Fault diagnosis method and system for operation inner leakage of high-pressure fire-resistant oil system

Abstract

The invention discloses a fault diagnosis method and system for internal leakage in operation of a high-pressure fire-resistant oil system. The method comprises the following steps: acquiring an actual flow value and related process parameter values from a distributed control system; obtaining an EH oil system expert knowledge base; establishing an EH oil system operation comprehensive fault diagnosis model; the EH oil system operation comprehensive fault diagnosis model is used for processing related process parameters, and a predicted flow value of the valve is obtained; inner leakage of the valve is determined, and early warning information is generated. The system comprises a parameter acquisition module, a data induction module, a model establishment module, a parameter processing module, a determination module and an early warning module. According to the invention, when EH in-oil leakage early warning occurs, an operator can monitor the remaining time of oil leakage and the current oil leakage rate in real time, so that the operator can know the most valuable index information when an accident occurs.

Description

Technical Field

[0001] The present application belongs to the field of intelligent control of high-pressure fire-resistant fuel oil systems in thermal power plants, and specifically relates to a fault diagnosis method and system for internal leakage in the operation of a high-pressure fire-resistant fuel oil system. Background Art

[0002] The function of the steam turbine EH oil system is to provide high-pressure fire-resistant oil, which drives the servo actuator, which responds to the command signal from the microprocessor controller to adjust the opening of each valve of the steam turbine. Among them, EH oil internal leakage is a common fault of the EH oil system in the steam turbine digital electro-hydraulic control system (DEH), which seriously affects the safe operation of the steam turbine.

[0003] The number of valves that need to be managed by modern coal chemical enterprises is increasing, and the various types of inspection and maintenance data generated are growing exponentially. However, the traditional management method has failed to give full play to the role of various valve maintenance information. It is only a ledger and record, which is manually queried when necessary, and then the technicians make personal judgments to form valve maintenance recommendations.

[0004] Currently, there are very few systems or means involved in judging the online operating status of relevant valves. Valve internal leakage is one of the main problems faced by relevant enterprises. There is a lack of online monitoring means, and it is impossible to monitor in real time and make predictive maintenance suggestions. When it is necessary to manually judge the internal leakage and take corresponding maintenance measures, the valve has been out of repair, resulting in a significant increase in maintenance costs, and even leading to abnormal production or unplanned shutdown of the device.

[0005] In the above-mentioned related technologies, process personnel can only judge whether there is an internal leakage by combining DCS parameters, or use temperature measurement methods to assist in on-site judgment of whether the valve is leaking. The detection is not timely and accurate, which may easily lead to abnormal production of chemical equipment. No effective solution has been proposed yet.

[0006] At present, the intelligent construction of coal-fired units has become a new hotspot in the development of power plants. Many scholars have begun to apply artificial intelligence technology, data mining technology, deep learning technology, etc. to coal-fired units, and have tried to establish various system models to perform status analysis and fault diagnosis on the units to improve the level of refined control of coal-fired units. Summary of the invention

[0007] In view of the above technical background, the object of the present invention is to provide a method and system for diagnosing internal leakage of a high-pressure fire-resistant oil system during operation.

[0008] In order to achieve the above object, the present invention adopts the following technical solution:

[0009] A method for diagnosing internal leakage of a high-pressure fire-resistant oil system, comprising:

[0010] 1) Obtaining the actual flow value and related process parameter values ​​from a distributed control system, wherein the distributed control system is a control system used in the field of industrial automation, the actual flow value is the real value obtained from the DCS, and the related process parameter value is the real value obtained from the DCS, which is used to predict the target parameters of the valve;

[0011] 2) Based on the expert experience and knowledge of more than 20 ultra-supercritical coal-fired units, the debugging results of the steam turbine professional during the capital construction and commissioning period and the operating data during the commercial operation period are summarized and summarized to obtain the EH oil system expert knowledge base;

[0012] 3) Based on the EH oil system expert knowledge base, a comprehensive fault diagnosis model for the EH oil system operation was established, including a diversity comparison model in the start-stop process under normal conditions, a steady-state and variable condition operation model during operation, and a data model for joint classification and identification of multiple faults in the start-operation-stop process under fault conditions;

[0013] 4) Inputting the relevant process parameter values ​​obtained from the distributed control system into the EH oil system operation comprehensive fault diagnosis model, so as to process the relevant process parameters using the EH oil system operation comprehensive fault diagnosis model to obtain the predicted flow value of the valve, wherein the EH oil system operation comprehensive fault diagnosis model is trained by machine learning using multiple sets of training data;

[0014] 5) When the difference between the predicted flow value and the actual flow value obtained from the distributed control system is greater than the deviation threshold, it is determined that the valve has internal leakage;

[0015] 6) After determining that the valve has internal leakage, generate warning information; wherein the warning information includes: valve position information and internal leakage information.

[0016] A further improvement of the present invention is that in step 1), the relevant process parameter values ​​include particle size, water content, acid value, foam characteristics, resistivity, temperature, flow coefficient of valves, parameters of process pipelines and volume of reaction equipment in oil.

[0017] A further improvement of the present invention is that, in step 2), based on the expert experience and knowledge of more than 20 ultra-supercritical coal-fired units, the debugging results of the steam turbine professional during the infrastructure commissioning period and the operating data during the commercial operation period are used to summarize and obtain the EH oil system expert knowledge base, including:

[0018] Data cleaning

[0019] Therefore, it is necessary to clean the original data and remove invalid data to ensure that all data is accepted by the computer. In order to improve the accuracy of the model and prevent downtime data and abnormal data from interfering with normal data, this part of the data is processed according to expert experience and reconstructed using Python programming to construct data for model training;

[0020] Feature Normalization

[0021] Eliminate the dimensional impact between indicators and perform data normalization to ensure the comparability between data indicators.

[0022] A further improvement of the present invention is that, in step 3), a comprehensive fault diagnosis model for the operation of the EH oil system is established based on the EH oil system expert knowledge base, including:

[0023] BPNN model design

[0024] A four-layer neural network is used to train the normal operation process of the EH oil system, where the number of input layer nodes is 5, the number of nodes in the first hidden layer is set to 26 after cross-validation and considering the computational efficiency, the number of nodes in the second hidden layer is 4, and the number of nodes in the output layer is 1 after the dynamic neuron activation layer, and the internal leakage is analyzed by analyzing the system operation characteristics;

[0025] Optimization Algorithm

[0026] The update formula used by the ADAM algorithm is as follows:

[0027] m0=v0=0

[0028]

[0029] Where m0 and v0 are the initialization parameter vector matrices, β1, β2, η, ε are the configuration parameters in the algorithm update process, and E and w are the error and weight in the algorithm update process;

[0030] Evaluation methods and indicators

[0031] Mean square error: the ratio of the sum of squares of the deviations between the observed value and the true value to the number of observations;

[0032]

[0033] The closer the MSE value is to 0, the more accurate the prediction model is in describing the experimental data.

[0034] A further improvement of the present invention is that in step 4), the EH oil system operation comprehensive fault diagnosis model is the flow coefficient of the valve and the parameters of the process pipeline; the flow coefficient is input into the EH oil system operation comprehensive fault diagnosis model to obtain the predicted flow value of the valve.

[0035] A further improvement of the present invention is that in step 5), determining whether the valve has internal leakage comprises:

[0036] Get the current status of the valve group, predict the flow value, and predict the parameter value;

[0037] When the actual flow value is greater than the flow threshold, it is determined that the valve has internal leakage.

[0038] A further improvement of the present invention is that in step 6), determining the position where the valve internal leakage occurs includes:

[0039] The current state of the regulating valve group is obtained; and the position of the internal leakage valve in the regulating valve group is determined according to the current state.

[0040] A further improvement of the present invention is that in step 6), generating early warning information includes:

[0041] The warning information is sent to the distributed control system so that the image display device of the distributed control system can be used to display the warning information and the broadcasting device of the distributed control system can be used to broadcast the warning information.

[0042] A fault diagnosis system for internal leakage of a high-pressure fire-resistant oil system, comprising:

[0043] A parameter acquisition module obtains the actual flow value and related process parameter values ​​from a distributed control system, wherein the distributed control system is a control system used in the field of industrial automation, the actual flow value is the real value obtained from the DCS, and the related process parameter value is the real value obtained from the DCS, which is used to predict the target parameters of the valve;

[0044] The data summarization module is based on the expert experience and knowledge of more than 20 ultra-supercritical coal-fired units, using the commissioning results of steam turbine professionals during the capital construction commissioning period and the operating data during the commercial operation period to summarize and obtain the EH oil system expert knowledge base;

[0045] The model building module builds a comprehensive fault diagnosis model for the EH oil system operation based on the EH oil system expert knowledge base, including a diversity comparison model in the start-stop process under normal conditions, a steady-state and variable condition operation model during operation, and a data model for joint classification and identification of multiple faults in the start-operation-stop process under fault conditions;

[0046] The parameter processing module inputs the relevant process parameter values ​​obtained from the distributed control system into the EH oil system operation comprehensive fault diagnosis model, so as to process the relevant process parameters using the EH oil system operation comprehensive fault diagnosis model to obtain the predicted flow value of the valve, wherein the EH oil system operation comprehensive fault diagnosis model is trained by machine learning using multiple sets of training data;

[0047] A determination module determines that the valve has internal leakage when the difference between the predicted flow value and the actual flow value obtained from the decentralized control system is greater than a deviation threshold;

[0048] The early warning module generates early warning information after determining that the valve has internal leakage; wherein the early warning information includes: valve position information and internal leakage information.

[0049] A further improvement of the present invention is that in the parameter acquisition module, the relevant process parameter values ​​include oil particle size, moisture, acid value, foam characteristics, resistivity, temperature, valve flow coefficient, process pipeline parameters and the volume of the reaction equipment.

[0050] Compared with the prior art, the present invention has at least the following beneficial technical effects:

[0051] The present invention provides a method and system for diagnosing internal leakage of a high-pressure fire-resistant oil system. Through data cleaning and screening, an EH oil system expert knowledge base is established. Through the EH oil system expert knowledge base, characteristic data is selected to establish a comprehensive fault diagnosis model for internal leakage of valves in the EH oil system. The comprehensive fault diagnosis model for internal leakage of valves in the EH oil system is used to obtain a valve group flow prediction value through neural network training. The internal leakage fault state is determined by the difference threshold between the predicted flow value and the actual flow value of the valve group, and early warning information is generated. The early warning information includes the position of the valve and the internal leakage information. The present invention allows operating personnel to monitor the remaining time of oil leakage and the current oil leakage rate in real time when an EH oil internal leakage warning occurs, so that operating personnel can understand the most valuable indicator information when an accident occurs. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 Schematic diagram of the fully connected network structure and neuron topology.

[0053] Figure 2 It is the effect diagram of the application example of the present invention.

[0054] Figure 3 It is a structural block diagram of the system of the present invention. DETAILED DESCRIPTION

[0055] In the following, only some exemplary embodiments are briefly described. As those skilled in the art will appreciate, the described embodiments may be modified in various ways without departing from the spirit or scope of the present invention. Therefore, the drawings and descriptions are considered to be exemplary and non-restrictive in nature.

[0056] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0057] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0058] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0059] Various structural schematic diagrams of the embodiments disclosed in the present invention are shown in the accompanying drawings. These figures are not drawn to scale, and some details are magnified and some details may be omitted for the purpose of clear expression. The shapes of various regions and layers shown in the figures and the relative sizes and positional relationships therebetween are only exemplary, and may deviate in practice due to manufacturing tolerances or technical limitations, and those skilled in the art may additionally design regions / layers with different shapes, sizes, and relative positions according to actual needs.

[0060] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0061] Example 1

[0062] The present invention provides a method for diagnosing internal leakage of a high-pressure fire-resistant oil system, comprising:

[0063] 1) Obtaining the actual flow value and related process parameter values ​​from a distributed control system, wherein the distributed control system is a control system used in the field of industrial automation, the actual flow value is the real value obtained from the DCS, and the related process parameter value is the real value obtained from the DCS, which is used to predict the target parameters of the valve;

[0064] 2) Based on the expert experience and knowledge of more than 20 ultra-supercritical coal-fired units, the debugging results of the steam turbine professional during the capital construction and commissioning period and the operating data during the commercial operation period are summarized and summarized to obtain the EH oil system expert knowledge base;

[0065] 3) Based on the EH oil system expert knowledge base, a comprehensive fault diagnosis model for the EH oil system operation was established, including a diversity comparison model in the start-stop process under normal conditions, a steady-state and variable condition operation model during operation, and a data model for joint classification and identification of multiple faults in the start-operation-stop process under fault conditions;

[0066] 4) Inputting the relevant process parameter values ​​obtained from the distributed control system into the EH oil system operation comprehensive fault diagnosis model, so as to process the relevant process parameters using the EH oil system operation comprehensive fault diagnosis model to obtain the predicted flow value of the valve, wherein the EH oil system operation comprehensive fault diagnosis model is trained by machine learning using multiple sets of training data;

[0067] 5) When the difference between the predicted flow value and the actual flow value obtained from the distributed control system is greater than the deviation threshold, it is determined that the valve has internal leakage;

[0068] 6) After determining that the valve has internal leakage, generate warning information; wherein the warning information includes: valve position information and internal leakage information.

[0069] In this embodiment, in step 1), the relevant process parameter values ​​include particle size, water content, acid value, foam characteristics, resistivity, temperature, flow coefficient of valves, parameters of process pipelines and volume of reaction equipment in oil.

[0070] In this embodiment, in step 2), based on the expert experience and knowledge of more than 20 ultra-supercritical coal-fired units, the debugging results of the steam turbine professional during the infrastructure commissioning period and the operating data during the commercial operation period are summarized and summarized to obtain the EH oil system expert knowledge base, including:

[0071] Data cleaning

[0072] Therefore, it is necessary to clean the original data and remove invalid data to ensure that all data is accepted by the computer. In order to improve the accuracy of the model and prevent downtime data and abnormal data from interfering with normal data, this part of the data is processed according to expert experience and reconstructed using Python programming to construct data for model training;

[0073] Feature Normalization

[0074] Eliminate the dimensional impact between indicators and perform data normalization to ensure the comparability between data indicators.

[0075] In this embodiment, in step 3), a comprehensive fault diagnosis model for the EH oil system operation is established based on the EH oil system expert knowledge base, including:

[0076] BPNN model design

[0077] like Figure 1 As shown, a four-layer neural network is used to train the normal operation process of the EH oil system, where the number of input layer nodes is 5. After cross-validation and considering the calculation efficiency, the number of nodes in the first hidden layer is set to 26, the number of nodes in the second hidden layer is 4, and the number of output layer nodes is 1 after the dynamic neuron activation layer. The internal leakage is analyzed by analyzing the system operation characteristics;

[0078] Optimization Algorithm

[0079] The update formula used by the ADAM algorithm is as follows:

[0080] m0=v0=0

[0081]

[0082] Where m0 and v0 are the initialization parameter vector matrices, β1, β2, η, ε are the configuration parameters in the algorithm update process, and E and w are the error and weight in the algorithm update process.

[0083] Evaluation methods and indicators

[0084] Mean square error: the ratio of the sum of squares of the deviations between the observed value and the true value to the number of observations;

[0085]

[0086] The closer the MSE value is to 0, the more accurate the prediction model is in describing the experimental data.

[0087] In this embodiment, in step 4), the relevant process parameter values ​​obtained from the distributed control system are input into the EH oil system operation comprehensive fault diagnosis model, so as to process the relevant process parameters using the EH oil system operation comprehensive fault diagnosis model to obtain the predicted flow value of the valve, wherein the EH oil system operation comprehensive fault diagnosis model is trained by machine learning using multiple sets of training data, including:

[0088] 1) The independent variable must have a significant impact on the dependent variable and be closely linearly correlated;

[0089] 2) The linear correlation between the independent and dependent variables must be real, not formal;

[0090] 3) The independent variables should have a certain degree of mutual exclusivity, that is, the correlation between the independent variables should not be higher than the correlation between the independent variables and the dependent variable;

[0091] 4) The independent variables should have complete statistical data and their predicted values ​​should be easy to determine.

[0092] In this embodiment, in step 4), the EH oil system operation comprehensive fault diagnosis model is the flow coefficient of the valve and the parameters of the process pipeline; the flow coefficient is input into the EH oil system operation comprehensive fault diagnosis model to obtain the predicted flow value of the valve.

[0093] In this embodiment, in step 5), determining whether the valve has internal leakage includes:

[0094] Get the current status of the valve group, predict the flow value, and predict the parameter value;

[0095] When the actual flow value is greater than the flow threshold, it is determined that the valve has internal leakage.

[0096] In this embodiment, in step 6), determining the location where the valve internal leakage occurs includes:

[0097] The current state of the regulating valve group is obtained; and the position of the internal leakage valve in the regulating valve group is determined according to the current state.

[0098] Generate early warning information, including:

[0099] The warning information is sent to the distributed control system so that the image display device of the distributed control system can be used to display the warning information and the broadcasting device of the distributed control system can be used to broadcast the warning information.

[0100] Example 2

[0101] like Figure 2 As shown in the figure, at 18:55, the oil level, tank temperature and main pipe oil pressure of the high-pressure fire-resistant oil tank were in a stable state, and then the oil level in the tank began to decrease, indicating that it was leaking. This model issued an early warning notification after 30 seconds, notified the relevant personnel of the fault condition in time, and went to the scene for processing.

[0102] Example 3

[0103] like Figure 3 As shown, the present invention provides a fault diagnosis system for internal leakage of a high-pressure fire-resistant oil system, which is characterized by comprising:

[0104] A parameter acquisition module obtains the actual flow value and related process parameter values ​​from a distributed control system, wherein the distributed control system is a control system used in the field of industrial automation, the actual flow value is the real value obtained from the DCS, and the related process parameter value is the real value obtained from the DCS, which is used to predict the target parameters of the valve;

[0105] The data summarization module is based on the expert experience and knowledge of more than 20 ultra-supercritical coal-fired units, using the commissioning results of steam turbine professionals during the capital construction commissioning period and the operating data during the commercial operation period to summarize and obtain the EH oil system expert knowledge base;

[0106] The model building module builds a comprehensive fault diagnosis model for the EH oil system operation based on the EH oil system expert knowledge base, including a diversity comparison model in the start-stop process under normal conditions, a steady-state and variable condition operation model during operation, and a data model for joint classification and identification of multiple faults in the start-operation-stop process under fault conditions;

[0107] The parameter processing module inputs the relevant process parameter values ​​obtained from the distributed control system into the EH oil system operation comprehensive fault diagnosis model, so as to process the relevant process parameters using the EH oil system operation comprehensive fault diagnosis model to obtain the predicted flow value of the valve, wherein the EH oil system operation comprehensive fault diagnosis model is trained by machine learning using multiple sets of training data;

[0108] A determination module determines that the valve has internal leakage when the difference between the predicted flow value and the actual flow value obtained from the decentralized control system is greater than a deviation threshold;

[0109] The early warning module generates early warning information after determining that the valve has internal leakage; wherein the early warning information includes: valve position information and internal leakage information.

[0110] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

[0111] In addition, it should be understood that although this specification is described in accordance with the implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation modes that can be understood by those skilled in the art. The above content is only to illustrate the technical idea of ​​the present invention, and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution according to the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A method for diagnosing internal leakage of a high-pressure fire-resistant oil system, characterized in that: include: 1) Obtaining the actual flow value and related process parameter values ​​from a distributed control system, wherein the distributed control system is a control system used in the field of industrial automation, the actual flow value is the real value obtained from the DCS, and the related process parameter value is the real value obtained from the DCS, which is used to predict the target parameters of the valve; 2) Based on the expert experience and knowledge of more than 20 ultra-supercritical coal-fired units, the debugging results of the steam turbine professional during the capital construction and commissioning period and the operating data during the commercial operation period are summarized and summarized to obtain the EH oil system expert knowledge base; 3) Based on the EH oil system expert knowledge base, a comprehensive fault diagnosis model for the EH oil system operation was established, including a diversity comparison model in the start-stop process under normal conditions, a steady-state and variable condition operation model during operation, and a data model for joint classification and identification of multiple faults in the start-operation-stop process under fault conditions; 4) Inputting the relevant process parameter values ​​obtained from the distributed control system into the EH oil system operation comprehensive fault diagnosis model, so as to process the relevant process parameters using the EH oil system operation comprehensive fault diagnosis model to obtain the predicted flow value of the valve, wherein the EH oil system operation comprehensive fault diagnosis model is trained by machine learning using multiple sets of training data; 5) When the difference between the predicted flow value and the actual flow value obtained from the distributed control system is greater than the deviation threshold, it is determined that the valve has internal leakage; 6) After determining that the valve has internal leakage, generate warning information; wherein the warning information includes: valve position information and internal leakage information.

2. A method for diagnosing internal leakage of a high-pressure fire-resistant oil system according to claim 1, characterized in that: In step 1), the relevant process parameter values ​​include particle size, water content, acid value, foam characteristics, resistivity, temperature, flow coefficient of valves, parameters of process pipelines and volume of reaction equipment in oil quality.

3. A fault diagnosis method for internal leakage of a high-pressure fire-resistant oil system according to claim 1, characterized in that: In step 2), based on the expert experience and knowledge of more than 20 ultra-supercritical coal-fired units, the commissioning results of the steam turbine professional during the capital construction commissioning period and the operating data during the commercial operation period were summarized and summarized to obtain the EH oil system expert knowledge base, including: Data cleaning Therefore, it is necessary to clean the original data and remove invalid data to ensure that all data is accepted by the computer. In order to improve the accuracy of the model and prevent downtime data and abnormal data from interfering with normal data, this part of the data is processed according to expert experience and reconstructed using Python programming to construct data for model training; Feature Normalization Eliminate the dimensional impact between indicators and perform data normalization to ensure the comparability between data indicators.

4. A method for diagnosing internal leakage in a high-pressure fire-resistant oil system according to claim 1, characterized in that: In step 3), based on the EH oil system expert knowledge base, a comprehensive fault diagnosis model for the EH oil system operation is established, including: BPNN model design A four-layer neural network is used to train the normal operation process of the EH oil system, where the number of input layer nodes is 5, the number of nodes in the first hidden layer is set to 26 after cross-validation and considering the computational efficiency, the number of nodes in the second hidden layer is 4, and the number of nodes in the output layer is 1 after the dynamic neuron activation layer, and the internal leakage is analyzed by analyzing the system operation characteristics; Optimization Algorithm The update formula used by the ADAM algorithm is as follows: m0=v0=0 Where m0 and v0 are the initialization parameter vector matrices, β1, β2, η, ε are the configuration parameters in the algorithm update process, and E and w are the error and weight in the algorithm update process; Evaluation methods and indicators Mean square error: the ratio of the sum of squares of the deviations between the observed value and the true value to the number of observations; The closer the MSE value is to 0, the more accurate the prediction model is in describing the experimental data.

5. A method for diagnosing internal leakage of a high-pressure fire-resistant oil system according to claim 1, characterized in that: In step 4), the EH oil system operation comprehensive fault diagnosis model is the flow coefficient of the valve and the parameters of the process pipeline; the flow coefficient is input into the EH oil system operation comprehensive fault diagnosis model to obtain the predicted flow value of the valve.

6. A method for diagnosing internal leakage in a high-pressure fire-resistant oil system according to claim 1, characterized in that: In step 5), it is determined that the valve has internal leakage, including: Get the current status of the valve group, predict the flow value, and predict the parameter value; When the actual flow value is greater than the flow threshold, it is determined that the valve has internal leakage.

7. A method for diagnosing internal leakage in a high-pressure fire-resistant oil system according to claim 1, characterized in that: In step 6), the location where the valve internal leakage occurs is determined, including: The current state of the regulating valve group is obtained; and the position of the internal leakage valve in the regulating valve group is determined according to the current state.

8. A method for diagnosing internal leakage in a high-pressure fire-resistant oil system according to claim 1, characterized in that: In step 6), early warning information is generated, including: The warning information is sent to the distributed control system so that the image display device of the distributed control system can be used to display the warning information and the broadcasting device of the distributed control system can be used to broadcast the warning information.

9. A fault diagnosis system for internal leakage of high-pressure fire-resistant oil system, characterized in that: include: A parameter acquisition module obtains the actual flow value and related process parameter values ​​from a distributed control system, wherein the distributed control system is a control system used in the field of industrial automation, the actual flow value is the real value obtained from the DCS, and the related process parameter value is the real value obtained from the DCS, which is used to predict the target parameters of the valve; The data summarization module is based on the expert experience and knowledge of more than 20 ultra-supercritical coal-fired units, using the commissioning results of steam turbine professionals during the capital construction commissioning period and the operating data during the commercial operation period to summarize and obtain the EH oil system expert knowledge base; The model building module builds a comprehensive fault diagnosis model for the EH oil system operation based on the EH oil system expert knowledge base, including a diversity comparison model in the start-stop process under normal conditions, a steady-state and variable condition operation model during operation, and a data model for joint classification and identification of multiple faults in the start-operation-stop process under fault conditions; The parameter processing module inputs the relevant process parameter values ​​obtained from the distributed control system into the EH oil system operation comprehensive fault diagnosis model, so as to process the relevant process parameters using the EH oil system operation comprehensive fault diagnosis model to obtain the predicted flow value of the valve, wherein the EH oil system operation comprehensive fault diagnosis model is trained by machine learning using multiple sets of training data; A determination module determines that the valve has internal leakage when the difference between the predicted flow value and the actual flow value obtained from the decentralized control system is greater than a deviation threshold; The early warning module generates early warning information after determining that the valve has internal leakage; wherein the early warning information includes: valve position information and internal leakage information.

10. A fault diagnosis system for internal leakage of a high-pressure fire-resistant oil system according to claim 9, characterized in that: In the parameter acquisition module, relevant process parameter values ​​include oil particle size, moisture, acid value, foam characteristics, resistivity, temperature, valve flow coefficient, process pipeline parameters and the volume of reaction equipment.

Images