Method and system for determining maintenance plan of air brake braking system of hydroelectric generating set and medium

Abstract

The invention discloses a method and system for determining a maintenance plan of an air brake braking system of a hydroelectric generating set and a medium. Relates to the technical field of unit maintenance. Characteristic data of the air brake braking system are collected to carry out state estimation on the air brake braking system, an overhaul strategy of the air brake braking system is made according to a state estimation result and a hydroelectric generating set work overhaul plan, faults of the air brake braking system are foreseen in time, and the overhaul strategy of the air brake braking system is made in combination with the hydroelectric generating set work overhaul plan; when the air brake braking system has the fault risk, the maintenance time of the air brake braking system and the working peak period of the hydroelectric generating set are staggered; on one hand, the maintenance cycle of the air brake braking system is reasonably arranged, frequent maintenance is avoided, and workload is reduced; on the other hand, the working peak period of the hydroelectric generating set is staggered by combining the working maintenance plan of the hydroelectric generating set, and the stability of the hydroelectric generating set is guaranteed.

Description

Technical Field

[0001] The invention relates to the technical field of unit maintenance, and in particular to a method, system and medium for determining a maintenance plan for a wind brake brake system of a hydropower unit. Background Art

[0002] The wind gate system is an important component of the mechanical braking system of the hydro-turbine generator set. Its main function is to provide braking torque and enable the unit to stop quickly at low speed to prevent the unit from rotating for a long time at a low speed, or from creeping due to water leakage from the turbine guide vanes, resulting in a tile burning accident. At the same time, it also undertakes the task of oil pressure jacking and supporting the rotating part of the unit during maintenance.

[0003] Since the wind brake brake system needs to be synchronized with the operating cycle of the hydro-turbine generator set, it is generally repaired along with the unit except when a fault occurs. However, the wind brake brake system is frequently used, and the frequency of faults gradually increases with the increase of service life. During maintenance, the unit needs to be partially lifted and supported. Frequent maintenance not only increases the workload, but also directly affects the normal operation of the hydro-turbine generator set. Summary of the invention

[0004] The technical problem to be solved by the present invention is: under normal circumstances, the wind brake brake system is maintained with the group except when a fault occurs; however, the wind brake brake system is frequently used, and with the increase of service life and fault frequency, not only the work tasks are increased, but also the normal operation of the hydro-turbine generator set is directly affected; the present scheme provides a method, system and medium for determining the maintenance plan of the wind brake brake system of the hydro-turbine generator set, collects characteristic data of the wind brake brake system to estimate the state of the wind brake brake system, and formulates a maintenance strategy for the wind brake brake system according to the state estimation result and the maintenance plan of the hydro-turbine generator set, timely predicts the fault of the wind brake brake system, and formulates a maintenance strategy for the wind brake brake system in combination with the maintenance plan of the hydro-turbine generator set; when there is a risk of failure in the wind brake brake system, the maintenance time of the wind brake brake system is staggered with the peak working period of the hydro-turbine generator set; on the one hand, the present scheme realizes the reasonable arrangement of the maintenance cycle of the wind brake brake system, avoids frequent maintenance and reduces the workload; on the other hand, in combination with the maintenance plan of the hydro-turbine generator set, the peak working period of the hydro-turbine generator set is staggered to ensure the stability of the hydro-turbine generator set.

[0005] The present invention is achieved through the following technical solutions:

[0006] This plan provides a method for determining the maintenance plan of the wind brake brake system of the hydropower unit, including:

[0007] Collect characteristic data of the wind brake brake system and pre-process the characteristic data; the characteristic data includes: environmental data, operating status data, and system attribute data;

[0008] The pre-processed characteristic data is input into the health status quantification model of the wind brake brake system to calculate the health status estimation value of the wind brake brake system;

[0009] Based on the estimated health status and combined with the maintenance plan of the hydropower unit, a maintenance time strategy for the wind gate brake system is formulated.

[0010] Working principle of this scheme: Since the wind brake brake system needs to keep synchronization with the operation cycle of the hydro-turbine generator set, under normal circumstances, the wind brake brake system is repaired along with the unit except when a fault occurs; however, the wind brake brake system is frequently used, and with the changes in time, environment and load, it is affected by factors such as component aging, mechanical wear of the wind brake platform and load impact. At the same time, the poor quality of the oil source of the wind brake oil circuit will cause the oil circuit to get stuck and the internal seal of the brake to wear. The frequency of failures gradually increases, and the unit needs to be shut down and partially jacked up and supported for maintenance. Frequent maintenance increases the workload and directly affects the normal operation of the hydro-turbine generator set. In view of this, the maintenance plan of the wind brake brake system of the hydro-turbine generator set in this scheme is indeed A method, system and medium are determined, characteristic data of the wind brake brake system are collected to estimate the state of the wind brake brake system, and a maintenance strategy for the wind brake brake system is formulated according to the state estimation result and the maintenance plan of the hydropower unit, so as to timely foresee the failure of the wind brake brake system and formulate a maintenance strategy for the wind brake brake system in combination with the maintenance plan of the hydropower unit; when there is a risk of failure in the wind brake brake system, the maintenance time of the wind brake brake system is staggered with the peak working period of the hydropower unit; on the one hand, this scheme realizes the reasonable arrangement of the maintenance cycle of the wind brake brake system, avoids frequent maintenance and reduces the workload; on the other hand, in combination with the maintenance plan of the hydropower unit, the peak working period of the hydropower unit is staggered to ensure the stability of the hydropower unit.

[0011] In addition, this solution also provides a new technical concept: this solution estimates the state of the wind brake brake system based on the environmental data, operating status data, and system attribute data of the wind brake brake system in multiple dimensions; compared with the traditional single-dimensional dimension of operating status data, the state estimation method of this solution is more accurate, especially for the continuous rainy season, high-altitude areas and other extreme environmental situations, the probability of the wind brake brake system failure frequency is higher, because the brake air tank and brake pipeline in the wind brake brake system are easily affected by the environmental humidity and environmental pressure, generating water vapor and rust to block the brake pipeline and the wind brake air inlet, resulting in poor air flow, which in turn causes the wind brake to get stuck when the wind brake is shut down. Estimating the state of the wind brake brake system based on the environmental data, operating status data, and system attribute data of the wind brake brake system in multiple dimensions can grasp the health status of the wind brake brake system in advance and provide timely warning and maintenance.

[0012] A further optimization scheme is that the characteristic data of the wind brake braking system is collected and preprocessed, including the following method:

[0013] Collect environmental data, operating status data and system attribute data of the air brake brake system; the environmental data include: environmental humidity excess △W, temperature excess △T and air pressure excess △P; operating status data include: average value J of the difference between the air brake action instruction and the air brake action, total number of air brake actions M; system attribute data include: total number of inspections N, inner surface area S of the metal brake pipe section;

[0014] The environmental data, operating status data and system attribute data are normalized to obtain the environmental representation quantity, the operating status representation quantity and the system attribute representation quantity.

[0015] A further optimization scheme is to normalize the environmental data, operating status data and system attribute data to obtain environmental representation, operating status representation and system attribute representation, including the following method:

[0016] The environmental indicator U is calculated according to the following formula:

[0017] The running status indication quantity H is calculated according to the following formula: H = JM / (J+M);

[0018] The system attribute representation quantity X is calculated according to the following formula: X = NS / (N + S)

[0019] Where f represents the environmental normalization coefficient.

[0020] A further optimization scheme is that the pre-processed characteristic data is input into the health status quantification model of the wind brake brake system to calculate the health status estimation value of the wind brake brake system, including the method:

[0021] The preprocessed feature data is used as a new sample set and merged with the old sample set to obtain a combined sample set, and the combined sample set is divided into a training set, a test set, and a validation set;

[0022] A quantitative model of the health status of the air outlet brake system is constructed based on the training set and the test set;

[0023] Input the validation set into the constructed health status quantification model of the air brake brake system to obtain the validation classification results;

[0024] The Mahalanobis distance between the verified classification result and the normal state of the wind brake brake system is calculated, and the Mahalanobis distance is normalized to obtain the estimated value of the health state of the wind brake brake system.

[0025] A further optimization scheme is to construct a quantitative model of the health status of the air outlet brake system based on the training set and the test set, including the following method:

[0026] Input the training set into the multi-source fusion network to extract high-dimensional fusion features;

[0027] Based on the nearest neighbor mean classifier, the comprehensive loss of the new sample set is considered to realize high-dimensional fusion features for health status classification;

[0028] Adjust the old sample data of each health status classification according to the health status classification results.

[0029] A further optimization scheme is to consider the comprehensive loss of the new sample set, including the following method:

[0030] The loss function L of the health status quantification model of the wind brake brake system is composed of the current training loss score and the old sample training loss:

[0031]

[0032] Where A represents the combined sample set; n represents the total number of health states; β1 represents the first indicative function; β2 represents the second indicative function; y 1i (x) represents the output of the i-th health state signal sample set in the current wind brake brake system health state quantification model; y 0i Represents the output of the old sample set of the i-th health state on the health state quantification model of the old wind brake brake system.

[0033] A further optimization scheme is that the Mahalanobis distance is normalized to obtain the estimated value of the health state of the air brake brake system, including the following method:

[0034] The estimated health status value G of the wind brake brake system is calculated according to the following formula:

[0035]

[0036] Among them, d represents the Mahalanobis distance between the verification classification result and the normal state of the wind brake brake system, and b represents the normalization parameter.

[0037] A further optimization scheme is to formulate a maintenance strategy for the wind brake brake system based on the estimated health status value and combined with the maintenance plan of the hydropower unit, including the following methods:

[0038] Obtain an estimate of the health status of the wind brake system, the downtime of the hydropower unit, and the last maintenance deadline;

[0039] Calculate the maintenance deadline of the wind brake brake system based on the estimated health status of the wind brake brake system;

[0040] If the last maintenance period of the hydropower unit is within the maintenance deadline of the wind brake system, the wind brake system shall be maintained along with the hydropower unit.

[0041] The present solution also provides a system for determining a maintenance plan for a wind brake system of a hydropower unit, which is used to implement the above-mentioned method for determining a maintenance plan for a wind brake system of a hydropower unit. The system includes:

[0042] The acquisition module is used to acquire characteristic data of the wind brake braking system and pre-process the characteristic data; the characteristic data includes: environmental data, operating status data, and system attribute data;

[0043] A preprocessing module, used for performing state estimation on the air brake braking system based on the preprocessed characteristic data;

[0044] The strategy formulation module is used to formulate a maintenance strategy for the wind gate brake system based on the state estimation results and combined with the maintenance plan of the hydropower unit.

[0045] The present solution also provides a computer-readable medium having a computer program stored thereon, and the computer program is executed by a processor to implement the method for determining the maintenance plan of the wind gate brake system of the hydropower unit as described above.

[0046] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0047] 1. The present invention provides a method, system and medium for determining a maintenance plan for a wind brake system of a hydropower unit; collecting characteristic data of the wind brake system to estimate the state of the wind brake system, and formulating a maintenance strategy for the wind brake system based on the state estimation result and the maintenance plan of the hydropower unit, timely predicting the failure of the wind brake system, and formulating a maintenance strategy for the wind brake system in combination with the maintenance plan of the hydropower unit; when there is a risk of failure in the wind brake system, the maintenance time of the wind brake system is staggered with the peak working period of the hydropower unit; on the one hand, it realizes a reasonable arrangement of the maintenance cycle of the wind brake system, avoids frequent maintenance, and reduces the workload; on the other hand, in combination with the maintenance plan of the hydropower unit, the peak working period of the hydropower unit is staggered to ensure the stability of the hydropower unit.

[0048] 2. The present invention provides a method, system and medium for determining a maintenance plan for a wind brake system of a hydropower unit; the state of the wind brake brake system is estimated based on multiple dimensions such as environmental data, operating status data and system attribute data of the wind brake brake system; compared with the traditional single-dimensional state estimation method from the operating status data, the state estimation method of this scheme is more accurate, especially for extreme environmental situations such as continuous rainy seasons and high-altitude areas, the health status of the wind brake brake system can be grasped in advance, and timely warning and maintenance can be issued.

[0049] 3. The present invention provides a method, system and medium for determining a maintenance plan for a wind brake system of a hydropower unit; after obtaining the health status type of the wind brake brake system through a quantification model of the health status of the wind brake brake system, the health level of the wind brake brake system is further understood, that is, the degree of its deviation from the normal state is standardized and measured to achieve preventive maintenance.

[0050] 3. The present invention provides a method, system and medium for determining a maintenance plan for a wind brake system of a hydropower unit; the preprocessed feature data is used as a new sample set and fused with the old sample set to obtain a combined sample set; in the process of using the combined sample set to train a quantitative model of the health status of the wind brake brake system, the comprehensive loss is considered to realize high-dimensional fusion features for health status classification, and for continuous features with continuous effects similar to the rainy season, incremental learning of continuous features is realized, and evaluation of fault increments is realized, thereby improving the evaluation accuracy of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, other relevant drawings can be obtained based on these drawings without creative work. In the drawings:

[0052] Figure 1 The present invention is a flowchart of the method for determining the maintenance plan for the wind gate brake system of a hydropower unit. DETAILED DESCRIPTION

[0053] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention.

[0054] Generally, the wind brake brake system is inspected by the group except when a fault occurs. However, the wind brake brake system is frequently used. As the service life and fault frequency increase, not only the operation tasks increase, but also the normal operation of the hydro-generator set is directly affected. In extreme environments such as the continuous rainy season and high altitude areas, the probability of the wind brake brake system failing is higher, because the brake air tank and brake pipeline in the wind brake brake system are easily affected by the environmental humidity and environmental pressure, generating water vapor and rust to block the brake pipeline and the wind brake air inlet, resulting in poor air flow, which in turn causes the wind brake to get stuck when the wind brake is shut down, and the wind brake brake system needs to be frequently repaired. In view of this, the present invention provides the following embodiments to solve the above technical problems:

[0055] Example 1

[0056] This embodiment provides a method for determining a maintenance plan for a wind brake system of a hydropower unit, such as Figure 1 As shown, including:

[0057] Step 1: Collect characteristic data of the wind brake braking system and pre-process the characteristic data; the characteristic data includes: environmental data, operating status data, and system attribute data; this step specifically includes the following method:

[0058] S11, collecting environmental data, operating status data and system attribute data of the wind brake brake system; the environmental data include: environmental humidity excess △W, temperature excess △T and air pressure excess △P; the operating status data include: the average value J of the difference between the wind brake action instruction and the wind brake action, the total number of wind brake actions M; the system attribute data include: the total number of maintenance times N, the inner surface area S of the metal brake pipe section; for areas with extreme climates such as frequent rainy seasons and extremely low or high air pressure, a comprehensive evaluation is performed through the environmental dimension data, operating status dimension data and system attribute dimension data, so that the constructed quantitative model of the health status of the wind brake brake system can be applied to a wider range of scenarios.

[0059] In some southern regions, the rainy season lasts for a long time, and the air humidity inside the factory is high. When the low-pressure machine compresses the air, the water vapor is compressed and condensed and gathered in the brake air tank (after testing, the water vapor content in the low-pressure air reaches 60%). When the unit is shut down and the brake is applied, the airflow carries the water vapor into the brake pipeline and the wind brake. With the long-term operation of the unit, there is a lot of rust inside the brake pipeline, and the mixed turbine oil forms oil stains, which block the brake pipeline and the wind brake air inlet hole. When the unit is shut down on a daily basis, the low-pressure air carries the oil stains inside the brake ring tube into the wind brake cylinder, causing the wind brake to get stuck when applying and removing the brake. Therefore, in addition to evaluating the health status in the dimension of the operating status of the wind brake brake system, this solution also considers the environmental dimension and the system attribute dimension for multi-dimensional feature extraction to improve the accuracy of the evaluation results.

[0060] The method for obtaining the average value J of the difference between the damper action command and the damper action includes:

[0061] Based on infrared distance measurement, multiple infrared sensors collect the lifting and lowering signals of the wind gate and send them to the host. The host calculates the lifting and lowering action distance data of the wind gate and compares and subtracts it with the wind gate action command, and finally obtains the average value J of the difference between multiple groups of wind gate action commands. Compared with the previous selection of infrared sensors, the influence of iron materials around the wind gate signal baffle is eliminated, and the distance measurement is purer; equipment fault inspection and maintenance are more convenient and simple to install; the lifting and lowering criteria of the wind gate are more rigorous, and the actual action stroke of the wind gate is judged by subtracting the data to increase the reliability of the data; the data can be displayed on the touch screen as needed.

[0062] S12, normalizing the environment data, operation status data and system attribute data to obtain environment representation, operation status representation and system attribute representation. This step specifically includes the following method:

[0063] The environmental indicator U is calculated according to the following formula:

[0064] The running status indication quantity H is calculated according to the following formula: H = JM / (J+M);

[0065] The system attribute representation quantity X is calculated according to the following formula: X = NS / (N + S)

[0066] Where f represents the environmental normalization coefficient.

[0067] Since the environmental data, operating status data and system attribute data contain relatively large parameters, in order to simplify the subsequent feature extraction and fusion process, this scheme uses the normalized environmental representation quantity, operating status representation quantity and system attribute representation quantity of each parameter to represent the environmental data, operating status data and system attribute data, thereby simplifying the system calculation process.

[0068] Step 2: Input the preprocessed characteristic data into the health status quantification model of the wind brake brake system to calculate the health status estimation value of the wind brake brake system; this step specifically includes the following method:

[0069] S21, taking the preprocessed feature data as a new sample set and fusing it with the old sample set to obtain a combined sample set, and dividing the combined sample set into a training set, a test set, and a validation set;

[0070] S22, build a quantitative model of the health status of the air outlet brake system based on the training set and the test set:

[0071] S221, input the training set into the multi-source fusion network to extract high-dimensional fusion features;

[0072] S222, based on the nearest neighbor mean classifier, the comprehensive loss of the new sample set is considered to realize high-dimensional fusion features for health status classification; specifically, this scheme uses the current training loss score and the old sample training loss to form the loss function L of the health status quantification model of the wind brake brake system:

[0073]

[0074] Where A represents the combined sample set; n represents the total number of health states; β1 represents the first indicative function; β2 represents the second indicative function; y 1i (x) represents the output of the i-th health state signal sample set in the current wind brake brake system health state quantification model; y 0i Represents the output of the old sample set of the i-th health state on the health state quantification model of the old wind brake brake system.

[0075] This scheme uses the preprocessed feature data as a new sample set and fuses it with the old sample set to obtain a combined sample set. Correspondingly, in the process of using the combined sample set to train the quantitative model of the health status of the wind brake brake system, the comprehensive loss is considered to realize high-dimensional fusion features for health status classification. For continuous features with continuous impacts such as the rainy season, incremental learning of continuous features is realized, and the evaluation of fault increments is realized, thereby improving the evaluation accuracy of the model.

[0076] S223, adjusting old sample data of each health status classification according to the health status classification result.

[0077] S23, inputting the verification set into the constructed air brake brake system health status quantification model to obtain a verification classification result;

[0078] S24, calculating the Mahalanobis distance between the verification classification result and the normal state of the wind brake brake system, and normalizing the Mahalanobis distance to obtain the estimated value of the health state of the wind brake brake system. The method of normalizing the Mahalanobis distance to obtain the estimated value of the health state of the wind brake brake system includes:

[0079] The estimated health status value G of the wind brake brake system is calculated according to the following formula:

[0080]

[0081] Among them, d represents the Mahalanobis distance between the verification classification result and the normal state of the wind brake braking system, and b represents the normalization parameter.

[0082] Step 3: Based on the estimated health status and combined with the maintenance plan of the hydropower unit, a maintenance time strategy for the wind brake brake system is formulated. This step specifically includes the following methods:

[0083] S31, obtaining the estimated health status of the wind brake braking system, the downtime of the hydropower unit and the last maintenance deadline;

[0084] S32, calculating the maintenance deadline of the wind brake brake system according to the estimated health status value of the wind brake brake system;

[0085] S33: If the last maintenance period of the hydropower unit is within the maintenance deadline of the wind brake brake system, the wind brake brake system shall be maintained along with the hydropower unit.

[0086] After obtaining the health status type of the wind brake brake system through the health status quantification model of the wind brake brake system, this scheme also further understands the health level of the wind brake brake system, that is, standardizes and measures the degree of its deviation from the normal state to achieve preventive maintenance.

[0087] Example 2

[0088] This embodiment provides a system for determining a maintenance plan for a wind brake system of a hydropower unit, which is used to implement the method for determining a maintenance plan for a wind brake system of a hydropower unit described in Embodiment 1. The system includes:

[0089] The acquisition module is used to acquire characteristic data of the wind brake braking system and pre-process the characteristic data; the characteristic data includes: environmental data, operating status data, and system attribute data;

[0090] A preprocessing module, used for performing state estimation on the air brake braking system based on the preprocessed characteristic data;

[0091] The strategy formulation module is used to formulate a maintenance strategy for the wind gate brake system based on the state estimation results and combined with the maintenance plan of the hydropower unit.

[0092] Example 3

[0093] This embodiment provides a computer-readable medium on which a computer program is stored. The computer program is executed by a processor to implement the method for determining a maintenance plan for a wind brake brake system of a hydropower unit as described in Embodiment 1. Specifically, the following steps are performed:

[0094] Step 1: Collect characteristic data of the wind brake brake system and pre-process the characteristic data; the characteristic data includes: environmental data, operating status data, and system attribute data;

[0095] Step 2, inputting the preprocessed characteristic data into the health status quantification model of the wind brake brake system to calculate the health status estimation value of the wind brake brake system;

[0096] Step three: Based on the estimated health status and combined with the maintenance plan of the hydropower unit, a maintenance time strategy for the wind gate brake system is formulated.

[0097] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for determining the maintenance plan of the wind brake brake system of a hydropower unit, characterized in that: include: Collecting characteristic data of the air brake braking system and preprocessing the characteristic data; The characteristic data includes: environmental data, operating status data, and system attribute data; The pre-processed characteristic data is input into the health status quantification model of the wind brake brake system to calculate the health status estimation value of the wind brake brake system; Based on the estimated health status and combined with the maintenance plan of the hydropower unit, a maintenance time strategy for the wind gate brake system is formulated.

2. The method for determining the maintenance plan of the wind brake brake system of a hydropower unit according to claim 1 is characterized in that: The collecting of characteristic data of the wind brake braking system and preprocessing of the characteristic data include: Collect environmental data, operating status data and system attribute data of the wind brake brake system; The environmental data include: environmental humidity excess △W, temperature excess △T and air pressure excess △P; the operating status data include: the average value J of the difference between the damper action instruction and the damper action, the total number of damper actions M; the system attribute data include: the total number of inspections N, the inner surface area S of the metal brake pipe section; The environmental data, operating status data and system attribute data are normalized to obtain the environmental representation quantity, the operating status representation quantity and the system attribute representation quantity.

3. The method for determining the maintenance plan of the wind brake brake system of a hydropower unit according to claim 2 is characterized in that: Normalizing the environmental data, the operating status data and the system attribute data to obtain the environmental representation quantity, the operating status representation quantity and the system attribute representation quantity, including the method: The environmental indicator U is calculated according to the following formula: The running status indication quantity H is calculated according to the following formula: H = JM / (J+M); The system attribute representation quantity X is calculated according to the following formula: X = NS / (N + S) Where f represents the environmental normalization coefficient.

4. The method for determining the maintenance plan of the wind brake brake system of a hydropower unit according to claim 2 is characterized in that: The method of inputting the pre-processed characteristic data into the health status quantification model of the air brake brake system to calculate the health status estimation value of the air brake brake system includes: The preprocessed feature data is used as a new sample set and merged with the old sample set to obtain a combined sample set, and the combined sample set is divided into a training set, a test set, and a validation set; A quantitative model of the health status of the air outlet brake system is constructed based on the training set and the test set; Input the validation set into the constructed health status quantification model of the air brake brake system to obtain the validation classification results; The Mahalanobis distance between the verified classification result and the normal state of the wind brake brake system is calculated, and the Mahalanobis distance is normalized to obtain the estimated value of the health state of the wind brake brake system.

5. The method for determining the maintenance plan of the wind brake brake system of a hydropower unit according to claim 4 is characterized in that: The method of constructing a health status quantification model of the air outlet brake system based on a training set and a test set includes: Input the training set into the multi-source fusion network to extract high-dimensional fusion features; Based on the nearest neighbor mean classifier, the comprehensive loss of the new sample set is considered to realize high-dimensional fusion features for health status classification; Adjust the old sample data of each health status classification according to the health status classification results.

6. The method for determining the maintenance plan of the wind brake brake system of a hydropower unit according to claim 5 is characterized in that: The comprehensive loss of the new sample set is considered, including the following method: The loss function L of the health status quantification model of the wind brake brake system is composed of the current training loss score and the old sample training loss: Where A represents the combined sample set; n represents the total number of health states; β1 represents the first indicative function; β2 represents the second indicative function; y 1i (x) represents the output of the i-th health state signal sample set in the current wind brake brake system health state quantification model; y 0i Represents the output of the old sample set of the i-th health state on the health state quantification model of the old wind brake brake system.

7. The method for determining the maintenance plan of the wind brake brake system of a hydropower unit according to claim 4 is characterized in that: The method of normalizing the Mahalanobis distance to obtain the estimated value of the health state of the air brake brake system includes: The estimated health status value G of the wind brake brake system is calculated according to the following formula: Among them, d represents the Mahalanobis distance between the verification classification result and the normal state of the wind brake brake system, and b represents the normalization parameter.

8. The method for determining the maintenance plan of the wind brake brake system of a hydropower unit according to claim 7, characterized in that: The maintenance strategy of the wind brake brake system is formulated based on the health status estimation value and combined with the maintenance plan of the hydropower unit, including the following methods: Obtain an estimate of the health status of the wind brake system, the downtime of the hydropower unit, and the last maintenance deadline; Calculate the maintenance deadline of the wind brake brake system based on the estimated health status of the wind brake brake system; If the last maintenance period of the hydropower unit is within the maintenance deadline of the wind brake system, the wind brake system will be maintained along with the hydropower unit; otherwise, the wind brake system will be maintained during any downtime of the hydropower unit within the maintenance deadline.

9. A system for determining a maintenance plan for a wind brake system of a hydropower unit, used to implement a method for determining a maintenance plan for a wind brake system of a hydropower unit according to any one of claims 1 to 8, the system comprising: A collection module, used for collecting characteristic data of the wind brake braking system and preprocessing the characteristic data; The characteristic data includes: environmental data, operating status data, and system attribute data; A preprocessing module, used for performing state estimation on the air brake braking system based on the preprocessed characteristic data; The strategy formulation module is used to formulate a maintenance strategy for the wind gate brake system based on the state estimation results and combined with the maintenance plan of the hydropower unit.

10. A computer readable medium having a computer program stored thereon, characterized in that: The computer program is executed by a processor to implement the method for determining the maintenance plan of the wind gate brake system of a hydropower unit as described in any one of claims 1-8.

Images