CN117646708B - Method for monitoring and analyzing states of offshore wind turbine generator - Google Patents
Method for monitoring and analyzing states of offshore wind turbine generator Download PDFInfo
- Publication number
- CN117646708B CN117646708B CN202410128274.2A CN202410128274A CN117646708B CN 117646708 B CN117646708 B CN 117646708B CN 202410128274 A CN202410128274 A CN 202410128274A CN 117646708 B CN117646708 B CN 117646708B
- Authority
- CN
- China
- Prior art keywords
- fan
- fitting
- blade
- bezier curve
- monitoring
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 68
- 238000012544 monitoring process Methods 0.000 title claims abstract description 50
- 230000008569 process Effects 0.000 claims abstract description 25
- 230000008859 change Effects 0.000 claims abstract description 23
- 238000004458 analytical method Methods 0.000 claims abstract description 18
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 17
- 238000012423 maintenance Methods 0.000 claims abstract description 10
- 230000033001 locomotion Effects 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- 230000001808 coupling effect Effects 0.000 claims description 4
- 238000003860 storage Methods 0.000 claims description 3
- 238000004260 weight control Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 6
- 238000012545 processing Methods 0.000 abstract description 3
- 230000005540 biological transmission Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000009795 derivation Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000010354 integration Effects 0.000 description 3
- 230000008447 perception Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000010365 information processing Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 238000010146 3D printing Methods 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000001934 delay Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000003534 oscillatory effect Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/72—Wind turbines with rotation axis in wind direction
Abstract
The invention discloses a method for monitoring and analyzing the state of an offshore wind turbine, which comprises the following steps: the method comprises the steps of respectively obtaining strain sensing information of the middle part and the tail end of a fan blade by using two strain sensors, and obtaining x-axis pose sensing information reflecting load change of the tail end of the fan blade by using pose sensors; forming a three-dimensional array according to two groups of strain sensing information and x-axis pose sensing information obtained from the middle and tail ends of the fan blade, and determining control points for performing Bezier curve fitting of the fan; fitting by using a random Bezier curve fitting algorithm according to the control points to generate a fan Bezier curve model; in the fitting process, random weights are introduced to adjust weight parameters of all control points, and the controllable fitting of the axial curve of the fan blade is realized through the adjustment of the weights. The method for monitoring and analyzing the states of the offshore wind turbine carries out intelligent comprehensive processing on the states and operation and maintenance information of various wind turbines, and achieves good offshore wind power state analysis and monitoring effects.
Description
Technical Field
The invention relates to the field of offshore wind power, in particular to a method for monitoring and analyzing states of an offshore wind turbine.
Background
In the energy crisis big background that the world non-renewable energy is gradually exhausted, the development of renewable energy is an important problem to be solved urgently. Among different renewable energy sources, wind power has great potential due to relatively high technical maturity, rich availability and relatively low environmental impact. Wind power plants may be categorized as either onshore or offshore wind farms depending on their location. With the rapid growth of wind power demand and the exhaustion of land resources in the past decade, offshore wind turbines have become an important point of wind power technology development. Compared with the land wind turbine, the offshore wind turbine has the advantages of rich wind resources, low turbulence, large erection space, small power transmission and distribution loss, small visual impact, small noise pollution and the like.
It is worth mentioning that the basic unit of operation of a wind farm is a wind turbine, whether it is an onshore wind turbine or an offshore wind turbine. The energy harvesting of conventional wind turbines is achieved by converting the kinetic energy of the wind into mechanical energy through blade rotation, and then into electrical energy through generators. Meanwhile, due to the influence of severe change of the offshore wind environment and difficulty in information monitoring, OWTs (oscillatory wave test systems) which cannot correspondingly adjust the severe change environment are extremely easy to generate low power generation efficiency and high component failure rate under an unreasonable fan strategy. Therefore, in order to improve the wind energy utilization rate of the offshore wind turbine to the environment, and reduce component faults caused by abnormal wind conditions, an accurate real-time environment and fan blade state monitoring system and an information supporting intelligent operation and maintenance system are very important. However, the existing state monitoring and environment sensing for the state analysis of the offshore wind turbine generator set face a great challenge, and the main reasons are that the monitoring system mostly depends on an internal electromechanical system, the information source is single, and the characteristics of the sensor network are limited, so that the information can be acquired and transmitted only by means of a severe offshore local area network, and the effect is poor in the actual offshore wind turbine generator set environment, so that the information monitoring is not specific and timely, and further various wind turbine generator faults and damage condition monitoring delays are serious. It should be noted that the information disclosed in the above background section is only for understanding the background of the present application and thus may include information that does not form the prior art that is already known to those of ordinary skill in the art.
Disclosure of Invention
The invention provides a method for monitoring and analyzing states of an offshore wind turbine, which aims to solve the problem that the information monitoring effect of the existing offshore wind turbine monitoring system is poor.
The technical problems of the invention are solved by the following technical scheme:
a monitoring and analyzing method for the state of an offshore wind turbine comprises the following steps: s1, strain sensing information of the middle part and the tail end of a fan blade is obtained by using two strain sensors respectively, and x-axis pose sensing information reflecting load change of the tail end of the fan blade is obtained by using pose sensors; s2, forming a three-dimensional array according to two groups of strain sensing information obtained from the middle and tail ends of the fan blade and the x-axis pose sensing information, and determining control points for performing Bezier curve fitting of the fan; s3, fitting is carried out by using a random Bezier curve fitting algorithm according to the control points, and a fan Bezier curve model is generated; in the fitting process, random weights are introduced to adjust weight parameters of all control points, and the controllable fitting of the axial curve of the fan blade is realized through the adjustment of the weights.
In some embodiments, the pose sensor comprises a gyroscope, or a combination of a gyroscope and an accelerometer.
In some embodiments, in step S3, according to the two sets of strain sensing information and the three-dimensional array formed by the x-axis pose sensing information, slope information of different control point positions of the bezier curve is reflected, a three-order bezier curve fitting algorithm is adopted to perform bezier curve fitting, and weight adjustment of the control points corresponding to each data dimension is performed through random weight control curve generation.
In some embodiments, the expression of the weight introduction method is as follows:
wherein B represents a Bezier curve, w i Weight coefficients representing different control point positions, b i For sensing information, t represents curve variable parameter, n represents the number of control points, B i,n (t) represents a Bezier curve at control point i; the maximum fitting of the axial curve of the fan blade is realized through weight adjustment.
In some embodiments, in step S3, the fan bezier curve model obtained by preliminary fitting is applied to an actual fan operation process, online model parameter adjustment iteration is performed through a random bezier curve algorithm, the fitting degree is determined in each iteration process, when the fitting degree error is greater than a predetermined threshold value, the weight of the control point corresponding to each data dimension is continuously adjusted according to the fitting degree gradient until the fitting degree error is determined to be less than the predetermined threshold value, and finally, the parameters of the fan bezier curve model are determined.
In some embodiments, the monitoring data for building and training a model obtained from the strain sensing information and the pose sensing information includes: the deformation of the leading edge of the blade, the tip offset of the blade, the degree of deflection of the blade relative to the plane of the fan blade, the degree of twist of the middle and end of the blade, the amount of twist of the entire blade, and the observation point location of the sensor.
In some embodiments, further comprising: and taking the y-axis and z-axis pose sensing information of the tail end of the fan blade obtained by the pose sensor as auxiliary operation and maintenance data of the model, and not directly participating in load analysis.
In some embodiments, further comprising: and analyzing the coupling effect of the blade and the wind environment from the time domain according to the time domain change generated by the signals received by the sensing caused by the movement of the fan blade.
In some embodiments, further comprising: and acquiring sensing information of a plurality of single fans of the offshore wind field, establishing a change relation of the integral offshore wind field according to the fan models and the position conditions and wind environment information among the fans, and determining the distribution of the integral offshore wind field.
A computer readable storage medium storing a computer program which, when executed by a processor, implements the above-mentioned method for monitoring and analyzing states of an offshore wind turbine.
Compared with the prior art, the invention has the beneficial effects that:
the invention provides a monitoring analysis method of the state of an offshore wind turbine, which comprises the steps of for a single fan blade, respectively acquiring strain sensing information of the middle part and the tail end of the fan blade by using two strain sensors, acquiring x-axis pose sensing information reflecting load change of the tail end of the fan blade by using pose sensors, and determining a control point for carrying out Bezier curve fitting of the fan by using a three-dimensional array formed by the sensing information; fitting is carried out by using a random Bezier curve fitting algorithm according to the control points, so as to generate a fan Bezier curve model; in the fitting process, the invention introduces random weights to adjust the weight parameters of each control point, and realizes the controllable fitting of the axial curve of the fan blade through the adjustment of the weights. Therefore, the invention realizes modeling and information deduction of the fan blade by combining the sensing information based on the fan Bezier curve model, completes complete state pose sensing of the single fan blade, and achieves good offshore wind power state analysis and monitoring effects.
In the invention, the strain sensor and the pose sensor are used as basic monitoring units of a sensing network, the whole pose and specific distortion condition of the blade can be sensed, the deduction can be carried out by means of the real-time observation state information obtained by the sensor on the boundary condition of the known fan blade characteristic, and the accurate modeling of the real-time fan state is completed by iterative fitting. The fitted Bezier curve can well represent pose information and deformation degree of a single blade, and the controllable fitting of the medial axis curve can be well realized by introducing random weights and adjusting the weights of different data dimensions. According to the method, the integral pose state is determined first, and then the accurate model is fitted, so that the prior information of the fan blade can be used for completing the model visualization process of the sensing information, the effectiveness and the readability of the sensing information are greatly improved, and good offshore wind power state analysis and monitoring effects are achieved according to the fan state and the operation and maintenance information, so that the safe and normal operation of an offshore fan is guaranteed.
The invention can establish a perfect state sensing network for a single fan, collect basic information of fan blades, and facilitate the iteration of the operation and sensing system of the offshore wind turbine. The integration of basic information is realized through the sensor network base station of the single fan, and the state of the single fan is accurately modeled in real time on the premise of the prior information of the fan blade. Meanwhile, the method can analyze the plane wind field of the fan blade according to the time-varying modeling results of the three blades of the fan. Based on the monitoring data of the sensing system, the real-time model monitoring of the fan blade is realized. Compared with the prior art, the invention is not limited to a simple communication integrated system, but utilizes the real-time state of the fan to carry out real-time accurate modeling and perception monitoring of the fan, and can effectively solve the problems of high operation and maintenance cost, difficult control, simple cluster fan state perception information, high error rate and the like of the offshore fan operation and maintenance system.
Other advantages of embodiments of the present invention are further described below.
Drawings
FIG. 1 is a flow chart of a method for monitoring and analyzing states of an offshore wind turbine in an embodiment of the invention.
Fig. 2 is a flow chart of a fan operation state analysis according to an embodiment of the present invention.
FIG. 3 is a schematic diagram of a sensing system for a fan blade in accordance with an embodiment of the present invention.
Fig. 4 is a flowchart of bezier curve fitting in an embodiment of the invention.
FIG. 5a is a side view of a fan blade in an embodiment of the present invention.
Fig. 5b and 5c are fan bezier curve modeling diagrams.
Fig. 6 is a schematic view of a fan blade rotation in an embodiment of the present invention.
FIG. 7 is a schematic view of a stroke environment derivation in accordance with an embodiment of the present invention.
Fig. 8 is a schematic diagram of a three-dimensional model of a wind turbine in accordance with an embodiment of the present invention.
FIG. 9 is a schematic diagram of a sensor system model of a single fan blade in an embodiment of the present invention.
FIG. 10 is a schematic diagram of a sensing system according to an embodiment of the present invention.
The reference numerals are as follows:
1 a strain sensor; 2 pose sensor.
Detailed Description
The invention will be further described with reference to the following drawings in conjunction with the preferred embodiments. It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other.
It should be noted that, in this embodiment, the terms of left, right, upper, lower, top, bottom, etc. are merely relative terms, or refer to the normal use state of the product, and should not be considered as limiting. The embodiment of the invention provides a method for monitoring and analyzing the state of an offshore wind turbine, which comprises the following steps: the method comprises the steps that a sensing system is formed by a strain sensor, a blade tip gyroscope and an accelerometer which are distributed in the upper middle of a fan blade, and the sensing system is used for sensing the state change of the fan blade caused by the offshore environment and acquiring original data; the sensing base station of the single fan is used for summarizing and analyzing sensing information, a Bezier curve model of the fan is used as a basis, modeling and information deduction of a plurality of blades of the fan are carried out by combining the sensing information, and complete state and pose sensing of the blades of the single fan is completed; further, a fan practical machine with the sensor is built in an equal-scale manner, and the effectiveness of the simulation verification method is simulated. The blade sensing method of the single fan can be popularized and used by establishing a wireless summarizing base station and the like, and further sensing and perception of large-area offshore wind power are realized.
The method for monitoring and analyzing the state of the offshore wind turbine provided by the embodiment of the invention is shown in fig. 1, and comprises the following steps:
s1, strain sensing information of the middle part and the tail end of a fan blade is obtained by using two strain sensors respectively, and x-axis pose sensing information reflecting load change of the tail end of the fan blade is obtained by using pose sensors; wherein the pose sensor comprises a gyroscope, or a combination of a gyroscope and an accelerometer.
S2, forming a three-dimensional array according to two groups of strain sensing information and x-axis pose sensing information obtained from the middle and tail ends of the fan blade, and determining control points for performing Bezier curve fitting of the fan;
s3, fitting is carried out by using a random Bezier curve fitting algorithm according to the control points, and a fan Bezier curve model is generated; in the fitting process, random weights are introduced to adjust weight parameters of all control points, and the controllable fitting of the axial curve of the fan blade is realized through the adjustment of the weights. Specifically, according to the three-dimensional array formed by the two groups of strain sensing information and the x-axis pose sensing information, slope information of different control point positions of the Bezier curve is reflected, a three-order Bezier curve fitting algorithm is adopted to perform Bezier curve fitting, and weight adjustment of control points corresponding to each data dimension is performed through random weight control curve generation. The expression of the weight introduction method is as follows:
wherein B represents a Bezier curve, w i Weight coefficients representing different control point positions, b i For sensing information, t represents curve variable parameter, n represents the number of control points, B i,n (t) represents a Bezier curve at control point i; the maximum fitting of the axial curve of the fan blade is realized through weight adjustment.
In addition, in step S3, the fan bezier curve model obtained by preliminary fitting is applied to the actual fan operation process, online model parameter adjustment iteration is performed through a random bezier curve algorithm, the fitting degree is determined in each iteration process, when the fitting degree error is greater than a predetermined threshold value, the weight of the control point corresponding to each data dimension is continuously adjusted according to the fitting degree gradient until the fitting degree error is determined to be less than the predetermined threshold value, and finally, the parameters of the fan bezier curve model are determined.
In some preferred embodiments, the monitoring data for modeling and training based on the strain sensing information and the pose sensing information includes: the deformation of the leading edge of the blade, the tip offset of the blade, the degree of deflection of the blade relative to the plane of the fan blade, the degree of twist of the middle and end of the blade, the amount of twist of the entire blade, and the observation point location of the sensor.
In some preferred embodiments, the pose sensor acquires the y-axis and z-axis pose sensing information of the tail end of the fan blade as auxiliary operation and maintenance data of the model without directly participating in load analysis.
In some preferred embodiments, the coupling of the blade to the wind environment is analyzed temporally, based on temporal changes in the motion of the fan blade resulting in the sensed received signal.
In some preferred embodiments, sensing information of a plurality of single fans of the offshore wind farm is collected, an overall offshore wind farm change relationship is established according to each fan model, position conditions among fans and wind environment information, and distribution of the overall offshore wind farm is determined.
The embodiment of the invention also provides a computer readable storage medium which stores a computer program, and when the computer program is executed by a processor, the method for monitoring and analyzing the state of the offshore wind turbine is realized.
The invention can perform multi-sensor information fusion based on the sensing network with priori fan characteristics in the micro-controller base station.
The embodiment of the invention provides a monitoring analysis method for the state of an offshore wind turbine generator set facing a wind field prototype machine, which is combined with an offshore wind turbine generator global monitoring system, is based on single-blade real-time modeling of a fan, and is used for information sensing and summarizing analysis of basic information of fan operation. The real-time fan state accurate modeling can be further completed by deducing the real-time observation state information obtained by the sensor on the boundary condition of the known fan blade characteristics. Further, intelligent comprehensive processing is carried out on the states and operation and maintenance information of various wind turbines, so that good offshore wind power state analysis and monitoring effects are achieved.
The embodiment of the invention provides a method for monitoring and analyzing the state of an offshore wind turbine by considering the actual running scene of a fan. On one hand, a complete fan blade state sensing system is built for a single fan, so that basic information of the fan blade is acquired, and the operation and sensing system iteration of the offshore wind turbine is facilitated. The integration of basic information is realized through the sensing base station of the single fan, and the state of the single fan is accurately modeled in real time on the premise of prior information of the fan blade. Meanwhile, the fan blade plane wind field is analyzed according to the time-varying modeling results of the three blades. Based on the monitoring data of the sensing system, the real-time model monitoring of the fan blade is realized. As shown in FIG. 2, the method for monitoring and analyzing the state of the offshore wind turbine generator provided by the embodiment of the invention can further perform information summarization analysis in terms of basic information processing, including multi-sensor fusion sensing and Bezier curve fitting solution. In the aspect of information conduction, a time-sharing transmission protocol can be set, so that the reliability of a data channel is improved; check bits can also be added to adjust information so as to ensure the accuracy and the integrity of the information. On the basis of a fan Bezier curve model obtained by Bezier curve fitting and solving, local disturbance rejection can be ensured through the design of multiple base stations of a distributed structure; the information quantity can be weakened, the redundancy of the information can be reduced, the information attenuation can be reduced, and the anti-interference capability can be improved.
The fine modeling of the offshore wind turbine is mainly aimed at the fan blades, so that the real-time modeling of the single blades of the fan is firstly carried out. The design of fan blades in the industrial field is initially to balance dynamic efficiency and structural design, and thrust borne by the blades drives the blades to rotate. The distribution of the thrust is proportional to the length of the blade, and the thrust born by the blade tip is greater than that born by the blade root. Thus, during operation of the wind turbine, the more significant state changes are the change in twist of the leading edge and the angular deviation of the blade tip from the ideal plane of the wind turbine blade. In comparison, the variation of the root of the blade is extremely small, and the model characteristics with smaller deformation can be confirmed through the prior model of the fan blade on the premise of not considering the deformation of the blade in the direction perpendicular to the blade. Information that accurately models a fan blade may include blade leading edge deformation, tip offset, and the like.
The sensing system of the embodiment of the present invention is shown in fig. 3, specifically, the sensing system layout of two strain sensors 1 and a pose sensor 2, preferably, the pose sensor 2 includes a gyroscope, or includes a combination of a gyroscope and an accelerometer, and the pose sensor 2 in this embodiment is a gyroscope and an accelerometer at the tip of a blade. Meanwhile, the twisting degree of the middle part and the tail end of the fan blade is observed through double strain gages (namely, two strain sensors 1), and meanwhile, the twisting amount of the whole fan blade is confirmed through the blade tip pose sensed by a gyroscope.
In some embodiments, when calculation is performed, firstly, the deflection degree of the blade relative to the plane of the fan blade is determined according to the sensing information provided by the pose sensor at the tail end (i.e. the blade tip) of the blade, then, based on the prior knowledge, the information obtained by the gyroscope and the three sensors of the two strain sensors (i.e. the double strain gauges) is input, a fan Bezier curve model is established, the distortion degree and the observation point position measured by the strain sensors at the middle part and the tail end of the blade are brought in, and fitting is performed on the premise of the existing Bezier curve.
Specifically, as shown in fig. 1, the method in the embodiment of the invention includes the following steps:
s1, strain sensing information of the middle part and the tail end of a fan blade is obtained by using two strain sensors respectively, and x-axis pose sensing information reflecting load change of the tail end of the fan blade is obtained by using pose sensors;
s2, forming a three-dimensional array according to two groups of strain sensing information and x-axis pose sensing information obtained from the middle and tail ends of the fan blade, and determining control points for performing Bezier curve fitting of the fan;
s3, fitting is carried out by using a random Bezier curve fitting algorithm according to the control points, and a fan Bezier curve model is generated; in the fitting process, random weights are introduced to adjust weight parameters of all control points, and the controllable fitting of the axial curve of the fan blade is realized through the adjustment of the weights.
When specific calculation is carried out, for the determined multi-point composite sensing system, it is easy to see that the existing accurate sensing information is always given to a plurality of nodes on the fan blade, and cannot be directly used for constructing a specific model of the fan blade. However, as the fan blade is of a relatively rigid sheet-shaped structure, the deformation of the fan blade can show a certain change rule under the structural limitation. How to reflect the change rule of the fan blade under the load condition by proper multi-point data. And modeling of the specific load deformation condition of the fan blade is performed by combining with the change rule, which is the key for sensing information processing.
1. Extracting and grouping information sources: as can be seen from the above construction of the sensing system, the sensing network at this time mainly includes two sets of strain sensing information (derived from two strain sensors on each fan blade, i.e. strain gauges, and one set of pose sensor information (specifically, three-dimensional pose information including three-axis pose changes, which is derived from a tip gyroscope)), so that the sensing information that can be provided actually is a time-varying five-dimensional array.
X＝[X 1 ，X 2 ，X x ，X y ，X z ]
Wherein X is x ，X y ，X z Different axial variable quantities of the proxy pose sensor are combined with the fact that the mounting mode of the sensor on the fan blade can be known, and the variable reflecting the load change is X x (i.e., x-axis pose sensing information).
2. Torsion component and extension component: x acquired by pose sensor y ，X z The corresponding y-axis and z-axis attitude sensing information reflecting the fan blade is taken as auxiliary operation data of the model, and the data does not directly participate in load analysis. From the wind turbine analysis report over the years, it has been found that for a relatively stable wind turbine blade structure, both the torsion and expansion components tend to be small, and that the deformation of the windward side is the main cause of the wind turbine load. Thus, embodiments of the present invention will introduce torsion and extension components separately and through X y And X z Independent twist and spread representations are made.
3. Bezier curve basic theory:
the smoothly varying blade information is almost impossible to directly derive from the sensed information, which requires a smooth modeling of the central axis of the fan blade with the sensed information as a control point. The control point forms a three-dimensional array to determine according to two groups of strain sensing information and x-axis pose sensing information obtained from the middle and tail ends of the fan blade, and a random Bezier curve algorithm is adopted for fitting a single fan blade basic sensing information summarizing link. For a given control point P 0 、P 1 The linear bezier curve is simply a straight line between two points. The resulting fan Bezier curve expression is as follows:
B(t)＝P 0 +(P 1 -P 0 )t＝(1-t)P 0 +tP 1 ，t∈[0，1]
wherein B represents a Bezier curve, P 0 、P 1 For two control points, t represents a curve variable parameter.
The basic sensing information source of the above formula curve fitting process is point location information. The problem of fitting the wind turbine blade is transferred to, and the whole movement process is reflected through point position information, but unlike the problem, in the fitting process of the wind turbine blade, the three-dimensional array reflects slope information of different control point positions, and particularly the three-dimensional array formed by two groups of strain sensing information and x-axis pose sensing information reflects slope information of different control point positions of a Bezier curve. And then combining the original points to obtain a fan blade state fitting method, and specifically adopting a third-order Bezier curve fitting algorithm to perform Bezier curve fitting. For n-dimensional data input of a fan blade, the basic Bezier curve fitting paradigm is as follows:
wherein B represents a Bezier curve, P 0 、P 1 For two control points, t represents a curve variable parameter, and n is a data dimension.
For a wireless sensor network applied to an offshore wind power system, the value of n is determined by considering actual conditions. In the embodiment of the invention, only three groups of data sources are adopted in order to ensure the reliability and the installation stability of the data. Thus, the basic method employs a three-dimensional Bezier curve fitting algorithm as follows:
B(t)＝P0(1-t) 3 +3P 1 t(1-t) 2 +3P 2 t 2 (1-t)+P 3 t 3 ，t∈[0，1]
wherein B represents a Bezier curve, P 0 、P 1 For two control points, t represents a curve variable parameter.
4. Variable weight random Bessel model construction combined with fan blade model
The fitting process of the Bezier curve follows the basic control point fitting curve rule, and accurate modeling of the fan blade cannot be achieved. Based on the sensing information of each control point, the generation rule of a specific generation curve is adjusted, so that the generation result fits the central axis of the fan blade as much as possible, and the generation rule is an important standard of Bezier curve fitting. Therefore, in the embodiment of the invention, after Bezier curve fitting, random weights are introduced to control a curve generation rule, and weight adjustment of control points corresponding to each data dimension is performed, wherein the expression of a specific weight introduction method is as follows:
wherein B represents a Bezier curve, w i Weight coefficients representing different control points, b i For sensing information, t represents curve variable parameter, n represents the number of control points, B i,n (t) represents a Bezier curve at the control point i. The controllable fitting of the medial axis curve can be realized through the weight adjustment of the control points corresponding to different data dimensions.
5. Model parameter fitting and specific results
After the method is clear, parameter fitting is carried out on the actual fan blade, and the purpose that the generated result fits the central axis of the fan blade as much as possible is achieved. For the parameters to be fitted, w= [ w ] 1 ,w 2 ,w 3 ]A random parameter table is generated, and fitting is performed by adopting a heuristic method of PSO (particle swarm optimization) fitting. The specific fitting procedure is shown in fig. 4: determined to be w= [0.92,0.65,0.34 ]]In the case of (2), a structure similar to the shape of an actual fan blade can be obtained. As shown in fig. 5a, 5b, 5 c. In fig. 5b, fmodel represents the central axis state of the fan blade, f is used to represent a bezier curve model, and for fig. 5b, the basic structure of Fmodel is derived from the section of the fan blade under pressure, and the overall shape of the central axis of the blade can reflect the loading state of the fan blade. The central axis can be completed in a multi-point fitting mode, namely a scattered point curve formed by small points. In an embodiment of the present invention, in the present invention,to complete the process of fitting the entire multi-scatter by as few points as possible, X is made x 、X 1 、X 2 Fitting according to corresponding rules, namely that the central axis state Fmodel of the fan blade is equal to the X of the Bezier curve model f x 、X 1 、X 2 Is a comprehensive calculation result of (1). Fig. 5c shows a procedure of a specific parameter fitting. In the deformation of the actual fan blade, the actual state obtained by sensing the sensor information is shown as a solid line in a broken line form in fig. 5c, but the broken line form cannot well reflect the state of the fan blade, and the embodiment of the invention carries out recursive fitting on the obtained state, so that a smoother fan blade state range is obtained. The specific fitting process is shown by the multiple broken lines of groups of black dots in fig. 5 c.
The value of the weight is also consistent with the actual environment, and the closer to the root of the blade, the smaller the deformation and the more stable the structure.
6. Model application policies
After the basic method is defined, the sensor network can complete the fitting of the fan blade main shaft information under the support of proper parameters. Furthermore, a random Bezier curve algorithm is required to be brought into an actual fan model to carry out parameter adjustment. For five-dimensional data acquired by the sensing network, three-dimensional data (namely sensing information of two strain sensors and X x ) As training set data, other data (i.e., X y ,X z ) As auxiliary operation and maintenance data. And performing online parameter adjustment iteration for optimizing weight parameter change generated in the online running process of the fan. Wherein the monitoring data for building and training the model obtained according to the strain sensing information and the pose sensing information can comprise: front edge deformation X of blade x Tip offset X of blade y Degree of deflection X of the blade relative to the plane of the fan blade z Degree of twist X in the middle and end of the blade 1 、X 2 The corresponding relation between specific point location information input and actual physical parameters can be according to the Bezier curve fitting process.
In the embodiment, when the fan Bezier curve model obtained by preliminary fitting is applied to the actual fan operation process, online model parameter adjustment iteration is performed through a random Bezier curve algorithm, fitting degree judgment is performed in each iteration process, when the fitting degree error is larger than a preset threshold value, the weight of a control point corresponding to each data dimension is continuously adjusted according to the fitting degree gradient until the fitting degree error is judged to be smaller than the preset threshold value, and finally, parameters of the fan Bezier curve model are determined.
The Bezier curve fitted according to the method can well represent pose information and deformation degree of the single blade, and further ensure safe and normal operation of the offshore wind turbine. By the mode of determining the overall pose state and fitting the accurate model, the single-blade sensing system can perform summarization analysis on the basis of the information of the conventional wireless sensing system, and the model visualization process of the sensing information is completed through the prior information of the fan blade, so that the effectiveness and the readability of the sensing information are greatly improved.
In addition to the state of the wind turbine itself, another major factor in offshore information monitoring is the variable wind environment. In order to perform more comprehensive analysis on wind environment information monitoring of a single fan, the embodiment of the invention also attempts to analyze the coupling effect of the blades and the wind environment from a time domain angle, so that analysis of single-machine position wind information is realized, and particularly, the coupling effect of the blades and the wind environment is analyzed from the time domain according to time domain change generated by signals received by sensing caused by movement of the blades of the fan. More specifically, by fan blade movement as shown in FIG. 6, the signals received by the sensing system also produce corresponding time domain signal changes. For a conventional three-blade fan as shown in fig. 5a, each independent blade structure only needs to rotate through an angle of 120 degrees to realize information sensing of the sensing system on the whole fan plane. Thus, embodiments of the present invention assume that the wind farm does not mutate during this time. And comprehensively analyzing the Bezier curve modeling result of each blade.
7. Under the support of the single-machine fan model and wind environment information, the integral marine wind field change relation can be established according to the position condition among the fans, the distribution of the integral marine wind field can be further determined, and particularly, the integral marine wind field change relation is established according to the position condition among the fans and the wind environment information by collecting the sensing information of a plurality of single fans of the marine wind field, and the distribution of the integral marine wind field is determined. Wind environment derivation as shown in fig. 7, the dashed line represents the fan state topology and the solid line represents the derivation of the global wind field. The specific deduction process is as follows: on the basis of fan state topology information, fan load information is synthesized, and the wind direction is deduced by combining the fan position orientation and the load to complete wind field deduction of each single machine position (namely, a single solid line); and further synthesizing to finish the derivation of the global wind field formed by multiple lines.
Sensing base station load information group input: and integrating the three-dimensional data set to be processed and the two-dimensional data set which can be directly used for observation, and summarizing and analyzing the information of the single blade. Further, the short-range data transmission such as UART, zigbee and the like is contacted to complete the summary of the sensing information of the sensing base station of the single fan. Finally, further total information summarization and integration of the whole sensing information are completed through means of long-distance data transmission, time-sharing communication and the like.
Experimental example:
in the example, the real machine verification of the single machine sensing system is performed, and before the model experiment is performed, a small-scale experimental model of the offshore wind turbine is built by a 3D printing technology and the like, and a three-model of the wind turbine built in the experimental example is shown in fig. 8. In order to ensure that the fan model can reflect the characteristics of an actual fan as much as possible, three-dimensional modeling software (such as SolidWorks 2020) is adopted to build and assemble a model for materializing the wind turbine.
The rotation condition and wind resistance of the model under the wind environment are verified through a wind tunnel load experiment, the suitability and representativeness of the model are ensured, and the running state of the offshore wind turbine under the ocean wind condition is restored to the greatest extent.
In terms of specific hardware design of the single-fan sensing system, the built single-fan sensing system model structure is shown in fig. 9, sensor units such as a strain sensor (namely a strain gauge 1 and a strain gauge 2), a gyroscope and an accelerometer (namely a pose sensor) are respectively laid out according to the method, and meanwhile, a mixed signal processor is used as a main control core and serves as a basic microcontroller base station, namely a sensing base station. The main control core is also connected with a display unit. The sensing system structure shown in fig. 10 is built, so that networking operation of a sensor, a sensing base station of each fan and a terminal server is realized.
Debugging codes, programming programs, ensuring the normal operation of time-sharing transmission protocol and microprocessor information summarization processing of a sensing system, and verifying a good accurate real-time monitoring function realized on a real-machine model by the method.
The foregoing is a further detailed description of the invention in connection with the preferred embodiments, and it is not intended that the invention be limited to the specific embodiments described. It will be apparent to those skilled in the art that several equivalent substitutions and obvious modifications can be made without departing from the spirit of the invention, and the same should be considered to be within the scope of the invention.
Claims (8)
1. The method for monitoring and analyzing the state of the offshore wind turbine generator is characterized by comprising the following steps of:
s1, strain sensing information of the middle part and the tail end of a fan blade is obtained by using two strain sensors respectively, and x-axis pose sensing information reflecting load change of the tail end of the fan blade is obtained by using pose sensors;
s2, forming a three-dimensional array according to two groups of strain sensing information obtained from the middle and tail ends of the fan blade and the x-axis pose sensing information, and determining control points for performing Bezier curve fitting of the fan;
s3, fitting is carried out by using a random Bezier curve fitting algorithm according to the control points, and a fan Bezier curve model is generated; in the fitting process, introducing random weights to adjust weight parameters of all control points, and realizing controllable fitting of the axial curve of the fan blade through the adjustment of the weights;
in step S3, according to the two groups of strain sensing information and the three-dimensional array formed by the x-axis pose sensing information, reflecting slope information of different control point positions of the Bezier curve, adopting a third-order Bezier curve fitting algorithm to perform Bezier curve fitting, generating a random weight control curve, and performing weight adjustment of the control points corresponding to each data dimension;
the expression of the weight introduction method is as follows:
wherein B represents a Bezier curve, w i Weight coefficients representing different control point positions, b i For sensing information, t represents curve variable parameter, n represents the number of control points, B i,n (t) represents a Bezier curve at control point i; the maximum fitting of the axial curve of the fan blade is realized through weight adjustment.
2. The method for monitoring and analyzing the state of an offshore wind turbine of claim 1, wherein the pose sensor comprises a gyroscope or a combination of a gyroscope and an accelerometer.
3. The method for monitoring and analyzing the state of the offshore wind turbine generator set according to claim 1, wherein in the step S3, the fan bezier curve model obtained by preliminary fitting is applied to an actual fan operation process, online model parameter adjustment iteration is performed through a random bezier curve fitting algorithm, fitting degree judgment is performed in each iteration process, when the fitting degree error is greater than a preset threshold value, the weight of a control point corresponding to each data dimension is continuously adjusted according to a fitting degree gradient until the fitting degree error is judged to be smaller than the preset threshold value, and finally parameters of the fan bezier curve model are determined.
4. A method for monitoring and analyzing states of an offshore wind turbine according to any one of claims 1 to 3, wherein the monitoring data for building and training a model obtained from the strain sensing information and the pose sensing information includes: the deformation of the leading edge of the blade, the tip offset of the blade, the degree of deflection of the blade relative to the plane of the fan blade, the degree of twist of the middle and end of the blade, the amount of twist of the entire blade, and the observation point location of the sensor.
5. A method for monitoring and analyzing the state of an offshore wind turbine according to any one of claims 1 to 3, further comprising: and taking the y-axis and z-axis pose sensing information of the tail end of the fan blade obtained by the pose sensor as auxiliary operation and maintenance data of the model, and not directly participating in load analysis.
6. A method for monitoring and analyzing the state of an offshore wind turbine according to any one of claims 1 to 3, further comprising: and analyzing the coupling effect of the blade and the wind environment from the time domain according to the time domain change generated by the signals received by the sensing caused by the movement of the fan blade.
7. A method for monitoring and analyzing the state of an offshore wind turbine according to any one of claims 1 to 3, further comprising: and acquiring sensing information of a plurality of single fans of the offshore wind field, establishing a change relation of the integral offshore wind field according to the fan models and the position conditions and wind environment information among the fans, and determining the distribution of the integral offshore wind field.
8. A computer readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements a method for monitoring and analyzing a state of an offshore wind turbine according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410128274.2A CN117646708B (en) | 2024-01-30 | 2024-01-30 | Method for monitoring and analyzing states of offshore wind turbine generator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410128274.2A CN117646708B (en) | 2024-01-30 | 2024-01-30 | Method for monitoring and analyzing states of offshore wind turbine generator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117646708A CN117646708A (en) | 2024-03-05 |
| CN117646708B true CN117646708B (en) | 2024-04-05 |
Family
ID=90048211
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410128274.2A Active CN117646708B (en) | 2024-01-30 | 2024-01-30 | Method for monitoring and analyzing states of offshore wind turbine generator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117646708B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB201020426D0 (en) * | 2010-12-02 | 2011-01-19 | Westland Helicopters | Aerofoil |
| BE1020627A4 (en) * | 2012-04-24 | 2014-02-04 | Citius Engineering S A | VERTICAL AXIS WIND MACHINE WITH SPHERICAL ROTOR. |
| CN106677979A (en) * | 2016-12-28 | 2017-05-17 | 中原工学院 | Coupling optimization method for blade aerodynamic configuration and main engine operation characteristics of wind turbine |
| CN113110422A (en) * | 2021-03-25 | 2021-07-13 | 贵州电网有限责任公司 | Path planning method based on Q-IGA dynamic fitting Bezier curve |
| JP2021154115A (en) * | 2020-02-16 | 2021-10-07 | オリジン ワイヤレス, インコーポレイテッドOrigin Wireless, Inc. | Method, apparatus, and system for wireless monitoring |
-
2024
- 2024-01-30 CN CN202410128274.2A patent/CN117646708B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB201020426D0 (en) * | 2010-12-02 | 2011-01-19 | Westland Helicopters | Aerofoil |
| BE1020627A4 (en) * | 2012-04-24 | 2014-02-04 | Citius Engineering S A | VERTICAL AXIS WIND MACHINE WITH SPHERICAL ROTOR. |
| CN106677979A (en) * | 2016-12-28 | 2017-05-17 | 中原工学院 | Coupling optimization method for blade aerodynamic configuration and main engine operation characteristics of wind turbine |
| JP2021154115A (en) * | 2020-02-16 | 2021-10-07 | オリジン ワイヤレス, インコーポレイテッドOrigin Wireless, Inc. | Method, apparatus, and system for wireless monitoring |
| CN113110422A (en) * | 2021-03-25 | 2021-07-13 | 贵州电网有限责任公司 | Path planning method based on Q-IGA dynamic fitting Bezier curve |
Non-Patent Citations (1)
| Title |
|---|
| 状态监测振动传感器优化布置理论及应用;王荣;贾民平;刘桂兴;;东南大学学报(自然科学版);20110120(第01期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117646708A (en) | 2024-03-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113236491A (en) | Wind power generation digital twin system | |
| CN111091298B (en) | Wind power plant flow field coupling characteristic evaluation and intelligent grouping method and system | |
| CN109543258B (en) | Wind turbine tower stress state calculation method based on modal superposition method | |
| Sessarego et al. | Development of an aeroelastic code based on three‐dimensional viscous–inviscid method for wind turbine computations | |
| CN109583069A (en) | Wind speed modeling method based on temporal correlation | |
| CN115146538A (en) | Power system state estimation method based on message passing graph neural network | |
| CN107784692B (en) | Three-dimensional skin modeling method and device for deformed blade | |
| CN117646708B (en) | Method for monitoring and analyzing states of offshore wind turbine generator | |
| Shubham et al. | Performance analysis of low Reynolds number vertical axis wind turbines using low-fidelity and mid-fidelity methods and wind conditions in the city of Nottingham | |
| CN109921472A (en) | Electric system equivalent inertia appraisal procedure based on particle swarm optimization algorithm | |
| Guo | Incorporating stochastic analysis in wind turbine design: data-driven random temporal-spatial parameterization and uncertainty quantification | |
| CN116596165A (en) | Wind power generation power prediction method and system | |
| CN114458516B (en) | On-line indirect measurement system and method for pitching and yawing moments of wind energy or tidal current energy generator set | |
| Cao et al. | Flow field distribution and structural strength performance evaluation of fixed offshore wind turbine based on digital twin technology | |
| Liu et al. | A coupled CFD/Multibody Dynamics analysis tool for offshore wind turbines with aeroelastic blades | |
| Wang et al. | Research on ultra-short term forecasting technology of wind power output based on wake model | |
| Arnold et al. | Simulation of fluid-structure-interaction on tidal current turbines based on coupled multibody and cfd methods | |
| Sun et al. | Numerical simulation of the unsteady flow and power of horizontal axis wind turbine using sliding mesh | |
| CN109802440B (en) | Offshore wind farm equivalence method, system and device based on wake effect factor | |
| EP3771819A1 (en) | A method for computer-implemented determination of a wind speed profile information of a wind field | |
| CN112861301A (en) | Wind power plant theoretical power intelligent calculation method based on real-time data of fans | |
| Cardell et al. | Estimating the system costs of wind power forecast uncertainty | |
| CN111950763A (en) | Method for predicting output power of distributed wind power station | |
| US8170813B2 (en) | Determining effects of turbine blades on fluid motion | |
| CN108712756A (en) | A kind of wind power generation unit blade status monitoring network reliability analysis method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |