Power transformer oil temperature prediction method and system based on multi-factor feature fusion

Abstract

The invention discloses a power transformer oil temperature prediction method and system based on multi-factor feature fusion, and the method comprises the steps: obtaining the load rate prediction data and the environment temperature prediction data of a power transformer at a future moment, and inputting the load rate prediction data and the environment temperature prediction data into a top oil temperature prediction model, the top oil temperature prediction model outputs a predicted top oil temperature value at a future moment; acquiring a second historical actual top oil temperature sequence within a period of historical time according to a preset extraction rule, and judging whether a difference value between the predicted top oil temperature value and each historical actual top oil temperature value in the first historical actual top oil temperature sequence is greater than a preset threshold value or not; if yes, the predicted top oil temperature value is corrected according to the average value of all the first historical actual top oil temperatures in the first historical actual top oil temperature sequence. The error of model prediction can be reduced, so that the corrected result is closer to a real value.

Description

Technical Field

[0001] The present invention belongs to the technical field of transformer oil temperature prediction, and in particular relates to a method and system for predicting the oil temperature of a power transformer based on multi-factor feature fusion. Background Art

[0002] Oil-immersed power transformers are a common type of equipment in power systems. The safety and reliability of their operation directly affect the safety and stability of the power grid. Oil-immersed power transformers require oil to cool and isolate the conductive components. Therefore, oil temperature is an important parameter. Whether its subsequent change trajectory during the operation of the power transformer can be accurately predicted is of great significance for timely fault detection and optimized load management of the power transformer.

[0003] The existing methods for predicting the top oil temperature of power transformers mainly include the empirical thermal model calculation method given in the IEEE and IEC guidelines and the equivalent thermal path method, as well as the prediction method based on artificial intelligence algorithms. In the empirical thermal model calculation method, the roughly evaluated fixed parameter values ​​cannot accurately reflect the current performance of the transformer, the degree of oil aging, and other information, thus affecting the prediction accuracy of the top oil temperature. Summary of the invention

[0004] The present invention provides a method and system for predicting the oil temperature of a power transformer based on multi-factor feature fusion, which are used to solve the technical problem of more accurately predicting the top oil temperature of a power transformer.

[0005] In a first aspect, the present invention provides a method for predicting oil temperature of a power transformer based on multi-factor feature fusion, comprising:

[0006] Acquire target historical data measured by the power transformer in a historical time period, wherein the target historical data includes load rate historical data and ambient temperature historical data;

[0007] Outputting at least one target historical data to a pre-built top layer oil temperature prediction model, the top layer oil temperature prediction model outputting the historical predicted top layer oil temperature of the power transformer corresponding to the at least one target historical data, and sorting the historical predicted top layer oil temperatures based on the time sequence to obtain a historical predicted top layer oil temperature sequence;

[0008] Obtaining load factor prediction data and ambient temperature prediction data of the power transformer at a future time, and inputting the load factor prediction data and ambient temperature prediction data into the top layer oil temperature prediction model, wherein the top layer oil temperature prediction model outputs a predicted top layer oil temperature value at a future time;

[0009] Acquire a second historical actual top oil temperature sequence within a historical period of time according to a preset extraction rule, and determine whether a difference between the predicted top oil temperature value and each historical actual top oil temperature value in the first historical actual top oil temperature sequence is greater than a preset threshold;

[0010] If it is greater than a preset threshold, the predicted top layer oil temperature value is corrected according to the average value of each first historical actual top layer oil temperature in the first historical actual top layer oil temperature sequence to obtain a final target predicted top layer oil temperature value.

[0011] In a second aspect, the present invention provides a power transformer oil temperature prediction system based on multi-factor feature fusion, comprising:

[0012] An acquisition module configured to acquire target historical data measured by the power transformer in a historical time period, wherein the target historical data includes load rate historical data and ambient temperature historical data;

[0013] A first output module is configured to output at least one target historical data to a pre-built top layer oil temperature prediction model, wherein the top layer oil temperature prediction model outputs a historical predicted top layer oil temperature of the power transformer corresponding to the at least one target historical data, and sorts each historical predicted top layer oil temperature based on a time sequence to obtain a historical predicted top layer oil temperature sequence;

[0014] A second output module is configured to obtain load rate prediction data and ambient temperature prediction data of the power transformer at a future time, and input the load rate prediction data and ambient temperature prediction data into the top layer oil temperature prediction model, and the top layer oil temperature prediction model outputs a predicted top layer oil temperature value at a future time;

[0015] a judgment module configured to obtain a second historical actual top oil temperature sequence within a historical period of time according to a preset extraction rule, and judge whether a difference between the predicted top oil temperature value and each historical actual top oil temperature value in the first historical actual top oil temperature sequence is greater than a preset threshold;

[0016] The correction module is configured to correct the predicted top oil temperature value according to the average value of each first historical actual top oil temperature in the first historical actual top oil temperature sequence if it is greater than a preset threshold, so as to obtain a final target predicted top oil temperature value.

[0017] According to a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of the method for predicting oil temperature of a power transformer based on multi-factor feature fusion according to any embodiment of the present invention.

[0018] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the steps of the method for predicting oil temperature of a power transformer based on multi-factor feature fusion according to any embodiment of the present invention.

[0019] The present application discloses a method and system for predicting the oil temperature of an electric transformer based on multi-factor feature fusion, which obtains load rate prediction data and ambient temperature prediction data of the electric transformer at a future moment, and inputs the load rate prediction data and ambient temperature prediction data into a top oil temperature prediction model, which outputs a predicted top oil temperature value at a future moment; obtains a second historical actual top oil temperature sequence within a historical period of time according to a preset extraction rule, and determines whether the difference between the predicted top oil temperature value and each historical actual top oil temperature value in the first historical actual top oil temperature sequence is greater than a preset threshold value; if it is greater than the preset threshold value, the predicted top oil temperature value is corrected according to the average value of each first historical actual top oil temperature in the first historical actual top oil temperature sequence, and the final target predicted top oil temperature value is obtained, which can reduce the error of the model prediction, thereby making the corrected result closer to the true value. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 A flowchart of a method for predicting oil temperature of a power transformer based on multi-factor feature fusion provided by an embodiment of the present invention;

[0022] Figure 2 A structural block diagram of a power transformer oil temperature prediction system based on multi-factor feature fusion provided by an embodiment of the present invention;

[0023] Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] See also Figure 1 , which shows a flow chart of a method for predicting oil temperature of a power transformer based on multi-factor feature fusion of the present application.

[0026] like Figure 1 As shown in FIG. 1 , the power transformer oil temperature prediction method based on multi-factor feature fusion specifically includes the following steps:

[0027] Step S101, obtaining target historical data measured by the power transformer in a historical time period, wherein the target historical data includes load rate historical data and ambient temperature historical data.

[0028] Step S102: output at least one target historical data to a pre-constructed top oil temperature prediction model, wherein the top oil temperature prediction model outputs the historical predicted top oil temperature of the power transformer corresponding to the at least one target historical data, and sorts each historical predicted top oil temperature based on time sequence to obtain a historical predicted top oil temperature sequence.

[0029] In this step, the top oil temperature prediction model is trained based on the target historical data of the power transformer, combined with the power transformer dynamic thermal model and the extended Kalman filter algorithm containing an adaptive noise parameter estimator. At least one target historical data is output to the pre-built top oil temperature prediction model, and the top oil temperature prediction model outputs the historical predicted top oil temperature of the power transformer corresponding to the at least one target historical data, and sorts each historical predicted top oil temperature based on the time sequence to obtain a historical predicted top oil temperature sequence.

[0030] Step S103, obtaining the load rate prediction data and the ambient temperature prediction data of the power transformer at a future time, and inputting the load rate prediction data and the ambient temperature prediction data into the top oil temperature prediction model, and the top oil temperature prediction model outputs the predicted top oil temperature value at a future time.

[0031] Step S104, obtaining a second historical actual top oil temperature sequence within a historical period of time according to a preset extraction rule, and determining whether the difference between the predicted top oil temperature value and each historical actual top oil temperature value in the first historical actual top oil temperature sequence is greater than a preset threshold.

[0032] In this step, the target historical data corresponding to the load rate prediction data and the ambient temperature prediction data are searched in the historical prediction top oil temperature sequence; the historical actual top oil temperature value corresponding to the target historical data whose load rate difference and ambient temperature difference are not greater than the preset target threshold value is extracted to obtain a second historical actual top oil temperature sequence within a historical period of time, wherein the load rate difference is the difference between the load rate historical data and the load rate prediction data, and the ambient temperature difference is the difference between the ambient temperature historical data and the ambient temperature prediction data.

[0033] It should be noted that after determining whether the difference between the predicted top layer oil temperature value and each historical actual top layer oil temperature value in the first historical actual top layer oil temperature sequence is greater than a preset threshold, if it is not greater than the preset threshold, the predicted top layer oil temperature value is directly used as the final target predicted top layer oil temperature value.

[0034] Step S105: if it is greater than a preset threshold, the predicted top layer oil temperature value is corrected according to the average value of each first historical actual top layer oil temperature in the first historical actual top layer oil temperature sequence to obtain a final target predicted top layer oil temperature value.

[0035] In this step, the average value of each first historical actual top oil temperature in the first historical actual top oil temperature sequence is obtained; the target difference between the predicted top oil temperature value and the average value is calculated; if the predicted top oil temperature value is greater than the average value, half of the target difference is subtracted from the predicted top oil temperature value to obtain the final target predicted top oil temperature value; if the predicted top oil temperature value is less than the average value, half of the target difference is added to the predicted top oil temperature value to obtain the final target predicted top oil temperature value.

[0036] In summary, the method of the present application obtains the load rate prediction data and the ambient temperature prediction data of the power transformer at a future moment, and inputs the load rate prediction data and the ambient temperature prediction data into the top oil temperature prediction model, and the top oil temperature prediction model outputs the predicted top oil temperature value at a future moment; according to the preset extraction rule, the second historical actual top oil temperature sequence within a historical period of time is obtained, and it is determined whether the difference between the predicted top oil temperature value and each historical actual top oil temperature value in the first historical actual top oil temperature sequence is greater than the preset threshold value; if it is greater than the preset threshold value, the predicted top oil temperature value is corrected according to the average value of each first historical actual top oil temperature in the first historical actual top oil temperature sequence to obtain the final target predicted top oil temperature value, which can reduce the error of the model prediction, thereby making the corrected result closer to the true value.

[0037] See also Figure 2 , which shows a structural block diagram of a power transformer oil temperature prediction system based on multi-factor feature fusion of the present application.

[0038] like Figure 2 As shown, the power transformer oil temperature prediction system 200 includes an acquisition module 210 , a first output module 220 , a second output module 230 , a judgment module 240 and a correction module 250 .

[0039] Among them, the acquisition module 210 is configured to obtain the target historical data measured by the power transformer in the historical time period, and the target historical data includes the load rate historical data and the ambient temperature historical data; the first output module 220 is configured to output at least one target historical data to a pre-built top oil temperature prediction model, and the top oil temperature prediction model outputs the historical predicted top oil temperature of the power transformer corresponding to the at least one target historical data, and sorts each historical predicted top oil temperature based on the time sequence to obtain a historical predicted top oil temperature sequence; the second output module 230 is configured to obtain the load rate prediction data and the ambient temperature prediction data of the power transformer at a future moment, and output the obtained top oil temperature sequence to the pre-built top oil temperature prediction model. The load rate prediction data and the ambient temperature prediction data are input into the top oil temperature prediction model, and the top oil temperature prediction model outputs the predicted top oil temperature value at a future moment; the judgment module 240 is configured to obtain a second historical actual top oil temperature sequence within a historical period of time according to a preset extraction rule, and to judge whether the difference between the predicted top oil temperature value and each historical actual top oil temperature value in the first historical actual top oil temperature sequence is greater than a preset threshold; the correction module 250 is configured to correct the predicted top oil temperature value according to the average value of each first historical actual top oil temperature in the first historical actual top oil temperature sequence if it is greater than the preset threshold, so as to obtain the final target predicted top oil temperature value.

[0040] It should be understood that Figure 2 Modules and references documented in Figure 1 Therefore, the operations and features described above for the method and the corresponding technical effects are also applicable to Figure 2 The modules in it will not be described in detail here.

[0041] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the power transformer oil temperature prediction method based on multi-factor feature fusion in any of the above method embodiments;

[0042] As an implementation mode, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows:

[0043] Acquire target historical data measured by the power transformer in a historical time period, wherein the target historical data includes load rate historical data and ambient temperature historical data;

[0044] Outputting at least one target historical data to a pre-built top layer oil temperature prediction model, the top layer oil temperature prediction model outputting the historical predicted top layer oil temperature of the power transformer corresponding to the at least one target historical data, and sorting the historical predicted top layer oil temperatures based on the time sequence to obtain a historical predicted top layer oil temperature sequence;

[0045] Obtaining load factor prediction data and ambient temperature prediction data of the power transformer at a future time, and inputting the load factor prediction data and ambient temperature prediction data into the top layer oil temperature prediction model, wherein the top layer oil temperature prediction model outputs a predicted top layer oil temperature value at a future time;

[0046] Acquire a second historical actual top oil temperature sequence within a historical period of time according to a preset extraction rule, and determine whether a difference between the predicted top oil temperature value and each historical actual top oil temperature value in the first historical actual top oil temperature sequence is greater than a preset threshold;

[0047] If it is greater than a preset threshold, the predicted top layer oil temperature value is corrected according to the average value of each first historical actual top layer oil temperature in the first historical actual top layer oil temperature sequence to obtain a final target predicted top layer oil temperature value.

[0048] The computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created according to the use of the power transformer oil temperature prediction system based on multi-factor feature fusion, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the power transformer oil temperature prediction system based on multi-factor feature fusion via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0049] Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3 In the example, the bus connection is used. The memory 320 is the above-mentioned computer-readable storage medium. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, the power transformer oil temperature prediction method based on multi-factor feature fusion of the above-mentioned method embodiment is realized. The input device 330 can receive input digital or character information, and generate key signal input related to user settings and function control of the power transformer oil temperature prediction system based on multi-factor feature fusion. The output device 340 may include display devices such as display screens.

[0050] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0051] As an implementation mode, the electronic device is applied to a power transformer oil temperature prediction system based on multi-factor feature fusion, and is used for a client, and includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can:

[0052] Acquire target historical data measured by the power transformer in a historical time period, wherein the target historical data includes load rate historical data and ambient temperature historical data;

[0053] Outputting at least one target historical data to a pre-built top layer oil temperature prediction model, the top layer oil temperature prediction model outputting the historical predicted top layer oil temperature of the power transformer corresponding to the at least one target historical data, and sorting the historical predicted top layer oil temperatures based on the time sequence to obtain a historical predicted top layer oil temperature sequence;

[0054] Obtaining load factor prediction data and ambient temperature prediction data of the power transformer at a future time, and inputting the load factor prediction data and ambient temperature prediction data into the top layer oil temperature prediction model, wherein the top layer oil temperature prediction model outputs a predicted top layer oil temperature value at a future time;

[0055] Acquire a second historical actual top oil temperature sequence within a historical period of time according to a preset extraction rule, and determine whether a difference between the predicted top oil temperature value and each historical actual top oil temperature value in the first historical actual top oil temperature sequence is greater than a preset threshold;

[0056] If it is greater than a preset threshold, the predicted top layer oil temperature value is corrected according to the average value of each first historical actual top layer oil temperature in the first historical actual top layer oil temperature sequence to obtain a final target predicted top layer oil temperature value.

[0057] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0058] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting oil temperature of power transformer based on multi-factor feature fusion, characterized in that: include: Acquire target historical data measured by the power transformer in a historical time period, wherein the target historical data includes load rate historical data and ambient temperature historical data; Outputting at least one target historical data to a pre-built top layer oil temperature prediction model, the top layer oil temperature prediction model outputting the historical predicted top layer oil temperature of the power transformer corresponding to the at least one target historical data, and sorting the historical predicted top layer oil temperatures based on the time sequence to obtain a historical predicted top layer oil temperature sequence; Obtaining load factor prediction data and ambient temperature prediction data of the power transformer at a future time, and inputting the load factor prediction data and ambient temperature prediction data into the top layer oil temperature prediction model, wherein the top layer oil temperature prediction model outputs a predicted top layer oil temperature value at a future time; Acquire a second historical actual top oil temperature sequence within a historical period of time according to a preset extraction rule, and determine whether a difference between the predicted top oil temperature value and each historical actual top oil temperature value in the first historical actual top oil temperature sequence is greater than a preset threshold; If it is greater than a preset threshold, the predicted top layer oil temperature value is corrected according to the average value of each first historical actual top layer oil temperature in the first historical actual top layer oil temperature sequence to obtain a final target predicted top layer oil temperature value.

2. The method for predicting oil temperature of a power transformer based on multi-factor feature fusion according to claim 1 is characterized in that: The top oil temperature prediction model is obtained by training based on target historical data of the power transformer, combined with a dynamic thermal model of the power transformer and an extended Kalman filter algorithm containing an adaptive noise parameter estimator.

3. The method for predicting oil temperature of a power transformer based on multi-factor feature fusion according to claim 1 is characterized in that: The step of obtaining a second historical actual top layer oil temperature sequence within a historical period of time according to a preset extraction rule includes: Searching for target historical data corresponding to the load rate prediction data and the ambient temperature prediction data in the historical prediction top oil temperature sequence; The historical actual top oil temperature values ​​corresponding to the target historical data whose load rate difference and ambient temperature difference are not greater than the preset target threshold are extracted to obtain a second historical actual top oil temperature sequence within a historical period of time, wherein the load rate difference is the difference between the load rate historical data and the load rate prediction data, and the ambient temperature difference is the difference between the ambient temperature historical data and the ambient temperature prediction data.

4. The method for predicting oil temperature of a power transformer based on multi-factor feature fusion according to claim 1 is characterized in that: After determining whether the difference between the predicted top layer oil temperature value and each historical actual top layer oil temperature value in the first historical actual top layer oil temperature sequence is greater than a preset threshold, the method further includes: If it is not greater than the preset threshold, the predicted top layer oil temperature value is directly used as the final target predicted top layer oil temperature value.

5. The method for predicting oil temperature of a power transformer based on multi-factor feature fusion according to claim 1 is characterized in that: The step of correcting the predicted top layer oil temperature value according to the average value of each first historical actual top layer oil temperature in the first historical actual top layer oil temperature sequence to obtain the final target predicted top layer oil temperature value includes: Obtaining an average value of each first historical actual top layer oil temperature in the first historical actual top layer oil temperature sequence; Calculating a target difference between the predicted top oil temperature value and the average value; Take half of the target difference and add or subtract it from the predicted top layer oil temperature value to obtain the final target predicted top layer oil temperature value.

6. A power transformer oil temperature prediction system based on multi-factor feature fusion, characterized in that: include: An acquisition module configured to acquire target historical data measured by the power transformer in a historical time period, wherein the target historical data includes load rate historical data and ambient temperature historical data; A first output module is configured to output at least one target historical data to a pre-built top layer oil temperature prediction model, wherein the top layer oil temperature prediction model outputs a historical predicted top layer oil temperature of the power transformer corresponding to the at least one target historical data, and sorts each historical predicted top layer oil temperature based on a time sequence to obtain a historical predicted top layer oil temperature sequence; A second output module is configured to obtain load rate prediction data and ambient temperature prediction data of the power transformer at a future time, and input the load rate prediction data and ambient temperature prediction data into the top layer oil temperature prediction model, and the top layer oil temperature prediction model outputs a predicted top layer oil temperature value at a future time; a judgment module configured to obtain a second historical actual top oil temperature sequence within a historical period of time according to a preset extraction rule, and judge whether a difference between the predicted top oil temperature value and each historical actual top oil temperature value in the first historical actual top oil temperature sequence is greater than a preset threshold; The correction module is configured to correct the predicted top oil temperature value according to the average value of each first historical actual top oil temperature in the first historical actual top oil temperature sequence if it is greater than a preset threshold, so as to obtain a final target predicted top oil temperature value.

7. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.

Images