A wind farm equipment whole life cycle health management system and method

Abstract

The present application relates to a kind of wind farm equipment full life cycle health management system, including wind turbine monitoring module, for obtaining and display the operating parameter of each wind turbine, the parameter of wind turbine subsystem;Alarm management module, for obtaining and display alarm event;Equipment health early warning module, for calculating the health score value of each wind turbine and wind turbine subsystem;Ledger management module is equipped with the equipment tree structure constructed according to equipment location and system affiliation, adds system, subsystem, equipment, component by equipment tree structure.

Description

Technical Field

[0001] This invention relates to a whole life cycle health management system and method for wind farm equipment, belonging to the field of wind farm equipment management. Background Technology

[0002] The external working environment of wind turbines is harsh and highly variable. Wind, as the power source, exhibits significant uncertainty in its speed and direction, causing frequent switching between different dynamic and static loads. Besides wind conditions, other climatic conditions such as rain, snow, and hail can severely impact the safety and reliability of wind turbines. The persistently high operating and maintenance costs of wind farms, thus weakening their economic benefits, have become a bottleneck for the development of the wind power industry.

[0003] Currently, there is considerable research on monitoring the operational status of key components of wind turbines, and this research is relatively mature. However, research on the entire lifecycle of all wind farm equipment, from individual equipment to overall turbine health assessment and intelligent development of operation and maintenance strategies, is relatively lacking. Compared to isolated systems for monitoring the status of critical equipment, wind farm equipment health management based on equipment lifecycle data faces more technical challenges. These challenges mainly manifest as imbalanced fault data, with the overall operating status of wind farm equipment being good and the proportion of faults being low. Furthermore, within the massive amount of equipment status data, the number of fault or anomaly samples is small, resulting in low data value density. Therefore, it is necessary to extract key parameters from the vast amount of operation and maintenance data to evaluate wind turbine health.

[0004] CN113610443A, "A Method for Evaluating the Health and Service Quality of Wind Turbines," discloses a method that comprehensively considers the severity, frequency, and duration of various faults in wind turbines, employs Fault Mode and Effects Analysis (FMEA), and defines and calculates the health status of wind turbines, achieving a comprehensive quantitative assessment of their health condition. This patent uses expert scoring to determine the severity level and weight of each fault, and its automation level needs further improvement. Summary of the Invention

[0005] To overcome the problems existing in the prior art, this invention designs a wind farm equipment full life cycle health management system and method. By comprehensively considering information such as the reliability, faults and defects, abnormal early warning, and power characteristic analysis of wind turbine equipment, the system automatically calculates the final result reflecting the health status of the equipment, guides the operation and maintenance management of wind farms, actively carries out corresponding maintenance work, controls the risks of wind turbine equipment, improves equipment reliability, and achieves the effect of lean maintenance.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] Technical Solution 1

[0008] A wind farm equipment lifecycle health management system, comprising:

[0009] The wind turbine monitoring module is used to acquire and display the operating parameters of each wind turbine and the parameters of the wind turbine subsystem;

[0010] The alarm management module is used to acquire and display alarm events;

[0011] The equipment health early warning module is used to calculate the health score of each fan and fan subsystem;

[0012] The ledger management module features a device tree structure built according to device location and system hierarchy, through which systems, subsystems, devices, and components can be added.

[0013] Further, the health score of the wind turbine is calculated, including the following steps:

[0014] Count the number of defects of different levels in each subsystem of the wind turbine;

[0015] Count the number of warnings at different levels for each subsystem of the wind turbine;

[0016] Calculate the compliance rate of the wind turbine power curve;

[0017] The health score of a single wind turbine is calculated based on the number of defects, the number of warnings, and the compliance rate.

[0018] Furthermore, the calculation of the health score of a single wind turbine is expressed by the following formula:

[0019] k = w1 k1 + w2 k2 + w3 k3

[0020] In the formula, w1, w2, and w3 represent weight values; k1 is the defect coefficient, with the initial value reduced by 30% for each Class A defect, 10% for each Class B defect, and 0.5% for each Class C defect; k2 is the early warning coefficient, with the initial value reduced by 1% for each Level 3 abnormal early warning event, 5% for each Level 2 abnormal early warning event, and 10% for each Level 3 abnormal early warning event; k3 is the power coefficient, with an initial value of 1. When the power curve compliance rate is greater than the threshold, the power coefficient is the initial value; when the power curve compliance rate is less than the threshold, the initial value is reduced by 1.0% for every 1% decrease.

[0021] Further, the health score of the wind turbine subsystem is calculated, including the following steps:

[0022] Select key equipment in the subsystem;

[0023] The number of defects of different levels in each key piece of equipment in the statistical subsystem;

[0024] The number of warnings at different levels for each key device in the statistical subsystem;

[0025] Calculate the failure rate of each key piece of equipment based on its continuous operating time and number of failures.

[0026] The health score of the subsystem is calculated based on the number of defects, the number of warnings, and the equipment failure rate.

[0027] Furthermore, the calculation of the failure rate includes:

[0028] If it is a disposable device, the device failure rate is calculated as 1-e. -t / MTTF MTBF is the cumulative trouble-free continuous operating time divided by the number of failures.

[0029] If not evaluated as a one-time device, the equipment failure rate is 1-e. -t / MTBF MTTF stands for Continuous Operating Time before equipment replacement.

[0030] Furthermore, the health score of the subsystem is calculated, expressed by the formula:

[0031] k = w1 k1 + w2 k2 + w3k3

[0032] In the formula, w1, w2, and w3 all represent weight values; k1 is the defect coefficient, with the initial value reduced by 30% for each Class A defect, 10% for each Class B defect, and 0.5% for each Class C defect; k2 is the early warning coefficient, with the initial value reduced by 1% for each Level 3 abnormal early warning event, 5% for each Level 2 abnormal early warning event, and 10% for each Level 3 abnormal early warning event; k3 is the equipment operation coefficient, which is assigned the initial value minus the equipment failure rate of each key piece of equipment.

[0033] Technical Solution Two

[0034] A method for full lifecycle health management of wind farm equipment includes the following steps:

[0035] The calculation of the health score of the wind turbine includes: counting the number of defects of different levels in each subsystem of the wind turbine; counting the number of early warnings of different levels in each subsystem of the wind turbine; calculating the compliance rate of the wind turbine power curve; and calculating the health score of a single wind turbine based on the number of defects, the number of early warnings, and the compliance rate.

[0036] The calculation of the health score of the wind turbine subsystem includes: selecting key equipment in the subsystem; counting the number of defects of different levels in each key equipment in the subsystem; counting the number of warnings of different levels in each key equipment in the subsystem; calculating the failure rate of each key equipment based on its continuous operating time and number of failures; and calculating the health score of the subsystem based on the number of defects, the number of warnings, and the equipment failure rate.

[0037] Furthermore, the calculation of the health score of a single wind turbine is expressed by the following formula:

[0038] k = w1 k1 + w2 k2 + w3 k3

[0039] In the formula, w1, w2, and w3 represent weight values; k1 is the defect coefficient, with the initial value reduced by 30% for each Class A defect, 10% for each Class B defect, and 0.5% for each Class C defect; k2 is the early warning coefficient, with the initial value reduced by 1% for each Level 3 abnormal early warning event, 5% for each Level 2 abnormal early warning event, and 10% for each Level 3 abnormal early warning event; k3 is the power coefficient, with an initial value of 1. When the power curve compliance rate is greater than the threshold, the power coefficient is the initial value; when the power curve compliance rate is less than the threshold, the initial value is reduced by 1.0% for every 1% decrease.

[0040] Furthermore, the calculation of the failure rate includes:

[0041] If it is a disposable device, the device failure rate is calculated as 1-e. -t / MTTF MTBF is the cumulative trouble-free continuous operating time divided by the number of failures.

[0042] If not evaluated as a one-time device, the equipment failure rate is 1-e. -t / MTBF MTTF stands for Continuous Operating Time before equipment replacement.

[0043] Furthermore, the health score of the subsystem is calculated, expressed by the formula:

[0044] k = w1 k1 + w2 k2 + w3k3

[0045] In the formula, w1, w2, and w3 all represent weight values; k1 is the defect coefficient, with the initial value reduced by 30% for each Class A defect, 10% for each Class B defect, and 0.5% for each Class C defect; k2 is the early warning coefficient, with the initial value reduced by 1% for each Level 3 abnormal early warning event, 5% for each Level 2 abnormal early warning event, and 10% for each Level 3 abnormal early warning event; k3 is the equipment operation coefficient, which is assigned the initial value minus the equipment failure rate of each key piece of equipment.

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

[0047] The final result, which reflects the health status of the equipment, is automatically calculated by comprehensively considering information such as the reliability, faults and defects, abnormal early warning, and power characteristic analysis of wind turbine equipment. This guides the operation and maintenance management of wind farms, encourages the active implementation of corresponding maintenance work, controls the risks of wind turbine equipment, improves equipment reliability, and achieves the effect of lean maintenance. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of the system described in this invention. Detailed Implementation

[0049] The present invention will now be described in more detail with reference to the embodiments.

[0050] Example 1

[0051] like Figure 1 As shown, a wind farm equipment lifecycle health management system includes the following steps:

[0052] The wind turbine monitoring module acquires and displays the operating parameters of each wind turbine and the parameters of the wind turbine subsystems. The operating parameters include wind turbine status, wind turbine model, rated power, daily power generation, utilization hours, mean time between failures (MTBF), mean time to repair (MTBF), unit alarms, active power, wind speed, etc. The wind turbine subsystems include the pitch control system, drive train, generator system, nacelle environment, etc.

[0053] The alarm management module acquires and displays alarm events, which are divided into four levels. The processing status of alarm events is divided into activated, cleared, confirmed, and unconfirmed.

[0054] The defect management module manages defect work orders, including: Defect Registration – The registrant has the right to modify or delete the defects they have registered. Defect Handling – The defect handler accepts the registered defect orders, investigates, and eliminates the defects. For handleable defects, a work order is generated to complete the elimination; for defects that cannot be handled immediately, a defect retention application is submitted; if the defect registration is incorrect, a defect cancellation application is submitted. Defect Acceptance – Registered defects are accepted by associated work orders; Defect Expiration – If a defect handler fails to handle a registered defect within 24 hours of its registration, they must accept the order and handle it within 24 hours; Defect Retention – If a registered defect cannot be shut down due to unit shutdown issues, lack of spare parts, or if it does not currently affect the unit's normal operation, the defect handler initiates a defect retention application, which can be added to planned maintenance for processing; Defect Cancellation – If a defect handler investigates a registered defect and finds that the actual situation does not match the registered defect description, they can initiate a defect cancellation application.

[0055] The equipment health early warning module is used to calculate the health score of each fan and each fan subsystem.

[0056] The operation and maintenance management module features a device tree structure built according to equipment location and system hierarchy, a maintenance record database, and an equipment change record database. Systems, subsystems, equipment, and components can be added through the device tree structure. For example: Wind Farm - #1 Wind Turbine - Yaw System - #1 Yaw Motor (#2 Yaw Motor, #3 Yaw Motor).

[0057] Example 2

[0058] Calculating the health score of the wind turbine includes the following steps:

[0059] Count the number of defects of different levels in each subsystem of the wind turbine;

[0060] Count the number of early warnings at different levels for each subsystem of the wind turbine;

[0061] Calculate the compliance rate of the wind turbine power curve;

[0062] Based on the number of defects, the number of warnings, and the compliance rate, the health score of a single wind turbine is calculated, expressed by the formula:

[0063] k = w1 k1 + w2 k2 + w3 k3

[0064] In the formula, w1, w2, and w3 all represent weight values; in this embodiment, w1 = 0.4, w2 = 0.4, and w3 = 0.2. k1 is the defect coefficient, with an initial value of 1. For each Class A defect, the initial value is reduced by 30%; for each Class B defect, the initial value is reduced by 10%; and for each Class C defect, the initial value is reduced by 0.5% (the corresponding defect is removed upon completion of defect elimination acceptance). k2 is the early warning coefficient, with an initial value of 1. For each Level 3 abnormal early warning event, the initial value is reduced by 1%; for each Level 2 abnormal early warning event, the initial value is reduced by 5%; and for each Level 3 abnormal early warning event, the initial value is reduced by 10%. k3 is the power coefficient, with an initial value of 1. When the power curve compliance rate is greater than 90%, the power coefficient remains at 1; when the power curve compliance rate is less than 90%, the initial value is reduced by 1.0% for every 1% decrease.

[0065] The health score of all wind turbines is calculated as the average of the individual wind turbine health scores.

[0066] Based on health scores, health levels are divided into four categories: healthy, sub-healthy, attentive, and high-risk. Specifically:

[0067] Health – Health score greater than 0.8;

[0068] Sub-health – Health level between 0.7 and 0.8;

[0069] Note – Health level should be between 0.6 and 0.7;

[0070] High risk – Health level less than 0.6.

[0071] Example 3

[0072] Calculating the health score of the subsystem includes the following steps:

[0073] The subsystem contains several devices; key devices are selected to participate in the subsystem's health assessment.

[0074] The number of defects of different levels in each key piece of equipment in the statistical subsystem;

[0075] The number of different warnings for each key device in the statistical subsystem;

[0076] Obtain the continuous operating time (i.e., the continuous operating time of the equipment within the interval between two failures (excluding downtime), in hours) and the number of failures for each key piece of equipment, and calculate the following indicators for each piece of equipment:

[0077] MTBF = Cumulative Trouble-Free Continuous Operation Time / Number of Failures, specifically the average value of similar equipment, used to assess repairable equipment, unit: hours;

[0078] MTTF = Continuous operating time before equipment replacement, specifically the average value of similar equipment, for evaluating one-time equipment, in hours;

[0079] If it is a one-time equipment assessment, then the equipment failure rate = 1 - e -t / MTTF If the equipment failure rate is not a one-time assessment, then the equipment failure rate = 1 - e -t / MTBF ;

[0080] Based on the number of defects, the number of warnings, and the equipment failure rate, the health score of the subsystem is calculated, expressed by the formula:

[0081] k = w1 k1 + w2 k2 + w3k3

[0082] In the formula, w1, w2, and w3 all represent weight values; in this embodiment, w1 = 0.4, w2 = 0.3, and w3 = 0.3. k1 is the defect coefficient, with an initial value of 1. For each Class A defect, the initial value is reduced by 30%; for each Class B defect, the initial value is reduced by 10%; and for each Class C defect, the initial value is reduced by 0.5% (the corresponding defect is removed upon completion of defect elimination acceptance). k2 is the early warning coefficient, with an initial value of 1. For each Level 3 abnormal early warning event, the initial value is reduced by 1%; for each Level 2 abnormal early warning event, the initial value is reduced by 5%; and for each Level 3 abnormal early warning event, the initial value is reduced by 10%. k3 is the equipment operation coefficient, with an initial value of 1, minus the equipment failure rate of each key piece of equipment participating in the evaluation.

[0083] Example 4

[0084] Based on the operational data of relevant measuring points of the equipment, determine the current expected value of a certain measuring point, and calculate the residual value between the expected value and the actual value of the measuring point. Based on the preset residual interval, issue an early warning and provide the warning level.

[0085] Taking the abnormal temperature warning of high-speed bearings in gearboxes as an example, a predictive model for the temperature of high-speed bearings in gearboxes is established using historical data on unit power, wind speed, nacelle temperature, generator speed, gear oil temperature, low-speed bearing temperature, and high-speed bearing temperature, employing machine learning algorithms. The normal distribution of the residuals in the prediction model is statistically analyzed, and a residual warning threshold of 6℃ is determined. The RMSE (Root Mean Square Error) of the residuals in the prediction model is statistically analyzed on a daily basis, and a value of 0.6℃ is determined. Under real-time conditions, a sliding window statistical analysis is used for the residuals. A Level 3 abnormality warning is issued if the residual exceeds 6℃ continuously; a Level 2 abnormality warning is issued if the daily residual RMSE exceeds 0.6℃; and a Level 1 abnormality warning is issued if the daily residual statistical analysis exceeds the RMSE for three consecutive days.

[0086] It should be noted that the functional units / modules in the various embodiments of the present invention can be integrated into one processing unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated into one unit / module. The integrated unit / module described above can be implemented in hardware or in the form of software functional units / modules.

[0087] From the above description of the embodiments, those skilled in the art will clearly understand that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code, or any suitable combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: Application-Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field-Programmable Gate Array (FPGA), processor, controller, microcontroller, microprocessor, other electronic units designed to implement the functions described herein, or combinations thereof. For software implementation, some or all of the processes of the embodiments can be implemented by a computer program instructing the associated hardware. During implementation, the program can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of a computer program from one place to another. Storage media can be any available medium accessible to a computer. Computer-readable media can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code having the form of instructions or data structures and accessible to a computer.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should be able to analyze that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the essence and scope of the technical solutions of the present invention.

Claims

1. A wind farm equipment full life cycle health management system, characterized in that, Comprise: Fan monitoring module, for obtaining and displaying the operating parameters of each fan, fan subsystem parameters; Alarm management module, for obtaining and displaying alarm events; Device health warning module, for calculating the health score value of each fan and fan subsystem; Ledger management module, provided with a device tree structure constructed according to the device location and system affiliation, and the system, subsystem, device and component are added through the device tree structure; Wherein, the health score value of the fan is calculated, comprising the following steps: Counting the number of different levels of defects of each subsystem of the fan; Counting the number of different levels of early warning of each subsystem of the fan; Calculate the compliance rate of the fan power curve; According to the number of defects, the number of early warnings and the compliance rate, the health score value of a single fan is calculated; Wherein, the health score value of a single fan is calculated, expressed in formula as: k = w1 k1+ w2 k2+ w3 k3; In the formula, w1, w2, w3 all represent weight values; k1 is the defect coefficient, the initial value is reduced by 30% for each A class defect fault; the initial value is reduced by 10% for each B class defect; the initial value is reduced by 0.5% for each C class defect; k2 is the early warning coefficient, the initial value is reduced by 1% for each three-level abnormal early warning event; the initial value is reduced by 5% for each two-level abnormal early warning event; the initial value is reduced by 10% for each one-level abnormal early warning event; k3 is the power coefficient, the initial coefficient is 1, when the power curve compliance rate is greater than the threshold value, the power coefficient is the initial value; when the power curve compliance rate value is less than the threshold value, the initial value is reduced by 1.0% for each 1%.

2. A wind farm equipment full life cycle health management system, characterized in that, Comprise: Fan monitoring module, for obtaining and displaying the operating parameters of each fan, fan subsystem parameters; Alarm management module, for obtaining and displaying alarm events; Device health warning module, for calculating the health score value of each fan and fan subsystem; Ledger management module, provided with a device tree structure constructed according to the device location and system affiliation, and the system, subsystem, device and component are added through the device tree structure; Wherein, the health score value of the fan subsystem is calculated, comprising the following steps: Selecting the key equipment in the subsystem; Counting the number of different levels of defects of each key equipment in the subsystem; Counting the number of different levels of early warning of each key equipment in the subsystem; According to the continuous operation time and the number of failures of each key equipment, the failure rate of each equipment is calculated; According to the number of defects, the number of early warnings and the failure rate of the equipment, the health score value of the subsystem is calculated; Wherein, the failure rate is calculated, comprising: If it is a single-use device, the computing device failure rate is 1 - e -t / MTTF , where MTTF is the continuous operating time before device replacement. If not a one-time device evaluation, then the device failure rate is 1 - e -t / MTBF , MTBF is the cumulative failure-free continuous operating time / failure number; Wherein, the health score value of the subsystem is calculated, expressed in formula as: k = w1 k1+ w2 k2+ w3k3; In the formula, w1, w2, w3 all represent weight values; k1 is a defect coefficient, and the initial value is reduced by 30% for each A-class defect fault; the initial value is reduced by 10% for each B-class defect; the initial value is reduced by 0.5% for each C-class defect; k2 is a pre-warning coefficient, and the initial value is reduced by 1% for each third-level abnormal pre-warning event; the initial value is reduced by 5% for each second-level abnormal pre-warning event; the initial value is reduced by 10% for each first-level abnormal pre-warning event; and k3 is a device operation coefficient, and is assigned as the initial value minus the device failure rate of each key device.

3. A wind farm equipment full life cycle health management method, characterized in that, The method comprises the following steps: calculating the health score value of the fan, comprising: counting the number of defects of different levels of each subsystem of the fan; counting the number of pre-warnings of different levels of each subsystem of the fan; calculating the compliance rate of the fan power curve; and calculating the health score value of a single fan according to the number of defects, the number of pre-warnings and the compliance rate; calculating the health score value of the fan subsystem, comprising: selecting key devices in the subsystem; counting the number of defects of different levels of each key device in the subsystem; counting the number of pre-warnings of different levels of each key device in the subsystem; calculating the failure rate of each device according to the continuous operation time and the number of faults of each key device; and calculating the health score value of the subsystem according to the number of defects, the number of pre-warnings and the device failure rate; wherein the failure rate is calculated, comprising: If it is a single-use device, the computing device failure rate is 1 - e -t / MTTF , where MTTF is the continuous operating time before device replacement. If not a one-time device evaluation, then the device failure rate is 1 - e -t / MTBF , MTBF is the cumulative failure-free continuous operating time / failure number; wherein the health score value of the subsystem is calculated, expressed by a formula as: k = w1 k1+ w2 k2+ w3k3; In the formula, w1, w2, w3 all represent weight values; k1 is a defect coefficient, and the initial value is reduced by 30% for each A-class defect fault; the initial value is reduced by 10% for each B-class defect; the initial value is reduced by 0.5% for each C-class defect; k2 is a pre-warning coefficient, and the initial value is reduced by 1% for each third-level abnormal pre-warning event; the initial value is reduced by 5% for each second-level abnormal pre-warning event; the initial value is reduced by 10% for each first-level abnormal pre-warning event; and k3 is a device operation coefficient, and is assigned as the initial value minus the device failure rate of each key device.

Images