Diagnostic evaluation method, system and equipment for expansion displacement of valve-side dry-type casing guide rod and medium

Abstract

The invention discloses a valve side dry type sleeve guide rod expansion displacement diagnosis and evaluation method, system and device and a medium, and relates to the technical field of power equipment state monitoring. The evaluation method comprises the following steps of data acquisition and preprocessing, data set division, displacement decomposition model construction, guide rod expansion displacement extraction, state evaluation and early warning and model optimization and updating. The evaluation system comprises a multi-source sensing unit, an intelligent analysis unit and a decision support unit, all the units achieve efficient cooperation through standardized data interfaces, and stable operation of the system under complex working conditions is ensured. The method solves the problem that a traditional method cannot distinguish displacement sources, online monitoring and accurate evaluation of the mechanical state of the bushing are achieved, potential faults such as abnormal expansion of the guide rod can be found in time, and the method is suitable for health state monitoring and fault early warning of the converter transformer bushing in a high-voltage direct-current power transmission system and has good application prospects. And a new technical means is provided for safe operation of power equipment.

Description

Technical Field

[0001] This invention relates to the field of power equipment condition monitoring technology, and in particular to a method, system, equipment and medium for diagnosing and evaluating the expansion displacement of a valve-side dry bushing guide rod. Background Technology

[0002] The valve-side dry bushing of the converter transformer is a core component of the high-voltage direct current transmission system, and its operating status directly affects the safety and stability of the entire power grid. During long-term operation, the bushing conductor is prone to cumulative deformation due to complex mechanical loads such as alternating electromagnetic forces and thermal expansion stress. In severe cases, this can lead to mechanical jamming, poor contact, or even equipment damage. Traditional monitoring methods mainly rely on periodic power outages for detection or indirect assessment using single electrical parameters such as temperature and partial discharge. However, these methods cannot directly reflect the mechanical deformation state of the conductor and suffer from problems such as long detection cycles and insufficient sensitivity. In recent years, although some research has attempted to use non-contact monitoring technologies such as vibration analysis or infrared thermometry, these methods still have significant drawbacks in practical applications, such as high false alarm rates and insufficient reliability, because they cannot effectively distinguish the sources of various mechanical displacements.

[0003] Existing displacement monitoring technologies mainly face three key technical bottlenecks:

[0004] First, in terms of measurement methods, conventional contact displacement sensors are limited in installation location and it is difficult to directly monitor the displacement of the guide rod, while non-contact measurement is easily affected by environmental factors.

[0005] Secondly, in terms of signal processing, the sleeve displacement signal contains multiple interference components such as core vibration (50-100Hz), mechanical resonance (>500Hz) and temperature deformation, which are difficult to effectively separate using traditional filtering algorithms;

[0006] Finally, in the condition assessment stage, the lack of a quantitative analysis model for the expansion characteristics of the guide rod makes it difficult to provide an accurate basis for operation and maintenance decisions.

[0007] While there are research reports internationally on advanced sensing technologies such as fiber optic gratings or laser vibration measurement, these solutions are often costly and have stringent requirements for installation environment and operation and maintenance, making them difficult to promote and apply on a large scale in the power industry. With the rapid development of ultra-high voltage direct current (UHVDC) projects, the mechanical stress problems faced by converter transformer bushings are becoming increasingly prominent, and the industry's demand for precise monitoring technology is becoming more urgent. Especially against the backdrop of large-scale grid integration of new energy sources, the increased fluctuations in grid load lead to bushings experiencing more complex mechanical-thermal coupling stresses, rendering traditional monitoring methods unable to meet the accuracy and real-time requirements of modern power systems for equipment status sensing.

[0008] Therefore, it is necessary to design a diagnostic and evaluation method, system, equipment, and medium for the expansion displacement of the valve-side dry bushing guide rod to solve the above problems. Summary of the Invention

[0009] The technical problem to be solved by this invention is to overcome the shortcomings of existing technologies and design a method and system for diagnosing and evaluating the expansion displacement of the guide rod in a valve-side dry bushing. This invention provides a method for diagnosing the expansion displacement of the guide rod in a converter transformer valve-side dry bushing based on random forest. The method includes installing a displacement sensor on the bushing end screen to acquire overall displacement data, while simultaneously collecting parameters such as vibration and temperature; constructing a displacement decomposition model using a random forest algorithm to separate core vibration and its own vibration interference, and extracting the true expansion displacement of the guide rod; establishing a state evaluation system based on the expansion displacement characteristics to achieve graded early warning of the bushing's mechanical state; and accurately measuring the guide rod's expansion displacement by installing a displacement sensor on the bushing end screen and combining vibration and temperature parameters. This innovative method, which uses line voltage and bus current, accurately extracts the true expansion displacement of the bushing rod. A displacement decomposition model is constructed using a random forest regression algorithm to effectively separate interference components such as core vibration, mechanical vibration, and temperature deformation. This accurately extracts the expansion displacement characteristics of the bushing rod, solving the problem of traditional methods failing to distinguish the source of displacement. It achieves precise measurement of the bushing rod's expansion displacement, providing a new technical means for bushing mechanical condition assessment. This effectively prevents equipment failures caused by abnormal bushing rod expansion, enabling online monitoring and accurate assessment of the bushing's mechanical condition. It can promptly detect potential faults such as abnormal expansion of the bushing rod, and is applicable to health status monitoring and fault early warning of converter transformer bushings in high-voltage direct current transmission systems, providing a new technical means for the safe operation of power equipment.

[0010] The solution adopted by this invention to solve its technical problem is as follows:

[0011] A method for diagnosing and evaluating the expansion displacement of the guide rod in a valve-side dry bushing.

[0012] Its features are,

[0013] Includes the following steps,

[0014] Step S1: Data acquisition and preprocessing,

[0015] Displacement sensors are installed on the end screen of the bushing to monitor the overall displacement of the bushing in real time. The temperature distribution of the bushing is monitored by an infrared camera. Vibration sensor data, temperature data, bus current and bus voltage are collected in real time, and the above data are processed for outliers and preprocessed.

[0016] Step S2: Dataset partitioning,

[0017] Displacement data monitored by displacement sensors is used as the target variable, and vibration data, temperature data, bus current and bus voltage are used as feature variables. The preprocessed data is divided into training set and test set in a 7:3 ratio.

[0018] Step S3: Construction of the displacement decomposition model.

[0019] A displacement decomposition model was established using the random forest regression algorithm, and the contribution of each disturbance factor to the total displacement was determined through feature importance analysis.

[0020] Step S4: Extraction of guide rod expansion displacement.

[0021] The data monitored in real time in step S1 is transmitted to the displacement decomposition model in step S3, and the decomposed guide rod real expansion displacement component is output through the displacement decomposition model.

[0022] Step S5: Status assessment and early warning.

[0023] Based on the historical trend of the guide rod expansion displacement, a threshold is set to establish a multi-level state assessment mechanism:

[0024] Step S6: Model optimization and update.

[0025] Regularly collect new data to revise the training model and dynamically adjust the warning threshold.

[0026] As a preferred embodiment of the present invention

[0027] Step S3 includes the following steps:

[0028] Step S3.1: Use random sampling with replacement to generate training subsets of multiple decision trees from the training set;

[0029] Step S3.2: Randomly select a subset of features for each decision tree and split the nodes based on the information gain ratio;

[0030] Step S3.3: Generate a decision tree using a depth-first approach until the preset maximum depth is reached.

[0031] As a preferred embodiment of the present invention

[0032] Step S4 includes the following steps:

[0033] Step S4.1: Simultaneously acquire the vibration sensor data, temperature data, bus current and bus voltage from step S1, complete signal conditioning, and input the conditioned data into the trained displacement decomposition model;

[0034] Step S4.2: Predict each component using the displacement decomposition model and calculate the guide rod displacement;

[0035] Step S4.3: Derive the displacement decomposition formula for the guide rod.

[0036] As a preferred embodiment of the present invention

[0037] The multi-level state assessment mechanism established in step S5 includes:

[0038] Level I: Displacement is below threshold L1, normal operation;

[0039] Level II: Displacement exceeds threshold L2;

[0040] As a preferred embodiment of the present invention

[0041] The information gain ratio in step S3.2 is calculated using information entropy, and the expression for information entropy is:

[0042]

[0043] in, Represents the dataset Information entropy Indicates the data set The proportion of samples, This represents the number of values ​​a random variable can take.

[0044] The information gain is calculated by subtracting the weighted average information entropy of the child nodes from the information entropy of the parent node. The calculation formula is as follows:

[0045]

[0046] Representation of features For dataset Information gain Representation of features The set of possible values, Represents the dataset The total number of samples in the middle Representation of features The value is The number of sample subsets;

[0047] The expression for the information gain ratio is as follows:

[0048]

[0049] in, , This indicates the degree to which feature A splits the dataset S, and its calculation formula is as follows:

[0050] .

[0051] As a preferred embodiment of the present invention

[0052] The displacement decomposition model optimization in step S6 includes the optimization of model parameters. The optimization of model parameters includes: adjusting the number of base learners in the random forest, the minimum number of samples for node splitting, the minimum number of samples for leaf nodes, and the feature subset sampling ratio, and using the Bayesian optimization algorithm to automatically search for the hyperparameter combination that optimizes the cross-validation accuracy.

[0053] As a preferred embodiment of the present invention

[0054] In step S3.3, the number of decision trees is 100-500, and the maximum depth of a single tree is set to 5-15 layers.

[0055] A diagnostic and evaluation system for the expansion displacement of a valve-side dry bushing guide rod.

[0056] Its features include:

[0057] The multi-source sensing unit integrates displacement, vibration, and temperature sensors to achieve synchronous acquisition of mechanical state parameters;

[0058] The intelligent analysis unit, equipped with an optimized random forest regression algorithm, processes sensor data in real time and extracts the true displacement components of the guide rod.

[0059] The decision support unit automatically generates a status assessment report based on displacement characteristic analysis and provides hierarchical early warning and operation and maintenance guidance.

[0060] A computer device, comprising:

[0061] Memory and processor

[0062] The memory and the processor are interconnected.

[0063] The memory stores computer instructions.

[0064] The processor executes the computer instructions to perform a diagnostic and evaluation method for the expansion displacement of a valve-side dry bushing guide rod.

[0065] A computer-readable storage medium,

[0066] The computer-readable storage medium stores computer instructions.

[0067] The computer instructions are used to cause the computer to execute the valve-side dry bushing guide rod expansion displacement diagnosis and evaluation method as described in claim 7.

[0068] Compared with the prior art, the present invention has the following beneficial effects:

[0069] This invention provides a method for diagnosing the expansion displacement of the guide rod in a converter transformer valve-side dry bushing based on random forest. The method includes installing a displacement sensor on the end screen of the bushing to acquire overall displacement data, while simultaneously collecting parameters such as vibration and temperature; constructing a displacement decomposition model using the random forest algorithm to separate core vibration and its own vibration interference, and extracting the true expansion displacement of the guide rod; and establishing a state assessment system based on the expansion displacement characteristics to achieve graded early warning of the bushing's mechanical state.

[0070] By installing a displacement sensor on the end screen of the bushing and combining it with parameters such as vibration and temperature, the expansion displacement of the guide rod can be accurately measured. By integrating displacement, vibration, temperature, bus voltage, and bus current, the accurate extraction of the true expansion displacement of the guide rod is innovatively achieved. A displacement decomposition model is constructed using a random forest regression algorithm to effectively separate interference components such as core vibration, mechanical vibration, and temperature deformation, and accurately extract the expansion displacement characteristics of the guide rod.

[0071] This invention solves the problem of traditional methods being unable to distinguish the source of displacement, achieving precise measurement of guide rod expansion displacement. It provides a new technical means for bushing mechanical condition assessment and can effectively prevent equipment failures caused by abnormal guide rod expansion. It enables online monitoring and precise assessment of bushing mechanical condition, allowing for timely detection of potential faults such as abnormal guide rod expansion. It is applicable to health status monitoring and fault early warning of converter transformer bushings in high-voltage direct current transmission systems, providing a new technical means for the safe operation of power equipment. It has the following advantages:

[0072] (1) By inputting parameters such as the temperature and mechanical vibration of the bushing into the random forest model, the expansion displacement of the guide rod can be separated, so as to realize early warning and accurate diagnosis of bushing guide rod faults and ensure the safe and stable operation of the power system.

[0073] (2) The displacement decomposition model created by the method of the present invention adopts a displacement decomposition mechanism driven by feature importance, which effectively distinguishes various interference components and effectively solves the technical problem that traditional methods are difficult to separate various interference components.

[0074] (3) The displacement decomposition model created by the method of the present invention supports online real-time calculation and dynamic early warning, and can accurately trigger three-level state assessment such as normal monitoring, abnormal early warning and emergency shutdown, providing a reliable basis for the mechanical condition diagnosis of the casing guide rod;

[0075] (4) The method of the present invention maintains the accuracy of the model through an adaptive learning algorithm, and can be extended to the mechanical vibration status monitoring of other power equipment, which has significant technical innovation and engineering practical value. Attached Figure Description

[0076] Figure 1 This is a flowchart of a valve-side dry bushing guide rod expansion displacement diagnosis and evaluation method proposed in this invention;

[0077] Figure 2 This is a flowchart of the overall model training process in the valve-side dry bushing guide rod expansion displacement diagnosis and evaluation method proposed in this invention.

[0078] Figure 3 This is the application process of the random forest model in the valve-side dry bushing guide rod expansion displacement diagnosis and evaluation method proposed in this invention;

[0079] Figure 4 This is a schematic diagram of the framework of a valve-side dry bushing guide rod expansion displacement diagnosis and evaluation system proposed in this invention.

[0080] Figure 5 This is a flowchart illustrating the process of constructing a displacement decomposition model in the valve-side dry bushing guide rod expansion displacement diagnosis and evaluation method proposed in this invention.

[0081] Figure 6 This is a flowchart illustrating the process of extracting the expansion displacement of the guide rod in the valve-side dry bushing guide rod expansion displacement diagnosis and evaluation method proposed in this invention.

[0082] Explanation of reference numerals in the attached figures:

[0083] 101. Multi-source sensing unit,

[0084] 102. Intelligent Analysis Unit

[0085] 103. Decision Support Unit. Detailed Implementation

[0086] The specific embodiments of the present invention are described below with reference to the accompanying drawings and examples. Although specific embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0087] It should be noted that the structures, colors, proportions, sizes, etc. shown in the accompanying drawings are only used to complement the content disclosed in the specification, so that those skilled in the art can understand and read them, and are not intended to limit the conditions under which the present invention can be implemented. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0088] Furthermore, certain terms are used in the specification to refer to specific components. Those skilled in the art will understand that different terms may be used to refer to the same component. This specification does not distinguish components based on differences in terminology, but rather on differences in their functions. The terms "comprising" or "including" used throughout the specification and claims are open-ended and should be interpreted as "comprising but not limited to." The following descriptions are preferred embodiments for carrying out the invention; however, these descriptions are for the purpose of understanding the general principles of the specification and are not intended to limit the scope of the invention. The scope of protection of this invention is determined by the appended claims.

[0089] To facilitate understanding of the embodiments of the present invention, further explanations and descriptions will be provided below with reference to the accompanying drawings and specific embodiments. The accompanying drawings do not constitute a limitation on the embodiments of the present invention.

[0090] In one embodiment, such as Figure 1 As shown, the method for diagnosing and evaluating the expansion displacement of the valve-side dry bushing guide rod includes the following steps:

[0091] Step S1: Data acquisition and preprocessing;

[0092] Step S2: Dataset partitioning;

[0093] Step S3: Construction of displacement decomposition model;

[0094] Step S4: Extraction of guide rod expansion displacement.

[0095] Step S5: Status assessment and early warning;

[0096] Step S6: Model optimization and update.

[0097] This invention collects relevant parameters of the guide rod's operation, establishes a displacement decomposition model, extracts the guide rod's expansion displacement, and establishes a condition assessment system based on the expansion displacement characteristics to achieve graded early warning of the casing's mechanical condition. It solves the problem of traditional methods being unable to distinguish the source of displacement, achieves accurate measurement of the guide rod's expansion displacement, provides a new technical means for casing mechanical condition assessment, and can effectively prevent equipment failures caused by abnormal guide rod expansion.

[0098] The displacement decomposition model is constructed using the random forest algorithm. The training and application process of the random forest model is as follows: Figure 2 and Figure 3 As shown, it includes the following steps:

[0099] Step S1: Data acquisition and preprocessing. A displacement sensor is installed on the end screen of the bushing to monitor the overall displacement of the bushing in real time. An infrared camera is used to monitor the overall temperature distribution of the bushing. Vibration sensor data, temperature data, bus current and bus voltage are collected in real time. The above data are then processed for outliers and preprocessed.

[0100] Vibration data is monitored directly using existing vibration sensors, while temperature data is monitored using an infrared camera to check the temperature of the bushing itself.

[0101] The specific handling of outliers is as follows: In the host computer, data that deviates significantly from the normal value is removed based on the 3σ principle.

[0102] Data preprocessing includes data cleaning, such as outlier removal and missing value imputation, as well as data transformation, such as normalization.

[0103] Step S2: Data set partitioning. The displacement data monitored by the displacement sensor is used as the target variable, and the vibration data, temperature data, bus current and bus voltage are used as feature variables to partition the dataset.

[0104] The specific partitioning process involves dividing the preprocessed data into a training set and a test set in a 7:3 ratio.

[0105] Step S3: Construction of the displacement decomposition model. A displacement decomposition model is established using the random forest regression algorithm. The RMSE of the guide rod expansion displacement output by the random forest model is used as the training objective. The contribution of each interference factor mentioned in step S1 to the total displacement is determined by feature importance analysis (an indicator used in machine learning to quantify the contribution of input features to the model's prediction results. It reflects the "influence" of each feature in the model's decision-making process—the higher the importance, the greater the influence of the feature on the model's output).

[0106] Root Mean Square Error (RMSE) is a commonly used statistical indicator to measure the difference between predicted and actual values. It reflects the accuracy of the prediction result; the smaller the value, the better the prediction effect. In this embodiment, RMSE is set to < 0.1 mm, and its calculation method is as follows:

[0107]

[0108] Where Yi is the actual guide rod displacement, and f(xi) is the guide rod displacement value output by the model.

[0109] Step S4: Extraction of guide rod expansion displacement. The data monitored in real time in step S1 is transmitted to the displacement decomposition model in step S3. The decomposed guide rod real expansion displacement components are output through the displacement decomposition model.

[0110] Step S5: Status assessment and early warning. Set thresholds based on the historical trend of guide rod expansion displacement and establish a multi-level status assessment mechanism.

[0111] Step S6: Model optimization and update. Regularly collect new data to correct the training model and dynamically adjust the warning threshold.

[0112] In addition, based on step S6, a continuous optimization scheme is adopted to dynamically update the training set according to the newly collected data, and the model is retrained every quarter to maintain prediction accuracy.

[0113] Overall displacement data is acquired by installing displacement sensors, while vibration, temperature, and other parameters are also collected. A displacement decomposition model is constructed using a random forest algorithm to separate core vibration from its own vibration interference and extract the true expansion displacement of the guide rod. Based on the expansion displacement characteristics, a state assessment system is established to achieve graded early warning of the bushing's mechanical condition, enabling online monitoring and accurate assessment of the bushing's mechanical condition. This system can promptly detect potential faults such as abnormal expansion of the guide rod and is suitable for health status monitoring and fault early warning of converter transformer bushings in high-voltage direct current transmission systems, providing a new technical means for the safe operation of power equipment.

[0114] like Figure 5 As shown, in another embodiment, step S3 includes the following steps:

[0115] Step S3.1: Use random sampling with replacement to generate training subsets of multiple decision trees from the training set;

[0116] Step S3.2: Randomly select a subset of features for each decision tree and split the nodes based on the information gain ratio;

[0117] Step S3.3: Generate a decision tree using a depth-first approach until the preset maximum depth is reached.

[0118] Specifically, the random forest regression algorithm displacement decomposition model is constructed as follows: Bootstrap sampling is used to extract multiple sample subsets with replacement from the training dataset, and each subset is used to construct an independent decision tree.

[0119] In the generation of each decision tree, a subset of features is first randomly selected from all features as candidate splitting attributes. Then, the optimal splitting feature and split point are selected based on the information gain ratio criterion. The growth of the decision tree adopts a depth-first strategy, recursively splitting nodes until one of the following termination conditions is met:

[0120] (1) Reach the preset maximum tree depth;

[0121] (2) The number of node samples is less than the set threshold;

[0122] (3) The information gain ratio is lower than the critical value.

[0123] like Figure 6 As shown, in another embodiment, step S4 includes the following steps:

[0124] Step S4.1: Simultaneously acquire the vibration sensor data, temperature data, bus current and bus voltage from step S1, complete signal conditioning, and input the conditioned data into the trained displacement decomposition model;

[0125] Step S4.2: Predict each component using the displacement decomposition model and calculate the guide rod displacement;

[0126] Step S4.3: Derive the displacement decomposition formula for the guide rod.

[0127] In another embodiment, the multi-level state evaluation mechanism established in step S5 includes:

[0128] Level I: Displacement is below threshold L1, normal operation;

[0129] Level II: Displacement exceeds threshold L2, requiring close monitoring and planned maintenance.

[0130] Level III: Displacement exceeds threshold L3, requiring mandatory maintenance and repair.

[0131] Among them, the thresholds L1, L2 and L3 used for displacement judgment are all obtained from historical operation data.

[0132] In another embodiment, the information gain ratio in step S3.2 is calculated using information entropy, which is a measure of the uncertainty or disorder of data. In decision tree algorithms, information entropy is typically used to measure the disorder of a dataset, and its expression is:

[0133]

[0134] in, Represents the dataset Information entropy Indicates the data set The proportion of samples, This represents the number of values ​​a random variable can take.

[0135] Information gain measures the improvement in purity that a feature brings to a classification task. In decision tree algorithms, information gain is used to select the best feature for node splitting. Information gain is calculated by subtracting the weighted average information entropy of the child nodes from the information entropy of the parent node. The formula is as follows:

[0136]

[0137] Representation of features For dataset Information gain Representation of features The set of possible values, Represents the dataset The total number of samples in the middle Representation of features The value is The number of sample subsets;

[0138] Information gain ratio (IGR) is a correction to information gain after considering the impact of the number of feature values. It can avoid over-reliance on features with a large number of values. The expression for the information gain ratio is as follows:

[0139]

[0140] in, The higher the value, the better the splitting effect. When the information gain ratio of a feature reaches its maximum value, it indicates that the feature has the greatest value for splitting nodes and is the optimal splitting feature; Split Information is used to measure the feature's value. The degree of splitting, in the formula using This indicates the degree to which the value of feature A splits the dataset S. The smaller the value, the more effectively feature A can divide the dataset S into smaller subsets. Therefore, the smaller the value, the greater the contribution of feature A to classification. The calculation formula is as follows:

[0141] .

[0142] In another embodiment, the displacement decomposition model optimization in step S6 includes the optimization of model parameters. The optimization of model parameters includes: adjusting the number of base learners in the random forest, the minimum number of samples for node splitting, the minimum number of samples for leaf nodes, and the feature subset sampling ratio, and using the Bayesian optimization algorithm to automatically search for the hyperparameter combination that optimizes the cross-validation accuracy.

[0143] In another embodiment, the number of decision trees in step S3.3 is 100-500, and the maximum depth of a single tree is set to 5-15 layers. Too many decision trees will lead to insufficient computing power, while too few will lead to an inaccurate model. For a single tree, too deep a tree will lead to overfitting, while too few will lead to poor generalization ability. The above data were calculated based on multiple experiments.

[0144] like Figure 4As shown, a valve-side dry bushing guide rod expansion displacement diagnostic and evaluation system includes:

[0145] The multi-source sensing unit integrates displacement, vibration, and temperature sensors to achieve synchronous acquisition of mechanical state parameters. After data acquisition, the data is sent to the intelligent analysis unit for processing and analysis.

[0146] The intelligent analysis unit, equipped with an optimized random forest regression algorithm, processes sensor data in real time and extracts the true displacement components of the guide rod. After analyzing the data, it sends the data to the decision support unit for evaluation.

[0147] The decision support unit automatically generates a status assessment report based on displacement characteristic analysis and provides hierarchical early warning and operation and maintenance guidance.

[0148] The aforementioned units achieve efficient collaboration through standardized data interfaces, ensuring the stable operation of the system under complex working conditions.

[0149] Further functional descriptions of the above-mentioned units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0150] A computer device having the above-mentioned features Figure 4 The valve-side dry casing guide rod expansion displacement diagnosis and evaluation system shown includes a memory and a processor, which are communicatively connected. The memory stores computer instructions, and the processor executes these computer instructions to perform the aforementioned valve-side dry casing guide rod expansion displacement diagnosis and evaluation method.

[0151] A computer-readable storage medium in which the methods described above according to embodiments of the present invention can be implemented in hardware, firmware, or implemented as recordable on a storage medium, or implemented as computer code originally stored on a remote storage medium or a non-transitory machine-readable storage medium and to be stored on a local storage medium after being downloaded over a network, so that the methods described herein can be processed by such software stored on a storage medium using a general-purpose computer, a special-purpose processor, or programmable or special-purpose hardware.

[0152] The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; furthermore, the storage medium can also include combinations of the above types of memory. It is understood that a computer, processor, microprocessor controller, or programmable hardware includes storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0153] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0154] The preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. The specific embodiments described above are merely illustrative and instructive, and not restrictive. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

[0155] Many other changes and modifications can be made without departing from the concept and scope of this invention. It should be understood that this invention is not limited to the specific embodiments, and the scope of this invention is defined by the appended claims.

Claims

1. A method for diagnosing and evaluating the expansion displacement of the guide rod in a valve-side dry bushing. Its features are, Includes the following steps, Step S1: Data acquisition and preprocessing, Displacement sensors are installed on the end screen of the bushing to monitor the overall displacement of the bushing in real time. The temperature distribution of the bushing is monitored by an infrared camera. Vibration sensor data, temperature data, bus current and bus voltage are collected in real time, and the above data are processed for outliers and preprocessed. Step S2: Dataset partitioning, Displacement data monitored by displacement sensors are used as target variables, and vibration data, temperature data, bus current and bus voltage are used as feature variables. The preprocessed data is divided into training set and test set in a 7:3 ratio. Step S3: Construction of the displacement decomposition model. A displacement decomposition model was established using the random forest regression algorithm, and the contribution of each disturbance factor to the total displacement was determined through feature importance analysis. Step S4: Extraction of guide rod expansion displacement. The data monitored in real time in step S1 is transmitted to the displacement decomposition model in step S3, and the decomposed guide rod real expansion displacement component is output through the displacement decomposition model. Step S5: Status assessment and early warning. Based on the historical trend of the guide rod expansion displacement, a threshold is set to establish a multi-level state assessment mechanism: Step S6: Model optimization and update. Regularly collect new data to revise the training model and dynamically adjust the warning threshold.

2. The method for diagnosing and evaluating the expansion displacement of the valve-side dry bushing guide rod as described in claim 1. Its features are, Step S3 includes the following steps: Step S3.1: Use random sampling with replacement to generate training subsets of multiple decision trees from the training set; Step S3.2: Randomly select a subset of features for each decision tree and split the nodes based on the information gain ratio; Step S3.3: Generate a decision tree using a depth-first approach until the preset maximum depth is reached.

3. The valve-side dry bushing guide rod expansion displacement diagnosis and evaluation method as described in claim 2. Its features are, Step S4 includes the following steps: Step S4.1: Simultaneously acquire the vibration sensor data, temperature data, bus current and bus voltage from step S1, complete signal conditioning, and input the conditioned data into the trained displacement decomposition model; Step S4.2: Predict each component using the displacement decomposition model and calculate the guide rod displacement; Step S4.3: Derive the displacement decomposition formula for the guide rod.

4. The method for diagnosing and evaluating the expansion displacement of the valve-side dry bushing guide rod as described in claim 3. Its features are, The multi-level state assessment mechanism established in step S5 includes: Level I: Displacement is below threshold L1, normal operation; Level II: Displacement exceeds threshold L2; Level III: Displacement exceeds threshold L3.

5. The method for diagnosing and evaluating the expansion displacement of the valve-side dry bushing guide rod as described in claim 4. Its features are, The information gain ratio in step S3.2 is calculated using information entropy, and the expression for information entropy is: ， in, Represents the dataset Information entropy Indicates the data set The sample proportion, This represents the number of values ​​a random variable can take. The information gain is calculated by subtracting the weighted average information entropy of the child nodes from the information entropy of the parent node. The calculation formula is as follows: ， Representation of features For dataset Information gain Representation of features The set of possible values, Represents the dataset The total number of samples in the middle Representation of features The value is The number of sample subsets; The expression for the information gain ratio is as follows: ， in, , This indicates the degree to which feature A splits the dataset S, and its calculation formula is as follows: 。 6. The method for diagnosing and evaluating the expansion displacement of the valve-side dry bushing guide rod as described in claim 5. Its features are, The displacement decomposition model optimization in step S6 includes the optimization of model parameters. The optimization of model parameters includes: adjusting the number of base learners in the random forest, the minimum number of samples for node splitting, the minimum number of samples for leaf nodes, and the feature subset sampling ratio, and using the Bayesian optimization algorithm to automatically search for the hyperparameter combination that optimizes the cross-validation accuracy.

7. The method for diagnosing and evaluating the expansion displacement of the valve-side dry bushing guide rod as described in claim 6. Its features are, In step S3.3, the number of decision trees is 100-500, and the maximum depth of a single tree is set to 5-15 layers.

8. A diagnostic and evaluation system for the expansion displacement of a valve-side dry bushing guide rod. Its features are, include: The multi-source sensing unit integrates displacement, vibration, and temperature sensors to achieve synchronous acquisition of mechanical state parameters; The intelligent analysis unit, equipped with an optimized random forest regression algorithm, processes sensor data in real time and extracts the true displacement components of the guide rod. The decision support unit automatically generates a status assessment report based on displacement characteristic analysis and provides hierarchical early warning and operation and maintenance guidance.

9. A computer device, Its features are, include: Memory and processor The memory and the processor are interconnected. The memory stores computer instructions. The processor executes the computer instructions to perform the valve-side dry bushing guide rod expansion displacement diagnosis and evaluation method as described in claim 7.

10. A computer-readable storage medium, Its features are, The computer-readable storage medium stores computer instructions. The computer instructions are used to cause the computer to execute the valve-side dry bushing guide rod expansion displacement diagnosis and evaluation method as described in claim 7.

Images