CN117436352B - Wind farm noise analysis method and system - Google Patents
Wind farm noise analysis method and system Download PDFInfo
- Publication number
- CN117436352B CN117436352B CN202311752941.6A CN202311752941A CN117436352B CN 117436352 B CN117436352 B CN 117436352B CN 202311752941 A CN202311752941 A CN 202311752941A CN 117436352 B CN117436352 B CN 117436352B
- Authority
- CN
- China
- Prior art keywords
- noise
- power plant
- wind power
- distribution
- cluster
- 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
- 238000004458 analytical method Methods 0.000 title claims abstract description 43
- 238000010248 power generation Methods 0.000 claims abstract description 77
- 238000007621 cluster analysis Methods 0.000 claims abstract description 38
- 238000000034 method Methods 0.000 claims abstract description 27
- 230000002093 peripheral effect Effects 0.000 claims abstract description 16
- 230000010485 coping Effects 0.000 claims abstract description 7
- 239000011159 matrix material Substances 0.000 claims description 80
- 238000001514 detection method Methods 0.000 claims description 34
- 238000011156 evaluation Methods 0.000 claims description 33
- 238000010586 diagram Methods 0.000 claims description 17
- 239000013598 vector Substances 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 11
- 238000012216 screening Methods 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 4
- 230000001186 cumulative effect Effects 0.000 claims description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 206010013954 Dysphoria Diseases 0.000 description 1
- 206010028813 Nausea Diseases 0.000 description 1
- 206010033557 Palpitations Diseases 0.000 description 1
- 208000009205 Tinnitus Diseases 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000036772 blood pressure Effects 0.000 description 1
- 239000004566 building material Substances 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 208000002173 dizziness Diseases 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000005802 health problem Effects 0.000 description 1
- 239000012774 insulation material Substances 0.000 description 1
- 230000008693 nausea Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 208000019116 sleep disease Diseases 0.000 description 1
- 208000020685 sleep-wake disease Diseases 0.000 description 1
- 231100000886 tinnitus Toxicity 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/06—Wind turbines or wind farms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/10—Noise analysis or noise optimisation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Water Supply & Treatment (AREA)
- Geometry (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Public Health (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Wind Motors (AREA)
Abstract
The invention provides a wind farm noise analysis method and system, which relate to the technical field of noise analysis and comprise the following steps: performing cluster analysis on historical operation data of the wind power plant and historical noise distribution data of surrounding areas of the ring wind power plant to determine a plurality of wind power plant power generation typical examples and a plurality of noise distribution typical examples of the surrounding areas of the ring wind power plant; performing joint analysis on all wind power plant operation typical examples and all noise distribution typical examples of surrounding areas of the ring wind power plant to obtain joint occurrence probability of a plurality of joint examples; determining a main noise hazard zone and a secondary noise hazard zone in a surrounding area of the ring wind power plant based on the joint occurrence probability and the maximum allowable noise threshold of all the joint examples; determining a noise countermeasure strategy of the wind power plant based on the primary noise hazard zone and the secondary noise hazard zone; the method is used for analyzing the damage degree of noise generated during the operation of the wind power plant to areas in different ranges in the peripheral area of the annular wind power plant, and further determining a reasonable noise coping strategy.
Description
Technical Field
The invention relates to the technical field of noise analysis, in particular to a wind farm noise analysis method and system.
Background
At present, due to the existence of noise pollution of wind power generation clusters, more and more noise annoyance complaints and civil disputes caused by wind power generation noise are generated in recent years. Wind turbine noise is mainly from aerodynamic noise, mechanical noise and structural noise. The mechanical noise and structural noise are mainly: the mechanical components such as gears, bearings and the like collide with each other and rub to generate vibration in the running process, so that noise is radiated through the solid structure; the unbalanced electromagnetic force makes the generator generate electromagnetic vibration and radiate electromagnetic noise through the solid structure; noise excited by periodic force is generated by a rotating mechanical member such as a rotating shaft. The aerodynamic noise of the wind turbine generator is derived from friction sound of rotating fan blades and air, the relation between the aerodynamic noise and the wind speed at the height of the hub is large, and the noise source of the fan is increased along with the increase of the wind speed. The damage of wind generating set noise, wind power noise may cause health problems including nausea, dizziness, tinnitus, palpitation, blood pressure rise, sleep disorder, dysphoria and the like according to related researches. According to the experiment, the noise influence range of each wind power equipment set in different running states can be measured.
However, because more than one wind turbine generator system device exists in the wind power plant, unless the wind power plant is detected in real time in the field, it is difficult to determine the noise hazard degree of a plurality of wind turbine generator system devices in the wind power plant to surrounding areas of the wind power plant or the areas affected by noise hazard of different degrees, and therefore a reasonable noise protection strategy is not easy to determine.
Therefore, the invention provides a wind farm noise analysis method and system.
Disclosure of Invention
The invention provides a wind power plant noise analysis method and system, which are used for determining the possibility of different noise distribution typical situations of a wind power plant peripheral area when the wind power plant is in different typical operation states through cluster analysis and joint analysis of historical operation data of the wind power plant and a plurality of noise distribution typical examples of the wind power plant peripheral area.
The invention provides a wind farm noise analysis method, which comprises the following steps:
s1: performing cluster analysis on historical operation data of the wind power plant and historical noise distribution data of surrounding areas of the ring wind power plant to determine a plurality of wind power plant power generation typical examples and a plurality of noise distribution typical examples of the surrounding areas of the ring wind power plant;
s2: performing joint analysis on all the wind power plant operation typical examples and all the noise distribution typical examples of surrounding areas of the ring wind power plant to obtain joint occurrence probability of a plurality of joint examples;
s3: determining a main noise hazard zone and a secondary noise hazard zone in a surrounding area of the ring wind power plant based on the joint occurrence probability and the maximum allowable noise threshold of all the joint examples;
s4: a noise handling strategy for the wind farm is determined based on the primary noise hazard zone and the secondary noise hazard zone.
Preferably, S1: performing cluster analysis on historical operation data of the wind power plant and historical noise distribution data of surrounding areas of the ring wind power plant to determine a plurality of wind power plant power generation typical examples and a plurality of noise distribution typical examples of the surrounding areas of the ring wind power plant, including:
s101: acquiring the distribution position of each wind turbine generator equipment in the wind power plant and the wind power plant in the wind power plant as the historical operation data of the wind power plant;
S102: generating a unit equipment distribution map based on the distribution position of each wind turbine equipment in the wind power plant;
s103: marking the single-day accumulated power generation amount of all wind turbine generator equipment in a wind power plant on the same day in a preset historical period in a turbine generator equipment distribution map to obtain a plurality of single-day equipment operation data distribution maps;
s104: performing cluster analysis on all the single-day equipment operation data distribution graphs to determine a plurality of wind power plant power generation typical examples;
s105: acquiring noise value curves of a plurality of noise detection positions in a peripheral area of a ring wind power plant in all single days in a preset historical period as historical noise distribution data;
s106: and carrying out cluster analysis on noise value curves of all noise detection positions in the surrounding area of the ring wind power plant in all single days in a preset history period, and determining a plurality of noise distribution typical examples of the surrounding area of the ring wind power plant.
Preferably, S104: performing cluster analysis on all the single-day equipment operation data distribution graphs to determine a plurality of wind power plant power generation typical examples, wherein the cluster analysis comprises the following steps:
sequencing all wind turbine generator system equipment in a generator system equipment distribution diagram based on a preset sequence to obtain an equipment sequence;
Generating a power generation amount sequence corresponding to the equipment sequence in each single-day equipment operation data distribution diagram based on the single-day accumulated power generation amount of all the wind turbine equipment in each single-day equipment operation data distribution diagram;
dividing all generating capacity sequences into sequence clusters with the number of each preset cluster based on the number of the plurality of preset clusters, and taking the sequence clusters with the number of each preset cluster as a division result of the number of each preset cluster;
calculating the sequence similarity between every two generated energy sequences in each sequence cluster, and taking the average value of all the sequence similarity of each sequence cluster as the overall similarity of the sequence clusters;
taking the average value of the overall similarity of all sequence clusters of the number of each preset cluster as the evaluation value of the dividing result of the number of each preset cluster;
based on the number of the preset clusters, continuously dividing all the generated energy sequences into new sequence clusters with the number of each preset cluster as new dividing results with the number of each preset cluster, calculating evaluation values of the new dividing results, and taking the latest dividing result corresponding to the largest evaluation value in the evaluation values of the latest dividing results obtained when the evaluation values of the dividing results with the number of all the preset clusters are maximized as a final dividing result when the evaluation values of the dividing results with the number of each preset cluster are maximized;
And determining a plurality of wind power plant power generation typical examples based on the final division result.
Preferably, the calculating the sequence similarity between the two power generation sequences in each sequence cluster includes:
;
in the method, in the process of the invention,for the sequence similarity between the two currently calculated generating sequences,/>For the total number of the single-day integrated power generation amount contained in each power generation amount sequence, <>For the value of the i-th daily cumulative power generation in the first of the two power generation sequences currently calculated, +.>The value of the i-th single-day integrated power generation amount in the second power generation amount sequence in the two power generation amount sequences calculated at present.
Preferably, determining a plurality of wind farm power generation typical examples based on the final division result includes:
acquiring a plurality of sequence clusters in a final dividing result as a final sequence cluster;
acquiring the similarity between each generated energy sequence in each final sequence cluster and each generated energy sequence in the final sequence cluster except for the current generated energy sequence, and taking the average value of all the similarities of the generated energy sequences in the final sequence cluster as the overall consistency of the generated energy sequences;
and taking a single-day equipment operation data distribution diagram corresponding to the power generation amount sequence with the maximum overall conformity degree in each final sequence cluster as a typical example of wind power plant power generation.
Preferably, S106: performing cluster analysis on noise value curves of all noise detection positions in a surrounding area of the ring wind power plant in all single days in a preset history period to determine a plurality of noise distribution typical examples of the surrounding area of the ring wind power plant, including:
extracting characteristic values of each preset category from noise value curves of all noise detection positions in a peripheral area of the ring wind power plant in all single days in a preset history period;
generating a characteristic value vector of each noise value curve based on all preset types of characteristic values in each noise value curve, and generating a noise characteristic distribution matrix of a single day based on the characteristic value vectors of all noise value curves of all noise detection positions in the single day;
and carrying out cluster analysis on the noise characteristic distribution matrixes of all single days in a preset historical period to determine a plurality of noise distribution typical examples of surrounding areas of the ring wind power plant.
Preferably, cluster analysis is performed on noise characteristic distribution matrixes of all single days in a preset history period, and a plurality of noise distribution typical examples of surrounding areas of the ring wind power plant are determined, including:
dividing noise characteristic distribution matrixes of all single days in a preset history period into matrix clusters of the number of each preset cluster based on the number of the plurality of preset clusters;
And calculating the similarity between every two noise characteristic distribution matrixes in the matrix cluster, wherein the method comprises the following steps:
;
in the method, in the process of the invention,for the similarity between the two noise characteristic distribution matrices currently calculated in the matrix cluster, +.>For the total number of rows of the noise characteristic distribution matrix, +.>For the total number of columns of the noise characteristic distribution matrix, +.>For the value of the j-th row and the k-th column in the first noise characteristic distribution matrix of the two noise characteristic distribution matrices currently calculated,/for>The value of the jth row and the kth column in the second noise characteristic distribution matrix in the two noise characteristic distribution matrices calculated at present;
taking the average value of all the similarities in each matrix cluster as the overall similarity of each matrix cluster;
taking the overall similarity of all matrix clusters of the number of each preset cluster as an evaluation value of the dividing result of the number of each preset cluster;
based on the number of the plurality of preset clusters, continuously dividing all noise characteristic distribution matrixes into new matrix clusters with the number of each preset cluster as new division results with the number of each preset cluster, calculating evaluation values of the new division results, and taking the plurality of matrix clusters in the latest division results corresponding to the maximum evaluation value in the evaluation values of the latest division results obtained when the maximum value occurs in the evaluation values of the division results with the number of all preset clusters as final matrix clusters;
Taking the average value of all the similarity of each noise characteristic distribution matrix in the final matrix cluster to be regarded as the overall consistency of each noise characteristic distribution matrix;
and taking the noise value curves of all noise detection positions corresponding to the noise characteristic distribution matrix corresponding to the maximum overall coincidence degree in each matrix cluster in a single day as a plurality of noise distribution typical examples of surrounding areas of the ring wind power plant.
Preferably, S2: performing joint analysis on all wind farm operation typical examples and all noise distribution typical examples of surrounding wind farm peripheral areas to obtain joint occurrence probability of a plurality of joint examples, wherein the joint occurrence probability comprises the following steps:
screening out wind power plant operation examples with similarity not smaller than a similarity threshold value from all wind power plant operation examples contained in wind power plant historical operation data, and taking the wind power plant operation examples with similarity smaller than the similarity threshold value from each wind power plant operation typical example as similar wind power plant operation examples of each wind power plant operation typical example;
taking the ratio of the total number of all similar wind farm operation examples of each wind farm operation typical example to the total number of all wind farm operation examples contained in the wind farm historical operation data as the occurrence probability of each wind farm operation typical example;
Screening out noise distribution typical examples with the similarity with each noise distribution typical example not smaller than a similarity threshold value from all noise distribution examples contained in the historical noise distribution data, and taking the noise distribution typical examples as similar noise distribution examples of each noise distribution typical example;
regarding the ratio of the total number of all similar noise distribution instances of each noise distribution typical instance to the total number of all noise distribution instances contained in the historical noise distribution data as the occurrence probability of each noise distribution typical instance of the surrounding area of the ring wind farm;
based on all wind farm operation typical examples and all noise distribution typical examples, a plurality of joint examples are built, and the product of the occurrence probability of the wind farm operation typical examples and the occurrence probability of the noise distribution typical examples contained in each joint example is taken as the joint occurrence probability of the corresponding joint example.
Preferably, S3: determining a primary noise hazard zone and a secondary noise hazard zone in a surrounding area of the ring wind farm based on joint occurrence probabilities and maximum allowable noise thresholds for all joint instances, comprising:
dividing a sub-primary noise hazard zone and a sub-secondary noise hazard zone of each noise distribution example in a circumferential area of the ring wind farm based on each noise distribution typical example and a maximum allowable noise threshold;
Determining the probability that each position point in the circumferential area of the ring wind power plant belongs to the main noise hazard area based on the joint occurrence probability of the joint instance and the sub-main noise hazard area and the sub-secondary noise hazard area of each noise distribution instance;
and taking an area surrounded by position points with probability of not less than fifty percent in the surrounding area of the ring wind power plant as a main noise hazard area, and taking an area surrounded by position points with probability of less than fifty percent in the surrounding area of the ring wind power plant as a secondary noise hazard area.
The invention provides a wind farm noise analysis system, which comprises:
the typical extraction module is used for carrying out cluster analysis on historical operation data of the wind power plant and historical noise distribution data of surrounding areas of the ring wind power plant, and determining a plurality of typical power generation examples of the wind power plant and a plurality of typical noise distribution examples of the surrounding areas of the ring wind power plant;
the joint analysis module is used for carrying out joint analysis on all the wind power plant operation typical examples and all the noise distribution typical examples of the surrounding areas of the ring wind power plant to obtain joint occurrence probability of a plurality of joint examples;
the primary and secondary division module is used for determining a primary noise hazard zone and a secondary noise hazard zone in a surrounding area of the ring wind power plant based on the joint occurrence probability and the maximum allowable noise threshold value of all the joint instances;
And the strategy determining module is used for determining the noise coping strategy of the wind power plant based on the main noise hazard zone and the secondary noise hazard zone.
Compared with the prior art, the invention has the following beneficial effects: through cluster analysis and joint analysis of historical operation data of the wind power plant and a plurality of noise distribution typical examples of surrounding areas of the ring wind power plant, the possibility of different noise distribution typical situations of the surrounding areas of the wind power plant caused by different typical operation states of the wind power plant, namely joint occurrence probability, is determined, based on the joint examples and the joint occurrence probability and a maximum allowable noise threshold value, a main noise hazard area and a secondary noise hazard area can be determined in the surrounding areas of the ring wind power plant with higher precision, namely the hazard degree of noise generated during operation of the wind power plant to areas in different ranges in the surrounding areas of the ring wind power plant is analyzed, and a reasonable noise coping strategy is determined.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a flow chart of a wind farm noise analysis method in an embodiment of the invention;
FIG. 2 is a flowchart of another wind farm noise analysis method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a wind farm noise analysis system in an embodiment of the invention.
Detailed Description
The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.
Example 1:
the invention provides a wind farm noise analysis method, referring to FIG. 1, comprising the following steps:
s1: performing cluster analysis on historical operation data of the wind farm (namely, data recorded with single-day accumulated power generation amounts of all wind turbine generator sets in the wind farm in all single days in a preset historical period (such as the last year) and data recorded with distribution positions of all wind turbine generator sets in the wind farm) and a surrounding area of the wind farm (namely, a surrounding area containing the surrounding wind farm, wherein the surrounding area of the wind farm is large enough to ensure that the surrounding area can contain a main noise hazard area and a secondary noise hazard area) (namely, data of noise value curves in all single days in a preset historical period detected by a plurality of noise detection positions in the surrounding area of the wind farm), and determining a plurality of typical examples of wind farm power generation (namely, data with single-day accumulated power generation amounts of all wind turbine generator sets with typical characteristics screened in the historical operation data of the wind farm) and a plurality of typical examples of noise distribution (namely, noise value curves with single-day all noise detection positions with typical characteristics screened in the historical noise distribution data of the wind farm);
S2: performing joint analysis on all wind farm operation typical examples and all noise distribution typical examples of surrounding areas of the ring wind farm (namely, calculating the probability of a certain noise typical example of the simultaneous wind farm operation typical examples and the surrounding areas of the ring wind farm), and obtaining joint occurrence probability (namely, the probability of the wind farm operation typical examples and the noise typical examples in the simultaneous joint examples) of a plurality of joint examples (namely, the assumption of the certain noise typical examples of the simultaneous wind farm operation typical examples and the surrounding areas of the ring wind farm);
s3: determining a primary noise hazard zone (namely, a zone which is seriously affected by the noise generated by the wind power plant) and a secondary noise hazard zone (a zone which is less severely affected by the noise generated by the wind power plant) in a surrounding area of the wind power plant based on the joint occurrence probability and the maximum allowable noise threshold value (namely, the maximum noise decibel value allowable for the area without noise protection) of all the joint examples;
s4: and determining a noise coping strategy of the wind power plant based on the main noise hazard zone and the secondary noise hazard zone (namely, a preset noise protection strategy corresponding to the preset main noise hazard zone and a preset noise protection strategy corresponding to the preset secondary noise hazard zone, wherein the noise protection strategy is used for preventing building of residential areas and the like or adopting a sound insulation material with a specified standard by building materials).
According to the technical scheme, through cluster analysis and joint analysis of historical operation data of the wind power plant and a plurality of noise distribution typical examples of surrounding areas of the wind power plant, the possibility of different noise distribution typical situations of the surrounding areas of the wind power plant caused by different typical operation states of the wind power plant, namely joint occurrence probability, is determined, based on the joint examples, the joint occurrence probability and a maximum allowable noise threshold value, a main noise hazard area and a secondary noise hazard area can be determined in the surrounding areas of the wind power plant with higher precision, namely the hazard degree of noise generated during operation of the wind power plant to the areas in different ranges in the surrounding areas of the wind power plant is analyzed, and a reasonable noise response strategy is determined.
Example 2:
based on example 1, S1: performing cluster analysis on historical operation data of the wind power plant and historical noise distribution data of surrounding areas of the ring wind power plant to determine a plurality of wind power plant power generation typical examples and a plurality of noise distribution typical examples of surrounding areas of the ring wind power plant, referring to fig. 2, including:
s101: acquiring all daily accumulated power generation amounts (namely, the accumulated power generation amount of single wind power plant equipment in a single day) of each wind power plant equipment (namely, the set equipment used for generating power in the wind power plant, such as a wind power generator set) in a preset historical period (namely, the historical period which is required to be covered by preset wind power plant historical operation data and historical noise distribution data) and the distribution position of each wind power plant equipment in the wind power plant, wherein the accumulated power generation amounts are the accumulated power generation amounts of the single wind power plant equipment in the single day are used as wind power plant historical operation data;
S102: generating a unit equipment distribution map based on the distribution position of each wind turbine equipment in the wind power plant (namely marking all wind turbine equipment in the wind power plant in an equal ratio plan view of the wind power plant based on the distribution position of each wind turbine equipment in the wind power plant to obtain the unit equipment distribution map, wherein the unit equipment distribution map is a schematic diagram representing the distribution position of all wind turbine equipment in the wind power plant;
s103: marking the single-day accumulated power generation amount of all wind turbine generator systems in the wind power plant on the same day in a preset historical period in a generator system equipment distribution diagram to obtain a plurality of single-day equipment operation data distribution diagrams (namely, a schematic diagram representing all accumulated power generation amounts of all wind turbine generator systems in the wind power plant on a certain day in the preset historical period);
s104: performing cluster analysis on all the single-day equipment operation data distribution graphs to determine a plurality of wind power plant power generation typical examples;
s105: acquiring noise value curves (namely curves representing the change process of environmental noise values detected by a certain noise detection position in real time in a single day) of all single days in a preset historical period as historical noise distribution data, wherein the noise detection positions in a peripheral area of a ring wind power plant (namely positions of noise detection equipment fixed by equipment are uniformly distributed in the peripheral area of the ring wind power plant);
S106: and carrying out cluster analysis on noise value curves of all noise detection positions in the surrounding area of the ring wind power plant in all single days in a preset history period, and determining a plurality of noise distribution typical examples of the surrounding area of the ring wind power plant.
According to the technical scheme, historical operation data of the wind power plant and historical noise distribution data of surrounding areas of the wind power plant are obtained, and a noise value curve which is provided with typical characteristics and contains all noise detection positions in a single day and is used as a wind power plant power generation typical example and a noise distribution typical example respectively is extracted through a clustering analysis process of the obtained data.
Example 3:
based on example 2, S104: performing cluster analysis on all the single-day equipment operation data distribution graphs to determine a plurality of wind power plant power generation typical examples, wherein the cluster analysis comprises the following steps:
sorting all wind turbine generator systems in the wind turbine generator system distribution map based on a preset sequence (for example, sorting the first row of wind turbine generator systems from the south to the north and from the east to the west, namely, sorting the first row of wind turbine generator systems adjacent to the first row and positioned at the north side of the first row from the south again according to the order from the east to the west, and pushing the first row of wind turbine generator systems in the first row, so as to obtain a device sequence (namely, a sequence comprising all wind turbine generator systems in the wind farm sorted according to the preset sequence);
Generating a power generation amount sequence corresponding to the equipment sequence in each single-day equipment operation data distribution diagram based on the single-day accumulated power generation amount of all the wind turbine equipment in each single-day equipment operation data distribution diagram (namely, a sequence obtained by correspondingly sequencing the single-day accumulated power generation amount of all the wind turbine equipment in the single-day equipment operation data distribution diagram according to the equipment sequence);
dividing all generating capacity sequences into sequence clusters with the number of each preset cluster (for example, the preset cluster number is 3, and dividing all generating capacity sequences into three sequence clusters, namely, the clusters containing a plurality of generating capacity sequences) based on the number of the preset clusters (the preset cluster number which needs to be tried in the dividing process of cluster analysis), and taking the sequence clusters as a dividing result of the number of each preset cluster (namely, a result of all sequence clusters containing the preset cluster number);
calculating the sequence similarity between every two generated energy sequences in each sequence cluster (namely, representing the similarity between two generated energy sequences), and taking the average value of all the sequence similarity of each sequence cluster as the overall similarity of the sequence cluster (namely, representing the numerical value of the overall similarity degree between all generated energy sequences in the sequence cluster);
Taking the average value of the overall similarity of all the sequence clusters of each preset cluster number as the evaluation value of the division result of each preset cluster number (namely, representing the degree that the corresponding division result meets the division requirement, the greater the corresponding evaluation value is, and the division requirement is that the average value of the overall similarity of all the sequence clusters in the division result is the largest);
based on the number of the preset clusters, continuously dividing all the generated energy sequences into new sequence clusters of the number of each preset cluster as new dividing results of the number of each preset cluster, calculating evaluation values of the new dividing results, and taking the most recent dividing result (the maximum value of the maximum values of the dividing results of the number of all preset clusters) as the final dividing result (namely the dividing result finally obtained in the whole cluster analysis process) when the maximum value of the evaluation values of the dividing results of the number of all preset clusters is found (the maximum value of the dividing results of the number of corresponding preset clusters) when the maximum value of the evaluation values of the dividing results of the number of each preset cluster is found (namely the maximum value of the dividing results of the number of all preset clusters);
And determining a plurality of wind power plant power generation typical examples based on the final division result.
According to the technical scheme, the generated energy sequence is used for representing a single-day equipment operation data distribution map, and multiple clustering analysis is carried out on all generated energy sequences based on the number of multiple preset clusters until multiple sequence clusters with the largest average value of the overall similarity of all sequence clusters in the dividing result are determined, and multiple wind power plant power generation typical examples are extracted based on the multiple sequence clusters, so that the determined representativeness of the multiple wind power plant power generation typical examples is ensured.
Example 4:
on the basis of embodiment 3, calculating the sequence similarity between the power generation amount sequences in each sequence cluster includes:
;
in the method, in the process of the invention,for the sequence similarity between the two currently calculated generating sequences,/>Accumulating for a single day contained in each power generation sequenceTotal amount of generated electricity>For the value of the i-th daily cumulative power generation in the first of the two power generation sequences currently calculated, +.>The value of the i-th single-day integrated power generation amount in the second power generation amount sequence in the two power generation amount sequences calculated at present.
The sequence similarity between every two generated energy sequences in each sequence cluster can be accurately calculated by using the formula.
Example 5:
on the basis of embodiment 3, a plurality of wind farm power generation typical examples are determined based on the final division result, including:
acquiring a plurality of sequence clusters in a final dividing result as a final sequence cluster;
acquiring the similarity between each generated energy sequence in each final sequence cluster and each generated energy sequence in the final sequence cluster except for the current generated energy sequence, and taking the average value of all the similarities of the generated energy sequences in the final sequence cluster as the overall consistency of the generated energy sequences (a numerical value representing the typical degree that the generated energy sequences can represent all the generated energy sequences in the sequence cluster);
and taking a single-day equipment operation data distribution diagram corresponding to the power generation amount sequence with the maximum overall conformity degree in each final sequence cluster as a typical example of wind power plant power generation.
According to the technical scheme, the similarity between each generated energy sequence in each final sequence cluster and each generated energy sequence in the final sequence cluster except for the current generated energy sequence is obtained, the average value of all the similarities of the generated energy sequences in the final sequence cluster is taken as the overall coincidence degree of the generated energy sequences, and the overall coincidence degree is taken as a screening reference, so that the representativeness of the determined power generation typical examples of the plurality of wind power plants is further ensured.
Example 6:
based on example 2, S106: performing cluster analysis on noise value curves of all noise detection positions in a surrounding area of the ring wind power plant in all single days in a preset history period to determine a plurality of noise distribution typical examples of the surrounding area of the ring wind power plant, including:
extracting characteristic values of each preset category from noise value curves of all noise detection positions in a peripheral area of the ring wind power plant in all single days in a preset history period (the characteristic values can comprise all maximum values, minimum values, maximum values, minimum values and the like of the noise value curves, or slope maximum values, slope minimum values and the like);
generating a characteristic value vector of each noise value curve (namely, ordering all characteristic values in each noise value curve according to a principle of ordering and fixing the characteristic values of each type) based on all preset types of characteristic values (namely, preset types of characteristic values of the noise value curve, such as maximum value, minimum value, maximum value and minimum value of the noise value curve) in each noise value curve, obtaining a characteristic value sequence, taking the characteristic value sequence as a row vector or a column vector of vector elements, namely, a characteristic value vector, generating a noise characteristic distribution matrix of a single day based on the characteristic value vectors of all the noise value curves in a single day (namely, when the characteristic value vectors are row (column) vectors, ordering the characteristic value vectors of the corresponding noise value curve in rows and columns based on a preset ordering sequence of the noise detection positions, obtaining a numerical array, taking the numerical array as matrix elements, and reserving a matrix generated by a relative position relation among the numerical array as the noise characteristic distribution matrix;
And carrying out cluster analysis on the noise characteristic distribution matrixes of all single days in a preset historical period to determine a plurality of noise distribution typical examples of surrounding areas of the ring wind power plant.
According to the technical scheme, the characteristic values of the noise value curves are extracted and sequenced, the vectorization representation of the noise value curves is realized, the characteristic value vectors of all the noise value curves of all the noise detection positions in a single day are sequenced and combined in a fixed sequencing mode of the noise detection positions, the matrixing representation of the noise distribution data in the single day is realized, the noise characteristic distribution matrix represented by the matrixing representation is used as a clustering object for clustering analysis, and the accurate extraction of the noise distribution typical examples is realized.
Example 7:
based on embodiment 6, a plurality of noise distribution typical examples of surrounding areas of the ring wind farm are determined by performing cluster analysis on noise feature distribution matrixes of all single days in a preset history period, including:
dividing the noise characteristic distribution matrix of all single days in a preset history period into matrix clusters (namely, clusters containing a plurality of noise characteristic distribution matrices) of each preset cluster number based on the number of the preset clusters;
and calculating the similarity between every two noise characteristic distribution matrixes in the matrix cluster, wherein the method comprises the following steps:
;
In the method, in the process of the invention,for the similarity between the two noise characteristic distribution matrices currently calculated in the matrix cluster, +.>For the total number of rows of the noise characteristic distribution matrix, +.>For the total number of columns of the noise characteristic distribution matrix, +.>For the value of the j-th row and the k-th column in the first noise characteristic distribution matrix of the two noise characteristic distribution matrices currently calculated,/for>The value of the jth row and the kth column in the second noise characteristic distribution matrix in the two noise characteristic distribution matrices calculated at present;
taking the average value of all the similarities in each matrix cluster (namely, the value representing the similarity degree between two noise characteristic distribution matrixes) as the overall similarity of each matrix cluster (namely, the value representing the similarity degree between all the noise characteristic distribution matrixes in the matrix cluster);
taking the overall similarity of all matrix clusters of each preset cluster number as an evaluation value of a division result of each preset cluster number (namely, representing the degree that the division result meets the division requirement, the greater the division requirement is, the greater the corresponding evaluation value is, and the division requirement is that the average value of the overall similarity of all matrix clusters in the division result is the largest);
based on the number of the preset clusters, continuously dividing all noise characteristic distribution matrixes into new matrix clusters of the number of each preset cluster as new dividing results of the number of each preset cluster, calculating evaluation values of the new dividing results, and taking the maximum value (the maximum value in the dividing results of the number of all preset clusters) of the maximum evaluation values (the maximum value in the dividing results of the number of all preset clusters) of the latest dividing results obtained when the evaluation values of the dividing results of the number of all preset clusters are maximum (the maximum value in the dividing results of the number of corresponding preset clusters) as the final matrix cluster when the maximum value of the evaluation values of the dividing results of the number of each preset cluster is present (namely, the maximum value in the dividing results of the number of all preset clusters);
Regarding the average value of all the similarity of each noise characteristic distribution matrix in the final matrix cluster to which the noise characteristic distribution matrix belongs as the overall coincidence degree of each noise characteristic distribution matrix (the numerical value representing the typical degree of all the noise characteristic distribution matrices in the matrix cluster to which the noise characteristic distribution matrix belongs;
and taking the noise value curves of all noise detection positions corresponding to the noise characteristic distribution matrix corresponding to the maximum overall coincidence degree in each matrix cluster in a single day as a plurality of noise distribution typical examples of surrounding areas of the ring wind power plant.
According to the technical scheme, clustering analysis is conducted on all the noise characteristic distribution matrixes for multiple times based on the number of the preset clusters until a plurality of matrix clusters with the largest average value of the overall similarity of all the matrix clusters in the dividing result are determined, a plurality of noise distribution typical examples are extracted based on the matrix clusters, the similarity between each noise characteristic distribution matrix in each final matrix cluster and each noise characteristic distribution matrix in the final matrix cluster except the current noise characteristic distribution matrix is obtained, the average value of all the similarities of the noise characteristic distribution matrix in the final matrix cluster is taken as the overall consistency of the noise characteristic distribution matrix, and the overall consistency is taken as a screening standard, so that the determined typical of the plurality of noise distribution typical examples is ensured.
Example 8:
based on example 1, S2: performing joint analysis on all wind farm operation typical examples and all noise distribution typical examples of surrounding wind farm peripheral areas to obtain joint occurrence probability of a plurality of joint examples, wherein the joint occurrence probability comprises the following steps:
screening out the wind farm operation examples (each wind farm operation example comprises data of single-day accumulated power generation capacity of all wind turbine generator equipment in a single day) which are contained in the wind farm historical operation data, wherein the similarity (namely, the sequence similarity between a power generation sequence corresponding to the wind farm operation typical example and a power generation sequence of a single wind farm operation example is regarded as the similarity between the wind farm operation example and the wind farm operation typical example, the calculation mode adopts the formula in the embodiment 4) is not smaller than a similarity threshold (namely, when the wind farm operation example is a preset minimum similarity which is required to be achieved by the similarity between the wind farm operation example and the wind farm operation typical example) from the wind farm operation typical example (namely, the wind farm operation example with higher similarity between the wind farm operation typical example and the wind farm operation typical example) which is used for screening out the wind farm operation typical example;
Regarding the ratio of the total number of all similar wind farm operation instances of each wind farm operation typical instance to the total number of all wind farm operation instances contained in the wind farm historical operation data as the occurrence probability of each wind farm operation typical instance (namely, determining the probability that each wind farm operation typical instance is likely to occur by taking a preset historical period as a sample);
among all the noise distribution examples included in the history noise distribution data (each noise distribution example includes data of a noise value curve of all the noise detection positions on a single day), a noise distribution typical example in which the similarity between the noise characteristic distribution matrix corresponding to each noise distribution typical example and the noise characteristic distribution matrix corresponding to the noise distribution typical example (in the same manner as embodiment 7) is not less than a similarity threshold (i.e., a preset minimum similarity to be reached when a preset noise distribution example for screening out the noise distribution typical example whose known similarity is required) is selected as a similar noise distribution example (i.e., a noise distribution example having a higher similarity to the noise distribution typical example) of each noise distribution typical example;
Taking the ratio of the total number of all similar noise distribution instances of each noise distribution typical instance to the total number of all noise distribution instances contained in the historical noise distribution data as the occurrence probability of each noise distribution typical instance of the surrounding area of the ring wind power plant (namely, determining the probability that each noise distribution typical instance is likely to occur by taking a preset historical period as a sample);
based on all wind farm operation typical examples and all noise distribution typical examples, a plurality of joint examples are built, and the product of the occurrence probability of the wind farm operation typical examples and the occurrence probability of the noise distribution typical examples contained in each joint example is taken as the joint occurrence probability of the corresponding joint example.
According to the technical scheme, the occurrence probability of each wind farm operation typical example is determined through screening the total number of the similar wind farm operation examples of each wind farm operation typical example, the occurrence probability of each noise distribution typical example is determined through screening the total number of the similar noise distribution examples of each noise distribution typical example, and the joint occurrence probability of each joint example is further calculated based on the occurrence probability.
Example 9:
Based on example 1, S3: determining a primary noise hazard zone and a secondary noise hazard zone in a surrounding area of the ring wind farm based on joint occurrence probabilities and maximum allowable noise thresholds for all joint instances, comprising:
dividing a sub-main noise hazard zone and a sub-sub noise hazard zone of each noise distribution instance in a surrounding wind farm peripheral area based on each noise distribution typical instance and a maximum allowable noise threshold (namely, regarding a noise detection position corresponding to a noise value curve in which a noise value exceeding the maximum allowable noise threshold appears in all noise value curves in the noise distribution typical instance as a first noise detection position, regarding a noise detection position corresponding to a noise value curve in which a noise value exceeding the maximum allowable noise threshold does not appear in all noise value curves in the noise distribution typical instance as a second noise detection position, regarding a midpoint between the second noise detection position closest to the first noise detection position and the corresponding first noise detection position as a dividing point, smoothly connecting all dividing points to obtain a dividing limit, regarding a region range including the first noise detection position in two region ranges divided by the dividing limit in the surrounding wind farm peripheral area as a sub-main noise hazard zone, and regarding the other region range as a sub-sub noise hazard zone);
Determining the probability that each position point in the peripheral area of the ring wind power plant belongs to the main noise hazard area based on the joint occurrence probability of the joint instance and the sub-main noise hazard area and the sub-sub noise hazard area of each noise distribution instance (taking the joint instance containing the maximum joint occurrence probability of all the joint instances of a single noise distribution instance as the decision joint instance of the noise distribution instance and taking the sum of the joint occurrence probabilities of the decision joint instances of all the noise distribution instances of which the single position point belongs to the main noise hazard area as the probability that the single position point belongs to the main noise hazard area);
and taking an area surrounded by position points with probability of not less than fifty percent in the surrounding area of the ring wind power plant as a main noise hazard area, and taking an area surrounded by position points with probability of less than fifty percent in the surrounding area of the ring wind power plant as a secondary noise hazard area.
According to the technical scheme, the surrounding area of the annular wind power plant is divided based on all the noise distribution examples and the maximum operation noise threshold value, and the main noise hazard area and the secondary noise hazard area are divided by combining the dividing result and the joint occurrence probability of the combined examples in a comprehensive wind power plant in various typical operation states.
Example 10:
the invention provides a wind farm noise analysis system, referring to FIG. 3, comprising:
the typical extraction module is used for carrying out cluster analysis on historical operation data of the wind power plant and historical noise distribution data of surrounding areas of the ring wind power plant, and determining a plurality of typical power generation examples of the wind power plant and a plurality of typical noise distribution examples of the surrounding areas of the ring wind power plant;
the joint analysis module is used for carrying out joint analysis on all the wind power plant operation typical examples and all the noise distribution typical examples of the surrounding areas of the ring wind power plant to obtain joint occurrence probability of a plurality of joint examples;
the primary and secondary division module is used for determining a primary noise hazard zone and a secondary noise hazard zone in a surrounding area of the ring wind power plant based on the joint occurrence probability and the maximum allowable noise threshold value of all the joint instances;
and the strategy determining module is used for determining the noise coping strategy of the wind power plant based on the main noise hazard zone and the secondary noise hazard zone.
Through cluster analysis and joint analysis of historical operation data of the wind power plant and a plurality of noise distribution typical examples of surrounding areas of the ring wind power plant, the possibility of different noise distribution typical situations of the surrounding areas of the wind power plant caused by different typical operation states of the wind power plant, namely joint occurrence probability, is determined, based on the joint examples and the joint occurrence probability and a maximum allowable noise threshold value, a main noise hazard area and a secondary noise hazard area can be determined in the surrounding areas of the ring wind power plant with higher precision, namely the hazard degree of noise generated during operation of the wind power plant to areas in different ranges in the surrounding areas of the ring wind power plant is analyzed, and a reasonable noise coping strategy is determined.
According to the technical scheme, through cluster analysis and joint analysis of historical operation data of the wind power plant and a plurality of noise distribution typical examples of surrounding areas of the wind power plant, the possibility of different noise distribution typical situations of the surrounding areas of the wind power plant caused by different typical operation states of the wind power plant, namely joint occurrence probability, is determined, based on the joint examples, the joint occurrence probability and a maximum allowable noise threshold value, a main noise hazard area and a secondary noise hazard area can be determined in the surrounding areas of the wind power plant with higher precision, namely the hazard degree of noise generated during operation of the wind power plant to the areas in different ranges in the surrounding areas of the wind power plant is analyzed, and a reasonable noise response strategy is determined.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (10)
1. A method for analyzing noise of a wind farm, comprising:
s1: performing cluster analysis on historical operation data of the wind power plant and historical noise distribution data of surrounding areas of the ring wind power plant to determine a plurality of wind power plant power generation typical examples and a plurality of noise distribution typical examples of the surrounding areas of the ring wind power plant;
S2: performing joint analysis on all the wind power plant operation typical examples and all the noise distribution typical examples of surrounding areas of the ring wind power plant to obtain joint occurrence probability of a plurality of joint examples;
s3: determining a main noise hazard zone and a secondary noise hazard zone in a surrounding area of the ring wind power plant based on the joint occurrence probability and the maximum allowable noise threshold of all the joint examples;
s4: a noise handling strategy for the wind farm is determined based on the primary noise hazard zone and the secondary noise hazard zone.
2. The wind farm noise analysis method according to claim 1, wherein S1: performing cluster analysis on historical operation data of the wind power plant and historical noise distribution data of surrounding areas of the ring wind power plant to determine a plurality of wind power plant power generation typical examples and a plurality of noise distribution typical examples of the surrounding areas of the ring wind power plant, including:
s101: acquiring the distribution position of each wind turbine generator equipment in the wind power plant and the wind power plant in the wind power plant as the historical operation data of the wind power plant;
s102: generating a unit equipment distribution map based on the distribution position of each wind turbine equipment in the wind power plant;
S103: marking the single-day accumulated power generation amount of all wind turbine generator equipment in a wind power plant on the same day in a preset historical period in a turbine generator equipment distribution map to obtain a plurality of single-day equipment operation data distribution maps;
s104: performing cluster analysis on all the single-day equipment operation data distribution graphs to determine a plurality of wind power plant power generation typical examples;
s105: acquiring noise value curves of a plurality of noise detection positions in a peripheral area of a ring wind power plant in all single days in a preset historical period as historical noise distribution data;
s106: and carrying out cluster analysis on noise value curves of all noise detection positions in the surrounding area of the ring wind power plant in all single days in a preset history period, and determining a plurality of noise distribution typical examples of the surrounding area of the ring wind power plant.
3. A method of analyzing wind farm noise according to claim 2, wherein S104: performing cluster analysis on all the single-day equipment operation data distribution graphs to determine a plurality of wind power plant power generation typical examples, wherein the cluster analysis comprises the following steps:
sequencing all wind turbine generator system equipment in a generator system equipment distribution diagram based on a preset sequence to obtain an equipment sequence;
generating a power generation amount sequence corresponding to the equipment sequence in each single-day equipment operation data distribution diagram based on the single-day accumulated power generation amount of all the wind turbine equipment in each single-day equipment operation data distribution diagram;
Dividing all generating capacity sequences into sequence clusters with the number of each preset cluster based on the number of the plurality of preset clusters, and taking the sequence clusters with the number of each preset cluster as a division result of the number of each preset cluster;
calculating the sequence similarity between every two generated energy sequences in each sequence cluster, and taking the average value of all the sequence similarity of each sequence cluster as the overall similarity of the sequence clusters;
taking the average value of the overall similarity of all sequence clusters of the number of each preset cluster as the evaluation value of the dividing result of the number of each preset cluster;
based on the number of the preset clusters, continuously dividing all the generated energy sequences into new sequence clusters with the number of each preset cluster as new dividing results with the number of each preset cluster, calculating evaluation values of the new dividing results, and taking the latest dividing result corresponding to the largest evaluation value in the evaluation values of the latest dividing results obtained when the evaluation values of the dividing results with the number of all the preset clusters are maximized as a final dividing result when the evaluation values of the dividing results with the number of each preset cluster are maximized;
and determining a plurality of wind power plant power generation typical examples based on the final division result.
4. A wind farm noise analysis method according to claim 3, wherein calculating the sequence similarity between the power generation sequences of each sequence cluster comprises:
;
In the method, in the process of the invention,for the sequence similarity between the two currently calculated generating sequences,/>For the total number of the single-day integrated power generation amount contained in each power generation amount sequence, <>For the value of the i-th daily cumulative power generation in the first of the two power generation sequences currently calculated, +.>The value of the i-th single-day integrated power generation amount in the second power generation amount sequence in the two power generation amount sequences calculated at present.
5. A wind farm noise analysis method according to claim 3, wherein determining a plurality of wind farm power generation representative examples based on the final division result comprises:
acquiring a plurality of sequence clusters in a final dividing result as a final sequence cluster;
acquiring the similarity between each generated energy sequence in each final sequence cluster and each generated energy sequence in the final sequence cluster except for the current generated energy sequence, and taking the average value of all the similarities of the generated energy sequences in the final sequence cluster as the overall consistency of the generated energy sequences;
and taking a single-day equipment operation data distribution diagram corresponding to the power generation amount sequence with the maximum overall conformity degree in each final sequence cluster as a typical example of wind power plant power generation.
6. A method of analyzing wind farm noise according to claim 2, wherein S106: performing cluster analysis on noise value curves of all noise detection positions in a surrounding area of the ring wind power plant in all single days in a preset history period to determine a plurality of noise distribution typical examples of the surrounding area of the ring wind power plant, including:
extracting characteristic values of each preset category from noise value curves of all noise detection positions in a peripheral area of the ring wind power plant in all single days in a preset history period;
generating a characteristic value vector of each noise value curve based on all preset types of characteristic values in each noise value curve, and generating a noise characteristic distribution matrix of a single day based on the characteristic value vectors of all noise value curves of all noise detection positions in the single day;
and carrying out cluster analysis on the noise characteristic distribution matrixes of all single days in a preset historical period to determine a plurality of noise distribution typical examples of surrounding areas of the ring wind power plant.
7. The method for analyzing noise of a wind farm according to claim 6, wherein performing cluster analysis on the noise characteristic distribution matrix of all single days in a preset history period to determine a plurality of noise distribution typical examples of surrounding areas of the wind farm comprises:
Dividing noise characteristic distribution matrixes of all single days in a preset history period into matrix clusters of the number of each preset cluster based on the number of the plurality of preset clusters;
and calculating the similarity between every two noise characteristic distribution matrixes in the matrix cluster, wherein the method comprises the following steps:
;
in the method, in the process of the invention,for the similarity between the two noise characteristic distribution matrices currently calculated in the matrix cluster, +.>For the total number of rows of the noise characteristic distribution matrix, +.>For the total number of columns of the noise characteristic distribution matrix, +.>For the value of the j-th row and the k-th column in the first noise characteristic distribution matrix of the two noise characteristic distribution matrices currently calculated,/for>The value of the jth row and the kth column in the second noise characteristic distribution matrix in the two noise characteristic distribution matrices calculated at present;
taking the average value of all the similarities in each matrix cluster as the overall similarity of each matrix cluster;
taking the average value of the overall similarity of all matrix clusters of the number of each preset cluster as the evaluation value of the dividing result of the number of each preset cluster;
based on the number of the plurality of preset clusters, continuously dividing all noise characteristic distribution matrixes into new matrix clusters with the number of each preset cluster as new division results with the number of each preset cluster, calculating evaluation values of the new division results, and taking the plurality of matrix clusters in the latest division results corresponding to the maximum evaluation value in the evaluation values of the latest division results obtained when the maximum value occurs in the evaluation values of the division results with the number of all preset clusters as final matrix clusters;
Taking the average value of all the similarity of each noise characteristic distribution matrix in the final matrix cluster to be regarded as the overall consistency of each noise characteristic distribution matrix;
and taking the noise value curves of all noise detection positions corresponding to the noise characteristic distribution matrix corresponding to the maximum overall coincidence degree in each matrix cluster in a single day as a plurality of noise distribution typical examples of surrounding areas of the ring wind power plant.
8. A method of wind farm noise analysis according to claim 1, wherein S2: performing joint analysis on all wind farm operation typical examples and all noise distribution typical examples of surrounding wind farm peripheral areas to obtain joint occurrence probability of a plurality of joint examples, wherein the joint occurrence probability comprises the following steps:
screening out wind power plant operation examples with similarity not smaller than a similarity threshold value from all wind power plant operation examples contained in wind power plant historical operation data, and taking the wind power plant operation examples with similarity smaller than the similarity threshold value from each wind power plant operation typical example as similar wind power plant operation examples of each wind power plant operation typical example;
taking the ratio of the total number of all similar wind farm operation examples of each wind farm operation typical example to the total number of all wind farm operation examples contained in the wind farm historical operation data as the occurrence probability of each wind farm operation typical example;
Screening out noise distribution typical examples with the similarity with each noise distribution typical example not smaller than a similarity threshold value from all noise distribution examples contained in the historical noise distribution data, and taking the noise distribution typical examples as similar noise distribution examples of each noise distribution typical example;
regarding the ratio of the total number of all similar noise distribution instances of each noise distribution typical instance to the total number of all noise distribution instances contained in the historical noise distribution data as the occurrence probability of each noise distribution typical instance of the surrounding area of the ring wind farm;
based on all wind farm operation typical examples and all noise distribution typical examples, a plurality of joint examples are built, and the product of the occurrence probability of the wind farm operation typical examples and the occurrence probability of the noise distribution typical examples contained in each joint example is taken as the joint occurrence probability of the corresponding joint example.
9. A method of wind farm noise analysis according to claim 1, wherein S3: determining a primary noise hazard zone and a secondary noise hazard zone in a surrounding area of the ring wind farm based on joint occurrence probabilities and maximum allowable noise thresholds for all joint instances, comprising:
dividing a sub-primary noise hazard zone and a sub-secondary noise hazard zone of each noise distribution example in a circumferential area of the ring wind farm based on each noise distribution typical example and a maximum allowable noise threshold;
Determining the probability that each position point in the circumferential area of the ring wind power plant belongs to the main noise hazard area based on the joint occurrence probability of the joint instance and the sub-main noise hazard area and the sub-secondary noise hazard area of each noise distribution instance;
and taking an area surrounded by position points with probability of not less than fifty percent in the surrounding area of the ring wind power plant as a main noise hazard area, and taking an area surrounded by position points with probability of less than fifty percent in the surrounding area of the ring wind power plant as a secondary noise hazard area.
10. A wind farm noise analysis system, comprising:
the typical extraction module is used for carrying out cluster analysis on historical operation data of the wind power plant and historical noise distribution data of surrounding areas of the ring wind power plant, and determining a plurality of typical power generation examples of the wind power plant and a plurality of typical noise distribution examples of the surrounding areas of the ring wind power plant;
the joint analysis module is used for carrying out joint analysis on all the wind power plant operation typical examples and all the noise distribution typical examples of the surrounding areas of the ring wind power plant to obtain joint occurrence probability of a plurality of joint examples;
the primary and secondary division module is used for determining a primary noise hazard zone and a secondary noise hazard zone in a surrounding area of the ring wind power plant based on the joint occurrence probability and the maximum allowable noise threshold value of all the joint instances;
And the strategy determining module is used for determining the noise coping strategy of the wind power plant based on the main noise hazard zone and the secondary noise hazard zone.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311752941.6A CN117436352B (en) | 2023-12-20 | 2023-12-20 | Wind farm noise analysis method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311752941.6A CN117436352B (en) | 2023-12-20 | 2023-12-20 | Wind farm noise analysis method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117436352A CN117436352A (en) | 2024-01-23 |
CN117436352B true CN117436352B (en) | 2024-03-22 |
Family
ID=89551940
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311752941.6A Active CN117436352B (en) | 2023-12-20 | 2023-12-20 | Wind farm noise analysis method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117436352B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106873359A (en) * | 2015-12-11 | 2017-06-20 | 中国电力科学研究院 | A kind of wind-powered electricity generation noise rating method based on cluster analysis and neutral net |
CN110533314A (en) * | 2019-08-23 | 2019-12-03 | 西安交通大学 | A kind of wind power plant exception unit recognition methods based on probability density distribution |
CN112581315A (en) * | 2020-12-25 | 2021-03-30 | 东北电力大学 | Wind power field clustering method based on extreme gradient dynamic density clustering |
CN114004512A (en) * | 2021-11-04 | 2022-02-01 | 西安热工研究院有限公司 | Multi-unit operation state outlier analysis method and system based on density clustering |
CN115345519A (en) * | 2022-09-13 | 2022-11-15 | 广东电网有限责任公司 | Offshore wind power risk assessment method, device, equipment and storage medium |
CN116861705A (en) * | 2023-09-04 | 2023-10-10 | 聚合电力工程设计(北京)股份有限公司 | Wind power plant noise and light and shadow influence analysis method based on computer simulation |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11188691B2 (en) * | 2018-12-21 | 2021-11-30 | Utopus Insights, Inc. | Scalable system and method for forecasting wind turbine failure using SCADA alarm and event logs |
EP4276303A1 (en) * | 2022-05-12 | 2023-11-15 | Vestas Wind Systems A/S | A method for controlling noise generated by a wind farm |
-
2023
- 2023-12-20 CN CN202311752941.6A patent/CN117436352B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106873359A (en) * | 2015-12-11 | 2017-06-20 | 中国电力科学研究院 | A kind of wind-powered electricity generation noise rating method based on cluster analysis and neutral net |
CN110533314A (en) * | 2019-08-23 | 2019-12-03 | 西安交通大学 | A kind of wind power plant exception unit recognition methods based on probability density distribution |
CN112581315A (en) * | 2020-12-25 | 2021-03-30 | 东北电力大学 | Wind power field clustering method based on extreme gradient dynamic density clustering |
CN114004512A (en) * | 2021-11-04 | 2022-02-01 | 西安热工研究院有限公司 | Multi-unit operation state outlier analysis method and system based on density clustering |
CN115345519A (en) * | 2022-09-13 | 2022-11-15 | 广东电网有限责任公司 | Offshore wind power risk assessment method, device, equipment and storage medium |
CN116861705A (en) * | 2023-09-04 | 2023-10-10 | 聚合电力工程设计(北京)股份有限公司 | Wind power plant noise and light and shadow influence analysis method based on computer simulation |
Non-Patent Citations (2)
Title |
---|
基于条件生成对抗网络的可再生能源日前场景生成方法;董骁翀;孙英云;蒲天骄;;中国电机工程学报;20200905;第40卷(第17期);5527-5536 * |
风机噪声环境影响预测研究;王雪莉;王志增;;环境科学与管理;20171015;第42卷(第10期);180-184 * |
Also Published As
Publication number | Publication date |
---|---|
CN117436352A (en) | 2024-01-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101592538B (en) | Method for computing steady-state output power of wind power station based on actual measured data | |
CN107909211B (en) | Wind field equivalent modeling and optimization control method based on fuzzy c-means clustering algorithm | |
CN110598726A (en) | Transmission tower bird damage risk prediction method based on random forest | |
CN104182594A (en) | Method for drawing power system wind area graph | |
CN110264002B (en) | Wind power plant micro-siting scheme evaluation method based on cluster analysis | |
Yu et al. | An efficient substation placement and sizing strategy based on GIS using semi-supervised learning | |
CN108448625A (en) | A kind of wind turbine incision wind speed computational methods based on data-driven | |
Martin et al. | Impact of near-future turbine technology on the wind power potential of low wind regions | |
CN113205210A (en) | Wind speed and power prediction method, system, equipment and storage medium for wind power plant with complex terrain | |
CN115173408B (en) | Distributed photovoltaic power station aggregation prediction method based on Pearson correlation coefficient | |
CN115358606A (en) | Method and system for evaluating energy efficiency of in-service wind power plant under gentle terrain | |
CN112700349B (en) | Method and device for selecting site of anemometer tower | |
CN107506856A (en) | Transmission line galloping situation method of discrimination and system based on wind field prediction | |
CN112200346A (en) | Short-term wind power prediction method for weather fluctuation process division and matching | |
CN110543540B (en) | Method and system for identifying ice-coated micro-terrain based on cyclic correlation | |
CN117436352B (en) | Wind farm noise analysis method and system | |
WO2024041409A1 (en) | Method and apparatus for determining representative wind generating set, and control method and apparatus | |
CN112211794A (en) | Cabin temperature abnormity early warning method, device, equipment and medium of wind turbine generator | |
CN108876060B (en) | Big data based prediction method for wind power output probability of sample collection | |
CN108932554A (en) | The method for optimizing configuration and device of a kind of wind power plant flow field measuring point | |
Huffman et al. | An exploratory analysis of wind patterns of Alberta, Canada | |
CN109816184B (en) | Topology planning method and device for large wind farm | |
CN113268552A (en) | Generator equipment hidden danger early warning method based on locality sensitive hashing | |
CN117911084B (en) | Comprehensive energy service market dividing system | |
Blazevic et al. | Assessment and analysis of wind energy potential in the mountain area around Sarajevo |
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 |