Generator set intelligent monitoring method, device and equipment and storage medium

Abstract

The invention provides an intelligent monitoring method, device and equipment for a generator set and a storage medium. The method comprises the following steps: acquiring historical operation data and historical fault data of the generator set based on a DCS system; constructing a first fault early warning model through unsupervised learning based on the historical operation data, constructing a second fault early warning model through supervised learning based on the historical fault data, and constructing a mechanism model of each key component of the generator set; according to the first fault early warning model, the second fault early warning model and the mechanism model, obtaining a first fault early warning result, a second fault early warning result and a third fault early warning result corresponding to the real-time operation data of the generator set; and fusing the first fault early warning result, the second fault early warning result and the third fault early warning result according to the similarity between the real-time operation data and the historical fault data to obtain an intelligent monitoring result of the generator set. The accuracy of the intelligent monitoring process of the generator set can be improved.

Description

Technical Field

[0001] This invention relates to the field of generator set data processing technology, and in particular to a generator set intelligent monitoring method, device, equipment and storage medium. Background Technology

[0002] Generator sets are core equipment in energy systems, and their failures can lead to significant economic losses. Moreover, with the development of generator sets, the possibility of failure and the severity of failures will further increase. Therefore, real-time analysis of various operating data of generator sets based on intelligent monitoring technology and automatic identification of potential faults can help improve the safe and stable operation of generator sets.

[0003] Currently, intelligent monitoring technologies for generator sets mainly include DCS-based hierarchical alarm technologies. These include alarms based on process parameter over-limits and system faults inherent in the DCS system itself, as well as hierarchical audible and visual alarms combining DCS algorithm blocks and screen configuration functions. However, DCS alarms often occur when the situation is already quite severe, typically causing equipment shutdowns and impacting the unit's safety and economy. With the development of digitalization and intelligence, big data analytics-based intelligent alarm technologies can process, analyze, and statistically analyze operational data through data modeling based on all measurement points on the DCS side to derive early warning information. This offers advantages such as simple modeling, rapid development, and adaptability to changing operating conditions. However, data modeling often relies on historical fault data, but unit equipment may have unknown or hidden faults, potentially leading to false alarms or missed alarms. Summary of the Invention

[0004] This invention provides a method, device, equipment, and storage medium for intelligent monitoring of generator sets, in order to solve the problem of insufficient early warning of unknown faults in the existing intelligent monitoring process of generator sets.

[0005] In a first aspect, embodiments of the present invention provide a method for intelligent monitoring of generator sets, including: Historical operating data of the generator set is acquired based on the DCS system, and historical fault data is extracted from the historical operating data. Based on the historical operating data, a first fault early warning model for the generator set is constructed through unsupervised learning. Based on the historical fault data, a second fault early warning model for the generator set is constructed through supervised learning. Mechanism models of various key components of the generator set are also constructed. Based on the first fault warning model, the second fault warning model and the mechanism model, the first fault warning result, the second fault warning result and the third fault warning result corresponding to the real-time operation data of the generator set are obtained respectively. The similarity between the real-time operating data and the historical fault data is calculated, and the first fault warning result, the second fault warning result and the third fault warning result are fused according to the similarity to obtain the intelligent monitoring result of the generator set.

[0006] In one possible implementation, calculating the similarity between the real-time operational data and the historical fault data includes: Cluster the historical fault data to obtain the cluster center for each fault type; The distance between the real-time running data and the cluster center of each fault type is calculated as the similarity between the real-time running data and each fault type.

[0007] In one possible implementation, the first fault warning result, the second fault warning result, and the third fault warning result are fused based on the similarity to obtain the intelligent monitoring result of the generator set, including: If the similarity between the real-time operating data and each type of fault exceeds a set similarity threshold, then the intelligent monitoring result of the generator set is obtained based on the second fault warning result. If there is no similarity higher than the set similarity threshold between the real-time operating data and each type of fault, then the intelligent monitoring result of the generator set is obtained based on the first fault warning result, the second fault warning result and the third fault warning result.

[0008] In one possible implementation, the intelligent monitoring results of the generator set are obtained based on the first fault warning result, the second fault warning result, and the third fault warning result, including: Determine whether the first fault warning result and the third fault warning result are consistent; If the first fault warning result and the third fault warning result are consistent, then the intelligent monitoring result of the generator set is obtained based on the first fault warning result or the third fault warning result. If the first fault warning result and the third fault warning result are inconsistent, then the higher confidence fault warning result between the first fault warning result and the third fault warning result is obtained; The intelligent monitoring results of the generator set are obtained based on the high-confidence fault early warning results.

[0009] In one possible implementation, obtaining the high-confidence fault warning result from the first fault warning result and the third fault warning result includes: Determine whether the first fault warning result is consistent with the second fault warning result, or whether the third warning result is consistent with the second fault warning result; If the first fault warning result is consistent with the second fault warning result, then the first fault warning result is determined as a high-confidence fault warning result; If the third fault warning result is consistent with the second fault warning result, then the third fault warning result is determined as a high-confidence fault warning result.

[0010] In one possible implementation, obtaining the high-confidence fault warning result from the first fault warning result and the third fault warning result includes: The previous period's operating data of the generator set and the corresponding intelligent monitoring results of the previous period are obtained by using a preset sliding window. Calculate the first degree of change of the real-time operating data relative to the operating data of the previous period, the second degree of change of the first fault warning result relative to the intelligent monitoring result of the previous period, and the third degree of change of the third fault warning result relative to the intelligent monitoring result of the previous period; If the second degree of change is relatively small compared to the first degree of change, then the first fault warning result is determined as a high-confidence fault warning result; If the third degree of change is relatively small compared to the first degree of change, then the third fault warning result is determined as a high-confidence fault warning result.

[0011] In one possible implementation, the historical fault data is clustered to obtain cluster centers for each fault type, including: The historical fault data is categorized according to operating conditions; Clustering is performed on historical fault data under each operating condition to obtain the cluster center for each fault type under each operating condition.

[0012] Secondly, embodiments of the present invention provide an intelligent monitoring device for generator sets, comprising: The data acquisition module is used to acquire historical operating data of the generator set based on the DCS system, and extract historical fault data from the historical operating data; The model building module is used to build a first fault early warning model of the generator set through unsupervised learning based on the historical operating data, and to build a second fault early warning model of the generator set through supervised learning based on the historical fault data, and to build a mechanism model of each key component of the generator set. The first fault monitoring module is used to obtain the first fault warning result, the second fault warning result and the third fault warning result corresponding to the real-time operation data of the generator set according to the first fault warning model, the second fault warning model and the mechanism model, respectively. The second fault monitoring module is used to calculate the similarity between the real-time operating data and the historical fault data, and to fuse the first fault warning result, the second fault warning result and the third fault warning result according to the similarity to obtain the intelligent monitoring result of the generator set.

[0013] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0014] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect or any possible implementation thereof.

[0015] In this embodiment of the invention, historical operating data of the generator set is acquired based on the DCS system, and historical fault data is extracted from the historical operating data. Then, based on the historical operating data, a first fault early warning model of the generator set is constructed through unsupervised learning. Based on the historical fault data, a second fault early warning model of the generator set is constructed through supervised learning. Mechanism models of various key components of the generator set are also constructed. Then, based on the first fault early warning model, the second fault early warning model, and the mechanism model, the first fault early warning result, the second fault early warning result, and the third fault early warning result corresponding to the real-time operating data of the generator set are obtained respectively. Then, the similarity between the real-time operating data and the historical fault data is calculated, and the first fault early warning result, the second fault early warning result, and the third fault early warning result are fused according to the similarity to obtain the intelligent monitoring result of the generator set. Thus, the final intelligent monitoring result is determined by comprehensively considering the generator set operating mechanism, the unsupervised model, and the supervised model, overcoming the problem of insufficient early warning of unknown faults in the intelligent monitoring process of the generator set and improving the accuracy of the intelligent monitoring process of the generator set. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the implementation of the intelligent monitoring method for generator sets provided in this embodiment of the invention. Figure 2 This is a schematic diagram of the intelligent monitoring device for generator sets provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0017] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0018] See Figure 1The document illustrates a flowchart of the implementation of the intelligent monitoring method for generator sets provided in an embodiment of the present invention, which is described in detail below: Step 101: Obtain historical operating data of the generator set based on the DCS system, and extract historical fault data from the historical operating data.

[0019] Distributed Control Systems (DCS) are key equipment in the operation and control of modern generator sets. They ensure power generation efficiency and safety by precisely monitoring and controlling various devices within the generator set. A DCS typically consists of three main parts: a data acquisition layer, a control layer, and a monitoring layer. The data acquisition layer acquires data from sensors and instruments, the control layer processes the data and issues control commands, and the monitoring layer provides operators with data visualization and analysis tools.

[0020] In this embodiment, historical operating data of the generator set during its historical operation can be collected based on the data acquisition layer of the DCS system. This includes electrical operating parameters such as generator active / reactive power, stator / rotor voltage and current, power factor, and frequency; thermal system parameters such as turbine speed, axial displacement, vibration value, bearing temperature, lubricating oil pressure, boiler main steam pressure / temperature, feedwater flow rate, and furnace negative pressure; circuit breaker and disconnector switch "open / close" position signals; relay protection device operation signals; fault alarms; and equipment status and alarm parameters such as the operation / stop / fault status of main auxiliary equipment (such as water pumps and fans).

[0021] Based on this, extracting historical fault data from historical operational data helps to build a data-driven fault early warning model based on historical fault data.

[0022] Step 102: Based on historical operating data, construct the first fault early warning model of the generator set through unsupervised learning; based on historical fault data, construct the second fault early warning model of the generator set through supervised learning; and construct the mechanism model of each key component of the generator set.

[0023] In this embodiment, considering the problem that traditional fault early warning models are insufficient for predicting unknown or latent faults during generator operation, in addition to constructing a second fault early warning model for the generator based on historical fault data through supervised learning, a first fault early warning model for the generator is also constructed based on historical operating data through unsupervised learning. Mechanism models of various key components of the generator are also constructed so that the final intelligent monitoring results can be determined by combining the generator operation mechanism and the unsupervised model, thereby improving the accuracy of the intelligent monitoring process of the generator.

[0024] For example, unsupervised learning can employ algorithms such as the Isolation Forest algorithm or clustering algorithm, while supervised learning can employ classification models (such as logistic regression, support vector machine, etc.) or regression models (such as linear regression, ridge regression, etc.). The mechanism models of each key component of the generator set can include the mechanism models of key components such as boilers and steam turbines, which can be determined according to actual needs.

[0025] Step 103: Based on the first fault warning model, the second fault warning model, and the mechanism model, obtain the first fault warning result, the second fault warning result, and the third fault warning result corresponding to the real-time operation data of the generator set.

[0026] For example, in this embodiment, the real-time operating data of the generator set can be input into the pre-constructed first fault warning model, second fault warning model and mechanism model respectively, so as to obtain the corresponding first fault warning result, second fault warning result and third fault warning result.

[0027] Step 104: Calculate the similarity between real-time operating data and historical fault data, and fuse the first fault warning result, the second fault warning result and the third fault warning result according to the similarity to obtain the intelligent monitoring result of the generator set.

[0028] In this embodiment, after obtaining the first, second, and third fault warning results, to obtain accurate intelligent monitoring results for the generator set, the similarity between real-time operating data and historical fault data can be calculated first. If the similarity between real-time operating data and historical fault data is high, it indicates that the fault corresponding to the real-time operating data is more likely to be a historically known fault. If the similarity between real-time operating data and historical fault data is low, it indicates that the fault corresponding to the real-time operating data is more likely to be an unknown or latent fault. Accordingly, the first, second, and third fault warning results can be fused in different ways to obtain the intelligent monitoring results for the generator set.

[0029] In one embodiment, calculating the similarity between real-time operational data and historical fault data includes: Cluster the historical fault data to obtain the cluster center for each fault type.

[0030] Calculate the distance between the real-time running data and the cluster center of each fault type, and use this distance as the similarity between the real-time running data and each fault type.

[0031] For example, clustering historical fault data to obtain cluster centers for each fault type includes: Historical fault data are categorized according to operating conditions.

[0032] Clustering is performed on historical fault data under each operating condition to obtain the cluster center for each fault type under each operating condition.

[0033] In this embodiment, to more accurately measure the similarity between real-time operating data and historical fault data, the fault types corresponding to the historical fault data can be statistically analyzed first. Specifically, considering the complexity of generator set operating conditions, a fault type classification can be performed for historical fault data under each operating condition, thereby allowing for a more refined determination of the similarity between real-time operating data and historical fault data.

[0034] In one embodiment, the first fault warning result, the second fault warning result, and the third fault warning result are fused based on similarity to obtain the intelligent monitoring result of the generator set, including: If the similarity between the real-time operating data and each type of fault exceeds the set similarity threshold, the intelligent monitoring result of the generator set is obtained based on the second fault warning result.

[0035] If there is no similarity higher than the set similarity threshold between the real-time operating data and each type of fault, then the intelligent monitoring result of the generator set is obtained based on the first fault warning result, the second fault warning result, and the third fault warning result.

[0036] In this embodiment, if the similarity between real-time operating data and historical fault data is high, it indicates that the fault corresponding to the real-time operating data is more likely to be a historically known fault. Therefore, the intelligent monitoring result of the generator set can be directly obtained based on the second fault warning result. If the similarity between real-time operating data and historical fault data is low, it indicates that the fault corresponding to the real-time operating data is more likely to be an unknown or latent fault. Therefore, the intelligent monitoring result of the generator set can be obtained by combining the first fault warning result, the second fault warning result, and the third fault warning result.

[0037] In one embodiment, the intelligent monitoring results of the generator set are obtained based on the first fault warning result, the second fault warning result, and the third fault warning result, including: Determine whether the first fault warning result and the third fault warning result are consistent.

[0038] If the first fault warning result and the third fault warning result are consistent, the intelligent monitoring result of the generator set can be obtained based on the first fault warning result or the third fault warning result.

[0039] If the first fault warning result and the third fault warning result are inconsistent, then the higher confidence fault warning result is obtained from the first fault warning result and the third fault warning result.

[0040] The intelligent monitoring results of the generator set are obtained based on the high-confidence fault early warning results.

[0041] In this embodiment, considering that if the similarity between real-time operating data and each fault type does not exceed a set similarity threshold, it indicates that the confidence level of the second fault warning result is low. Therefore, the judgment can be based primarily on the first and third fault warning results. If the first and third fault warning results are consistent, it indicates that the confidence levels of both the first and third fault warning results are high. Therefore, the intelligent monitoring result of the generator set can be directly obtained based on either the first or third fault warning result. If the first and third fault warning results are inconsistent, it is necessary to further determine the high-confidence fault warning result among the first and third fault warning results to obtain the intelligent monitoring result of the generator set based on the high-confidence fault warning result.

[0042] In one embodiment, obtaining the high-confidence fault warning result from the first fault warning result and the third fault warning result includes: Determine whether the first fault warning result is consistent with the second fault warning result, or whether the third warning result is consistent with the second fault warning result.

[0043] If the first fault warning result is consistent with the second fault warning result, then the first fault warning result is determined as a high-confidence fault warning result.

[0044] If the third fault warning result is consistent with the second fault warning result, then the third fault warning result will be determined as a high-confidence fault warning result.

[0045] In this embodiment, when determining the high-confidence fault warning result among the first and third fault warning results, the second fault warning result is considered for auxiliary judgment. If the first fault warning result is consistent with the second fault warning result, it indicates that the confidence of the first fault warning result is higher. If the third fault warning result is consistent with the second fault warning result, it indicates that the confidence of the third fault warning result is higher. Accordingly, when the generator set is suspected of having an unknown fault or a hidden fault, a high-confidence fault warning result can be obtained more accurately.

[0046] In one embodiment, obtaining the high-confidence fault warning result from the first fault warning result and the third fault warning result includes: The generator set's previous operating data and the corresponding intelligent monitoring results for the previous period are obtained using a preset sliding window.

[0047] The calculation includes the first degree of change of the real-time operating data relative to the operating data of the previous period, the second degree of change of the first fault warning result relative to the intelligent monitoring result of the previous period, and the third degree of change of the third fault warning result relative to the intelligent monitoring result of the previous period.

[0048] If the second degree of change is relatively small compared to the first degree of change, then the first fault warning result is determined as a high-confidence fault warning result.

[0049] If the third degree of change is relatively small compared to the first degree of change, then the third fault warning result is determined as a high-confidence fault warning result.

[0050] In this embodiment, when determining the high-confidence fault warning result among the first and third fault warning results, the system considers combining the previous period's operating data and its corresponding previous period's intelligent monitoring results for auxiliary judgment. Specifically, considering the continuity between generator set operating data and the causal relationship between generator set operating data and intelligent monitoring results, generally, the intelligent monitoring results of the generator set should correspond to its operating data. Therefore, the system calculates the first degree of change of real-time operating data relative to the previous period's operating data, the second degree of change of the first fault warning result relative to the previous period's intelligent monitoring results, and the third degree of change of the third fault warning result relative to the previous period's intelligent monitoring results. The more reliable fault warning result is determined by comparing the degree of change in the intelligent monitoring results with the degree of change in the operating data.

[0051] For example, when calculating the first degree of change of real-time running data relative to the running data of the previous period, the difference between the real-time running data and each group of running data in the previous period can be calculated, and the average of the differences can be used as the first degree of change of real-time running data relative to the running data of the previous period. In this way, the sudden change characteristics of real-time running data can be accurately extracted when the running data of the previous period is relatively stable.

[0052] Alternatively, the first difference between the real-time running data and the running data at the previous moment can be calculated, and the second difference between each pair of adjacent sets of running data in the previous period can be calculated. The first degree of change of the real-time running data relative to the running data of the previous period can be obtained based on the degree of change of the first difference relative to each second difference. Thus, when the running data of the previous period changes continuously, the continuous change trend characteristics of the real-time running data can be accurately extracted.

[0053] Based on this, considering that intelligent monitoring results, first fault warning results, and third fault warning results generally indicate whether a fault exists, or represent the trend of fault changes such as early, middle, and late stages of a fault, the degree of inconsistency between the first fault warning result or the third fault warning result and the intelligent monitoring results of the previous period can be calculated when the intelligent monitoring results, first fault warning results, and third fault warning results generally indicate whether a fault exists. This degree of inconsistency can be used as the corresponding second degree of change. For example, if there are 7 normal and 3 fault results in the intelligent monitoring results of the previous period, and if the first fault warning result or the third fault warning result indicates a fault, the degree of inconsistency of the first fault warning result or the third fault warning result can be quantified as 1 / 3. If the first fault warning result or the third fault warning result indicates a normal result, the degree of inconsistency of the first fault warning result or the third fault warning result can be quantified as 1 / 7.

[0054] Alternatively, when the intelligent monitoring results, the first fault warning results, and the third fault warning results generally represent the fault change trend, the degree of inconsistency between the fault change trend represented by the first fault warning result or the fault change trend represented by the third fault warning result and the fault change trend represented by the intelligent monitoring results of the previous period is calculated, and this degree of inconsistency is taken as the corresponding third degree of change.

[0055] Based on this, if the second degree of change is relatively smaller than the first degree of change, it indicates that the first fault warning result is more consistent with the degree of change corresponding to the real-time operating data, and the first fault warning result is determined as a high-confidence fault warning result. If the third degree of change is relatively smaller than the first degree of change, it indicates that the third fault warning result is more consistent with the degree of change corresponding to the real-time operating data, and the third fault warning result is determined as a high-confidence fault warning result.

[0056] This invention acquires historical operating data of a generator set based on a DCS system and extracts historical fault data from this data. Then, based on this historical operating data, a first fault warning model for the generator set is constructed using unsupervised learning. Based on the historical fault data, a second fault warning model is constructed using supervised learning. Mechanism models of various key components of the generator set are also constructed. Then, based on the first, second, and mechanistic models, first, second, and third fault warning results corresponding to the real-time operating data of the generator set are obtained, respectively. The similarity between the real-time operating data and the historical fault data is calculated, and the first, second, and third fault warning results are fused based on the similarity to obtain the intelligent monitoring result of the generator set. This comprehensively considers the generator set's operating mechanism, the unsupervised model, and the supervised model to determine the final intelligent monitoring result, overcoming the problem of insufficient early warning for unknown faults in the intelligent monitoring process of the generator set and improving the accuracy of the intelligent monitoring process.

[0057] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0058] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0059] Figure 2 A schematic diagram of the intelligent monitoring device for generator sets provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below: like Figure 2 As shown, the intelligent monitoring device for generator sets includes: The data acquisition module 21 is used to acquire historical operating data of the generator set based on the DCS system and extract historical fault data from the historical operating data.

[0060] The model building module 22 is used to build a first fault early warning model of the generator set based on historical operating data through unsupervised learning, and to build a second fault early warning model of the generator set based on historical fault data through supervised learning, and to build the mechanism model of each key component of the generator set.

[0061] The first fault monitoring module 23 is used to obtain the first fault warning result, the second fault warning result and the third fault warning result corresponding to the real-time operation data of the generator set according to the first fault warning model, the second fault warning model and the mechanism model.

[0062] The second fault monitoring module 24 is used to calculate the similarity between real-time operating data and historical fault data, and to fuse the first fault warning result, the second fault warning result and the third fault warning result based on the similarity to obtain the intelligent monitoring result of the generator set.

[0063] In one possible implementation, the second fault monitoring module 24 is specifically used for: Cluster the historical fault data to obtain the cluster center for each fault type.

[0064] Calculate the distance between the real-time running data and the cluster center of each fault type, and use this distance as the similarity between the real-time running data and each fault type.

[0065] In one possible implementation, the second fault monitoring module 24 is specifically used for: If the similarity between the real-time operating data and each type of fault exceeds the set similarity threshold, the intelligent monitoring result of the generator set is obtained based on the second fault warning result.

[0066] If there is no similarity higher than the set similarity threshold between the real-time operating data and each type of fault, then the intelligent monitoring result of the generator set is obtained based on the first fault warning result, the second fault warning result, and the third fault warning result.

[0067] In one possible implementation, the second fault monitoring module 24 is specifically used for: Determine whether the first fault warning result and the third fault warning result are consistent.

[0068] If the first fault warning result and the third fault warning result are consistent, the intelligent monitoring result of the generator set can be obtained based on the first fault warning result or the third fault warning result.

[0069] If the first fault warning result and the third fault warning result are inconsistent, then the higher confidence fault warning result is obtained from the first fault warning result and the third fault warning result.

[0070] The intelligent monitoring results of the generator set are obtained based on the high-confidence fault early warning results.

[0071] In one possible implementation, the second fault monitoring module 24 is specifically used for: Determine whether the first fault warning result is consistent with the second fault warning result, or whether the third warning result is consistent with the second fault warning result.

[0072] If the first fault warning result is consistent with the second fault warning result, then the first fault warning result is determined as a high-confidence fault warning result.

[0073] If the third fault warning result is consistent with the second fault warning result, then the third fault warning result will be determined as a high-confidence fault warning result.

[0074] In one possible implementation, the second fault monitoring module 24 is specifically used for: The generator set's previous operating data and the corresponding intelligent monitoring results for the previous period are obtained using a preset sliding window.

[0075] The calculation includes the first degree of change of the real-time operating data relative to the operating data of the previous period, the second degree of change of the first fault warning result relative to the intelligent monitoring result of the previous period, and the third degree of change of the third fault warning result relative to the intelligent monitoring result of the previous period.

[0076] If the second degree of change is relatively small compared to the first degree of change, then the first fault warning result is determined as a high-confidence fault warning result.

[0077] If the third degree of change is relatively small compared to the first degree of change, then the third fault warning result is determined as a high-confidence fault warning result.

[0078] In one possible implementation, the second fault monitoring module 24 is specifically used for: Historical fault data are categorized according to operating conditions.

[0079] Clustering is performed on historical fault data under each operating condition to obtain the cluster center for each fault type under each operating condition.

[0080] Figure 3 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 3 As shown, the electronic device 3 of this embodiment includes a processor 30 and a memory 31. The memory 31 stores a computer program 32. When the processor 30 executes the computer program 32, it implements the steps in the various method embodiments described above. Alternatively, when the processor 30 executes the computer program 32, it implements the functions of each module / unit in the various device embodiments described above.

[0081] For example, computer program 32 may be divided into one or more modules / units, which are stored in memory 31 and executed by processor 30 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 32 in electronic device 3.

[0082] Electronic device 3 may include, but is not limited to, processor 30 and memory 31. Those skilled in the art will understand that... Figure 3This is merely an example of electronic device 3 and does not constitute a limitation on electronic device 3. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 3 may also include input / output devices, network access devices, buses, etc.

[0083] The processor 30 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0084] The memory 31 can be an internal storage unit of the electronic device 3, such as a hard disk or memory of the electronic device 3. The memory 31 can also be an external storage device of the electronic device 3, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 3. Furthermore, the memory 31 can include both internal and external storage units of the electronic device 3. The memory 31 is used to store the computer program 32 and other programs and data required by the electronic device 3. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0085] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0086] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the methods described in the above-described method embodiments.

[0087] Computer programs include computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. Computer-readable media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0088] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0089] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for intelligent monitoring of generator sets, characterized in that, include: Historical operating data of the generator set is acquired based on the DCS system, and historical fault data is extracted from the historical operating data. Based on the historical operating data, a first fault early warning model for the generator set is constructed through unsupervised learning. Based on the historical fault data, a second fault early warning model for the generator set is constructed through supervised learning. Mechanism models of various key components of the generator set are also constructed. Based on the first fault warning model, the second fault warning model and the mechanism model, the first fault warning result, the second fault warning result and the third fault warning result corresponding to the real-time operation data of the generator set are obtained respectively. The similarity between the real-time operating data and the historical fault data is calculated, and the first fault warning result, the second fault warning result and the third fault warning result are fused according to the similarity to obtain the intelligent monitoring result of the generator set.

2. The intelligent monitoring method for generator sets according to claim 1, characterized in that, Calculating the similarity between the real-time operational data and the historical fault data includes: Cluster the historical fault data to obtain the cluster center for each fault type; The distance between the real-time running data and the cluster center of each fault type is calculated as the similarity between the real-time running data and each fault type.

3. The intelligent monitoring method for generator sets according to claim 2, characterized in that, The first fault warning result, the second fault warning result, and the third fault warning result are fused based on the similarity to obtain the intelligent monitoring result of the generator set, including: If the similarity between the real-time operating data and each type of fault exceeds a set similarity threshold, then the intelligent monitoring result of the generator set is obtained based on the second fault warning result. If there is no similarity higher than the set similarity threshold between the real-time operating data and each type of fault, then the intelligent monitoring result of the generator set is obtained based on the first fault warning result, the second fault warning result and the third fault warning result.

4. The intelligent monitoring method for generator sets according to claim 3, characterized in that, The intelligent monitoring results of the generator set are obtained based on the first fault warning result, the second fault warning result, and the third fault warning result, including: Determine whether the first fault warning result and the third fault warning result are consistent; If the first fault warning result and the third fault warning result are consistent, then the intelligent monitoring result of the generator set is obtained based on the first fault warning result or the third fault warning result. If the first fault warning result and the third fault warning result are inconsistent, then the higher confidence fault warning result between the first fault warning result and the third fault warning result is obtained; The intelligent monitoring results of the generator set are obtained based on the high-confidence fault early warning results.

5. The intelligent monitoring method for generator sets according to claim 4, characterized in that, Obtaining the high-confidence fault warning result from the first fault warning result and the third fault warning result includes: Determine whether the first fault warning result is consistent with the second fault warning result, or whether the third warning result is consistent with the second fault warning result; If the first fault warning result is consistent with the second fault warning result, then the first fault warning result is determined as a high-confidence fault warning result; If the third fault warning result is consistent with the second fault warning result, then the third fault warning result is determined as a high-confidence fault warning result.

6. The intelligent monitoring method for generator sets according to claim 4, characterized in that, Obtaining the high-confidence fault warning result from the first fault warning result and the third fault warning result includes: The previous period's operating data of the generator set and the corresponding intelligent monitoring results of the previous period are obtained by using a preset sliding window. Calculate the first degree of change of the real-time operating data relative to the operating data of the previous period, the second degree of change of the first fault warning result relative to the intelligent monitoring result of the previous period, and the third degree of change of the third fault warning result relative to the intelligent monitoring result of the previous period; If the second degree of change is relatively small compared to the first degree of change, then the first fault warning result is determined as a high-confidence fault warning result; If the third degree of change is relatively small compared to the first degree of change, then the third fault warning result is determined as a high-confidence fault warning result.

7. The intelligent monitoring method for generator sets according to claim 2, characterized in that, Clustering is performed on the historical fault data to obtain cluster centers for each fault type, including: The historical fault data is categorized according to operating conditions; Clustering is performed on historical fault data under each operating condition to obtain the cluster center for each fault type under each operating condition.

8. An intelligent monitoring device for generator sets, characterized in that, include: The data acquisition module is used to acquire historical operating data of the generator set based on the DCS system, and extract historical fault data from the historical operating data; The model building module is used to build a first fault early warning model of the generator set through unsupervised learning based on the historical operating data, and to build a second fault early warning model of the generator set through supervised learning based on the historical fault data, and to build a mechanism model of each key component of the generator set. The first fault monitoring module is used to obtain the first fault warning result, the second fault warning result and the third fault warning result corresponding to the real-time operation data of the generator set according to the first fault warning model, the second fault warning model and the mechanism model, respectively. The second fault monitoring module is used to calculate the similarity between the real-time operating data and the historical fault data, and to fuse the first fault warning result, the second fault warning result and the third fault warning result according to the similarity to obtain the intelligent monitoring result of the generator set.

9. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 6.

Images