Intelligent power plant data management method and system based on big data

Abstract

The invention provides an intelligent power plant data management method and system based on big data, and the method comprises the steps: carrying out the primary clustering analysis of historical power plant data, and dividing the power plant data into a plurality of types of primary data according to the primary clustering analysis result; carrying out secondary clustering analysis on the plurality of primary data so as to divide the plurality of primary data into a plurality of secondary data according to a secondary clustering classification result; performing anomaly detection on each type of secondary data, extracting abnormal data from each type of secondary data according to an anomaly detection result, selecting an observation time period, generating a sample data sequence, and constructing an anomaly detection model of each type of secondary data according to the sample data sequence; and obtaining target power plant data of the target power plant every first preset time, inputting the target power plant data into the anomaly detection model to obtain a detection result of each type of secondary data in each period, and uploading the detection result. According to the invention, mass power plant data generated in real time can be rapidly processed, and detection results can be orderly stored.

Description

Technical Field

[0001] The present invention relates to the technical field of power plant data management, and in particular to a smart power plant data management method and system based on big data. Background Art

[0002] With the rapid development of the energy industry, power plants are the core link of energy supply, and the collection, processing and analysis of their operating data are crucial to ensuring power plant safety and improving operational efficiency.

[0003] However, with the widespread application of technologies such as the Internet of Things and big data, the amount of data generated by power plants has exploded. Traditional power plant data management methods have problems such as low data processing efficiency, insufficient data value mining, and untimely abnormal warnings. It is difficult to meet the current needs of intelligent and refined management of power plants. Based on this, how to effectively manage this data, identify abnormal data, and build an anomaly detection model based on it to achieve timely discovery and feedback of power plant problems has become a technical problem that needs to be urgently solved in the current power plant data management field. Summary of the invention

[0004] The purpose of the present invention is to provide a smart power plant data management method and system based on big data, aiming to solve the problems of low data processing efficiency, insufficient data value mining, and untimely abnormal warning existing in traditional technologies.

[0005] In a first aspect, the present invention provides a smart power plant data management method based on big data, the method comprising:

[0006] Acquire historical power plant data related to the target power plant, and perform a cluster analysis on the historical power plant data, so as to divide the power plant data into a plurality of primary data according to the cluster analysis result;

[0007] Performing secondary cluster analysis on the multiple types of primary data to divide the multiple types of primary data into multiple types of secondary data according to the secondary cluster classification result;

[0008] Performing anomaly detection on each type of secondary data to extract abnormal data from each type of secondary data according to the anomaly detection result, selecting an observation period to generate a sample data sequence corresponding to each type of secondary data according to the abnormal data in the observation period, and constructing an anomaly detection model for each type of secondary data according to the sample data sequence;

[0009] The target power plant data of the target power plant is obtained at first preset time intervals, and the target power plant data is divided into multiple target secondary data, so that the target secondary data is input into the anomaly detection model, the detection results of each secondary data in each cycle are obtained, and the detection results are uploaded.

[0010] Furthermore, the step of acquiring historical power plant data related to the target power plant and performing a cluster analysis on the historical power plant data to divide the power plant data into multiple types of primary data according to the cluster analysis result includes:

[0011] Initialize K1 cluster centers randomly, calculate the Euclidean distance from each historical power plant data to each cluster center, and select the minimum Euclidean distance from all Euclidean distances under the same historical power plant data to divide the historical power plant data into the cluster to which the minimum Euclidean distance belongs;

[0012] In this way, all historical power plant data are traversed to obtain K1 classified clusters, each cluster corresponding to a type of primary data.

[0013] Furthermore, the step of performing secondary cluster analysis on the plurality of primary data to divide the plurality of primary data into a plurality of secondary data according to the secondary cluster classification result includes:

[0014] Randomly initialize K2 cluster centers to obtain K2 classified clusters.

[0015] The step of calculating the Euclidean distance from each historical power plant data to each cluster center includes:

[0016] The Euclidean distance from each historical power plant data to the cluster center is calculated according to the following formula:

[0017]

[0018] Among them, d(X i , Y j ) represents the Euclidean distance between the i-th historical power plant data and the j-th cluster center, x in represents the value of the i-th historical power plant data in the n-th dimension, y jn Represents the value of the j-th cluster center in the n-th dimension.

[0019] Furthermore, the step of performing anomaly detection on each type of secondary data to extract abnormal data from each type of secondary data according to the anomaly detection result includes:

[0020] Obtaining a data mean and a data standard deviation for each type of secondary data, and setting a detection threshold according to the data standard deviation;

[0021] Obtaining the absolute value of the difference between the secondary data and the corresponding data mean, and determining whether the absolute value of the difference is greater than a detection threshold;

[0022] If the absolute value of the difference is greater than the detection threshold, the secondary data is determined to be abnormal data;

[0023] If the absolute value of the difference is less than or equal to the detection threshold, it is determined that the secondary data is normal data.

[0024] Further, the step of selecting the observation period to generate a sample data sequence corresponding to each type of secondary data based on the abnormal data in the observation period and constructing an abnormal detection model for each type of secondary data includes:

[0025] Assume the initial time t 0 , then taking the time period from time t 0 - t 1 to time t 0 as the observation period, the generated sample data sequence is E = (E 1 , E 2 , …, E a ), where E a represents the a-th abnormal data in the observation period;

[0026] Obtain at least one fault type related to each sample data sequence, and divide all the sample data sequences according to the fault type to obtain at least one target sample data sequence under each fault type;

[0027] Train the initial abnormal detection model under this fault type according to the at least one target sample data sequence to obtain multiple final abnormal detection models respectively corresponding to each fault type.

[0028] Further, the step of obtaining the absolute value of the difference between the secondary data and the corresponding data mean and determining whether the absolute value of the difference is greater than the detection threshold includes:

[0029] Judge whether the secondary data is abnormal data according to the following formula:

[0030] |Z ij - μ j | > 3σ j

[0031] where Z ij represents the i-th secondary data under the j-th cluster, μ j represents the data mean of all secondary data under the j-th cluster, and σ j represents the standard deviation of all secondary data under the j-th cluster.

[0032] Further, the step of obtaining the target power plant data of the target power plant every first preset time, dividing the target power plant data into multiple target secondary data, inputting the target secondary data into the abnormal detection model to obtain the detection results of each type of secondary data in each cycle, and uploading the detection results includes:

[0033] When a fault exists, the detection result is the fault type in each cycle;

[0034] The detection result and abnormal data related to the detection result are uploaded to the corresponding storage space according to the fault type.

[0035] In a second aspect, the present invention provides a smart power plant data management system based on big data, the system comprising:

[0036] A primary classification module, used for acquiring historical power plant data related to a target power plant, and performing a primary cluster analysis on the historical power plant data, so as to classify the power plant data into a plurality of primary data according to the result of the primary cluster analysis;

[0037] A secondary classification module, used for performing secondary clustering analysis on the plurality of primary data, so as to divide the plurality of primary data into a plurality of secondary data according to the secondary clustering classification result;

[0038] A detection model building module, used to perform anomaly detection on each type of secondary data, extract abnormal data from each type of secondary data according to the anomaly detection result, select an observation period, generate a sample data sequence corresponding to each type of secondary data according to the abnormal data in the observation period, and build an anomaly detection model for each type of secondary data according to the sample data sequence;

[0039] A data storage module is used to obtain the target power plant data of the target power plant at intervals of a first preset time, and divide the target power plant data into multiple target secondary data, so as to input the target secondary data into the anomaly detection model, obtain the detection results of each secondary data in each cycle, and upload the detection results.

[0040] In a third aspect, the present invention provides a storage medium, which stores one or more programs, which, when executed by a processor, implement the above-mentioned big data-based smart power plant data management method.

[0041] In a fourth aspect, the present invention provides an electronic device, the electronic device comprising a memory and a processor, wherein:

[0042] The memory is used to store computer programs;

[0043] When the processor is used to execute the computer program stored in the memory, the above-mentioned smart power plant data management method based on big data is implemented.

[0044] Compared with the prior art, the present invention has the following advantages:

[0045] 1. The present invention conducts preliminary clustering analysis and secondary clustering analysis on a large amount of historical power plant data, and then realizes the precise classification of historical power plant data, providing precise data support for subsequent precise analysis of fault types, enabling the historical power plant data to achieve the division of primary data such as operation data, status data, and environmental data. The operation data includes secondary data such as power generation, power supply coal consumption, main steam temperature, pressure, reheat steam temperature, flue gas temperature, turbine heat consumption rate, power generation make-up water rate, etc. The status data includes secondary data such as voltage, current, and temperature. The environmental data includes secondary data such as temperature, humidity, wind speed, wind direction, rainfall, and sunshine intensity.

[0046] 2. The present invention conducts anomaly detection on the secondary data, thereby generating a sample data sequence related to each type of secondary data, that is, an anomaly data set, and then precisely locks the secondary data related to each fault type based on each fault type, thereby efficiently and accurately completing the training of the anomaly detection model, so as to be able to quickly process the large amount of power plant data continuously generated by subsequent real-time detection, and can store the detection results in an orderly manner, so that in case of a fault, timely feedback and traceability can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 is a flowchart of a big data-based intelligent power plant data management method proposed in an embodiment of the present invention;

[0048] Figure 2 is a schematic structural diagram of a big data-based intelligent power plant data management system proposed in an embodiment of the present invention.

[0049] The following specific embodiments will further illustrate the present invention in conjunction with the above-mentioned drawings. SPECIFIC EMBODIMENTS

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. Unless otherwise defined, the technical terms or scientific terms used herein shall have the ordinary meaning as understood by those of ordinary skill in the art in the field to which the present invention belongs. The words such as "including" used herein mean that the elements or items appearing before this word cover the elements or items listed after this word and their equivalents, without excluding other elements or items.

[0051] As Figure 1 shown, an embodiment of the present invention provides a big data-based intelligent power plant data management method, which includes steps S101 to S104, where:

[0052] Step S101: acquiring historical power plant data related to a target power plant, and performing a cluster analysis on the historical power plant data, so as to divide the power plant data into multiple types of primary data according to the cluster analysis result;

[0053] Step S102: performing secondary cluster analysis on the multiple types of primary data, so as to divide the multiple types of primary data into multiple types of secondary data according to the secondary cluster classification result;

[0054] It should be noted that the amount of data of the historical power plant data is very large, including primary data such as operation data, status data, and environmental data. At the same time, the operation data includes secondary data such as power generation, power supply coal consumption, main steam temperature, pressure, reheat steam temperature, exhaust temperature, turbine heat rate, power generation feed water rate, etc. The status data includes secondary data such as voltage, current, and temperature. The environmental data includes secondary data such as temperature, humidity, wind speed, wind direction, rainfall, and sunshine intensity. Based on this, it is necessary to perform two clustering analyses on these data in order to obtain the bottom-level data.

[0055] Specifically, in some embodiments, it is first necessary to randomly initialize K1 cluster centers, calculate the Euclidean distance from each historical power plant data to each cluster center, and select the minimum Euclidean distance from all Euclidean distances under the same historical power plant data to divide the historical power plant data into the cluster to which the minimum Euclidean distance belongs; in this way, all historical power plant data are traversed to obtain K1 classified clusters, each cluster corresponding to a type of primary data.

[0056] It should be pointed out that the method of secondary cluster analysis is basically the same as that of primary cluster analysis. The difference lies in the number of selected cluster centers and the objects of cluster analysis. Therefore, in secondary clustering, K2 cluster centers will be randomly initialized to obtain K2 classified clusters.

[0057] More specifically, the Euclidean distance from each historical power plant data to the cluster center is calculated according to the following formula:

[0058]

[0059] Among them, d(X i , Y j ) represents the Euclidean distance between the i-th historical power plant data and the j-th cluster center, x in represents the value of the i-th historical power plant data in the n-th dimension, y jn Represents the value of the j-th cluster center in the n-th dimension.

[0060] It should also be noted that, in some embodiments, the number of times selected for the two clustering analyses is generally related to the number of data clusters required to be divided, that is, they are generally equal.

[0061] Step S103: performing anomaly detection on each type of secondary data to extract abnormal data from each type of secondary data according to the anomaly detection result, selecting an observation period to generate a sample data sequence corresponding to each type of secondary data according to the abnormal data in the observation period, and constructing an anomaly detection model for each type of secondary data according to the sample data sequence;

[0062] It should be pointed out that in the anomaly detection process, it is first necessary to obtain the data mean and data standard deviation of each type of secondary data, and set the detection threshold according to the data standard deviation; then obtain the absolute value of the difference between the secondary data and the corresponding data mean, and determine whether the absolute value of the difference is greater than the detection threshold; if the absolute value of the difference is greater than the detection threshold, the secondary data is determined to be abnormal data; if the absolute value of the difference is less than or equal to the detection threshold, the secondary data is determined to be normal data.

[0063] Specifically, whether the secondary data is abnormal data is determined according to the following formula:

[0064] |Z ij -μ j |>3σ j

[0065] Among them, Z ij represents the i-th secondary data under the j-th cluster, μ j represents the data mean of all secondary data under the jth cluster, σ j Represents the standard deviation of all secondary data under the jth cluster.

[0066] In addition, in some embodiments, it is assumed that the initialization time t 0 , then at time t 0 -t 1 At time t 0 is the observation period, then the generated sample data sequence is E=(E 1 、E 2 ,…,E a ), where E aIndicates the ath abnormal data in the observation period; obtains at least one fault type related to each sample data sequence, and divides all sample data sequences according to the fault type to obtain at least one target sample data sequence under each fault type; trains the initial abnormality detection model under the fault type according to the at least one target sample data sequence to obtain multiple final abnormality detection models corresponding to each fault type. It should be noted that there are many types of fault types, such as short circuit faults, involving secondary data such as voltage and current data.

[0067] Step S104: acquiring the target power plant data of the target power plant at first preset time intervals, and dividing the target power plant data into a plurality of target secondary data, so as to input the target secondary data into the anomaly detection model, obtain the detection result of each secondary data in each cycle, and upload the detection result.

[0068] It should be noted that after the anomaly detection model related to each fault type is constructed, the target power plant data of the target power plant is collected in real time to monitor the target power plant in real time. The method of dividing it into multiple target secondary data is the same as the method of processing historical power plant data. Then, based on the fault type, one or more target secondary data are input into the corresponding anomaly detection model to obtain the detection result of the fault type. When a fault exists, the detection result is the fault type in each cycle; according to the fault type, the detection result and the abnormal data related to the detection result are uploaded to the corresponding storage space so that subsequent management personnel can trace the source at the same time and issue an alarm message at the same time.

[0069] In summary, the above-mentioned smart power plant data management method based on big data has the following advantages:

[0070] 1. The present invention performs preliminary cluster analysis and secondary cluster analysis on massive amounts of historical power plant data, thereby achieving accurate classification of historical power plant data, providing accurate data support for subsequent accurate analysis of fault types, so that historical power plant data can be divided into primary data such as operation data, status data, and environmental data. The operation data includes secondary data such as power generation, power supply coal consumption, main steam temperature, pressure, reheat steam temperature, exhaust temperature, turbine heat rate, power generation feed water rate, etc. The status data includes secondary data such as voltage, current, and temperature. The environmental data includes secondary data such as temperature, humidity, wind speed, wind direction, rainfall, and sunshine intensity.

[0071] 2. The present invention performs anomaly detection on the secondary data to generate a sample data sequence related to each secondary data, namely, an abnormal data set, and then accurately locks the secondary data related to each fault type, thereby efficiently and accurately training the complete anomaly detection model, so as to quickly process the massive power plant data continuously generated by subsequent real-time detection, and store the detection results in an orderly manner, so as to enable timely feedback and tracing when a fault occurs.

[0072] like Figure 2 As shown, an embodiment of the present invention further provides a smart power plant data management system based on big data, the system comprising:

[0073] A primary classification module 10, for acquiring historical power plant data related to a target power plant, and performing a primary cluster analysis on the historical power plant data, so as to classify the power plant data into a plurality of primary data according to the result of the primary cluster analysis;

[0074] A secondary classification module 20, configured to perform secondary clustering analysis on the plurality of primary data, so as to divide the plurality of primary data into a plurality of secondary data according to the secondary clustering classification result;

[0075] The detection model building module 30 is used to perform anomaly detection on each type of secondary data, extract abnormal data from each type of secondary data according to the anomaly detection result, select an observation period, generate a sample data sequence corresponding to each type of secondary data according to the abnormal data in the observation period, and build an anomaly detection model for each type of secondary data according to the sample data sequence;

[0076] The data storage module 40 is used to obtain the target power plant data of the target power plant at intervals of a first preset time, and divide the target power plant data into multiple target secondary data, so as to input the target secondary data into the anomaly detection model, obtain the detection results of each secondary data in each cycle, and upload the detection results.

[0077] On the other hand, the present invention further proposes a storage medium on which one or more programs are stored, and when the program is executed by a processor, the above-mentioned smart power plant data management method based on big data is implemented.

[0078] On the other hand, the present invention also proposes an electronic device, including a memory and a processor, wherein the memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory to implement the above-mentioned smart power plant data management method based on big data.

[0079] Those skilled in the art will appreciate that the logic and / or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable instructions for implementing logical functions, and may be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For purposes of this specification, "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0080] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0081] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or a combination thereof: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0082] Although the embodiments of the present invention are described in detail above, it is obvious to those skilled in the art that various modifications and variations can be made to these embodiments. However, it should be understood that such modifications and variations are within the scope and spirit of the present invention as described in the claims. Moreover, the present invention described herein may have other embodiments and may be implemented or realized in a variety of ways.

Claims

1. A smart power plant data management method based on big data, characterized in that: The method comprises: Acquire historical power plant data related to the target power plant, and perform a cluster analysis on the historical power plant data, so as to divide the power plant data into a plurality of primary data according to the cluster analysis result; Performing secondary cluster analysis on the multiple types of primary data to divide the multiple types of primary data into multiple types of secondary data according to the secondary cluster classification result; Performing anomaly detection on each type of secondary data to extract abnormal data from each type of secondary data according to the anomaly detection result, selecting an observation period to generate a sample data sequence corresponding to each type of secondary data according to the abnormal data in the observation period, and constructing an anomaly detection model for each type of secondary data according to the sample data sequence; The target power plant data of the target power plant is obtained at first preset time intervals, and the target power plant data is divided into multiple target secondary data, so that the target secondary data is input into the anomaly detection model, the detection results of each secondary data in each cycle are obtained, and the detection results are uploaded.

2. The method for managing smart power plant data based on big data according to claim 1 is characterized in that: The step of acquiring historical power plant data related to the target power plant and performing a cluster analysis on the historical power plant data to divide the power plant data into multiple primary data according to the cluster analysis result includes: Initialize K1 cluster centers randomly, calculate the Euclidean distance from each historical power plant data to each cluster center, and select the minimum Euclidean distance from all Euclidean distances under the same historical power plant data to divide the historical power plant data into the cluster to which the minimum Euclidean distance belongs; In this way, all historical power plant data are traversed to obtain K1 classified clusters, each cluster corresponding to a type of primary data.

3. The method for managing smart power plant data based on big data according to claim 2 is characterized in that: The step of performing secondary cluster analysis on the plurality of primary data to divide the plurality of primary data into a plurality of secondary data according to the secondary cluster classification result comprises: Randomly initialize K2 cluster centers to obtain K2 classified clusters. The step of calculating the Euclidean distance from each historical power plant data to each cluster center includes: The Euclidean distance from each historical power plant data to the cluster center is calculated according to the following formula: Among them, d(X i , Y j ) represents the Euclidean distance between the i-th historical power plant data and the j-th cluster center, x in represents the value of the i-th historical power plant data in the n-th dimension, y jn Represents the value of the j-th cluster center in the n-th dimension.

4. The smart power plant data management method based on big data according to claim 3 is characterized in that: The step of performing anomaly detection on each type of secondary data to extract abnormal data from each type of secondary data according to the anomaly detection result includes: Obtaining a data mean and a data standard deviation for each type of secondary data, and setting a detection threshold according to the data standard deviation; Obtaining the absolute value of the difference between the secondary data and the corresponding data mean, and determining whether the absolute value of the difference is greater than a detection threshold; If the absolute value of the difference is greater than the detection threshold, the secondary data is determined to be abnormal data; If the absolute value of the difference is less than or equal to the detection threshold, the secondary data is determined to be normal data.

5. The method for managing smart power plant data based on big data according to claim 4 is characterized in that: The step of selecting an observation period to generate a sample data sequence corresponding to each type of secondary data according to the abnormal data in the observation period, and constructing an abnormality detection model for each type of secondary data according to the sample data sequence includes: Assuming the initialization time is t0, the observation period is from time t0-t1 to time t0, and the generated sample data sequence is E=(E1, E2, ..., E a ), where E a Indicates the ath abnormal data in the observation period; Acquire at least one fault type associated with each sample data sequence, and divide all sample data sequences according to the fault type to obtain at least one target sample data sequence under each fault type; The initial anomaly detection model under the fault type is trained according to the at least one target sample data sequence to obtain a plurality of final anomaly detection models corresponding to each fault type.

6. The method for managing smart power plant data based on big data according to claim 4 is characterized in that: The step of obtaining the absolute value of the difference between the secondary data and the corresponding data mean, and determining whether the absolute value of the difference is greater than a detection threshold comprises: Determine whether the secondary data is abnormal data according to the following formula: |Z ij -m j |>3s j Among them, Z ij represents the i-th secondary data under the j-th cluster, μ j represents the data mean of all secondary data under the jth cluster, σ j Represents the standard deviation of all secondary data under the jth cluster.

7. The method for managing smart power plant data based on big data according to claim 1, characterized in that: The step of acquiring target power plant data of the target power plant at first preset intervals, dividing the target power plant data into a plurality of target secondary data, inputting the target secondary data into the anomaly detection model, obtaining the detection result of each secondary data in each cycle, and uploading the detection result comprises: When a fault exists, the detection result is the fault type in each cycle; The detection result and abnormal data related to the detection result are uploaded to the corresponding storage space according to the fault type.

8. A smart power plant data management system based on big data, characterized in that: The system comprises: A primary classification module, used for acquiring historical power plant data related to a target power plant, and performing a primary cluster analysis on the historical power plant data, so as to classify the power plant data into a plurality of primary data according to the result of the primary cluster analysis; A secondary classification module, used for performing secondary clustering analysis on the plurality of primary data, so as to divide the plurality of primary data into a plurality of secondary data according to the secondary clustering classification result; A detection model building module, used to perform anomaly detection on each type of secondary data, extract abnormal data from each type of secondary data according to the anomaly detection result, select an observation period, generate a sample data sequence corresponding to each type of secondary data according to the abnormal data in the observation period, and build an anomaly detection model for each type of secondary data according to the sample data sequence; A data storage module is used to obtain the target power plant data of the target power plant at intervals of a first preset time, and divide the target power plant data into multiple target secondary data, so as to input the target secondary data into the anomaly detection model, obtain the detection results of each secondary data in each cycle, and upload the detection results.

9. A storage medium, characterized in that: The storage medium stores one or more programs, which, when executed by the processor, implement the big data-based smart power plant data management method as described in any one of claims 1 to 7.

10. An electronic device, comprising a memory and a processor, wherein: The memory is used to store computer programs; When the processor is used to execute the computer program stored in the memory, it implements the big data-based smart power plant data management method as described in any one of claims 1-7.

Images