Fan overhauling system based on digitalization of overhauling knowledge

Abstract

The application relates to the technical field of fan testing and overhauling, and discloses a fan overhauling system based on overhauling knowledge digitization, which comprises an overhauling knowledge module, an overhauling sequence module, an overhauling simulation module and a testing and overhauling module; wherein: the overhauling knowledge module is used for collecting fan overhauling data and performing digitization processing of overhauling knowledge; the overhauling sequence module is used for calculating the priority of each part to be overhauled and constructing an overhauling sequence; the overhauling simulation module is used for simulating fan overhauling through a digital twin system and acquiring an overhauling consumption factor of each part; and the testing and overhauling module is used for controlling testing and overhauling of the parts and controlling updating of the overhauling sequence; the application integrates advanced technologies such as overhauling knowledge digitization, overhauling task serialization and digital twin system simulation operation, and realizes fine and efficient execution of fan testing and overhauling work.

Description

Technical Field

[0001] This invention relates to the technical field of wind turbine testing and maintenance, and specifically to a wind turbine maintenance system based on the digitalization of maintenance knowledge. Background Technology

[0002] With the rapid development of the wind power industry, wind turbines, as core equipment, directly impact the economic benefits and energy supply security of the entire wind farm through their stability and operation and maintenance efficiency. However, in the actual operation and maintenance of wind turbines, efficient and reasonable testing and maintenance has become a crucial and complex task. Currently, there are many challenges in the execution of wind turbine testing and maintenance tasks. Due to the lack of an accurate component health assessment mechanism and a scientific and reasonable method for prioritizing maintenance tasks, wind turbine testing and maintenance work is often in a reactive state. The traditional wind turbine testing and maintenance plan is mostly based on experience and manual judgment, which limits the accuracy of assessing component wear, failure probability, and maintenance priority, and easily overlooks some hidden fault factors. Especially in large and complex modern wind turbine units, there are numerous components, and the interrelationships between these components are intricate, making it difficult to achieve comprehensive and accurate testing and maintenance decisions solely based on manual experience. The disassembly and maintenance sequence of components often fails to fully utilize data analysis and intelligent algorithms for optimization. An unreasonable testing and maintenance sequence can not only increase unnecessary disassembly and assembly costs and time consumption, but also trigger a chain reaction, causing originally healthy components to fail prematurely due to misoperation or secondary damage during the disassembly and assembly process. Before implementing testing and maintenance tasks, risk prediction and cost analysis for each component are insufficient. Current assessment systems often fail to comprehensively quantify direct maintenance losses (such as spare parts costs and labor costs) and indirect maintenance losses (such as downtime loss and reduced power generation), and fail to combine these with factors such as wind turbine operating status and historical fault records to form a dynamic risk and cost assessment plan.

[0003] For example, patent CN112665884A discloses an online maintenance system for parallel-operated wind turbines. Utilizing the negative pressure at the inlet of the parallel-operated wind turbines, a maintenance connection pipe is installed behind the inlet baffle and in front of the outlet baffle of one wind turbine. This allows for the extraction of leaked flue gas and hot air from the outlet baffle of the wind turbine requiring unilateral isolation, achieving reliable isolation and maintenance for parallel-operated wind turbines online, thus solving the problem of difficult isolation during online maintenance of parallel-operated wind turbines. It also addresses the issue that in most power plants, the inlet and outlet baffles of parallel-operated wind turbines cannot be completely and tightly closed, causing isolation difficulties. A failure of one wind turbine can even lead to unplanned unit shutdowns. The system structure is simple and can be flexibly configured according to the actual equipment conditions; the equipment operates reliably for a long time without affecting wind turbine efficiency due to leakage; the system requires minimal maintenance and effectively ensures the personal safety of maintenance personnel. However, it suffers from the problems mentioned in the background technology: a lack of accurate component health assessment mechanisms and insufficient risk prediction and cost analysis for each component.

[0004] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide a wind turbine maintenance system based on the digitization of maintenance knowledge. It integrates advanced technologies such as the digitization of maintenance knowledge, the serialization of maintenance tasks, and the simulation operation of digital twin system, so as to realize the refined and efficient execution of wind turbine testing and maintenance work.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] The wind turbine maintenance system based on digitalized maintenance knowledge includes a maintenance knowledge module, a maintenance sequence module, a maintenance simulation module, and a test and maintenance module; among which:

[0008] The maintenance knowledge module is used to collect wind turbine maintenance data and digitize the maintenance knowledge.

[0009] The maintenance sequence module calculates the priority of each part to be inspected based on the digitally processed maintenance knowledge and constructs the maintenance sequence.

[0010] The maintenance simulation module is used to simulate wind turbine maintenance through a digital twin system and obtain the maintenance wear factor for each component.

[0011] The test and maintenance module is used to control the testing and maintenance of parts, and to control the updating of the maintenance sequence.

[0012] As a preferred embodiment of the wind turbine maintenance system based on digital maintenance knowledge as described in this invention, the formula for calculating the priority of each part to be maintained by the maintenance sequence module is as follows:

[0013] P i =(1+exp(-α·(F)) i +β·L i +D i ))) -1 ;

[0014] Among them, P i This indicates the priority of the i-th part; the value of i ranges from 1, 2, ..., n; n is the number of parts to be inspected.

[0015] F i L represents the failure frequency of the i-th component; i This represents the potential loss value of the i-th part;

[0016] α and β are both weighting factors;

[0017] D i This represents the abnormal risk rate of the i-th part.

[0018] As a preferred embodiment of the wind turbine maintenance system based on digital maintenance knowledge described in this invention, the maintenance sequence module includes a data acquisition unit. The data acquisition unit collects the wind turbine's temperature and noise intensity through sensors to calculate the anomaly risk rate, as follows:

[0019] Divide the fan into M regions, where M is a positive integer; collect the temperature and noise intensity of the M regions;

[0020] If the region containing the i-th part does not contain a connected region, the formula for calculating the anomaly risk rate is as follows:

[0021]

[0022] Among them, t i This represents the real-time temperature of the region where the i-th part is located. v represents the historical average temperature of the region where the i-th part is located; i This represents the real-time noise intensity of the area where the i-th part is located. The historical average noise intensity of the region where the i-th part is located is represented; the connected region is specifically defined as: if any two regions are adjacent and physically connected, then the two regions are connected regions to each other;

[0023] If the region containing the i-th part has no connected regions, then the region containing the i-th part is divided into m sub-regions, and each sub-region is adjacent to and physically connected to one connected region of the region containing the i-th part; where m equals the number of connected regions in the region containing the i-th part; the formula for calculating the anomaly risk rate is as follows:

[0024]

[0025] Among them, t l This represents the real-time temperature of the connected region of the i-th component. v represents the historical average temperature of the connected region of the i-th component; l This represents the real-time noise intensity of the connected region of the i-th component. w represents the historical average noise intensity of the connected region of the i-th component; t w represents the temperature correction factor. v The noise correction coefficient is represented by ; the connected region of the i-th part is the region that is adjacent to and physically connected to the sub-region of the i-th part within the connected region of the i-th part.

[0026] As a preferred embodiment of the wind turbine maintenance system based on digital maintenance knowledge described in this invention, the maintenance knowledge module collects maintenance knowledge including fuzzy rules, fault frequency, potential loss value, maintenance logic sequence, maintenance tools, precautions, and process descriptions.

[0027] As a preferred embodiment of the wind turbine maintenance system based on digital maintenance knowledge described in this invention, the maintenance sequence module constructs the maintenance sequence as follows: All parts are sorted in descending order of priority to form a maintenance sequence; the maintenance sequence is adjusted based on the constraints of each part; the constraints of each part are as follows: For any part A, the constraint is that in the maintenance sequence, all parts in the constraint part set of part A are placed before part A; if part A does not meet the constraint, the last position of all parts in the constraint part set of part A in the maintenance sequence is read, and the position of part A is adjusted to after the last position; the constraint part set of part A is constructed based on maintenance knowledge, and the construction method is as follows: Traverse all parts except part A; if any part B has a maintenance logical order with part A and its maintenance logical order is before part A, then add part B to the constraint part set of part A.

[0028] As a preferred embodiment of the wind turbine maintenance system based on digital maintenance knowledge described in this invention, the method by which the test and maintenance module controls the updating of the maintenance sequence is as follows:

[0029] Remove the parts that have completed maintenance from the maintenance sequence;

[0030] Update the priority of each part in the maintenance sequence and calculate the comprehensive ranking index of each part in the maintenance sequence;

[0031] The parts in the maintenance sequence are sorted in descending order of their comprehensive ranking index.

[0032] The maintenance sequence is adjusted based on the constraints of each part.

[0033] As a preferred embodiment of the wind turbine maintenance system based on digital maintenance knowledge described in this invention, the formula for calculating the comprehensive ranking index is as follows:

[0034]

[0035] Among them, K i ε represents the overall ranking index of the i-th part; ε is the cost adjustment factor, and η is the smoothing factor;

[0036] J i Let represent the overall wear and tear index of the i-th part, calculated using the following formula:

[0037] J i =w1·C i1 +w2·C i2 +w3·C i3 +w4·C i4 ;

[0038] Among them, w1, w2, w3, and w4 are all weight parameters;

[0039] C i1 C represents the machining loss factor of the i-th part; i2 This represents the consumable consumption factor for the i-th part;

[0040] C i3 C represents the manual inspection and wear factor of the i-th part; i4 This represents the production stoppage and maintenance factor for the i-th part;

[0041] P i ' represents the updated priority value of the i-th part; it is obtained by assigning the priority P of the i-th part based on the fuzzy rule. i The update is performed as follows: set fuzzy rules, and based on the maintenance status of the already inspected parts and the fuzzy rules, update the priority of the parts in the uninspected maintenance sequence.

[0042] As a preferred embodiment of the wind turbine maintenance system based on digital maintenance knowledge described in this invention, the method by which the test and maintenance module controls the testing and maintenance of parts is as follows:

[0043] S100: Obtain the wear factor of each part through the maintenance simulation module;

[0044] The maintenance simulation module uses a digital twin system to simulate the complete wind turbine maintenance process according to the maintenance sequence and obtains the maintenance consumption factor for each component.

[0045] S200: Controls the maintenance of the first part in the maintenance sequence;

[0046] S300: Update the maintenance sequence based on the digitally processed maintenance knowledge and the maintenance consumption factor;

[0047] S400: Repeat steps S100-S300 until the maintenance sequence is empty, completing the maintenance of the fan.

[0048] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0049] This invention transforms maintenance knowledge into calculable data, enabling dynamic and accurate assessment of the maintenance priority of each component, thereby achieving precise prediction and on-demand testing and maintenance. By comprehensively considering the risk index and wear factor of the components, the disassembly and maintenance sequence is scientifically and rationally arranged, reducing ineffective labor and resource waste, while also minimizing additional losses caused by improper disassembly and assembly sequences.

[0050] Through deep learning and analysis of massive amounts of operation and maintenance data, the risk level of components can be predicted in real time, comprehensively considering both direct and indirect maintenance wear and tear, providing strong support for testing and maintenance decisions, and facilitating cost optimization and risk prevention from a global perspective. By using a digital twin system to simulate actual maintenance operations in a virtual environment, potential problems can be identified and resolved in advance, significantly reducing uncertainty in the actual maintenance process and improving maintenance efficiency and safety. Attached Figure Description

[0051] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0052] Figure 1 A schematic diagram of the structure of the wind turbine maintenance system based on digital maintenance knowledge provided by the present invention;

[0053] Figure 2 A flowchart of the maintenance sequence update method provided by the present invention. Detailed Implementation

[0054] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0055] This embodiment introduces a wind turbine maintenance system based on digital maintenance knowledge, referring to... Figure 1 The system includes a maintenance knowledge module, a maintenance sequence module, a maintenance simulation module, and a test and maintenance module.

[0056] The maintenance knowledge module is used to collect wind turbine maintenance data and digitize the maintenance knowledge.

[0057] The maintenance knowledge module collects maintenance knowledge including fuzzy rules, fault frequency, potential loss value, maintenance logic sequence, maintenance tools, precautions, and process descriptions.

[0058] The digital processing of fuzzy rules is as follows:

[0059] Fuzzy rule definition: First, fuzzy rules regarding the adjustment of wind turbine maintenance priority are extracted from expert experience and actual cases, such as "if the repaired parts are not faulty, the priority of the unrepaired parts is reduced" and "if the repaired parts are faulty, the priority of the unrepaired parts is increased".

[0060] Fuzzy set construction: Convert the concepts in maintenance knowledge into fuzzy sets, such as setting up fuzzy language variables such as "no fault", "minor fault" and "major fault", and defining the corresponding membership functions.

[0061] Fuzzy rule digitization: Transform the defined fuzzy rules into mathematical form, such as: "IF the status of the repaired part is 'no fault' THEN the priority of the unrepaired part is 'low'".

[0062] The remaining digital processing of maintenance knowledge includes: digitizing fault data, collecting wind turbine operation data, and statistically analyzing the frequency of fault occurrence; quantifying potential losses by calculating the economic losses that may result from the failure of each component, achieving digital management of loss values; streamlining maintenance logic and procedures by outlining and developing clear maintenance steps based on maintenance practice, and converting them into digital flowcharts or algorithms for easy computer understanding and execution; digitizing maintenance tools and methods by recording the usage methods and applicable scenarios of various maintenance tools and storing them in digital form for easy retrieval and application; and standardizing process instructions and precautions by converting various process requirements, operating procedures, and precautions during maintenance into structured text or multimedia materials and inputting them into the system for maintenance personnel to refer to and learn from.

[0063] Maintenance knowledge is the foundation of equipment maintenance digitization, providing data support for the digitization of the equipment maintenance process. Equipment maintenance knowledge primarily exists in the factory disassembly and assembly manuals and maintenance operation guidelines provided by equipment manufacturers, as well as in the maintenance experience continuously explored, summarized, and improved by technical personnel in related fields during maintenance practice. This form of maintenance knowledge is not easily understood, processed, and applied by computers. To achieve more convenient digital services through computers, it must be converted into a form that is easy for computers to understand and manage. Based on the characteristic that the equipment maintenance process is based on the disassembly and maintenance of each component as the basic unit, a maintenance knowledge digitization approach is adopted, with the maintenance of each component as the smallest unit.

[0064] The maintenance sequence module calculates the priority of each part to be inspected based on the digitally processed maintenance knowledge and constructs the maintenance sequence.

[0065] The maintenance sequence module calculates the priority of each part to be inspected using the following formula:

[0066] P i =(1+exp(-α·(F)) i +β·L i +D i ))) -1 ;

[0067] Among them, P i P represents the priority of the i-th part; i The value range of P is (0,1), and P i The closer the value of i is to 1, the higher the priority; the value of i ranges from 1, 2, ..., n; n is the number of parts to be inspected;

[0068] F i F represents the failure frequency of the i-th component; that is, how frequently the i-th component fails. i The larger the value, the more frequent the failures; the failure rate is obtained by counting the actual number of failures of the i-th part over a period of time and dividing it by the total number of operating hours or usage cycles.

[0069] L i α represents the potential loss value of the i-th part; β represents the downtime loss amount caused by the failure of the i-th part; α is calculated by taking the statistical average of historical loss data caused by each part. The analysis examines the potential production interruption losses that might result from a fan shutdown due to a part failure. For example, critical components essential for continuous operation would cause significant production stoppage losses if they fail; such components should be given higher priority. α and β are weighting factors, set based on experience.

[0070] D iThis represents the abnormal risk rate of the i-th part; it also represents the degree to which the operating state of the i-th part deviates from the normal range, with a larger value indicating a more severe abnormality.

[0071] The method for calculating the abnormal risk rate is as follows:

[0072] The fan is divided into M regions, where M is a positive integer; the temperature and noise intensity of the M regions are continuously collected by sensors.

[0073] If the region containing the i-th part does not contain a connected region, the formula for calculating the anomaly risk rate is as follows:

[0074]

[0075] Among them, t i This represents the real-time temperature of the region where the i-th part is located. v represents the historical average temperature of the region where the i-th part is located; i This represents the real-time noise intensity of the area where the i-th part is located. The historical average noise intensity of the region where the i-th part is located is represented; the connected region is specifically defined as: if any two regions are adjacent and physically connected, then the two regions are connected regions to each other;

[0076] If the region containing the i-th part has no connected regions, then the region containing the i-th part is divided into m sub-regions, and each sub-region is adjacent to and physically connected to one connected region of the region containing the i-th part; where m equals the number of connected regions in the region containing the i-th part; the formula for calculating the anomaly risk rate is as follows:

[0077]

[0078] Among them, t l This represents the real-time temperature of the connected region of the i-th component. v represents the historical average temperature of the connected region of the i-th component; l This represents the real-time noise intensity of the connected region of the i-th component. w represents the historical average noise intensity of the connected region of the i-th component; t w represents the temperature correction factor. v The noise correction coefficients are set by those skilled in the art based on actual needs; the connected region of the i-th part is the region that is adjacent to and physically connected to the sub-region of the i-th part within the connected region of the i-th part.

[0079] Due to the interconnected nature of regions, the temperature and noise intensity of components within any given region are affected by the connected regions. Therefore, by setting temperature and noise correction coefficients and incorporating the temperature and noise intensity of connected regions into the calculation, the abnormal risk rate of components can be assessed more accurately.

[0080] The method for constructing the maintenance sequence is as follows: Sort all parts in descending order of priority and form a maintenance sequence; adjust the maintenance sequence based on the constraints of each part; the constraints of each part are as follows: For any part A, the constraint is that in the maintenance sequence, all parts in the constraint part set of part A are placed before part A; if part A does not meet the constraint, read the last position of all parts in the constraint part set of part A in the maintenance sequence and adjust the position of part A after the last position; the constraint part set of part A is constructed based on maintenance knowledge, and the construction method is as follows: Traverse all parts except part A; if any part B has a maintenance logical order with part A and its maintenance logical order is before part A, then add part B to the constraint part set of part A.

[0081] Equipment maintenance is constrained by the assembly relationships between each component. To complete maintenance tasks, the disassembly and handling of components must adhere to these constraints. For specific maintenance tasks, a maintenance sequence must be established based on the constraint information between equipment components to determine the order in which components are operated.

[0082] The maintenance simulation module is used to simulate the complete wind turbine maintenance process according to the maintenance sequence through the digital twin system, and to obtain the maintenance consumption factor of each part, including processing maintenance consumption factor, consumable maintenance consumption factor, labor maintenance consumption factor, and shutdown maintenance consumption factor.

[0083] The equipment maintenance process is transformed from its original form into a visualized, virtual, and dynamic one through 3D visualization. The actual maintenance process and knowledge are reconstructed in a 3D virtual environment; the simulated maintenance includes 3D visualization simulations of tool operation, instrument use, and processing techniques, preparing for the actual maintenance process.

[0084] The test and maintenance module is used to control the testing and maintenance of parts, and to control the updating of the maintenance sequence.

[0085] Reference Figure 2 The method for controlling the updating of the maintenance sequence is as follows:

[0086] Remove the parts that have completed maintenance from the maintenance sequence;

[0087] Update the priority of each part in the maintenance sequence and calculate the comprehensive ranking index of each part in the maintenance sequence; the formula is as follows:

[0088]

[0089] Among them, K i ε represents the overall ranking index of the i-th part; ε is the cost adjustment factor, and η is the smoothing factor, both set based on experience.

[0090] J i Let represent the overall wear and tear index of the i-th part, calculated using the following formula:

[0091] J i =w1·C i1 +w2·C i2 +w3·C i3 +w4·C i4 ;

[0092] Among them, w1, w2, w3, and w4 are all weight parameters, which are set based on experience;

[0093] C i1 Let represent the processing and maintenance consumption factor of the i-th part; represent the cost of using maintenance equipment and transportation costs during the maintenance process.

[0094] C i2 Represents the consumable maintenance factor for the i-th part; represents the cost of electrical components and consumable materials replaced during the maintenance process.

[0095] C i3 This represents the labor consumption factor for the i-th part, including the labor costs of maintenance personnel and auxiliary maintenance personnel.

[0096] C i4 represents the downtime maintenance loss factor for the i-th part; it represents the planned loss caused by downtime maintenance of the part, which can be calculated by multiplying the maintenance time of the part by the unit time revenue of the blower operation.

[0097] P i ' represents the updated priority value of the i-th part; the method for obtaining it is to assign the priority P of the i-th part based on the fuzzy rule. i To update, follow these steps:

[0098] First, fuzzy rules are set based on expert experience. Then, the priority of parts in the un-repaired repair sequence is updated according to the repair status of the repaired parts. For example, the repair status of the repaired parts can be set to three types: scrapped, faulty, and intact. An example of setting fuzzy rules is as follows: if part A is scrapped, the priority of parts B and C is updated to 1; if part A is intact, the priority of parts B and D is reduced by 50%.

[0099] The parts in the maintenance sequence are sorted in descending order of their comprehensive ranking index.

[0100] The maintenance sequence is adjusted based on the constraints of each part.

[0101] The methods for controlling the testing and maintenance of parts are as follows:

[0102] S100: Obtain the wear factor of each part through the maintenance simulation module;

[0103] The maintenance simulation module uses a digital twin system to simulate the complete wind turbine maintenance process according to the maintenance sequence and obtains the maintenance consumption factor for each component.

[0104] S200: Controls the maintenance of the first part in the maintenance sequence;

[0105] S300: Update the maintenance sequence based on the digitally processed maintenance knowledge and the maintenance consumption factor;

[0106] S400: Repeat steps S100-S300 until the maintenance sequence is empty, completing the maintenance of the fan.

[0107] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0108] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A wind turbine maintenance system based on digital maintenance knowledge, characterized in that: It includes a maintenance knowledge module, a maintenance sequence module, a maintenance simulation module, and a test and maintenance module; among which: The maintenance knowledge module is used to collect wind turbine maintenance data and digitize the maintenance knowledge. The maintenance sequence module calculates the priority of each part to be inspected based on the digitally processed maintenance knowledge and constructs the maintenance sequence. The maintenance simulation module is used to simulate wind turbine maintenance through a digital twin system and obtain the maintenance wear factor for each component. The test and maintenance module is used to control the testing and maintenance of parts, and to control the updating of the maintenance sequence. The maintenance sequence module calculates the priority of each part to be inspected using the following formula: ； in, This indicates the priority of the i-th part; the value of i ranges from 1, 2, ..., n; n is the number of parts to be inspected. This represents the failure frequency of the i-th component; This represents the potential loss value of the i-th part; , All are weighting factors; This represents the abnormal risk rate of the i-th part; The maintenance sequence module includes a data acquisition unit, which collects the temperature and noise intensity of the fan through sensors to calculate the anomaly risk rate, as follows: Divide the fan into M regions, where M is a positive integer; collect the temperature and noise intensity of the M regions; If the region containing the i-th part does not contain a connected region, the formula for calculating the anomaly risk rate is as follows: ； in, This represents the real-time temperature of the region where the i-th part is located. This represents the historical average temperature of the region where the i-th part is located; This represents the real-time noise intensity of the area where the i-th part is located. The historical average noise intensity of the region where the i-th part is located is represented; the connected region is specifically defined as: if any two regions are adjacent and physically connected, then the two regions are connected regions to each other; If the region containing the i-th part has a connected region, then the region containing the i-th part is divided into m sub-regions, and each sub-region is adjacent to and physically connected to a connected region in the region containing the i-th part; where m equals the number of connected regions in the region containing the i-th part; the formula for calculating the anomaly risk rate is as follows: ； in, This represents the real-time temperature of the connected region of the i-th component. This represents the historical average temperature of the connected region of the i-th component; This represents the real-time noise intensity of the connected region of the i-th component. This represents the historical average noise intensity of the connected region of the i-th component; This represents the temperature correction factor. The noise correction coefficient is represented by ; the connected region of the i-th part is the region that is adjacent to and physically connected to the sub-region of the i-th part within the connected region of the i-th part.

2. The wind turbine maintenance system based on digital maintenance knowledge as described in claim 1, characterized in that: The maintenance knowledge module collects maintenance knowledge including fuzzy rules, fault frequency, potential loss value, maintenance logic sequence, maintenance tools, precautions, and process descriptions.

3. The wind turbine maintenance system based on digital maintenance knowledge as described in claim 2, characterized in that, The method for constructing the maintenance sequence module is as follows: All parts are sorted in descending order of priority to form a maintenance sequence; the maintenance sequence is adjusted based on the constraints of each part; the constraints of each part are as follows: For any part A, the constraint is that in the maintenance sequence, all parts in the constraint part set of part A are placed before part A; if part A does not meet the constraint, the last position of all parts in the constraint part set of part A in the maintenance sequence is read, and the position of part A is adjusted to after the last position; the constraint part set of part A is constructed based on maintenance knowledge, and the construction method is as follows: Traverse all parts except part A; if any part B has a maintenance logical order with part A and its maintenance logical order is before part A, then add part B to the constraint part set of part A.

4. The wind turbine maintenance system based on digital maintenance knowledge as described in claim 3, characterized in that, The method by which the test and maintenance module controls the updating of the maintenance sequence is as follows: Remove the parts that have completed maintenance from the maintenance sequence; Update the priority of each part in the maintenance sequence and calculate the comprehensive ranking index of each part in the maintenance sequence; The parts in the maintenance sequence are sorted in descending order of their comprehensive ranking index. The maintenance sequence is adjusted based on the constraints of each part.

5. The wind turbine maintenance system based on digital maintenance knowledge as described in claim 4, characterized in that, The formula for calculating the comprehensive ranking index is as follows: ； in, This represents the overall ranking index of the i-th part; As a cost adjustment factor, It is a smoothing factor; Let represent the overall wear and tear index of the i-th part, calculated using the following formula: ； in, , , , All are weighted parameters; This represents the processing loss factor for the i-th part; This represents the consumable consumption factor for the i-th part; This represents the manual inspection and wear factor for the i-th part; This represents the production stoppage and maintenance factor for the i-th part; This represents the updated priority value of the i-th part; it is obtained by assigning priority to the i-th part based on the fuzzy rule. The update is performed as follows: set fuzzy rules, and based on the maintenance status of the already inspected parts and the fuzzy rules, update the priority of the parts in the uninspected maintenance sequence.

6. The wind turbine maintenance system based on digital maintenance knowledge as described in claim 5, characterized in that, The testing and maintenance module controls the testing and maintenance of parts in the following way: S100: Obtain the wear factor of each part through the maintenance simulation module; The maintenance simulation module uses a digital twin system to simulate the complete wind turbine maintenance process according to the maintenance sequence and obtains the maintenance consumption factor for each component. S200: Controls the maintenance of the first part in the maintenance sequence; S300: Update the maintenance sequence based on the digitally processed maintenance knowledge and the maintenance consumption factor; S400: Repeat steps S100-S300 until the maintenance sequence is empty, completing the maintenance of the fan.

7. The wind turbine maintenance system based on digital maintenance knowledge as described in claim 6, characterized in that, When executing step S200, if an emergency update mechanism trigger event is detected, the emergency update mechanism of the maintenance sequence is triggered. The emergency update mechanism of the maintenance sequence specifically includes: Suspend the current maintenance process; Re-collect temperature and noise intensity data for the area where the part is located; The anomaly risk rate was recalculated based on the updated sensor data; The priority of subsequent parts is dynamically adjusted based on the latest wear and tear factors; After generating a new maintenance sequence, continue with step S300.

8. The wind turbine maintenance system based on digital maintenance knowledge as described in claim 7, characterized in that, The emergency update mechanism is triggered by any one of the following conditions: The actual wear factor of the currently inspected parts exceeds the first preset threshold; The temperature change rate of the part currently under maintenance exceeds the second preset threshold. The noise intensity change rate of the currently inspected part exceeds the third preset threshold.

Images