Transformer monitoring method and system based on analysis of gas dissolved in oil, terminal equipment and storage medium

Abstract

The invention discloses a transformer monitoring method and system based on analysis of gas dissolved in oil, terminal equipment and a storage medium. According to the method, preliminary fault type identification is carried out through a trained decision tree model according to the gas concentration of each dissolved gas, the characteristic ratio of a plurality of gas concentrations and the gas concentration change rate of each dissolved gas obtained through calculation of chromatographic data of each dissolved gas, so that a preliminary fault prediction identification result of the transformer is obtained; and carrying out secondary fault identification on the transformer through the trained support vector machine classifier to obtain a final fault identification result of the transformer to be detected. According to the invention, preliminary fault identification and secondary fault identification are carried out through the decision tree model and the support vector classifier, and refined fault identification is carried out on the transformer, so that the accuracy of fault identification is improved; the problem that the accuracy of fault identification is low due to the fact that the fault type cannot be finely identified by single fault identification operation in the existing transformer monitoring technology is solved.

Description

Technical Field

[0001] The present invention relates to the field of transformer monitoring, and in particular to a transformer monitoring method, system, terminal device and storage medium based on dissolved gas analysis in oil. Background Art

[0002] As a crucial device in the power system, the failure of a transformer will not only affect the stability of power supply, but also pose a threat to the safety of the power system. Therefore, it is particularly important to identify transformer faults and control the operating state.

[0003] In terms of fault identification, with the development of technology, the existing commonly used fault identification technology is to combine oil chromatography monitoring with a machine learning algorithm model to identify the primary fault type of a transformer. However, there are different types of transformer faults, such as discharge faults, overheating faults, and component aging faults, etc., and each fault type is further divided into several sub-fault types. It is easy to result in low accuracy of fault identification because the fault type cannot be accurately identified by a single fault identification operation. Therefore, the existing transformer monitoring technology has the problem of low accuracy of fault identification because the fault type cannot be accurately identified by a single fault identification operation. Summary of the Invention

[0004] The present invention provides a transformer monitoring method, system, terminal device and storage medium based on dissolved gas analysis in oil, which can solve the problem of low accuracy of fault identification in the existing transformer monitoring technology due to the inability to accurately identify the fault type by a single fault identification operation.

[0005] To solve the above technical problems, an embodiment of the present invention provides a transformer monitoring method based on dissolved gas analysis in oil, including:

[0006] Obtain the chromatographic data and theoretical concentration data of various dissolved gases in the transformer oil to be measured; wherein, the theoretical concentration data includes the normal concentration and warning concentration of the gas;

[0007] Determine the peak value and peak area of the chromatographic data of various dissolved gases according to the chromatographic data of various dissolved gases;

[0008] Calculate the gas concentration of various dissolved gases according to the peak value and peak area of the chromatographic data of various dissolved gases;

[0009] Calculate several gas concentration characteristic ratios according to the gas concentration of various dissolved gases;

[0010] Perform preliminary fault prediction and identification on the transformer to be measured through the trained decision tree model according to the gas concentration of various dissolved gases and several gas concentration characteristic ratios, and obtain the preliminary fault prediction and identification result of the transformer to be measured;

[0011] Based on the preliminary fault prediction and identification results of the transformer to be measured, the gas concentrations of various dissolved gases, and several gas concentration characteristic ratios, the trained support vector machine classifier performs secondary fault prediction and identification to obtain the final fault prediction and identification results of the transformer to be measured.

[0012] Furthermore, the calculation formula for the gas concentration of various dissolved gases is:

[0013] x i = a 1 H i + a 2 S i ;

[0014] Wherein, x i is the gas concentration of the i-th dissolved gas; H i is the peak value of the chromatographic data of the i-th dissolved gas; S i is the peak area of the chromatographic data of the i-th dissolved gas; i is the type number of the dissolved gas; a 1 and a 2 are the preset peak weight and peak area weight respectively.

[0015] Furthermore, after obtaining the final fault prediction and identification results of the transformer to be measured, it further includes:

[0016] When the final fault prediction and identification result is that the transformer to be measured has a fault type, calculate the fault severity index according to the gas concentrations of various dissolved gases and the theoretical concentration data;

[0017] Compare the fault severity index with the preset threshold;

[0018] In the case where the fault severity index is greater than the preset threshold, obtain the power outage loss cost of the area controlled by the transformer to be measured; determine whether the power outage loss cost is greater than the preset loss threshold, if so, start the transformer oil emergency replacement system to replace the oil body of the transformer to be measured; if not, control the transformer to be measured to stop;

[0019] In the case where the fault severity index is not greater than the preset threshold, keep the current operating state of the transformer to be measured.

[0020] Furthermore, the calculation formula for the fault severity index is:

[0021]

[0022] Wherein, FSI is the fault severity index; x i is the gas concentration of the i-th dissolved gas; i is the type number of the dissolved gas; n is the total number of types of dissolved gases; wi is the fault weight coefficient of the i-th gas; is the normal concentration of the i-th dissolved gas; is the warning concentration of the i-th dissolved gas.

[0023] Furthermore, the model training of the decision tree model includes:

[0024] Obtain the historical gas concentrations of various dissolved gases in transformer oil with true fault type labels, and calculate several historical gas concentration characteristic ratios according to the historical gas concentrations of various dissolved gases;

[0025] Divide the historical gas concentrations of various dissolved gases and several historical gas concentration characteristic ratios into a training set, a validation set, and a test set according to a certain proportion;

[0026] According to the preset decision tree splitting criterion, repeatedly perform subset partitioning on the training set until the preset stop condition is met to obtain the decision tree model;

[0027] Input the validation set into the decision tree model for evaluation to obtain the preliminary fault prediction and identification results, and calculate the evaluation index according to the preliminary fault prediction and identification results and the true fault type labels. Adjust the parameters of the decision tree model according to the evaluation index until the evaluation index reaches the preset standard to obtain the decision tree model with adjusted parameters;

[0028] Input the test set into the decision tree model with adjusted parameters for evaluation to obtain the final fault prediction and identification results, and calculate the evaluation index according to the final fault prediction and identification results and the true fault type labels. Re-adjust the parameters of the decision tree model according to the evaluation index until the evaluation index reaches the preset standard to obtain the trained decision tree model.

[0029] Based on the above method embodiment, the present invention correspondingly provides a system embodiment;

[0030] An embodiment of the present invention provides a transformer monitoring system based on dissolved gas analysis in oil, including: a data acquisition module, a gas concentration calculation module, a gas concentration characteristic ratio calculation module, a preliminary fault prediction and identification module, and a secondary fault prediction and identification module;

[0031] The data acquisition module is used to acquire the chromatographic data and theoretical concentration data of various dissolved gases in the transformer oil to be measured; wherein, the theoretical concentration data includes the normal concentration and warning concentration of the gas;

[0032] The gas concentration calculation module is used to determine the peak value and peak area of the chromatographic data of various dissolved gases according to the chromatographic data of various dissolved gases, and calculate the gas concentration of various dissolved gases according to the peak value and peak area of the chromatographic data of various dissolved gases;

[0033] The gas concentration characteristic ratio calculation module is used to calculate a number of gas concentration characteristic ratios according to the gas concentrations of various dissolved gases;

[0034] The preliminary fault prediction and identification module is used to perform preliminary fault prediction and identification through a trained decision tree model based on the gas concentrations of various dissolved gases and a number of gas concentration characteristic ratios, and obtain the preliminary fault prediction and identification result of the transformer to be measured;

[0035] The secondary fault prediction and identification module is used to perform secondary fault prediction and identification through a trained support vector machine classifier based on the preliminary fault prediction and identification result of the transformer to be measured, the gas concentrations of various dissolved gases, and a number of gas concentration characteristic ratios, and obtain the final fault prediction and identification result of the transformer to be measured.

[0036] Further, after the secondary fault prediction and identification module, there is also included: a transformer control module; the transformer control module includes a fault severity index calculation unit, a fault severity judgment unit, a first control unit, and a second control unit;

[0037] The fault severity index calculation unit is used to calculate the fault severity index according to the gas concentrations of various dissolved gases and the theoretical concentration data when the final fault prediction and identification result is that the transformer to be measured has a fault type;

[0038] The fault severity judgment unit is used to compare the fault severity index with a preset threshold;

[0039] The first control unit is used to, when the fault severity index is greater than the preset threshold, obtain the power outage loss cost of the area controlled by the transformer to be measured; judge whether the power outage loss cost is greater than the preset loss threshold, and if so, start the transformer oil emergency replacement system to replace the oil body of the transformer to be measured; if not, control the transformer to be measured to stop;

[0040] The second control unit is used to keep the current operating state of the transformer to be measured when the fault severity index is not greater than the preset threshold.

[0041] Further, the calculation formula of the fault severity index is:

[0042]

[0043] where FSI is the fault severity index; x i is the gas concentration of the i-th dissolved gas; i is the type number of the dissolved gas; n is the total number of types of dissolved gases; w i is the fault weight coefficient of the i-th gas; is the normal concentration of the i-th dissolved gas; is the warning concentration of the i-th dissolved gas.

[0044] Based on the above method item embodiments, the present invention correspondingly provides terminal device item embodiments, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements a transformer monitoring method based on dissolved gas analysis in oil as described in the present invention.

[0045] Based on the above method item embodiments, the present invention correspondingly provides computer-readable storage medium item embodiments, including: a stored computer program. When the computer program runs, it controls the device where the computer-readable storage medium is located to execute a transformer monitoring method based on dissolved gas analysis in oil as described in the present invention.

[0046] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0047] The present invention first calculates the gas concentration of each dissolved gas according to the chromatographic data of various dissolved gases in the oil of the transformer, and calculates the characteristic ratios of several gas concentrations and the gas concentration change rates of various dissolved gases. Secondly, through the trained decision tree model, it performs preliminary fault type identification based on the gas concentration of each dissolved gas, the characteristic ratios of several gas concentrations, and the gas concentration change rates of various dissolved gases, and obtains the preliminary fault prediction and identification results of the transformer. Finally, through the trained support vector machine classifier, it performs secondary fault identification on the transformer to obtain the final fault identification results of the transformer to be measured. That is, the present invention performs preliminary fault identification and secondary fault identification through the decision tree model and the support vector classifier respectively, performs refined fault identification on the transformer, improves the accuracy of fault identification, and solves the problem that the existing transformer monitoring technology has low accuracy of fault identification due to the inability to finely identify the fault type by a single fault identification operation. Description of the Drawings

[0048] Figure 1 : is the step flow chart of a transformer monitoring method based on dissolved gas analysis in oil provided by the embodiment of the present invention;

[0049] Figure 2 : is the system structure diagram of a transformer monitoring system based on dissolved gas analysis in oil provided by the embodiment of the present invention. Detailed Embodiments

[0050] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0051] In the description of the present invention, it should be understood that the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features.

[0052] Embodiment 1:

[0053] Referring to Figure 1 , which is a step flow chart of a transformer monitoring method based on dissolved gas analysis in oil provided by an embodiment of the present invention; the method at least includes the following steps:

[0054] Step S1: Obtain chromatographic data and theoretical concentration data of various dissolved gases in the transformer oil to be measured; wherein, the theoretical concentration data includes the normal concentration and warning concentration of the gas;

[0055] In this embodiment, the method for obtaining chromatographic data of various dissolved gases in the transformer oil to be measured is to perform oil-gas separation on the oil sample to be monitored, and perform chromatographic analysis on various gases in the oil sample to be monitored by a gas chromatograph to obtain chromatographic data of various gases in the oil sample.

[0056] In this embodiment, the theoretical concentration data of various dissolved gases in the transformer oil to be measured can refer to the domestic power industry standard "Guide for the Analysis and Judgment of Dissolved Gases in Transformer Oil" (DL / T 722 2000) or the international standard IEC60599 standard, or can also be obtained from the equipment manual provided by the transformer manufacturer.

[0057] In this embodiment, the various dissolved gases in the transformer oil to be measured mainly include hydrogen H 2 , methane CH 4 , ethane C 2 H 6 , ethylene C 2 H 4 , acetylene C 2 H 2 , carbon monoxide CO and carbon dioxide CO 2 .

[0058] Step S2: Determine the peak value and peak area of the chromatographic data of various dissolved gases according to the chromatographic data of various dissolved gases;

[0059] In this embodiment, according to the chromatographic data of various dissolved gases, the ordinate corresponding to the vertex of the chromatographic peak of each dissolved gas is used as the peak value of the chromatographic data of each dissolved gas, and the chromatographic peak area within a preset range before and after the appearance time of the chromatographic peak of each dissolved gas is calculated; wherein, the preset range before and after the appearance time of the chromatographic peak can be set by those skilled in the art according to the actual situation.

[0060] Step S3: Calculate the gas concentration of each dissolved gas according to the peak value and peak area of the chromatographic data of each dissolved gas;

[0061] In this embodiment, the calculation formula for the gas concentration of each dissolved gas is:

[0062] x i =a 1 H i +a 2 S i ;

[0063] wherein, x i is the gas concentration of the i-th dissolved gas; H i is the peak value of the chromatographic data of the i-th dissolved gas; S i is the peak area of the chromatographic data of the i-th dissolved gas; i is the type number of the dissolved gas; a 1 and a 2 are the preset peak value weight and peak area weight respectively.

[0064] In this embodiment, after calculating the gas concentration of each dissolved gas, it further includes:

[0065] Performing data preprocessing on the calculated gas concentration of each dissolved gas respectively to obtain the processed gas concentration of each dissolved gas; wherein, the data preprocessing includes but is not limited to min-max normalization and standardization processing.

[0066] Step S4: Calculate several gas concentration characteristic ratios according to the gas concentration of each dissolved gas;

[0067] In this embodiment, the gas concentration characteristic ratios include the gas concentration ratio of acetylene to ethylene the gas concentration ratio of methane to hydrogen the gas concentration ratio of ethylene to ethane the gas concentration ratio of carbon monoxide to carbon dioxide and the gas concentration ratio of hydrogen to the total concentration of hydrocarbon gases wherein, the total concentration of hydrocarbon gases is the sum of the gas concentrations of methane, ethane, ethylene and acetylene;

[0068] In this embodiment, the gas concentration ratio of acetylene to ethylene It can be used to judge the discharge energy and fault severity of the transformer; Exemplarily, when the ratio is less than 0.1, it is judged that the transformer generally has an overheating fault. When the ratio is greater than 0.1, it is judged that the transformer generally has a discharge fault. When the ratio increases from 0.1 to 3, it indicates that the spark discharge intensity of the transformer increases. When the ratio is greater than 3, it indicates that the transformer has a high-energy discharge fault;

[0069] In this embodiment, the gas concentration ratio of methane to hydrogen It can be used to judge the discharge type. Different ratio ranges correspond to different discharge types. The discharge types include partial discharge, low-energy discharge and high-energy discharge; Exemplarily, the ratio range of partial discharge is 0.1 - 0.5. In the initial stage of partial discharge, the production amount of hydrogen is relatively large and methane is relatively small, so this ratio is relatively small. The ratio range of low-energy discharge is 0.5 - 1.5. As the discharge energy increases but has not reached the high-energy level, the production amount of methane will increase relatively compared with that in partial discharge, making this ratio increase. The ratio range of high-energy discharge is greater than 1.5. In high-energy discharge, the discharge energy is large, which will promote more complex chemical reactions to occur. The proportion of methane production relative to hydrogen further increases, resulting in this ratio rising to this range;

[0070] In this embodiment, the gas concentration ratio of ethylene to ethane It can be used to judge the overheating fault type of the transformer. Different ratio ranges correspond to different overheating fault types. The overheating fault types include low-temperature overheating fault, medium-temperature overheating fault and high-temperature overheating fault; Exemplarily, the ratio range of low-temperature overheating fault is [1, 3), the range of medium-temperature overheating fault is [3, 10), and the ratio range of high-temperature overheating fault is [10, +∞). As the temperature rises, the chemical reaction becomes more intense, the production amount of ethylene increases significantly, and the proportion relative to ethane further increases, resulting in this ratio increasing as the temperature rises;

[0071] In this embodiment, the gas concentration ratio of carbon monoxide to carbon dioxide It can be used to judge the aging fault type of the equipment insulation according to the level of the ratio. The aging fault types include low-risk aging fault, medium-risk aging fault and high-risk fault. The higher the ratio, the higher the risk level of the aging fault of the equipment insulation;

[0072] In this embodiment, the gas concentration ratio of hydrogen to the total concentration of hydrocarbon gases It can be used to determine the type of fault. Different ratio ranges correspond to different types of faults. Exemplarily, when the ratio is in the range of [0.9, +∞), it is preliminarily considered that the transformer may have a partial discharge fault. At this time, the content of hydrocarbon gases such as methane is relatively small, and hydrogen is one of the main characteristic gases. When the ratio is in the range of [0.6, 0.9], it is preliminarily considered that the transformer may have a spark discharge fault. Spark discharge is generally caused by a relatively large potential difference inside the transformer, resulting in sparks at some gaps or poor contact parts, decomposing the insulating oil and solid insulating materials, generating more hydrogen and a certain amount of acetylene. When the ratio is in the range of [0.3, 0.6), it is preliminarily considered that the transformer may have an arc discharge fault. When the ratio is in the range of (0, 0.3), it is preliminarily considered that the transformer may have an overheating fault;

[0073] Step S5: Perform preliminary fault prediction and identification on the transformer to be tested according to the gas concentrations of various dissolved gases and several gas concentration characteristic ratios by using the trained decision tree model, and obtain the preliminary fault prediction and identification result of the transformer to be tested;

[0074] In this embodiment, the model training of the decision tree model includes:

[0075] Obtain the historical gas concentrations of various dissolved gases in the transformer oil with true fault type labels, and calculate several historical gas concentration characteristic ratios according to the historical gas concentrations of various dissolved gases;

[0076] Divide the historical gas concentrations of various dissolved gases and several historical gas concentration characteristic ratios into a training set, a validation set and a test set according to a certain proportion;

[0077] According to the preset decision tree splitting criterion, repeatedly perform subset partitioning on the training set until the preset stop condition is met, and obtain the decision tree model; wherein, the decision tree splitting criterion includes but is not limited to information gain and Gini index; the preset stop condition includes but is not limited to reaching the maximum tree depth and the minimum number of samples;

[0078] Input the validation set into the decision tree model for evaluation, obtain the preliminary fault prediction and identification result, and calculate the evaluation index according to the preliminary fault prediction and identification result and the true fault type label. Adjust the parameters of the decision tree model according to the evaluation index until the evaluation index reaches the preset standard, and obtain the decision tree model with adjusted parameters; wherein, the evaluation index includes but is not limited to accuracy rate, recall rate and F1 score;

[0079] The test set is input into the decision tree model with adjusted parameters for evaluation to obtain the final fault prediction and identification results. Then, according to the final fault prediction and identification results and the true fault type labels, evaluation metrics are calculated. Based on the evaluation metrics, the parameters of the decision tree model are readjusted until the evaluation metrics reach the preset standard, and a trained decision tree model is obtained. Among them, the evaluation metrics include but are not limited to accuracy, recall rate, and F1 score.

[0080] Step S6: The trained support vector machine classifier performs secondary fault prediction and identification based on the preliminary fault prediction and identification results of the transformer to be tested, the gas concentrations of various dissolved gases, and several gas concentration characteristic ratios, and obtains the final fault prediction and identification results of the transformer to be tested.

[0081] In this embodiment, the trained support vector machine classifier performs secondary fault prediction and identification based on the preliminary fault prediction and identification results of the transformer to be tested, the gas concentrations of various dissolved gases, and several gas concentration characteristic ratios, and obtains the final fault prediction and identification results of the transformer to be tested, specifically as follows:

[0082] Numerically transform the preliminary fault prediction and identification results of the transformer to be tested to obtain numerically transformed preliminary fault prediction and identification results.

[0083] Combine the numerically transformed preliminary fault prediction and identification results, the gas concentrations of various dissolved gases, and several gas concentration characteristic ratios into a feature vector.

[0084] According to the type of the trained support vector machine classifier, determine the decision function, and input the feature vector into the decision function for solution to obtain a decision value. Among them, different decision values correspond to different final fault prediction and identification results. The types of the trained support vector machine classifier include linear support vector machine and non-linear support vector machine.

[0085] According to the decision value, determine the final fault prediction and identification results of the transformer to be tested. Among them, the final fault prediction and identification results are the fault subclasses of the preliminary fault prediction and identification results.

[0086] In this embodiment, after obtaining the final fault prediction and identification results of the transformer to be tested, it further includes:

[0087] When the final fault prediction and identification results indicate that the transformer to be tested has a fault type, calculate the fault severity index according to the gas concentrations of various dissolved gases and the theoretical concentration data.

[0088] In this embodiment, the calculation formula of the fault severity index is:

[0089]

[0090] Among them, FSI is the fault severity index; x i is the gas concentration of the i-th dissolved gas; i is the type number of the dissolved gas; n is the total number of types of dissolved gases; w i is the fault weight coefficient of the i-th gas; is the normal concentration of the i-th dissolved gas; is the warning concentration of the i-th dissolved gas;

[0091] Compare the fault severity index with a preset threshold; among them, the preset threshold is set to 0.7 and can be set by technicians according to the actual situation;

[0092] When the fault severity index is greater than the preset threshold, obtain the power outage loss cost of the area controlled by the transformer to be measured; judge whether the power outage loss cost is greater than the preset loss threshold. If so, start the transformer oil emergency replacement system to replace the oil of the transformer to be measured; if not, control the transformer to be measured to stop;

[0093] In this embodiment, the power outage loss cost of the area controlled by the transformer to be measured is the total amount of power outage losses pre-stored in the database by all users in the area controlled by the transformer to be measured; the preset loss threshold is set by technicians according to the actual situation;

[0094] When the fault severity index is not greater than the preset threshold, keep the current operating state of the transformer to be measured.

[0095] In this embodiment, after respectively performing preliminary fault identification and secondary fault identification through a decision tree model and a support vector classifier to perform refined fault identification on the transformer and improve the accuracy of fault identification, when it is determined that the transformer has a fault according to the fault identification result, calculate the fault severity index, and combine the fault severity index with the power outage loss cost for judgment to determine the operation control strategy of the transformer, so as to combine the transformer fault identification result with the actual power outage loss cost and improve the rationality of the transformer operation control strategy.

[0096] Embodiment 2:

[0097] Refer to Figure 2 , which is the system structure diagram of a transformer monitoring system based on dissolved gas analysis in oil provided by an embodiment of the present invention. The system at least includes: a data acquisition module, a gas concentration calculation module, a gas concentration characteristic ratio calculation module, a preliminary fault prediction and identification module, and a secondary fault prediction and identification module;

[0098] The data acquisition module is used to acquire the chromatographic data and theoretical concentration data of various dissolved gases in the oil of the transformer to be measured; among them, the theoretical concentration data includes the normal concentration and warning concentration of the gas;

[0099] The gas concentration calculation module is used to determine the peak value and peak area of the chromatographic data of various dissolved gases according to the chromatographic data of various dissolved gases, and calculate the gas concentration of various dissolved gases according to the peak value and peak area of the chromatographic data of various dissolved gases;

[0100] The gas concentration characteristic ratio calculation module is used to calculate several gas concentration characteristic ratios according to the gas concentration of various dissolved gases;

[0101] The preliminary fault prediction and identification module is used to perform preliminary fault prediction and identification through a trained decision tree model according to the gas concentration of various dissolved gases and several gas concentration characteristic ratios, and obtain the preliminary fault prediction and identification result of the transformer to be measured;

[0102] The secondary fault prediction and identification module is used to perform secondary fault prediction and identification through a trained support vector machine classifier according to the preliminary fault prediction and identification result of the transformer to be measured, the gas concentration of various dissolved gases and several gas concentration characteristic ratios, and obtain the final fault prediction and identification result of the transformer to be measured.

[0103] In this embodiment, after the secondary fault prediction and identification module, there is further included: a transformer control module; the transformer control module includes a fault severity index calculation unit, a fault severity judgment unit, a first control unit and a second control unit;

[0104] The fault severity index calculation unit is used to calculate the fault severity index according to the gas concentration of various dissolved gases and the theoretical concentration data when the final fault prediction and identification result is that the transformer to be measured has a fault type;

[0105] The fault severity judgment unit is used to compare the fault severity index with a preset threshold;

[0106] The first control unit is used to obtain the power-off loss cost of the area controlled by the transformer to be measured when the fault severity index is greater than the preset threshold; judge whether the power-off loss cost is greater than the preset loss threshold, if so, start the transformer oil emergency replacement system to replace the oil body of the transformer to be measured; if not, control the transformer to be measured to stop;

[0107] The second control unit is used to keep the current operating state of the transformer to be measured when the fault severity index is not greater than the preset threshold.

[0108] In this embodiment, the calculation formula of the fault severity index is:

[0109]

[0110] Among them, FSI is the fault severity index; x i is the gas concentration of the i-th dissolved gas; i is the type number of the dissolved gas; n is the total number of types of dissolved gases; w i is the fault weight coefficient of the i-th gas; is the normal concentration of the i-th dissolved gas; is the warning concentration of the i-th dissolved gas.

[0111] Based on the above method item embodiments, another embodiment is provided;

[0112] A transformer monitoring terminal device based on dissolved gas analysis in oil provided by another embodiment of the present invention includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements a transformer monitoring method based on dissolved gas analysis in oil described in any one of the above method item embodiments of the present invention.

[0113] Exemplarily, the computer program can be divided into one or more modules / units. The one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the transformer monitoring terminal device based on dissolved gas analysis in oil.

[0114] The transformer monitoring terminal device based on dissolved gas analysis in oil can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The transformer monitoring terminal device based on dissolved gas analysis in oil may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that, for example, the transformer monitoring terminal device based on dissolved gas analysis in oil may further include input / output devices, network access devices, a bus, etc.

[0115] The so-called processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the transformer monitoring terminal device based on dissolved gas analysis in oil, and connects various parts of the entire transformer monitoring terminal device based on dissolved gas analysis in oil through various interfaces and lines.

[0116] The memory can be used to store the computer programs and / or modules. The processor realizes various functions of the transformer monitoring terminal device based on dissolved gas analysis in oil by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, at least one magnetic disk storage device, flash device, or other volatile solid-state storage devices.

[0117] Based on the above method item embodiments, another embodiment is provided;

[0118] A storage medium provided by another embodiment of the present invention includes a stored computer program, wherein when the computer program runs, it controls the device where the storage medium is located to execute the method for monitoring a transformer based on dissolved gas analysis in oil described in any one of the above method item embodiments of the present invention.

[0119] Among them, the above storage medium is a computer-readable storage medium. When the module / unit integrated with the transformer monitoring system / terminal device based on dissolved gas analysis in oil is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-mentioned embodiment methods of the present invention, it can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0120] It should be noted that the above terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above terminal device is only an example and does not constitute a limitation on the terminal device. It may include more or fewer components, or combine some components, or different components.

[0121] The specific embodiments described above further elaborate on the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only the specific embodiments of the present invention and is not used to limit the protection scope of the present invention. In particular, it is pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A transformer monitoring method based on dissolved gas analysis in oil, characterized in that: include: Obtaining chromatographic data and theoretical concentration data of various dissolved gases in the transformer oil to be tested; wherein the theoretical concentration data includes normal concentration and warning concentration of the gas; According to the chromatographic data of various dissolved gases, the peak values ​​and peak areas of the chromatographic data of various dissolved gases are determined; Calculate the gas concentration of various dissolved gases based on the peak values ​​and peak areas of the chromatographic data of various dissolved gases; According to the gas concentration of various dissolved gases, several gas concentration characteristic ratios are calculated; The trained decision tree model is used to perform preliminary fault prediction and identification based on the gas concentrations of various dissolved gases and several gas concentration characteristic ratios, and the preliminary fault prediction and identification results of the transformer on the waiting side are obtained; Through the trained support vector machine classifier, secondary fault prediction and identification are performed according to the preliminary fault prediction and identification results of the waiting transformer, the gas concentrations of various dissolved gases and several gas concentration characteristic ratios to obtain the final fault prediction and identification results of the waiting transformer.

2. A transformer monitoring method based on dissolved gas analysis in oil according to claim 1, characterized in that: The calculation formula for the gas concentration of the various dissolved gases is: x i =a1H i +a2S i ; Among them, x i is the gas concentration of the i-th dissolved gas; H i is the peak value of the chromatographic data of the i-th dissolved gas; S i is the peak area of ​​the chromatographic data of the i-th dissolved gas; i is the type number of the dissolved gas; a1 and a2 are the preset peak weight and peak area weight, respectively.

3. A transformer monitoring method based on dissolved gas analysis in oil according to claim 2, characterized in that: After obtaining the final fault prediction and identification result of the transformer on the waiting side, it also includes: When the final fault prediction identification result is that the transformer to be tested has a fault type, the fault severity index is calculated based on the gas concentration and theoretical concentration data of various dissolved gases; comparing the fault severity index with a preset threshold; When the fault severity index is greater than a preset threshold, the power outage loss cost of the area controlled by the transformer under test is obtained; it is determined whether the power outage loss cost is greater than the preset loss threshold. If so, the transformer oil emergency replacement system is started to replace the oil of the transformer under test; if not, the transformer under test is controlled to shut down; When the fault severity index is not greater than a preset threshold, the current operating state of the transformer to be tested is maintained.

4. A transformer monitoring method based on dissolved gas analysis in oil according to claim 3, characterized in that: The calculation formula of the fault severity index is: Where FSI is the fault severity index; x i is the gas concentration of the i-th dissolved gas; i is the type number of the dissolved gas; n is the total number of dissolved gas types; w i is the fault weight coefficient of the i-th gas; is the normal concentration of the ith dissolved gas; is the warning concentration of the ith dissolved gas.

5. A transformer monitoring method based on dissolved gas analysis in oil according to claim 4, characterized in that: The model training of the decision tree model includes: Obtain the historical gas concentration of each dissolved gas in the transformer oil with the real fault type label, and calculate several historical gas concentration characteristic ratios according to the historical gas concentration of each dissolved gas; The historical gas concentration of each dissolved gas and several historical gas concentration characteristic ratios are divided into a training set, a validation set and a test set according to a certain ratio; According to the preset decision tree splitting criteria, the training set is repeatedly divided into subsets until the preset stopping condition is met to obtain a decision tree model; Input the validation set into the decision tree model for evaluation to obtain preliminary fault prediction and identification results, and calculate the evaluation index based on the preliminary fault prediction and identification results and the actual fault type label. Adjust the parameters of the decision tree model according to the evaluation index until the evaluation index reaches the preset standard, and obtain the decision tree model with adjusted parameters. The test set is input into the decision tree model with adjusted parameters for evaluation to obtain the final fault prediction and recognition result. The evaluation index is calculated based on the final fault prediction and recognition result and the actual fault type label. The parameters of the decision tree model are readjusted according to the evaluation index until the evaluation index reaches the preset standard to obtain a trained decision tree model.

6. A transformer monitoring system based on dissolved gas analysis in oil, characterized in that: include: Data acquisition module, gas concentration calculation module, gas concentration characteristic ratio calculation module, preliminary fault prediction and identification module and secondary fault prediction and identification module; The data acquisition module is used to acquire chromatographic data and theoretical concentration data of various dissolved gases in the transformer oil to be tested; wherein the theoretical concentration data includes normal concentration and warning concentration of the gas; The gas concentration calculation module is used to determine the peak values ​​and peak areas of the chromatographic data of various dissolved gases according to the chromatographic data of various dissolved gases, and calculate the gas concentrations of various dissolved gases according to the peak values ​​and peak areas of the chromatographic data of various dissolved gases; The gas concentration characteristic ratio calculation module is used to calculate a number of gas concentration characteristic ratios according to the gas concentrations of various dissolved gases; The preliminary fault prediction and identification module is used to perform preliminary fault prediction and identification according to the gas concentrations of various dissolved gases and a number of gas concentration characteristic ratios through a trained decision tree model to obtain a preliminary fault prediction and identification result of the transformer on the waiting side; The secondary fault prediction and identification module is used to perform secondary fault prediction and identification based on the preliminary fault prediction and identification result of the waiting transformer, the gas concentration of various dissolved gases and several gas concentration characteristic ratios through a trained support vector machine classifier to obtain the final fault prediction and identification result of the waiting transformer.

7. A transformer monitoring system based on dissolved gas analysis in oil according to claim 6, characterized in that: After the secondary fault prediction and identification module, it also includes: a transformer control module; the transformer control module includes a fault severity index calculation unit, a fault severity judgment unit, a first control unit and a second control unit; The fault severity index calculation unit is used to calculate the fault severity index according to the gas concentration and theoretical concentration data of various dissolved gases when the final fault prediction identification result is that the transformer to be tested has a fault type; The fault severity determination unit is used to compare the fault severity index with a preset threshold value; The first control unit is used to obtain the power outage loss cost of the area controlled by the transformer under test when the fault severity index is greater than a preset threshold; determine whether the power outage loss cost is greater than the preset loss threshold, and if so, start the transformer oil emergency replacement system to replace the oil of the transformer under test; if not, control the transformer under test to shut down; The second control unit is used to maintain the current operating state of the transformer to be tested when the fault severity index is not greater than a preset threshold.

8. A transformer monitoring system based on dissolved gas analysis in oil according to claim 7, characterized in that: The calculation formula of the fault severity index is: Where FSI is the fault severity index; x i is the gas concentration of the i-th dissolved gas; i is the type number of the dissolved gas; n is the total number of dissolved gas types; w i is the fault weight coefficient of the i-th gas; is the normal concentration of the ith dissolved gas; is the warning concentration of the ith dissolved gas.

9. A transformer monitoring terminal device based on dissolved gas analysis in oil, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the transformer monitoring method based on dissolved gas analysis in oil as described in any one of claims 1 to 5 is implemented.

10. A storage medium, characterized in that: The storage medium includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute a transformer monitoring method based on dissolved gas analysis in oil as described in any one of claims 1 to 5.

Images