Operation and maintenance collaborative optimization control method and system based on historical fault data of draught fan

Abstract

The invention discloses an operation and maintenance collaborative optimization control method and system based on historical fault data of a fan, and the method comprises the steps: S1, obtaining the historical fault data of a wind turbine generator, obtaining a historical shutdown fault record of the wind turbine generator after data preprocessing, and storing the historical shutdown fault record in a database; establishing a Weibull distribution wind turbine generator set fault model according to the wind turbine generator set shutdown fault historical record; s2, predicting the future service trend of the wind turbine generator based on a Weibull distribution wind turbine generator fault model, and distributing a power generation reference instruction for each wind turbine generator according to the maintenance plan; and S3, based on the power generation reference instruction of each wind turbine generator and the power generation feasible region of the wind turbine generator, constructing a wind turbine generator group service quality multi-time scale collaborative optimization regulation and control framework so as to ensure safe and stable operation of a wind power plant and improve the power generation capacity at the same time. The method can ensure safe and stable operation of the wind power plant and improve the generating capacity.

Description

Technical Field

[0001] The present invention mainly relates to the field of wind power technology, and in particular to an operation and maintenance collaborative optimization control method and system based on historical fault data of a wind turbine. Background Art

[0002] As one of the most economically valuable green energy sources in today's society, wind energy has received widespread attention and vigorous development from countries around the world. The rapid development of wind power has brought huge market opportunities to the wind power equipment manufacturing industry. While wind power companies have achieved great development, they are also facing serious operation and maintenance difficulties. With the increase in the service life of the unit, wind turbine accidents have occurred frequently in recent years, and more and more problems such as wind turbine fires, tower collapses, and blade breakage have been exposed. Due to the diversity and complexity of the unit parts, its operation and maintenance are still difficult. In order to better ensure the stable, safe and economical operation of wind turbines, it is imperative to study the failure and operation and maintenance strategies of wind turbines. At the same time, it is also very important to reduce the unit's per-kilowatt-hour cost, which is related to the competitiveness of the wind power industry in the current energy structure. For the units that have been put into operation, increasing the power generation is one of the most effective ways to reduce the per-kilowatt-hour cost. In summary, repairing wind turbines and increasing the power generation of wind turbines are both important control objectives of wind turbines. However, natural wind has the characteristics of random fluctuations, which will cause fluctuations in wind power. If wind turbines are repaired when the wind power is high, it will cause power generation losses. Therefore, research on the optimization of wind turbine operation and maintenance plans that incorporate constraints on increasing the power generation of wind turbines plays an important role in increasing the power generation of wind turbines and ensuring the safe operation of wind turbines.

[0003] At present, most studies only consider the maintenance tasks of wind turbines within a certain period of time, and have not yet considered the loss of power generation due to the maintenance of wind turbines, resulting in varying degrees of over-maintenance and under-maintenance. Summary of the invention

[0004] In view of the technical problems existing in the prior art, the present invention provides an operation and maintenance collaborative optimization control method and system based on historical fault data of wind turbines, which can ensure the safe and stable operation of wind farms while improving power generation.

[0005] In order to solve the above technical problems, the technical solution proposed by the present invention is: An operation and maintenance collaborative optimization control method based on historical fault data of a wind turbine comprises the following steps: S1. Obtain historical fault data of wind turbines, obtain historical records of wind turbine shutdown faults after data preprocessing, and then establish a Weibull distribution wind turbine fault model based on the historical records of wind turbine shutdown faults; S2, predict the future service trend of wind turbines based on the wind turbine fault model of Weibull distribution, and then assign power generation reference instructions to each wind turbine according to the maintenance plan; S3. Based on the reference instructions for power generation of each wind turbine and the feasible domain of power generation of the wind turbine, a multi-time scale collaborative optimization and control framework for the service quality of the wind turbine group is constructed to ensure the safe and stable operation of the wind farm while increasing power generation.

[0006] Preferably, in step S1, in the Weibull distribution wind turbine fault model, the failure rate of the wind turbine is defined according to the two-parameter Weibull theory: for:

[0007] in, For time, is the shape parameter, is the size parameter, the shape parameter of the Weibull distribution and size parameters It is estimated by fitting historical parameters; When the failure rate Over time Decreasing, the equipment is operating in the early failure period; when When the failure rate Over time Increasing, the equipment is operating in the performance degradation period, which is suitable for fault loss modeling; When the failure rate Not over time Changes, the equipment operates in a period of occasional failures.

[0008] Preferably, in step S1, in the Weibull distribution wind turbine fault model, the Weibull failure distribution function of the wind turbine is for:

[0009] The unknown parameters in the Weibull failure distribution function are estimated by least squares estimation.

[0010] Preferably, the specific process of estimating the unknown parameters in the Weibull failure distribution function by least squares parameter estimation theory is: Linearize the Weibull failure distribution function and deform the failure distribution function left and right:

[0011] Take the natural logarithm of both sides of the equation:

[0012] make , , , ; According to the least squares parameter estimation theory, the coefficient and The calculation formula is:

[0013] in , , is the total number of samples, for The corresponding Sample values, for The corresponding Sample values.

[0014] Preferably, in step S2, the specific process of allocating a power generation reference instruction to each wind turbine is as follows: If the cluster stability margin Longer than planned maintenance cycle , indicating that the stability of the fleet is good, and power generation is given priority; If the cluster stability margin Shorter than planned maintenance cycle , but greater than the fault threshold , indicating that the stability of the fleet is poor and load reduction is required; If the cluster stability margin Shorter than planned maintenance cycle , and is less than the fault threshold , indicating that the fleet is about to fail, and the maintenance plan needs to be advanced while avoiding wind energy loss; If the average wind power in the forecast time window Greater than high wind power threshold , then arrange maintenance after the forecast time window to avoid wind energy loss; if the average wind power in the forecast time window Less than high wind power threshold , then arrange maintenance immediately.

[0015] Preferably, if the cluster stability margin Longer than planned maintenance cycle , the corresponding lower limit of wind turbine power output is:

[0016] in, is the active power of the wind turbine under the maximum power point tracking strategy, Active power demand of the power grid; The calculation formula for the power output upper limit reference is:

[0017] Among them, the increase coefficient ; The upper limit of wind turbine power output is: .

[0018] Preferably, if the cluster stability margin Shorter than planned maintenance cycle , but greater than the fault threshold , the corresponding upper limit of wind turbine power output is:

[0019] The lower limit of wind turbine power output is:

[0020] Among them, the load reduction factor .

[0021] Preferably, in step S3, the multi-time scale collaborative optimization control framework of the service quality of the wind turbine group includes a medium- and long-term scale and a short-term scale; Medium- and long-term scale: Based on all historical data from the start of operation of the wind farm to the present, the service status of the wind farm is evaluated and the power generation strategy is planned every m days, and the evaluation results are considered to remain accurate within the next evaluation time window; Short-term scale: The service quality of the wind farm is regulated every n seconds. Based on the feasible domain of wind turbine power generation, a short-time scale control method for service quality for tower vibration suppression is established.

[0022] Preferably, in the short-time scale control method for service quality of tower vibration suppression, the displacement model of the tower top is:

[0023] in is the horizontal displacement of the tower top, is the lateral displacement of the tower top, is the tower height, is the horizontal thrust, is the concentrated stress in the tower height direction, is the wind wheel gravity, is the cabin gravity, is the aerodynamic torque, E is the elastic modulus of the tower structure material, is the section inertia moment of the tower structure at a distance Z; Z is the distance from the bottom of the tower structure; The objective function and constraints are:

[0024]

[0025] in is the active power of the wind turbine, is the active power reference value of the wind turbine, is the reactive power of the wind turbine, is the reactive power reference value of the wind turbine, is the horizontal displacement increment of the wind turbine, is the lateral displacement increment of the wind turbine, The power obtained by the lower health controller, and are the minimum reactive capacity and maximum reactive capacity of the lower health controller, is the number of wind turbines, is the total number of sampling steps within the forecast time horizon, The wind turbine group number, For the A sampling moment.

[0026] The present invention also discloses an operation and maintenance collaborative optimization control system based on historical fault data of wind turbines, comprising an interconnected memory and a processor, wherein a computer program is stored on the memory, and when the computer program is run by the processor, the steps of the method described above are executed.

[0027] Compared with the prior art, the advantages of the present invention are: The present invention establishes a Weibull distributed wind turbine fault model based on the historical records of wind turbine shutdown failures obtained after data preprocessing; predicts the future service trend of the monitored wind turbine based on the Weibull failure distribution curve, and allocates power generation reference instructions to each wind turbine according to the maintenance plan; based on the feasible domain of wind turbine power generation, establishes a multi-time scale collaborative optimization and control framework for the service quality of wind turbine groups, to ensure the safe and stable operation of wind farms while increasing power generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a flow chart of an embodiment of the operation and maintenance collaborative optimization control method based on historical fault data of wind turbines of the present invention.

[0029] Figure 2 A flow chart of obtaining power generation reference instructions considering service status and maintenance plan in the present invention.

[0030] Figure 3 It is a schematic diagram of the multi-time scale collaborative optimization and control framework of the service quality of a wind turbine group in the present invention. DETAILED DESCRIPTION

[0031] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0032] like Figure 1 As shown, the operation and maintenance collaborative optimization control method based on historical fault data of wind turbines provided by the embodiment of the present invention includes the following steps: S1. Obtain historical fault data of wind turbines, obtain the historical records of wind turbine shutdown faults after data preprocessing, and establish a Weibull distributed wind turbine fault model; Specifically, based on the historical fault data of wind turbines, data preprocessing is performed to remove abnormal data. Because wind turbines will generate a large amount of fault data during operation, some of which are normal fault data, and there will also be many abnormal data. In order to prevent abnormal data from interfering with the fault database, it is necessary to remove abnormal data and form a historical record of wind turbine shutdown faults. Based on the historical records of wind turbine shutdown failures, a Weibull distributed wind turbine failure model is established; Defining the failure rate of wind turbines based on the two-parameter Weibull theory for:

[0033] in, For time, is the shape parameter, is the size parameter, the shape parameter of the Weibull distribution and size parameters It is estimated by fitting historical parameters. When the failure rate Over time Decreasing, the equipment is operating in the early failure period; when When the failure rate Over time Increasing, the equipment is operating in the performance degradation period, which is suitable for fault loss modeling; When the failure rate Not over time Changes, the equipment operates in a period of occasional failures.

[0034] The Weibull failure distribution function of a wind turbine is:

[0035] Based on the wind turbine fault model of Weibull distribution, the least squares estimation is applied to the parameter estimation of the two-parameter Weibull distribution to estimate the unknown parameters in the linear function. Specifically, the Weibull failure distribution function is linearized and the failure distribution function is deformed left and right:

[0036] Take the natural logarithm of both sides of the equation:

[0037] make , , , , then the above formula can be rewritten as According to the least squares parameter estimation theory, the coefficients of the above linear regression equation are and The calculation formula is:

[0038] in , , is the total number of samples, for The corresponding Sample values, for The corresponding Sample values.

[0039] For small sample historical fault data, the empirical distribution function is obtained by the approximate median rank method:

[0040] in, The time when the fault occurred.

[0041] Based on the fault occurrence time and the empirical distribution function, a regression line with the minimum deviation from the historical discrete data points is obtained through parameter estimation to fit the fault shutdown trend of the wind turbine.

[0042] S2, predicting the future service trend of the wind turbine based on the Weibull failure distribution curve, and then assigning power generation reference instructions to each wind turbine according to the maintenance plan; Based on the Weibull failure distribution curve, the future service trend of the monitored wind turbine is predicted. The Weibull failure distribution curve contains both the historical records within the sampling time and the future prediction trend, which can provide guidance for the formulation of maintenance strategies; when the shape coefficient of the failure distribution curve is greater than 1, it means that the wind turbine system is in the failure degradation period; based on the shape coefficient of the failure distribution curve, a maintenance threshold is set to arrange the maintenance time of the wind turbine group. The service stability margin of the wind turbine group is the period before the shape coefficient rises to the maintenance threshold, and a fault threshold is set. If the shape coefficient of the curve rises to the fault threshold, it means that the wind turbine group is about to fail.

[0043] like Figure 2 As shown in Figure 2, based on the service stability margin of the wind turbine group and maintenance planning cycles , assign power generation reference instructions to each wind turbine; if the third-party operation and maintenance company sets the next scheduled maintenance time within the planned maintenance cycle after comprehensively analyzing factors such as logistics support and weather conditions After that, the power generation reference instructions are assigned to each wind turbine according to the following logic: 1) If the cluster stability margin Longer than planned maintenance cycle , indicating that the stability of the fleet is good, and the power generation is given priority. The lower limit of wind turbine power output is:

[0044] in, is the active power of the wind turbine under the maximum power point tracking (MPPT) strategy, Active power demand of the power grid.

[0045] The calculation formula for the power output upper limit reference is:

[0046] Among them, the increase coefficient .

[0047] The upper limit of wind turbine power output is:

[0048] 2) If the cluster stability margin Shorter than planned maintenance cycle , but greater than the fault threshold , indicating that the stability of the fleet is poor and load reduction is required.

[0049] The upper limit of wind turbine power output is:

[0050] The lower limit of wind turbine power output is:

[0051] Among them, the load reduction factor .

[0052] If the stability margin of the fleet is lower than the fault threshold, it means that the fleet is about to fail and the maintenance plan needs to be advanced while avoiding wind energy loss.

[0053] If the average wind power in the forecast time window Greater than high wind power threshold , then arrange maintenance after the forecast time window to avoid wind energy loss; if the average wind power in the forecast time window Less than high wind power threshold , then arrange maintenance immediately.

[0054] S3. Based on the feasible domain of wind turbine power generation, a multi-time scale collaborative optimization and control framework for the service quality of wind turbine groups is constructed to ensure the safe and stable operation of wind farms while increasing power generation.

[0055] like Figure 3 As shown in the figure, a multi-time scale collaborative optimization control framework for the service quality of wind turbine groups is proposed based on the wind turbine generation reference instructions and maintenance plans, including: Medium- and long-term scale: The central controller conducts the above-mentioned wind farm service status assessment and power generation strategy planning every 240 hours (10 days) based on all historical data from the start of operation of the wind farm to the present, and considers that the assessment results remain accurate within the next assessment time window; Short-term scale: The central controller controls the service quality of the wind farm every 30 seconds. It mainly establishes a short-term scale control method for service quality for tower vibration suppression based on the feasible domain of wind turbine power generation. The displacement model of the tower top is:

[0056] in is the horizontal displacement of the tower top, is the lateral displacement of the tower top, is the tower height, is the horizontal thrust, is the concentrated stress in the tower height direction, is the wind wheel gravity, is the cabin gravity, is the aerodynamic torque, E is the elastic modulus of the tower structure material, is the section inertia moment of the tower structure at a distance Z; Z is the distance from the bottom of the tower structure.

[0057] The objective function and constraints are:

[0058]

[0059] in is the active power of the wind turbine, is the active power reference value of the wind turbine, is the reactive power of the wind turbine, is the reactive power reference value of the wind turbine, is the horizontal displacement increment of the wind turbine, is the lateral displacement increment of the wind turbine, The power obtained by the lower health controller, and are the minimum reactive capacity and maximum reactive capacity of the lower health controller, is the number of wind turbines, is the total number of sampling steps within the forecast time horizon; The wind turbine group number, For the A sampling moment.

[0060] The globally optimal power output is solved, and the reference power generation range of the wind turbine set formulated by medium- and long-term service quality control is used as the constraint condition of the optimization problem to meet the safe and stable operation requirements of the wind farm.

[0061] The present invention establishes a Weibull distributed wind turbine fault model based on the historical records of wind turbine shutdown failures obtained after data preprocessing; predicts the future service trend of the monitored wind turbine based on the Weibull failure distribution curve, and allocates power generation reference instructions to each wind turbine according to the maintenance plan; based on the feasible domain of wind turbine power generation, establishes a multi-time scale collaborative optimization and control framework for the service quality of wind turbine groups, to ensure the safe and stable operation of wind farms while increasing power generation.

[0062] An embodiment of the present invention also provides an operation and maintenance collaborative optimization control system based on historical fault data of a wind turbine, comprising an interconnected memory and a processor, wherein a computer program is stored on the memory, and when the computer program is run by the processor, the steps of the method described above are executed.

[0063] The present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by hardware related to computer program instructions. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned method embodiment can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. Computer-readable storage media include: any entity or device that can carry computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. The memory is used to store computer programs and / or modules. The processor implements various functions by running or executing computer programs and / or modules stored in the memory, and calling data stored in the memory. The memory may include a high-speed random access memory and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0064] The above are only preferred embodiments of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should be regarded as the protection scope of the present invention.

Claims

1. An operation and maintenance collaborative optimization control method based on historical fault data of wind turbines, characterized in that: Includes steps: S1. Obtain historical fault data of wind turbines, obtain historical records of wind turbine shutdown faults after data preprocessing, and then establish a Weibull distribution wind turbine fault model based on the historical records of wind turbine shutdown faults; S2, predict the future service trend of wind turbines based on the wind turbine fault model of Weibull distribution, and then assign power generation reference instructions to each wind turbine according to the maintenance plan; S3. Based on the reference instructions for power generation of each wind turbine and the feasible domain of power generation of the wind turbine, a multi-time scale collaborative optimization and control framework for the service quality of the wind turbine group is constructed to ensure the safe and stable operation of the wind farm while increasing power generation.

2. The operation and maintenance collaborative optimization control method based on wind turbine historical fault data according to claim 1 is characterized in that: In step S1, in the Weibull distribution wind turbine fault model, the failure rate of the wind turbine is defined according to the two-parameter Weibull theory: for: in, For time, is the shape parameter, is the size parameter, the shape parameter of the Weibull distribution and size parameters It is estimated by fitting historical parameters; When the failure rate Over time Decreasing, the equipment is operating in the early failure period; when When the failure rate Over time Increasing, the equipment is operating in the performance degradation period, which is suitable for fault loss modeling; When the failure rate Not over time Changes, the equipment operates in a period of occasional failures.

3. The operation and maintenance collaborative optimization control method based on wind turbine historical fault data according to claim 2 is characterized in that: In step S1, in the Weibull distribution wind turbine failure model, the Weibull failure distribution function of the wind turbine is for: The unknown parameters in the Weibull failure distribution function are estimated by least squares estimation.

4. The operation and maintenance collaborative optimization control method based on wind turbine historical fault data according to claim 3 is characterized in that: The specific process of estimating the unknown parameters in the Weibull failure distribution function by using the least squares parameter estimation theory is: Linearize the Weibull failure distribution function and deform the failure distribution function left and right: Take the natural logarithm of both sides of the equation: make , , , ; According to the least squares parameter estimation theory, the coefficient and The calculation formula is: in , , is the total number of samples, for The corresponding Sample values, for The corresponding Sample values.

5. The operation and maintenance collaborative optimization control method based on wind turbine historical fault data according to any one of claims 1 to 4, characterized in that: In step S2, the specific process of allocating power generation reference instructions to each wind turbine is as follows: If the cluster stability margin Longer than planned maintenance cycle , indicating that the stability of the fleet is good, and power generation is given priority; If the cluster stability margin Shorter than planned maintenance cycle , but greater than the fault threshold , indicating that the stability of the fleet is poor and load reduction is required; If the cluster stability margin Shorter than planned maintenance cycle , and is less than the fault threshold , indicating that the fleet is about to fail, and the maintenance plan needs to be advanced while avoiding wind energy loss; If the average wind power in the forecast time window Greater than high wind power threshold , then arrange maintenance after the forecast time window to avoid wind energy loss; if the average wind power in the forecast time window Less than high wind power threshold , then arrange maintenance immediately.

6. The operation and maintenance collaborative optimization control method based on wind turbine historical fault data according to claim 5 is characterized in that: If the cluster stability margin Longer than planned maintenance cycle , the corresponding lower limit of wind turbine power output is: in, is the active power of the wind turbine under the maximum power point tracking strategy, Active power demand of the power grid; The calculation formula for the power output upper limit reference is: Among them, the increase coefficient ; The upper limit of wind turbine power output is: 。 7. The operation and maintenance collaborative optimization control method based on wind turbine historical fault data according to claim 6 is characterized in that: If the cluster stability margin Shorter than planned maintenance cycle , but greater than the fault threshold , the corresponding upper limit of wind turbine power output is: The lower limit of wind turbine power output is: Among them, the load reduction factor .

8. The operation and maintenance collaborative optimization control method based on wind turbine historical fault data according to any one of claims 1 to 4, characterized in that: In step S3, the multi-time scale collaborative optimization control framework of the wind turbine group service quality includes medium- and long-term scales and short-term scales; Medium- and long-term scale: Based on all historical data from the start of operation of the wind farm to the present, the service status of the wind farm is evaluated and the power generation strategy is planned every m days, and the evaluation results are considered to remain accurate within the next evaluation time window; Short-term scale: The service quality of the wind farm is regulated every n seconds. Based on the feasible domain of wind turbine power generation, a short-time scale control method for service quality for tower vibration suppression is established.

9. The operation and maintenance collaborative optimization control method based on wind turbine historical fault data according to claim 8 is characterized in that: In the short-time scale control method of service quality for tower vibration suppression, the displacement model of the tower top is: in is the horizontal displacement of the tower top, is the lateral displacement of the tower top, is the tower height, is the horizontal thrust, is the concentrated stress in the tower height direction, is the wind wheel gravity, is the cabin gravity, is the aerodynamic torque, E is the elastic modulus of the tower structure material, is the section inertia moment of the tower structure at a distance Z; Z is the distance from the bottom of the tower structure; The objective function and constraints are: in is the active power of the wind turbine, is the active power reference value of the wind turbine, is the reactive power of the wind turbine, is the reactive power reference value of the wind turbine, is the horizontal displacement increment of the wind turbine, is the lateral displacement increment of the wind turbine, The power obtained by the lower health controller, and are the minimum reactive capacity and maximum reactive capacity of the lower health controller, is the number of wind turbines, is the total number of sampling steps within the forecast time horizon, The wind turbine group number, For the A sampling moment.

10. An operation and maintenance collaborative optimization control system based on historical fault data of a wind turbine, comprising a memory and a processor connected to each other, wherein a computer program is stored in the memory, characterized in that: When the computer program is executed by a processor, the computer program performs the steps of the method according to any one of claims 1 to 9.

Images