A method, medium and system for monitoring health margin of key components of a converter valve
Through principal component analysis, feature extraction and health margin calculation of the monitoring data of UHV DC transmission converter valves is solved, and the problem of large data volume and difficult analysis is realized, and the global health status quantization and monitoring efficiency of key components of the converter valve are achieved. promote.
Patent Information
- Application Number
- CN202111222791.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-10-20
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2041-10-20
AI Technical Summary
In UHV DC transmission projects, it is difficult to achieve comprehensive monitoring of data monitoring of converter valves and their key components, making it difficult to accurately analyze and obtain effective health status indicators.
The monitoring data is extracted using principal component analysis method, and by judging whether the characteristic data of each attribute type reaches the fluctuation limit, the current health margin of abnormal attribute type is calculated, thereby giving the overall health status of the key components of the converter valve.
Compressed data through principal component analysis, solve the problem of large amounts of data and difficult to analyze, provide multi-dimensional health status analysis, and improve monitoring efficiency and reliability.
Smart Images

Figure CN114384348B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of converter valves, and in particular to a method, medium and system for monitoring the health margin of key components of a converter valve. Background Art
[0002] The UHV DC converter valve is one of the most important core equipment in DC transmission, and its operating status monitoring is of great significance to the safety and stability of the power system. Online monitoring of the operating status parameters of the UHV DC converter valve and its key components, and then establishing the UHV DC converter valve information status health index, can timely discover potential safety hazards of the equipment, effectively improve the reliability of the converter valve operation, save operation and maintenance costs, and has great research and engineering application value.
[0003] It is generally difficult to achieve comprehensive data monitoring of converter valves and their key components in UHV DC transmission projects. Some areas have completed multiple types of online data monitoring, but due to the huge amount of data and complex data features, it is generally difficult to accurately analyze and extract features. Therefore, the traditional method has the problem of being difficult to analyze a large amount of data features in a timely and accurate manner, and on the other hand, it is unable to provide effective indicators of the health status of the converter valve. Summary of the invention
[0004] The embodiments of the present invention provide a method, medium and system for monitoring the health margin of key components of a converter valve, so as to solve the problem that it is difficult to timely and accurately analyze whether the key components of the converter valve are healthy and provide effective health status indicators in the prior art.
[0005] The embodiment of the present invention discloses the following technical solution:
[0006] In a first aspect, a method for monitoring the health margin of a key component of a converter valve is provided, comprising:
[0007] Collect monitoring data of key components of the converter valve at preset intervals within a preset time period;
[0008] Classifying the monitoring data according to the attribute types of the monitoring data;
[0009] Extracting features from the monitoring data using principal component analysis to obtain feature data;
[0010] Determining whether the characteristic data of each attribute type reaches a fluctuation limit;
[0011] If the characteristic data of a certain attribute type reaches the fluctuation limit, the attribute type is determined as an abnormal attribute type;
[0012] The percentage of the difference between the current value in the monitoring data of each abnormal attribute type and the protection set value corresponding to each abnormal attribute type is calculated to obtain the current health margin of each abnormal attribute type.
[0013] In a second aspect, a computer-readable storage medium is provided, on which computer program instructions are stored; when the computer program instructions are executed by a processor, a method for monitoring the health margin of key components of a converter valve as described in an embodiment of the first aspect is implemented.
[0014] In a third aspect, a system for monitoring the health margin of key components of a converter valve is provided, comprising: a computer-readable storage medium as described in the embodiment of the second aspect.
[0015] In this way, the embodiment of the present invention compresses the acquired monitoring data by principal component analysis, thereby solving the problem that it is difficult to obtain feature data due to the large amount of data, and analyzes the health of key components of the converter valve from multiple dimensions by calculating the current health margin of abnormal attribute types and the overall health margin of all attribute types, and gives a global overall health status, quantifies the health of the converter valve, and improves the monitoring efficiency of the UHV DC transmission converter valve and the reliability of UHV DC transmission. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.
[0017] Figure 1 is a flow chart of a method for monitoring the health margin of key components of a converter valve according to Embodiment 1 of the present invention;
[0018] Figure 2 It is a flow chart of a method for monitoring the health margin of key components of a converter valve according to Example 2 of the present invention. DETAILED DESCRIPTION
[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0020] Example 1
[0021] Embodiment 1 of the present invention discloses a method for monitoring the health margin of key components of a converter valve. Figure 1 As shown, the method includes the following process:
[0022] Step S101: collecting monitoring data of key components of the converter valve at preset intervals within a preset time period.
[0023] The monitoring data is data collected in real time and is sorted in chronological order. The preset duration and the preset time can be selected based on experience and actual conditions. It should be understood that the last monitoring data among the multiple monitoring data collected within the preset duration is the current value of the monitoring data.
[0024] Step S102: Classify the monitoring data according to the attribute types of the monitoring data.
[0025] The attribute type of the collected monitoring data can be selected according to whether the type of abnormality needs to be monitored.
[0026] For example, in a specific embodiment of the present invention, the attribute types of the monitoring data include: converter valve assembly current, converter valve assembly voltage, converter valve assembly current change rate, converter valve assembly voltage change rate, converter valve equipment component pressure, converter valve valve cooling system flow, converter valve temperature, converter valve abnormal discharge data, thyristor level operating status data and converter valve transmission power.
[0027] The above-mentioned attribute types can be classified into three categories: converter valve monitoring data, thyristor monitoring data and valve cooling system monitoring data. Thus, the converter valve monitoring data includes: converter valve component current, converter valve component voltage, converter valve component current change rate, converter valve component voltage change rate, converter valve temperature, converter valve abnormal discharge data and converter valve transmission power; thyristor monitoring data includes: thyristor level operation status data; valve cooling system monitoring data includes: converter valve equipment component pressure and converter valve valve cooling system flow.
[0028] Step S103: extracting features from the monitoring data using principal component analysis to obtain feature data.
[0029] Specifically, this step includes the following process:
[0030] (1) Obtain the first matrix composed of monitoring data.
[0031] Among them, the elements of the first matrix are monitoring data, and the row vectors of the first matrix are sorted according to the order of acquisition time. The column vectors of the first matrix are sorted according to the order of the preset monitoring data types, and the sorting can be determined based on experience and actual conditions. For example, in one embodiment of the present invention, the sorting is as follows: converter valve component current, converter valve component voltage, converter valve component current change rate, converter valve component voltage change rate, converter valve equipment component pressure, converter valve valve cooling system flow, converter valve temperature, converter valve abnormal discharge data, thyristor stage operating status data and converter valve transmission power. The first matrix X can be expressed as [x1 x2 … x i …x D ], m represents the number of attribute types of monitoring data, and x represents the data vector of monitoring data of attribute type with order i.
[0032] (2) Calculate the product of the transposed matrix of the first matrix and the first matrix to obtain the second matrix.
[0033] The second matrix is: X T X=[x1 x2 … x m ] T [x1 x2 … x m ]. Therefore, the second matrix is an m×m square matrix.
[0034] (3) Calculate the eigenvalues of the second matrix, and sort the eigenvectors corresponding to the eigenvalues in descending order to obtain the third matrix.
[0035] The calculation is performed using an existing method for calculating matrix eigenvalues, which will not be described in detail here. Since the second matrix is a square matrix, the number of eigenvalues is m.
[0036] (4) Calculate the data fidelity of the eigenvalues after sorting them in descending order.
[0037] The calculation formula for data fidelity is:
[0038]
[0039] Among them, Fid represents data fidelity, λ i represents the eigenvalues of i in descending order, λ total represents the sum of all eigenvalues, that is D is the data dimension, that is, the number of columns of the matrix.
[0040] (5) Determine whether the data fidelity is not less than the data fidelity threshold.
[0041] Let γ represent the data fidelity threshold, and compare the relationship between Fid and γ.
[0042] (6) If the data fidelity is less than the data fidelity threshold, the data dimension is increased by one, and the step of calculating the data fidelity is returned to repeat the steps of calculating the data fidelity and judging until the data fidelity is not less than the data fidelity threshold.
[0043] The initial values of data fidelity and data dimension are both 0. If Fid < γ, then D + 1. The key categories are determined by gradually expanding the dimension.
[0044] (7) If the data fidelity is not less than the data fidelity threshold, a fourth matrix is obtained from the third matrix according to the current data dimension.
[0045] That is, the fourth matrix is a matrix with D columns. P represents the fourth matrix, P = [e1 e2 … e i …e D ],e i Denotes the eigenvalue λ i The corresponding feature vector.
[0046] (8) Calculate the product of the transposed matrix of the fourth matrix and the first matrix and then transpose it to obtain feature data.
[0047] Specifically, the calculation formula of the characteristic data is:
[0048] Y=[P T X] T =[[e1 e2 … e i … e D ] T X] T .
[0049] Among them, Y represents the feature data, and the feature data is the output matrix [y1 y2 … y i … y D ],y i A data vector representing the monitoring data of the attribute category of order i.
[0050] It should be understood that, since the collected monitoring data changes with time, step S4 also continuously analyzes new monitoring data, that is, the input monitoring data will change with time.
[0051] Step S104: Determine whether the characteristic data of each attribute type reaches the fluctuation limit.
[0052] Specifically, this step includes the following process:
[0053] (1) Calculate the standard deviation of feature data of a certain attribute type within a preset time range.
[0054] This calculation method is a prior art and will not be described in detail here. The preset time range can be selected according to the test and actual situation, for example, 100ms before the current value of the monitoring data. Generally, the preset time range can be the same as the aforementioned preset time length, so that the result is more accurate.
[0055] (2) When the standard deviation of the characteristic data of a certain attribute type exceeds the fluctuation threshold, it is determined that the characteristic data of the attribute type has reached the fluctuation limit.
[0056] The fluctuation threshold value may also be selected based on experience and actual conditions, such as 30%. If the fluctuation threshold value is exceeded, it is considered that the fluctuation limit is exceeded.
[0057] Step S105: If the characteristic data of a certain attribute type reaches the fluctuation limit, the attribute type is determined as an abnormal attribute type.
[0058] For example, if the characteristic data of the converter valve assembly current exceeds the fluctuation limit, the converter valve assembly current is of an abnormal attribute type.
[0059] Step S106: Calculate the percentage of the difference between the current value in the monitoring data of each abnormal attribute type and the protection set value corresponding to each abnormal attribute type to obtain the current health margin of each abnormal attribute type.
[0060] Specifically, the protection setting value can be preset according to experience and actual conditions. The current value is the last monitoring data among the multiple monitoring data collected within the preset time as mentioned above. The current health margin of each abnormal attribute type can be classified and output according to the three categories of converter valves, thyristors and valve cooling systems divided above.
[0061] Through the above steps, embodiment 1 of the present invention can determine the current health margin of the abnormal attribute type and output it to the user.
[0062] Example 2
[0063] Embodiment 2 of the present invention discloses a method for monitoring the health margin of key components of a converter valve. Figure 2 As shown, the method includes the following process:
[0064] Step S201: collecting monitoring data of key components of the converter valve at preset intervals within a preset time period.
[0065] Step S202: Classify the monitoring data according to the attribute types of the monitoring data.
[0066] Steps S201 to S202 are the same as the aforementioned steps S101 to S102 and will not be described in detail herein.
[0067] Step S203: Calculate the current health margin of each attribute type.
[0068] Specifically, the step includes:
[0069] Calculate the percentage of the difference between the current value of the monitoring data of each attribute type and the protection set value corresponding to each attribute type to obtain the current health margin of each attribute type.
[0070] Similarly, the protection setting value can be determined based on experience and actual conditions.
[0071] Step S204: Calculate the sum of the product of the current health margin of each attribute type and the corresponding health coefficient to obtain the overall health margin of the key components of the converter valve.
[0072] The calculation formula can be expressed as:
[0073] M safe =K safe1 ×S1+K safe2 ×S2+…+K safei ×S i +…K safen ×S m
[0074] Among them, M safe is the overall health margin, S i represents the current health margin corresponding to attribute type i, K safei Indicates the health factor of the attribute type.
[0075] For example, when the attribute types include: converter valve component current, converter valve component voltage, converter valve component current change rate, converter valve component voltage change rate, converter valve equipment component pressure, converter valve valve cooling system flow, converter valve temperature, converter valve abnormal discharge data, thyristor level operation status data and converter valve transmission power, then S1 to S m They correspond to the current health margin of the converter valve assembly current, the current health margin of the converter valve assembly voltage, the current health margin of the converter valve assembly current change rate, the current health margin of the converter valve assembly voltage change rate, the current health margin of the converter valve equipment component pressure, the current health margin of the converter valve valve cooling system flow, the current health margin of the converter valve temperature, the current health margin of the converter valve abnormal discharge data, the current health margin of the thyristor level operation status data and the current health margin of the converter valve transmission power; K safe1 To K safen They correspond respectively to the current health factor of the converter valve assembly, the voltage health factor of the converter valve assembly, the current change rate health factor of the converter valve assembly, the voltage change rate health factor of the converter valve assembly, the pressure health factor of the converter valve equipment assembly, the flow health factor of the converter valve cooling system, the temperature health factor of the converter valve, the abnormal discharge data health factor of the converter valve, the thyristor level operation status data health factor and the converter valve transmission power health factor.
[0076] Embodiment 2 of the present invention can determine the overall health margin of key components of the converter valve through the above steps and output it to the user.
[0077] Example 3
[0078] Embodiment 3 of the present invention discloses a computer-readable storage medium, on which computer program instructions are stored; when the computer program instructions are executed by a processor, the method for monitoring the health margin of key components of a converter valve described in any of the above embodiments 1 and 2 is implemented.
[0079] Example 4
[0080] Embodiment 4 of the present invention discloses a system for monitoring the health margin of key components of a converter valve, including: a computer-readable storage medium as described in the above embodiment 3. The monitoring system can be executed on various devices such as computers, servers, and microprocessors.
[0081] In summary, the embodiments of the present invention compress the acquired monitoring data by principal component analysis, thereby solving the problem that it is difficult to obtain feature data due to a large amount of data, and by calculating the current health margin of abnormal attribute types and the overall health margin of all attribute types, the health of key components of the converter valve is analyzed from multiple dimensions, and the global overall health status is given, the health of the converter valve is quantified, and the monitoring efficiency of the UHV DC transmission converter valve and the reliability of UHV DC transmission are improved.
[0082] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. A method for monitoring the health margin of key components of a converter valve, characterized in that: include: Collect monitoring data of key components of the converter valve at preset intervals within a preset time period; Classifying the monitoring data according to the attribute types of the monitoring data; Extracting features from the monitoring data using principal component analysis to obtain feature data; Determining whether the characteristic data of each attribute type reaches a fluctuation limit; If the characteristic data of a certain attribute type reaches the fluctuation limit, the attribute type is determined as an abnormal attribute type; Calculate the percentage of the difference between the current value in the monitoring data of each abnormal attribute type and the protection set value corresponding to each abnormal attribute type to obtain the current health margin of each abnormal attribute type; The step of extracting features from the monitoring data by principal component analysis to obtain feature data comprises: Acquire a first matrix composed of monitoring data, wherein the elements of the first matrix are monitoring data, the row vectors of the first matrix are sorted according to the order of acquisition time, and the column vectors of the first matrix are sorted according to the order of the attribute types of the monitoring data preset; Calculate the product of the transposed matrix of the first matrix and the first matrix to obtain a second matrix; Calculate the eigenvalues of the second matrix, and sort the eigenvectors corresponding to the eigenvalues in descending order of the eigenvalues to obtain a third matrix; Calculating the data fidelity of the eigenvalues after sorting the eigenvalues from large to small; Determining whether the data fidelity is not less than a data fidelity threshold; If the data fidelity is less than the data fidelity threshold, the data dimension is increased by one, and the step of calculating the data fidelity is returned to repeat the steps of calculating the data fidelity and judging until the data fidelity is not less than the data fidelity threshold, wherein the initial values of the data fidelity and the data dimension are both 0; If the data fidelity is not less than the data fidelity threshold, a fourth matrix is obtained by truncating the third matrix according to the current data dimension; Calculate the product of the transposed matrix of the fourth matrix and the first matrix and then transpose it to obtain the feature data; The calculation formula of the data fidelity is: ; in, Fid represents the data fidelity, λ i The order of sorting in descending order is i The characteristic value of λ total represents the sum of all the eigenvalues, D is the data dimension; The calculation formula of the characteristic data is: ; in, Y Represents the feature data, which is the output matrix , y i The order is i The data vector of the monitoring data of the attribute type, P represents the fourth matrix, , e i Represents eigenvalue λ i The corresponding eigenvector, X Denotes the first matrix, X= , m Indicates the number of attribute types of the monitoring data.
2. The method for monitoring the health margin of key components of a converter valve according to claim 1, characterized in that: The step of judging whether the characteristic data of each attribute type reaches the fluctuation limit comprises: Calculate the standard deviation of the characteristic data of a certain attribute type within a preset time range; When the standard deviation of the characteristic data of a certain attribute type exceeds the fluctuation threshold, it is determined that the characteristic data of the attribute type reaches the fluctuation limit.
3. The method for monitoring the health margin of key components of a converter valve according to claim 1, characterized in that: After the step of classifying the monitoring data according to the attribute types of the monitoring data, the method further comprises: calculating a current health margin for each of said attribute categories; The sum of the products of the current health margin of each attribute type and the corresponding health coefficient is calculated to obtain the overall health margin of the key components of the converter valve.
4. The method for monitoring the health margin of key components of a converter valve according to claim 3, characterized in that: The step of calculating the current health margin of each attribute type comprises: The percentage of the difference between the current value of the monitoring data of each attribute type and the protection set value corresponding to each attribute type is calculated to obtain the current health margin of each attribute type.
5. The method for monitoring the health margin of key components of a converter valve according to claim 1, characterized in that: The attribute types include: converter valve assembly current, converter valve assembly voltage, converter valve assembly current change rate, converter valve assembly voltage change rate, converter valve equipment component pressure, converter valve valve cooling system flow, converter valve temperature, converter valve abnormal discharge data, thyristor level operating status data and converter valve transmission power.
6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions; when the computer program instructions are executed by the processor, the method for monitoring the health margin of the key components of the converter valve according to any one of claims 1 to 5 is implemented.
7. A monitoring system for the health margin of key components of a converter valve, characterized in that: include: The computer readable storage medium of claim 6.
