Waveform classification optimization method and system for electric power secondary circuit hidden danger early warning

Abstract

The invention discloses a waveform classification optimization method and system for hidden danger early warning of a power secondary circuit. The method comprises the following steps: obtaining three-phase current recording file window data; classifying and labeling the three-phase current recording file window data based on a preset numerical value judgment principle to obtain labeled target three-phase current recording file window data; converting window data of the target three-phase current recording file into two-dimensional images through a GASF (Geometric Amplitude Shift Function) and a GASF (Geometric Amplitude Shift Function), and combining the two-dimensional images into at least one image unit; inputting the at least one image unit into a preset VGG16 convolutional neural network for iterative training to obtain a current secondary circuit neutral line fault diagnosis model; and inputting the real-time fault recording time sequence data into the current secondary circuit neutral line fault diagnosis model, and outputting to obtain a waveform classification result at a three-phase current circuit neutral line. The problem that efficient and accurate online detection and judgment are difficult to realize at present when the neutral line of the current secondary circuit is virtually connected or broken is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of sampling circuit anomaly diagnosis, and in particular relates to a waveform classification optimization method and system for early warning of hidden dangers in power secondary circuits. Background Art

[0002] Relay protection devices should sensitively and reliably respond to faults and abnormal operating conditions of various electrical equipment in the power system, and selectively trip related circuit breakers quickly, thus playing a significant role in safeguarding the safety of the power grid. When the sampling circuit of the protection device fails, it is often easy to cause power grid or equipment accidents. Existing secondary circuit anomaly diagnosis methods and systems can currently realize and warn of most types of secondary circuit operation hazards, but they rely heavily on relatively complete fault recording data sets and still have low recognition rates for some circuit hazards, such as accurately determining whether the neutral line of the current circuit is loosely connected or broken.

[0003] It is well known that the health status of the neutral line of the current secondary circuit cannot be reflected under the condition of three-phase balance. Even in the case of slight external interference from the large power grid, the unbalanced characteristics of the three-phase current are difficult to effectively distinguish from the normal power grid disturbance state. Considering the instability and time-varying nature of the neutral line transition resistance, as well as the imbalance caused by different secondary circuit loads, it is usually difficult to establish an accurate and reasonable simulation model for the current circuit neutral line virtual connection or disconnection scenario and summarize and refine the mechanism formula covering all scenarios. The current commonly used practice of refining formulas based on experience has an accurate warning rate of less than 50% in the actual production and operation of a power grid company in a certain province. The virtual connection or disconnection of the neutral line of the current circuit occurs frequently in actual production and can easily directly cause the relevant protection to malfunction, bringing immeasurable risks to the high-quality operation of the new power system. It is urgent to carry out in-depth research and solutions to the hidden dangers of this type of secondary circuit. Summary of the Invention

[0004] The present invention provides a waveform classification optimization method and system for warning of hidden dangers in power secondary circuits, which is used to solve the technical problem that it is currently difficult to achieve efficient and accurate online detection and judgment of virtual connection or disconnection of the neutral line in the current secondary circuit, thereby resulting in a decrease in the accuracy of abnormal diagnosis of the current sampling circuit and the existence of diagnostic blind spots.

[0005] In a first aspect, the present invention provides a waveform classification optimization method for electric secondary circuit hidden danger warning, comprising: Constructing a large power grid simulation model based on RTDS, wherein the large power grid simulation model includes a single-phase high-resistance grounding fault simulation model and a neutral line fault simulation model; Performing ergodic simulation on the single-phase high-resistance grounding fault simulation model and the neutral line fault simulation model to obtain three-phase current recording file window data during the period of superimposed neutral line fault under the condition of single-phase fault of the transmission line; Classify and label the three-phase current recording file window data based on a preset numerical judgment principle to obtain labeled target three-phase current recording file window data; Converting the target three-phase current recording file window data into a two-dimensional image through the Gram angle and field GASF, and combining each two-dimensional image into at least one image unit; Inputting the at least one image unit into a preset VGG16 convolutional neural network for iterative training to obtain a current secondary circuit neutral line fault diagnosis model; Based on the fault recording master station platform, real-time fault recording time series data during a single-phase fault in the station-end transmission line is obtained, and the real-time fault recording time series data is input into the current secondary circuit neutral line fault diagnosis model. The current secondary circuit neutral line fault diagnosis model outputs the waveform classification result at the neutral line of the three-phase current circuit.

[0006] In a second aspect, the present invention provides a waveform classification optimization system for electric secondary circuit hidden danger warning, comprising: A construction module is configured to construct a large power grid simulation model based on RTDS, wherein the large power grid simulation model includes a single-phase high-resistance grounding fault simulation model and a neutral line fault simulation model; A simulation module configured to perform ergodic simulation on the single-phase high-resistance grounding fault simulation model and the neutral line fault simulation model, and obtain three-phase current recording file window data under the condition of a single-phase fault of the transmission line and a period of superimposed neutral line fault; a marking module configured to classify and mark the three-phase current recording file window data based on a preset numerical judgment principle to obtain marked target three-phase current recording file window data; a conversion module configured to convert the target three-phase current recording file window data into a two-dimensional image through a Gram angle and field GASF, and combine each two-dimensional image into at least one image unit; A training module is configured to input the at least one image unit into a preset VGG16 convolutional neural network for iterative training to obtain a current secondary circuit neutral line fault diagnosis model; The output module is configured to obtain real-time fault recording time series data during a single-phase fault in a station-end transmission line based on a fault recording master station platform, and input the real-time fault recording time series data into the current secondary circuit neutral line fault diagnosis model. The current secondary circuit neutral line fault diagnosis model outputs a waveform classification result at the neutral line of the three-phase current circuit.

[0007] 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 execute the steps of the waveform classification optimization method for power secondary circuit hidden danger warning according to any embodiment of the present invention.

[0008] In a fourth aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon. When the program instructions are executed by a processor, the processor executes the steps of the waveform classification optimization method for power secondary circuit hidden danger warning of any embodiment of the present invention.

[0009] The waveform classification optimization method and system for electric power secondary circuit hidden danger warning of the present application are specifically designed to identify the situations of virtual connection and disconnection of the neutral line of the current circuit. It is mainly used in the electric power secondary circuit hidden danger warning device when facing the situation of primary equipment failure in the power system itself. At the same time, it superimposes the rapid classification of the extremely special abnormal situation of virtual connection or disconnection of the neutral line of the current secondary circuit, filling the gap in the rapid adaptive identification of abnormal situations of large-scale electric power secondary circuit data collection, breaking through the limitations of many traditional monitoring methods and conditions such as traditional reliance on mathematical mechanism derivation, difficulty in comprehensive testing and verification of new detection principles, and homologous comparison, and making enhanced supplements to the adaptive diagnosis of secondary circuit abnormalities, so that the abnormality identification method can better meet on-site needs and have universality. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following is a brief introduction to 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 any creative work.

[0011] Figure 1 A flowchart of a waveform classification optimization method for electric secondary circuit hidden danger warning provided by one embodiment of the present invention; Figure 2 A structural block diagram of a waveform classification optimization system for electric power secondary circuit hidden danger warning provided by one embodiment of the present invention; Figure 3 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0012] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0013] See also Figure 1 , which shows a flow chart of a waveform classification optimization method for power secondary circuit hidden danger warning according to the present application. The waveform classification optimization method for power secondary circuit hidden danger warning according to the present invention specifically monitors and provides early warning for weak and broken neutral lines in three-phase current circuits. It can be used in power generation plants (hydropower, thermal power, and renewable energy) and substations deployed in directly grounded systems at voltage levels of 110kV and above, containing various types and models of relay protection devices.

[0014] like Figure 1 As shown, the waveform classification optimization method for electric power secondary circuit hidden danger warning specifically includes the following steps: Step S101 : constructing a large power grid simulation model based on RTDS, wherein the large power grid simulation model includes a single-phase high-resistance grounding fault simulation model and a neutral line fault simulation model.

[0015] In this step, a large power grid simulation model is built through RTDS, and a direct grounding method is adopted to simulate the simulation of single-phase metallic grounding fault and single-phase high-resistance grounding fault of the transmission line grid fault. Among them, the single-phase high-resistance grounding fault simulation model sets the fault grounding resistance growth step to a simulation step of 0.1 ohm, and the grounding transition resistance range is from 0.1 to 300 ohms; a neutral line fault simulation model is built that is transmitted to the current loop of the relay protection device through a high-voltage mutual inductor to simulate the neutral line fault of the three-phase current loop. Among them, the neutral line fault simulation model sets the neutral line transition resistance growth step to a simulation step of 0.1 ohm, and the resistance range is from 0.1 to 1000 ohms.

[0016] It should be noted that during the RTDS fault simulation process, noise conditions with ≤3 consecutive loss points and a total loss rate ≤10% were proactively added to ensure that normal waveform data accounted for 90%, waveform data with noise conditions with consecutive loss points accounted for 5%, and waveform data with noise conditions with a total loss rate accounted for 5%. The specific distribution of consecutive loss point samples should ensure that they occur in the first 30%, middle 40%, and last 30% of the fault waveform data, with the same distribution ratio. Samples with a total loss rate ≤10% were set according to a random occurrence time period.

[0017] Step S102 , performing ergodic simulation on the single-phase high-resistance grounding fault simulation model and the neutral line fault simulation model, and obtaining three-phase current waveform file window data of the period when the transmission line is under the condition of a single-phase fault and a neutral line fault.

[0018] In this step, ergodic simulation is performed on the single-phase high-resistance grounding fault simulation model and the neutral line fault simulation model to simulate the three-phase current recording file window data (including 5 ms before the fault start time to 5 ms after the fault end time) during the period of superimposed neutral line fault under the condition of single-phase fault of the transmission line.

[0019] Step S103 , classifying and labeling the three-phase current recording file window data based on a preset numerical value determination principle to obtain labeled target three-phase current recording file window data.

[0020] In this step, the RTDS simulation result data is extracted and converted into a Comrade format waveform file. At the same time, the neutral line fault type is marked for all the above waveform files (divided into: the neutral line transition resistance range of 0.1 ohm to 0.5 ohm is normal, the resistance range of 0.6 ohm to 10 ohm is high resistance, the resistance range of 11 ohm to 100 ohm is too high resistance, and the resistance range of 101 ohm to 1000 ohm is completely broken).

[0021] For the current loops of line protection, busbar protection branches, and circuit breaker protection devices, the three-phase current recording file window data is classified and labeled according to the first numerical judgment sub-principle to obtain the labeled target three-phase current recording file window data.

[0022] Specifically, determine whether the data in a certain three-phase current recording file window meets the first preset condition, wherein the first preset condition is: under the condition of a single-phase short circuit fault, there is a non-fault phase current that is less than 10% of the fault phase current and in the opposite direction, and the corresponding non-fault phase second harmonic content is less than 10%, and the calculation formula for the second harmonic content of the non-fault phase is: the effective value of the second harmonic of the non-fault phase current divided by the effective value of the fundamental wave of the non-fault phase current; if the first preset condition is met, the data in the window of the three-phase current recording file is marked as a normal situation type; determine whether the data in a certain three-phase current recording file window meets the second preset condition, wherein the second preset condition is: under the condition of a single-phase short circuit fault, there is a non-fault phase current that is greater than 10% of the fault phase current and in the opposite direction, and the corresponding non-fault phase second harmonic content is greater than 15%, and the duration is less than 20ms; if the second preset condition is met, the data in the window of the three-phase current recording file is marked as a normal situation type; determine whether the data in the window of the three-phase current recording file meets the second preset condition, wherein the second preset condition is: under the condition of a single-phase short circuit fault, there is a non-fault phase current that is greater than 10% of the fault phase current and in the opposite direction, and the corresponding non-fault phase second harmonic content is greater than 15%, and the duration is less than 20ms; if the second preset condition is met, the three-phase current The data in the recording file window is marked as a large resistance type; whether the data in a certain three-phase current recording file window meets the third preset condition, wherein the third preset condition is: under the condition of a single-phase short circuit fault, there is a non-fault phase current that is greater than 15% of the fault phase current and in the opposite direction, and the corresponding non-fault phase second harmonic content is greater than 15%, and the duration is greater than 20ms; if the third preset condition is met, the data in the three-phase current recording file window is marked as an excessive resistance type; whether the data in a certain three-phase current recording file window meets the fourth preset condition, wherein the fourth preset condition is: under the condition of a single-phase short circuit fault, there is a non-fault phase current that is greater than 15% of the fault phase current and in the opposite direction, and the corresponding non-fault phase second harmonic content is greater than 30%, and the duration is greater than 20ms; if the fourth preset condition is met, the data in the three-phase current recording file window is marked as a complete disconnection type.

[0023] For main transformer protection, when a single-phase grounding fault occurs in the high-voltage side power grid of the main transformer, and a virtual connection or disconnection of the neutral line occurs in the medium-voltage side current secondary circuit of the main transformer protection device, the three-phase current recording file window data is classified and labeled according to the second numerical judgment sub-principle to obtain the labeled target three-phase current recording file window data.

[0024] Specifically, if the main transformer adopts YNynd11, the following judgment process is performed: when in the secondary circuit recording of the medium voltage side current of the main transformer in a certain three-phase current recording file window data, there is a non-fault phase current smaller than the fault phase current And the direction is opposite, and the third harmonic content of the corresponding non-fault phase is less than 10%, then the data of a three-phase current recording file window is marked as normal type; when in the secondary circuit recording of the medium voltage side current of the main transformer of a three-phase current recording file window data, the non-fault phase current is greater than the fault phase current And the direction is opposite, the third harmonic content of the corresponding non-fault phase is greater than 15%, and the duration is less than 20ms, then the data of a three-phase current recording file window is marked as a large resistance type; when the secondary circuit recording of the medium voltage side current of the main transformer in a three-phase current recording file window data has a non-fault phase current greater than the fault phase current And the direction is opposite, the third harmonic content of the corresponding non-fault phase is greater than 15%, and the duration is greater than 20ms, then the data of a three-phase current recording file window is marked as the type of excessive resistance; when the secondary circuit recording of the medium-voltage side current of the main transformer in a three-phase current recording file window data has a non-fault phase current greater than the fault phase current And the direction is opposite, the third harmonic content of the corresponding non-fault phase is greater than 30%, and the duration is greater than 20ms, then the data in a three-phase current recording file window is marked as a complete disconnection type.

[0025] If the main transformer adopts YNyd11, the following judgment process is performed: when the non-fault phase current is less than 30% of the fault phase current and the phase leads 160° in the secondary circuit recording of the medium-voltage side current of a certain three-phase current recording file window data, and the corresponding non-fault phase fifth harmonic content is less than 10%, then the three-phase current recording file window data is marked as normal type; when the non-fault phase current is greater than 30% of the fault phase current and the phase leads 160° in the secondary circuit recording of the medium-voltage side current of a certain three-phase current recording file window data, and the corresponding non-fault phase fifth harmonic content is greater than 15%, and the duration is less than 20ms, then the three-phase current recording file window data is marked as resistance Large type; when in the secondary circuit recording of the medium-voltage side current of the main transformer in a certain three-phase current recording file window data, the non-fault phase current is greater than 30% of the fault phase current and the phase leads 160°, the corresponding non-fault phase fifth harmonic content is greater than 15%, and the duration is greater than 20ms, then the three-phase current recording file window data is marked as the excessive resistance type; when in the secondary circuit recording of the medium-voltage side current of the main transformer in a certain three-phase current recording file window data, the non-fault phase current is greater than 30% of the fault phase current and the phase leads 160°, the corresponding non-fault phase fifth harmonic content is greater than 30%, and the duration is greater than 20ms, then the three-phase current recording file window data is marked as the complete disconnection type.

[0026] Step S104 : converting the target three-phase current recording file window data into a two-dimensional image through the Gram angle and field GASF, and combining each two-dimensional image into at least one image unit.

[0027] In this step, the target three-phase current recording file window data is normalized according to a preset data processing rule to obtain the normalized target three-phase current recording file window data, wherein the data processing rule is: (window current sampling point - window average sampling point) / window sampling point standard deviation or (window current sampling point - window minimum sampling point) / (window maximum sampling point - window minimum sampling point); the normalized target three-phase current recording file window data is converted into a two-dimensional image through the Gram angle and field GASF, and each two-dimensional image is combined into at least one image unit.

[0028] It should be noted that after obtaining all the image units of normal / abnormal neutral lines (large resistance, excessive resistance, and complete disconnection) of the current loop to be trained, verified, and tested, the setting of the field GASF built-in parameter image_size should be dynamically adjusted according to the fault recording sampling rate to maintain consistent time resolution. Specifically: The actual fault recording data of large power grids comes from different fault recording devices and protection devices. The sampling rates of the device recordings may vary. Data with low sampling rates are smoother and more blurred, especially in the processing of high-frequency components, resulting in different image granularity. The model may regard the image texture changes caused by the sampling rate difference as classification characteristics rather than real fault modes, ultimately reducing model performance.

[0029] image_size=(fault recording sampling rate Time window) / base sampling rate, where the fault recording sampling rate is generally divided into three types: 1200 Hz, 4000 Hz, and 10000 Hz. The time window is automatically calculated based on the specific duration of each power grid fault, generally ranging from 20 to 30 (+10ms), and the base sampling rate is 4000 Hz.

[0030] Step S105: inputting the at least one image unit into a preset VGG16 convolutional neural network for iterative training to obtain a current secondary circuit neutral line fault diagnosis model.

[0031] In this step, the pre-trained VGG16 convolutional neural network is used for fault mode training, and data-enhanced feature extraction is used (horizontal flipping is applied to 50% of randomly selected images; the images are enlarged or reduced, and the scaling ratio is randomly selected in the range of [−10% to +10%]). Finally, a current secondary circuit neutral line fault diagnosis model is obtained through training.

[0032] Step S106: Based on the fault recording master station platform, real-time fault recording time series data during the period when a single-phase fault occurs in the station-end transmission line is obtained, and the real-time fault recording time series data is input into the current secondary circuit neutral line fault diagnosis model. The current secondary circuit neutral line fault diagnosis model outputs the waveform classification result at the neutral line of the three-phase current circuit.

[0033] In summary, the method of the present application requires that the electrical and physical distance between the occurrence points of single-phase grounding faults in the power grid and the occurrence points of virtual connection and disconnection of the neutral line of the three-phase current loop shall not exceed two substations of the same voltage level. It can specifically identify the situations of virtual connection and disconnection of the neutral line of the current loop. It is mainly used for the power secondary circuit hidden danger early warning device when facing the situation of primary equipment failure in the power system itself. At the same time, it superimposes the rapid classification of the extremely special abnormal situation of virtual connection or disconnection of the neutral line of the current secondary circuit, filling the gap in the rapid adaptive identification of abnormal situations of large-scale power secondary circuit data collection, breaking through the limitations of many traditional monitoring methods such as traditional reliance on mathematical mechanism derivation, difficulty in comprehensive testing and verification of new detection principles, and homologous comparison, and making enhanced supplements to the adaptive diagnosis of secondary circuit abnormalities, so that the abnormality identification method can better meet on-site needs and be universal.

[0034] See also Figure 2 , which shows a structural block diagram of a waveform classification optimization system for electric power secondary circuit hidden danger warning of the present application.

[0035] like Figure 2 As shown, the waveform classification optimization system 200 includes a construction module 210 , a simulation module 220 , a labeling module 230 , a conversion module 240 , a training module 250 and an output module 260 .

[0036] Among them, the construction module 210 is configured to construct a large power grid simulation model based on RTDS, wherein the large power grid simulation model includes a single-phase high-resistance grounding fault simulation model and a neutral line fault simulation model; the simulation module 220 is configured to perform ergodic simulation on the single-phase high-resistance grounding fault simulation model and the neutral line fault simulation model, and simulate to obtain the three-phase current recording file window data under the condition of a single-phase fault of the transmission line and a superimposed neutral line fault period; the marking module 230 is configured to classify and mark the three-phase current recording file window data based on a preset numerical judgment principle, and obtain the marked target three-phase current recording file window data; the conversion module 240 is configured to convert the target three-phase current The recording file window data is converted into a two-dimensional image through the Gram angle and field GASF, and each two-dimensional image is combined into at least one image unit; the training module 250 is configured to input the at least one image unit into a preset VGG16 convolutional neural network for iterative training to obtain a current secondary circuit neutral line fault diagnosis model; the output module 260 is configured to obtain real-time fault recording time series data during a single-phase fault on the station-end transmission line based on the fault recording master station platform, and input the real-time fault recording time series data into the current secondary circuit neutral line fault diagnosis model, and the current secondary circuit neutral line fault diagnosis model outputs a waveform classification result at the neutral line of the three-phase current circuit.

[0037] 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 also apply to Figure 2 The modules in it will not be described in detail here.

[0038] In other embodiments, embodiments of the present invention further provide a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor is caused to execute the waveform classification optimization method for electric power secondary circuit hidden danger warning in any of the above method embodiments; As an embodiment, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows: Constructing a large power grid simulation model based on RTDS, wherein the large power grid simulation model includes a single-phase high-resistance grounding fault simulation model and a neutral line fault simulation model; Performing ergodic simulation on the single-phase high-resistance grounding fault simulation model and the neutral line fault simulation model to obtain three-phase current recording file window data during the period of superimposed neutral line fault under the condition of single-phase fault of the transmission line; Classify and label the three-phase current recording file window data based on a preset numerical judgment principle to obtain labeled target three-phase current recording file window data; Converting the target three-phase current recording file window data into a two-dimensional image through the Gram angle and field GASF, and combining each two-dimensional image into at least one image unit; Inputting the at least one image unit into a preset VGG16 convolutional neural network for iterative training to obtain a current secondary circuit neutral line fault diagnosis model; Based on the fault recording master station platform, real-time fault recording time series data during a single-phase fault in the station-end transmission line is obtained, and the real-time fault recording time series data is input into the current secondary circuit neutral line fault diagnosis model. The current secondary circuit neutral line fault diagnosis model outputs the waveform classification result at the neutral line of the three-phase current circuit.

[0039] 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 based on the use of the waveform classification optimization system for power secondary circuit hidden danger warning, 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 memory device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely located relative to the processor, and these remote memories may be connected to the waveform classification optimization system for power secondary circuit hidden danger warning 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.

[0040] 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 3The example of the bus connection is taken as an example. The memory 320 is the computer-readable storage medium mentioned above. 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, implements the waveform classification optimization method for electric power secondary circuit hidden danger warning of the above method embodiment. 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 waveform classification optimization system for electric power secondary circuit hidden danger warning. The output device 340 may include a display device such as a display screen.

[0041] 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 fully described in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0042] As an embodiment, the electronic device is applied to a waveform classification optimization system for power secondary circuit hidden danger warning, and is used for a client, and includes: 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: Constructing a large power grid simulation model based on RTDS, wherein the large power grid simulation model includes a single-phase high-resistance grounding fault simulation model and a neutral line fault simulation model; Performing ergodic simulation on the single-phase high-resistance grounding fault simulation model and the neutral line fault simulation model to obtain three-phase current recording file window data during the period of superimposed neutral line fault under the condition of single-phase fault of the transmission line; Classify and label the three-phase current recording file window data based on a preset numerical judgment principle to obtain labeled target three-phase current recording file window data; Converting the target three-phase current recording file window data into a two-dimensional image through the Gram angle and field GASF, and combining each two-dimensional image into at least one image unit; Inputting the at least one image unit into a preset VGG16 convolutional neural network for iterative training to obtain a current secondary circuit neutral line fault diagnosis model; Based on the fault recording master station platform, real-time fault recording time series data during a single-phase fault in the station-end transmission line is obtained, and the real-time fault recording time series data is input into the current secondary circuit neutral line fault diagnosis model. The current secondary circuit neutral line fault diagnosis model outputs the waveform classification result at the neutral line of the three-phase current circuit.

[0043] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods of each embodiment or certain portions of the embodiments.

[0044] 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 various embodiments of the present invention.

Claims

1. A waveform classification optimization method for electric secondary circuit hidden danger early warning, characterized in that: include: Constructing a large power grid simulation model based on RTDS, wherein the large power grid simulation model includes a single-phase high-resistance grounding fault simulation model and a neutral line fault simulation model; Performing ergodic simulation on the single-phase high-resistance grounding fault simulation model and the neutral line fault simulation model to obtain three-phase current recording file window data during the period of superimposed neutral line fault under the condition of single-phase fault of the transmission line; Classify and label the three-phase current recording file window data based on a preset numerical judgment principle to obtain labeled target three-phase current recording file window data; Converting the target three-phase current recording file window data into a two-dimensional image through the Gram angle and field GASF, and combining each two-dimensional image into at least one image unit; Inputting the at least one image unit into a preset VGG16 convolutional neural network for iterative training to obtain a current secondary circuit neutral line fault diagnosis model; Based on the fault recording master station platform, real-time fault recording time series data during a single-phase fault in the station-end transmission line is obtained, and the real-time fault recording time series data is input into the current secondary circuit neutral line fault diagnosis model. The current secondary circuit neutral line fault diagnosis model outputs the waveform classification result at the neutral line of the three-phase current circuit.

2. The waveform classification optimization method for electric power secondary circuit hidden danger early warning according to claim 1 is characterized in that: The step of converting the target three-phase current recording file window data into a two-dimensional image through the Gram angle and field GASF, and combining each two-dimensional image into at least one image unit comprises: Normalizing the target three-phase current recording file window data according to a preset data processing rule to obtain normalized target three-phase current recording file window data, wherein the data processing rule is: (window current sampling point - window average sampling point) / window sampling point standard deviation or (window current sampling point - window minimum sampling point) / (window maximum sampling point - window minimum sampling point); The normalized target three-phase current recording file window data is converted into a two-dimensional image through the Gram angle and field GASF, and each two-dimensional image is combined into at least one image unit.

3. The waveform classification optimization method for electric power secondary circuit hidden danger early warning according to claim 1 is characterized in that: The three-phase current recording file window data is classified and labeled based on the preset numerical judgment principle, and the labeled target three-phase current recording file window data includes: For the current loops of the line protection, busbar protection branches, and circuit breaker protection devices, classify and label the three-phase current recording file window data according to the first numerical determination sub-principle to obtain labeled target three-phase current recording file window data; For main transformer protection, when a single-phase grounding fault occurs in the high-voltage side power grid of the main transformer, and a virtual connection or disconnection of the neutral line occurs in the medium-voltage side current secondary circuit of the main transformer protection device, the three-phase current recording file window data is classified and labeled according to the second numerical judgment sub-principle to obtain the labeled target three-phase current recording file window data.

4. The waveform classification optimization method for electric power secondary circuit hidden danger early warning according to claim 3 is characterized in that: The three-phase current recording file window data is classified and labeled according to the first numerical determination sub-principle, and the labeled target three-phase current recording file window data is obtained, including: Determine whether the data in a three-phase current recording file window meets a first preset condition, wherein the first preset condition is: under a single-phase short circuit fault condition, there is a non-fault phase current that is 10% less than the fault phase current and in the opposite direction, and the second harmonic content of the corresponding non-fault phase is less than 10%; If the first preset condition is met, the data in the three-phase current recording file window is marked as a normal situation type; Determine whether the data in a three-phase current recording file window meets a second preset condition, wherein the second preset condition is: under the condition of a single-phase short circuit fault, there is a non-fault phase current that is 10% greater than the fault phase current and in the opposite direction, and the corresponding non-fault phase second harmonic content is greater than 15% and the duration is less than 20ms; If the second preset condition is met, the data in the three-phase current recording file window is marked as a relatively large resistance type, wherein the resistance range of the relatively large resistance type is 0.6 ohms to 10 ohms; Determine whether the data in a three-phase current recording file window meets a third preset condition, wherein the third preset condition is: under a single-phase short circuit fault condition, there is a non-fault phase current that is 15% greater than the fault phase current and in the opposite direction, and the corresponding non-fault phase second harmonic content is greater than 15% and lasts for more than 20ms; If the third preset condition is met, the data in the three-phase current recording file window is marked as an excessive resistance type; Determine whether the data in a three-phase current recording file window meets a fourth preset condition, wherein the fourth preset condition is: under a single-phase short circuit fault condition, there is a non-fault phase current that is 15% greater than the fault phase current and in the opposite direction, and the corresponding non-fault phase second harmonic content is greater than 30% and lasts for more than 20ms; If the fourth preset condition is met, the data in the three-phase current recording file window is marked as a complete disconnection type.

5. The waveform classification optimization method for electric secondary circuit hidden danger warning according to claim 3 is characterized in that: The three-phase current recording file window data is classified and labeled according to the second numerical determination sub-principle, and the labeled target three-phase current recording file window data is obtained, including: If the main transformer adopts YNynd11, the following judgment process is performed: When the current of the secondary circuit of the medium voltage side of the main transformer in a certain three-phase current recording file window data is smaller than the current of the fault phase, And the direction is opposite, when the third harmonic content of the corresponding non-fault phase is less than 10%, the data in a three-phase current recording file window is marked as normal type; When the current of the secondary circuit of the medium voltage side of the main transformer in a certain three-phase current recording file window data is greater than the current of the fault phase, And the direction is opposite, the third harmonic content of the corresponding non-fault phase is greater than 15%, and the duration is less than 20ms, then the data in a three-phase current recording file window is marked as a large resistance type, where the resistance range of the large resistance type is 0.6 ohms-10 ohms; When the current of the secondary circuit of the medium voltage side of the main transformer in a certain three-phase current recording file window data is greater than the current of the fault phase, And the direction is opposite, the third harmonic content of the corresponding non-fault phase is greater than 15%, and the duration is greater than 20ms, then the data in a three-phase current recording file window is marked as excessive resistance type; When the current of the secondary circuit of the medium voltage side of the main transformer in a certain three-phase current recording file window data is greater than the current of the fault phase, And the direction is opposite, the third harmonic content of the corresponding non-fault phase is greater than 30%, and the duration is greater than 20ms, then the data in a three-phase current recording file window is marked as a complete disconnection type.

6. The waveform classification optimization method for electric power secondary circuit hidden danger early warning according to claim 3 is characterized in that: The three-phase current recording file window data is classified and labeled according to the second numerical determination sub-principle, and the labeled target three-phase current recording file window data is obtained, including: If the main transformer adopts YNyd11, the following judgment process is performed: When the secondary circuit current recording of the medium voltage side of the main transformer in a certain three-phase current recording file window shows that the current of the non-fault phase is less than 30% of the fault phase current and the phase leads by 160 degrees, and the fifth harmonic content of the corresponding non-fault phase is less than 10%, the data in the three-phase current recording file window is marked as normal; If, in the secondary circuit recording of the medium-voltage side current of a main transformer in a certain three-phase current recording file window data, the non-fault phase current is 30% greater than the fault phase current and the phase leads by 160°, and the corresponding fifth harmonic content of the non-fault phase is greater than 15% and the duration is less than 20ms, then the data in the three-phase current recording file window data is marked as a large resistance type, where the resistance range of the large resistance type is 0.6 ohms to 10 ohms; If the secondary circuit current recording on the medium voltage side of the main transformer in a certain three-phase current recording file window shows that the non-fault phase current is 30% greater than the fault phase current and the phase leads by 160°, and the corresponding fifth harmonic content of the non-fault phase is greater than 15% and the duration is greater than 20ms, then the data in the three-phase current recording file window will be marked as the excessive resistance type; When the secondary circuit current recording on the medium-voltage side of the main transformer in a certain three-phase current recording file window data shows that the non-fault phase current is 30% greater than the fault phase current and the phase leads by 160°, and the corresponding fifth harmonic content of the non-fault phase is greater than 30% and the duration is greater than 20ms, then the three-phase current recording file window data is marked as a complete line break type.

7. A waveform classification optimization system for electric secondary circuit hidden danger warning, characterized in that: include: A construction module is configured to construct a large power grid simulation model based on RTDS, wherein the large power grid simulation model includes a single-phase high-resistance grounding fault simulation model and a neutral line fault simulation model; A simulation module configured to perform ergodic simulation on the single-phase high-resistance grounding fault simulation model and the neutral line fault simulation model, and obtain three-phase current recording file window data under the condition of a single-phase fault of the transmission line and a period of superimposed neutral line fault; a marking module configured to classify and mark the three-phase current recording file window data based on a preset numerical judgment principle to obtain marked target three-phase current recording file window data; a conversion module configured to convert the target three-phase current recording file window data into a two-dimensional image through a Gram angle and field GASF, and combine each two-dimensional image into at least one image unit; A training module is configured to input the at least one image unit into a preset VGG16 convolutional neural network for iterative training to obtain a current secondary circuit neutral line fault diagnosis model; The output module is configured to obtain real-time fault recording time series data during a single-phase fault in a station-end transmission line based on a fault recording master station platform, and input the real-time fault recording time series data into the current secondary circuit neutral line fault diagnosis model. The current secondary circuit neutral line fault diagnosis model outputs a waveform classification result at the neutral line of the three-phase current circuit.

8. 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 perform the method according to any one of claims 1 to 6.

9. 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 6 is implemented.

Images