (not included in the compressor components). Figure 1 (as shown in the figure) and used to provide measurements analyzed according to the method according to the invention. Data acquired by sensors 1, 1', 1'' is sent to system 3 of the invention, which processes data according to the method of the invention. Herein, system 3 is adapted to adapt counter 5, managed by system 3, to an evaluation of the generated alarm. Herein, the measured value is evaluated based on a specified threshold range, ultimately triggering an alarm if the measured value falls outside the threshold range. Upon such an alarm action being triggered, an evaluation request is sent to the operator, requesting evaluation of the alarm based on at least three options. These options are provided to the operator as choices from which the operator may select. Herein, the data provided to the operator also includes historical data and simulation data to support his decision-making. The corresponding data is obtained from a historical database 8 and a simulation unit 7, which are locally available at the industrial plant 6. The historical database 8 and the simulation unit 7 are further connected to a remote database. Historical database 8 is adapted to exchange information with a historical remote database 10 to retrieve additional data to be used during the evaluation. Such a feature is particularly advantageous for providing a basic amount of historical data to be utilized after upgrades or modifications to existing continuous flow engines 2, 2 that render the available historical data partially unavailable. Communication between historical database 8 and historical remote database 10 is triggered on demand.

[0077] Simulation unit 7 and simulation remote database 9 exchange data periodically or triggered by simulation remote database 9. For example, if certain upgrades to the existing model used in simulation unit 7 are available, such as reducing the required processing power or increasing simulation efficiency or reliability, the corresponding data is sent to simulation unit 7 to improve its performance. However, if simulation unit 7 itself has noticed a deficiency, a request to simulation remote database 9 can also be triggered directly on that side.

[0078] If the operator evaluates an alarm action as false, the threshold range is increased to compensate for threshold ranges that are no longer appropriate. If the alarm action is evaluated as correct, the counter is decreased by a specific amount, where the counter value does not fall below zero. If the alarm action is considered potentially correct, the counter value is increased as a specified amount is added to the counter until the counter limit is reached. Upon reaching the counter limit, an increase in the threshold range is triggered. In particular, leveraging the expertise of third parties (like continuous flow engine manufacturers and / or companies specializing in such continuous flow engines) provides significant benefits. Decades of in-depth internal and extensive expertise can be used to specifically customize and optimize corresponding models for continuous flow engines used in such simulations.

[0079] In this paper, the operator can also provide an evaluation containing the dataset sent from interface 4 to system 3 to reset the counter. System 3, as shown, also provides the possibility of manually requesting automatic counter reset upon providing an evaluation based on option ii. This is surprisingly beneficial for customizing a general-purpose system 3 specifically to meet the specific needs of the corresponding continuous streaming engine.

[0080] If the counter limit is reached, the threshold range adaptation is calculated according to the following formula (II):

[0081] A = (ML)·y (II),

[0082] Where A = adjustment value,

[0083] M = measured value,

[0084] L = the limit value of the threshold range,

[0085] y is a predefined fit factor.

[0086] The threshold range is increased by A where the corresponding limit value of the measured value exceeds the measured value. It must be understood that such an increase in the threshold range can also be equal to a decrease in the corresponding limit of the threshold range if the limit is a lower limit and the measured value is below that value. In this case, the term (ML) becomes negative and is added to the corresponding end of the threshold range, resulting in an overall increase in the threshold range. In this paper, the operator specifically assigns the adaptation factor to each characteristic of the measurement on a case-by-case basis, where, for the system shown, the suggested adaptation factor is in the range of 0.2 to 0.08. For the first implementation, a higher adaptation factor is chosen to provide a faster initial adaptation to the current system. After the working system is established, the adaptation factor is reduced, considering that the required further adaptation should be significantly lower. In cases where a large number of adaptations are subsequently needed, this may indicate further problems with the continuous flow engine, thus triggering different kinds of evaluations and considerations.

[0087] For typical applications, such a system that provides automatic adaptation of threshold ranges offers significant benefits. For example, such a system can be used to adapt isolated systems typically separated from the distributed system, thereby providing continuous improvement in the threshold range. However, such a system can also be beneficially utilized with connected systems that receive data from the distributed system, where, for example, such local automatic direct adjustment is used in a first step to quickly respond to the current situation, while general adaptation of the data based on such distributed system data occurs periodically.

[0088] Furthermore, the data acquired during the process of the present invention and the currently utilized threshold range is not simply stored locally. The corresponding data is also forwarded to and stored in two remote databases (not shown in the figure). One database belongs to the owner of industrial plant 6, who is informed and maintains the state of the continuous streaming engine utilized during tracking. The second database belongs to the maintenance provider, who is responsible for performing maintenance and scheduling it accordingly. In this document, the service is provided on a flexible basis, where the corresponding requested actions are delivered on demand. The forwarded corresponding data not only allows for some internal assessment of whether maintenance is required but also allows for quantifying the benefits of performing maintenance earlier or later, providing the owner with evidence of whether the corresponding maintenance actions are appropriate.

[0089] like Figure 1 System 3, as shown, is included as an upgrade in industrial plant 6. This paper presents the final solution provided to the owner of industrial plant 6. The upgrade kit, including System 3, along with historical database 8 and simulation unit 7, is connected to existing data in the infrastructure of industrial plant 6. System 3 is configured to automatically scan for relevant measurements available in the data infrastructure for continuous flow engines 2 and 2'. Based on this assessment, field personnel are requested to install the corresponding upgrades. For example, they are requested to install additional sensors 1, 1', and 1″ in continuous flow engines 2 and 2' and provide certain connections in the data infrastructure. After connecting to the historical remote database via the Internet and then to the simulation remote database 9, the corresponding updates are automatically loaded and installed in the system. Furthermore, a feedback loop with user interface 4 is tested and established.

[0090] Figure 2The scheme of the method of the present invention is illustrated. In this document, sensor data 21 is forwarded to perform step 22, which includes evaluating the sensor data using the system of the present invention. If an evaluation by an operator is requested, step 25, i.e., retrieval of historical data, is performed. If it is noted that the available historical data is insufficient, an additional step 27 is performed, including retrieving historical data from a remote database. Further, step 26 includes retrieving simulation data from simulation unit 7. If it is noted that the model is insufficient to provide a reasonable simulation, an optional step 28, including retrieving simulation data from a remote database, is performed.

[0091] During step 24, the corresponding data is forwarded to the user interface, allowing the operator to view the corresponding alarm action. If the alarm action is deemed potentially correct, step 29 is executed, where the change in the counter value is quantified into a specific condition column. Subsequently, according to step 30, the counter is adapted accordingly and stored in step 23, and the corresponding database is utilized by the system of the present invention. If the corresponding counter value exceeds a threshold, step 32, including a modification of the threshold, is triggered. Such a change is also stored in the database utilized by the system of the present invention and is considered for future evaluations, such as measurements forwarded by sensors.

[0092] The invention has been described in further detail for illustrative purposes only. However, the invention should not be construed as limited to these embodiments, as they represent embodiments that provide additional benefits to solve a particular problem or meet a particular need. The scope of protection should be understood to be defined only by the appended claims.

Claims

1. A method for monitoring an industrial plant (6) containing at least one continuous flow engine (2, 2'), wherein the method comprises the following steps - Obtain measurement values ​​from at least one sensor (1, 1', 1"), - Assess whether the measured value falls within the threshold range. - If the measured value falls outside the threshold range, an alarm action is triggered, wherein the alarm action includes sending an evaluation request to the operator. - Receive evaluation data from the operator, wherein the evaluation data is based on at least three options, wherein the options include: i. Assess the alarm action as false; Such evaluation data triggers an increase in the threshold range. ii. Assess the alarm action as correct. iii. The assessment that the alarm action may be correct, wherein such assessment data triggers the addition of a counter (5), wherein if the counter (5) reaches the counter limit, the threshold range is increased.

2. The method of claim 1, wherein the evaluation data is based on option ii. Such evaluation data reduces the counter (5). The minimum counter value is zero.

3. The method according to any one of claims 1 to 2, wherein the adjustment of the threshold range is calculated according to formula (I). T new =((T·(1+x1))+(T+x2)) / 2 (I), Where T new =Adjusted threshold range, T = threshold range, x1 is the specified increment factor. x2 specifies the increment value.

4. The method according to claim 3, wherein T new - T is at most B, where B is the adaptation constraint.

5. The method according to any one of claims 1 to 2, wherein the measured value is a measured value of the compressor component of the continuous flow engine (2, 2').

6. The method according to any one of claims 1 to 2, wherein the simulated value is provided to the operator together with the measured value. The simulated values ​​are based on a model of the continuous flow engine (2, 2') or a model of a portion of the continuous flow engine.

7. The method according to any one of claims 1 to 2, wherein historical values ​​are provided to the operator together with the measured values. The historical values ​​are retrieved from the historical database based on the current state of the continuous stream engine (2, 2').

8. The method according to any one of claims 1 to 2, wherein the continuous flow engine (2, 2') is a gas turbine, a steam turbine or a compressor, more preferably a gas turbine, and even more preferably a gas turbine including a compressor unit.

9. The method according to any one of claims 1 to 2, wherein the method comprises the step of sending the dataset to a distributed database. The dataset includes data containing the threshold range and / or data related to the evaluation data.

10. The method of claim 9, wherein the distributed database includes smart contracts triggered when the threshold range falls outside the specified range and / or the operator's assessment differs from reference data regarding the operator's assessment.

11. The method according to any one of claims 1 to 2, wherein the method comprises obtaining at least three, more preferably at least four, or even more preferably at least seven measurements for evaluation. The measured values ​​are those of the compressor components of the continuous flow engine (2, 2').

12. A system (3) suitable for use in the method according to any one of claims 1 to 11, comprising a processor and a non-transitory computer-readable medium, the non-transitory computer-readable medium comprising computer-executable instructions, which, when executed by the processor, cause the system (3) to perform operations including: - Obtain measurement values ​​from at least one sensor (1, 1', 1"), - Assess whether the measured value falls within the threshold range. - If the measured value falls outside the threshold range, an alarm action is triggered, wherein the alarm action includes sending an evaluation request to the operator. - Receive evaluation data from the operator, wherein the evaluation data is based on at least three options, wherein the options include iv. Evaluate the alarm action as false; wherein such evaluation data triggers an increase in the threshold range, v. Assess the alarm action as correct. vi. Evaluate that the alarm action may be correct, wherein such evaluation data triggers the addition of a counter (5), wherein if the counter (5) reaches the counter limit, an increase in the threshold range is triggered.

13. An upgrade kit comprising at least the system (3) according to claim 12.

14. A computer program product tangibly embodied in a machine-readable storage medium, including instructions operable to cause a computing entity to perform the method according to any one of claims 1 to 11.

15. The use of the method according to any one of claims 1 to 11, the system (3) according to claim 12, the upgrade kit according to claim 13, or the computer program product according to claim 14 for upgrading an industrial plant (6) comprising at least one continuous flow engine (2, 2').

Images