Wind turbine optimization control method and system based on temperature perception
By analyzing the position and life of the temperature sensor of the wind turbine, determining the extraction frequency and inputting the temperature perception model, the problem of low reliability of the temperature sensor is solved, and the optimization control effect of the wind turbine is improved.
Patent Information
- Application Number
- CN202310910532.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-07-24
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2043-07-24
AI Technical Summary
In the prior art, the working reliability of the temperature sensor cannot be guaranteed, resulting in poor control effect of the wind turbine, lack of in-depth analysis and low reliability of temperature perception results.
By obtaining the layout information of the temperature sensor, position influence coefficient and reliability analysis, the extraction frequency is determined, the temperature perception model is input to obtain an optimization control scheme, and the optimization control of the wind turbine is carried out.
It improves the control quality and control accuracy of the wind turbine unit, and improves the optimization and control effect of the wind turbine unit.
Smart Images

Figure CN116771601B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of optimization control, and in particular to a method and system for optimizing control of a wind turbine generator set based on temperature perception. Background Art
[0002] During the wind power generation process, the power output of the unit needs to be adjusted according to the temperature of the unit to achieve the goal of high efficiency and energy saving. At present, the temperature is mainly collected by installing temperature sensors on the wind measuring bracket. However, there are differences in the working performance of temperature sensors from different manufacturers and with different matching, and with the use of sensors, the working reliability of temperature sensors cannot be guaranteed, resulting in the control of wind turbines cannot achieve the expected effect. The existing technology has the technical problems of lack of in-depth analysis of temperature sensors, low reliability of temperature sensing results, and poor control effect of wind turbines. Summary of the invention
[0003] The present application provides a wind turbine optimization control method and system based on temperature perception, which is used to solve the technical problems in the prior art of lack of in-depth analysis of temperature sensors, low reliability of temperature perception results, and poor wind turbine control effect.
[0004] In view of the above problems, the present application provides a wind turbine optimization control method and system based on temperature perception.
[0005] In a first aspect of the present application, a wind turbine optimization control method based on temperature perception is provided, the method comprising:
[0006] Acquire N temperature sensor layout information of the target wind turbine group, wherein the N temperature sensor layout information includes N temperature sensor position information and N temperature sensor configuration information;
[0007] Determine N position influence coefficients based on the position information of the N temperature sensors;
[0008] Extracting the service life of the sensor and the preset service life information from the N temperature sensor configuration information respectively, to obtain the service life information of the N sensors and the preset service life information of the N sensors;
[0009] Collecting fault information of the N temperature sensors occurring within a preset historical time period to obtain N fault information;
[0010] Performing reliability analysis based on the N fault information, the N sensor service life information and the N sensor preset life information to obtain N reliability coefficients;
[0011] Determine N first extraction frequencies according to the N position influence coefficients, and determine N second extraction frequencies according to the N reliability coefficients;
[0012] Collect data information of N temperature sensors within a preset time window to obtain a data set of N temperature sensors;
[0013] Inputting the N first extraction frequencies, the N second extraction frequencies and the N temperature sensor data sets into a temperature perception model to obtain a temperature perception result;
[0014] The temperature sensing result is used as an index, and a search is performed based on a temperature-unit control scheme mapping relationship to obtain an optimized control scheme for the wind turbine unit, and the target wind turbine unit is optimally controlled according to the optimized control scheme for the wind turbine unit.
[0015] The second aspect of the present application provides a wind turbine optimization control system based on temperature perception, the system comprising:
[0016] A layout information acquisition module, wherein the layout information acquisition module is used to acquire layout information of N temperature sensors of a target wind turbine group, wherein the layout information of the N temperature sensors includes location information of the N temperature sensors and configuration information of the N temperature sensors;
[0017] A position influence coefficient determination module, the position influence coefficient determination module is used to determine N position influence coefficients based on the position information of the N temperature sensors;
[0018] A preset life acquisition module, the preset life acquisition module is used to extract the sensor service life and preset life information from the N temperature sensor configuration information, and obtain the N sensor service life information and the N sensor preset life information;
[0019] A fault information acquisition module, the fault information acquisition module is used to collect fault information of the N temperature sensors occurring within a preset historical time period to obtain N fault information;
[0020] A reliability coefficient obtaining module, the reliability coefficient obtaining module is used to perform reliability analysis based on the N fault information, the N sensor service life information and the N sensor preset life information to obtain N reliability coefficients;
[0021] An extraction frequency determination module, the extraction frequency determination module is used to determine N first extraction frequencies according to the N position influence coefficients, and determine N second extraction frequencies according to the N reliability coefficients;
[0022] A sensor data acquisition module, wherein the sensor data acquisition module is used to collect data information of N temperature sensors within a preset time window to obtain a set of N temperature sensor data;
[0023] A temperature sensing result obtaining module, wherein the temperature sensing result obtaining module is used to input the N first extraction frequencies, the N second extraction frequencies and the N temperature sensor data sets into a temperature sensing model to obtain a temperature sensing result;
[0024] The optimization control module is used to use the temperature sensing result as an index, search based on the temperature-unit control scheme mapping relationship, obtain the wind turbine unit optimization control scheme, and optimize the control of the target wind turbine unit according to the wind turbine unit optimization control scheme.
[0025] One or more technical solutions provided in this application have at least the following technical effects or advantages:
[0026] The present application obtains N temperature sensor layout information of the target wind turbine group, wherein the N temperature sensor layout information includes N temperature sensor location information and N temperature sensor configuration information, and then determines N position influence coefficients based on the N temperature sensor location information, and then respectively advances the sensor service life and preset life information from the N temperature sensor configuration information to obtain N sensor service life information and N sensor preset life information, and obtains N fault information by collecting fault information occurring in the N temperature sensors within a preset historical time period, and then obtains N fault information based on the N fault information, the N sensor service life information and the N sensor preset life information. Reliability analysis is performed based on the information to obtain N reliability coefficients, and then N first extraction frequencies are determined based on the N position influence coefficients, and N second extraction frequencies are determined based on the N reliability coefficients. By collecting data information of N temperature sensors in a preset time window, N temperature sensor data sets are obtained, and then the N first extraction frequencies, N second extraction frequencies and N temperature sensor data sets are input into the temperature perception model to obtain temperature perception results, and then the temperature perception results are used as indexes to retrieve based on the mapping relationship between temperature and unit control schemes to obtain the wind turbine optimization control scheme, and the target wind turbine is optimized and controlled according to the wind turbine optimization control scheme. The technical effect of improving the control quality of wind turbines and improving control accuracy is achieved. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0028] Figure 1 A schematic flow chart of a wind turbine optimization control method based on temperature perception provided in an embodiment of the present application;
[0029] Figure 2 A schematic diagram of a flow chart for obtaining N position influence coefficients in a wind turbine optimization control method based on temperature perception provided in an embodiment of the present application;
[0030] Figure 3 A schematic diagram of a process for obtaining N second extraction frequencies in a wind turbine optimization control method based on temperature perception provided in an embodiment of the present application;
[0031] Figure 4 A schematic diagram of the structure of a wind turbine optimization control system based on temperature perception provided in an embodiment of the present application.
[0032] Explanation of the accompanying drawings: layout information acquisition module 11, position influence coefficient determination module 12, preset life acquisition module 13, fault information acquisition module 14, reliability coefficient acquisition module 15, extraction frequency determination module 16, sensor data acquisition module 17, temperature sensing result acquisition module 18, optimization control module 19. DETAILED DESCRIPTION
[0033] The present application provides a wind turbine optimization control method and system based on temperature perception, which is used to solve the technical problems in the prior art of lack of in-depth analysis of temperature sensors, low reliability of temperature perception results, and poor wind turbine control effect.
[0034] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0035] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.
[0036] Embodiment 1
[0037] like Figure 1 As shown, the present application provides a wind turbine optimization control method based on temperature perception, wherein the method comprises:
[0038] Step S100: acquiring N temperature sensor layout information of a target wind turbine group, wherein the N temperature sensor layout information includes N temperature sensor position information and N temperature sensor configuration information;
[0039] In an embodiment of the present application, the target wind turbine group is any wind turbine group that needs to be optimized and controlled based on the results of temperature perception. The N temperature sensor layout information is used to describe the layout of the temperature sensors in the target wind turbine group, including N temperature sensor position information and N temperature sensor configuration information. Among them, the N temperature sensor position information describes the corresponding positions of the N temperature sensors in the target wind turbine group. The N temperature sensor configuration information describes the configuration features of the N temperature sensors that are different from other temperature sensors, including service life, design life, manufacturer, etc.
[0040] Step S200: determining N position influence coefficients based on the position information of the N temperature sensors;
[0041] Further, such as Figure 2 As shown, step S200 of the embodiment of the present application also includes:
[0042] Step S210: obtaining the unit area boundary of the target wind turbine unit;
[0043] Step S220: traverse the position information of the N temperature sensors, determine the distance values between the N temperature sensors and the boundary of the unit area, and obtain N longitudinal distance sets and N transverse distance sets;
[0044] Step S230: determining N first position influence coefficients based on the N longitudinal distance sets;
[0045] Step S240: determining N second position influence coefficients based on the N lateral distance sets;
[0046] Step S250: performing weighted calculations on the N first position influence coefficients and the N second position influence coefficients according to preset weight ratios to obtain N position influence coefficients.
[0047] In a possible embodiment, the influence of different positions of the N temperature sensors in the target wind turbine on the data collected by the temperature sensors is determined according to the position information of the N temperature sensors, that is, the N position influence coefficients. Thus, the goal of analyzing the accuracy of the sensor data collected from the perspective of the sensor position is achieved.
[0048] Specifically, the unit area boundary of the target wind turbine group is the boundary obtained by connecting the position of the unit at the edge of the target wind turbine group in series. The position points of the N temperature sensors in the target wind turbine group are determined by the position information of the N temperature sensors, and the horizontal and vertical distance values of the position points of the N temperature sensors in the target wind turbine group from the unit area boundary of the target wind turbine group are calculated respectively, and N longitudinal distance sets and N transverse distance sets are obtained. Among them, each longitudinal distance set includes two distance values of each temperature sensor from the unit area boundaries on both sides of the longitudinal direction of the target wind turbine group. Each transverse distance set includes two distance values of each temperature sensor from the unit area boundaries on both sides of the transverse direction of the target wind turbine group. Select the smaller distance value in the N longitudinal distance sets respectively, compare it with the longitudinal distance value of the unit area boundary of the target wind turbine group, and use the ratio as the N first position influence coefficients. The larger the ratio, the larger the corresponding N first position influence coefficients, and the greater the degree to which the temperature sensor is affected by the position. Select the smaller distance value from the N lateral distance sets respectively, compare it with the lateral distance value of the unit area boundary of the target wind turbine unit, and use the ratio as the N second position influence coefficients. The larger the ratio, the larger the corresponding N second position influence coefficients, and the greater the degree to which the temperature sensor is affected by the position. Among them, the preset weight ratio is a weight ratio value pre-set by a technician in this field, and the N first position influence coefficients and the N second position influence coefficients are weightedly calculated according to the preset weight ratio to obtain the N position influence coefficients. The technical effect of improving the accuracy of the position influence coefficient calculation and thus improving the accuracy of temperature perception is achieved.
[0049] Step S300: extracting the service life of the sensor and the preset service life information from the N temperature sensor configuration information respectively, to obtain the service life information of the N sensors and the preset service life information of the N sensors;
[0050] Step S400: collecting fault information of the N temperature sensors within a preset historical time period to obtain N fault information;
[0051] In a possible embodiment, the sensor service life and preset service life information are respectively extracted from the N temperature sensor configuration information to obtain N sensor service life information and N sensor preset service life information. The N sensor service life information is the length of time the N sensors have been used. The N sensor preset service life information is the length of time the N sensors can be used without failure, which is preset.
[0052] In one embodiment, N fault information is obtained by collecting fault conditions of N temperature sensors within a preset historical time period, wherein the N fault information is used to record faults of the N temperature sensors, including fault time, fault type, fault degree, etc.
[0053] Step S500: performing reliability analysis based on the N fault information, the N sensor service life information and the N sensor preset life information to obtain N reliability coefficients;
[0054] Furthermore, step S500 of the embodiment of the present application further includes:
[0055] Step S510: The reliability formula is:
[0056]
[0057] Among them, G i is the reliability coefficient of the ith temperature sensor, x1 is the proportion of the temperature sensor service life affecting the reliability, x2 is the proportion of the temperature sensor failure affecting the reliability, x1+x2=1, t 0i is the preset life of the ith temperature sensor, t 1i is the service life of the ith temperature sensor, i is the number of temperature sensors, n is the total number of failures of the ith temperature sensor, j is the failure sequence value of the ith temperature sensor in the preset historical time period, i is an integer greater than or equal to 1, n is an integer greater than or equal to 1, j is an integer greater than or equal to 1, y ij is the damage of the i-th temperature sensor when the j-th failure occurs, Y ij is the performance value of the sensor when the i-th temperature sensor has the j-th fault, and α is the empirical coefficient for fault impact analysis;
[0058] Step S520: inputting the N fault information, the N sensor service life information and the N sensor preset life information into the reliability analysis formula respectively to perform reliability analysis and obtain the N reliability coefficients.
[0059] Furthermore, step S500 of the embodiment of the present application further includes:
[0060] Step S530: acquiring a jth performance data set by collecting performance data of the i-th temperature sensor before the j-th failure occurs, wherein the j-th performance data set includes a data transmission speed and a temperature data collection lag time;
[0061] Step S540: Evaluate the j-th performance data set using an expert survey method to generate a sensor performance value when the i-th temperature sensor has the j-th failure.
[0062] Specifically, by using the reliability formula to quantitatively analyze the influence of life and failure on N temperature sensors, N reliability coefficients are obtained, wherein the N reliability coefficients reflect the reliability of the N temperature sensors after use and after failure repair.
[0063] Specifically, the N fault information, the N sensor service life information and the N sensor preset life information are respectively input into the reliability analysis formula to perform reliability analysis, so as to obtain the N reliability coefficients.
[0064] Specifically, the jth performance data set is obtained by collecting the performance data of the i-th temperature sensor before the j-th failure occurs, wherein the j-th performance data set includes the data transmission speed and the temperature data collection lag time. The j-th performance data set is evaluated by the expert survey method to generate the performance value of the sensor when the i-th temperature sensor has the j-th failure. Exemplarily, 10 experts are used to evaluate the j-th performance data set respectively to obtain 10 evaluation values, and the 10 evaluation values are averaged, and the processed results are used as the performance value of the sensor when the i-th temperature sensor has the j-th failure.
[0065] Step S600: determining N first extraction frequencies according to the N position influence coefficients, and determining N second extraction frequencies according to the N reliability coefficients;
[0066] Further, such as Figure 3 As shown, step S600 of the embodiment of the present application also includes:
[0067] Step S610: obtaining a preset extraction frequency;
[0068] Step S620: respectively divide the N position influence coefficients by the total value of the N position influence coefficients, and multiply the ratio by the preset extraction frequency to obtain N first extraction frequencies;
[0069] Step S630: respectively divide the N reliability coefficients by the total value of the N reliability coefficients, and multiply the ratio by the preset extraction frequency to obtain N second extraction frequencies.
[0070] In a possible embodiment, N first extraction frequencies are determined according to the N position influence coefficients, wherein the first extraction frequency is the interval time for collecting data from the temperature sensor. N second extraction frequencies are determined according to the N reliability coefficients, wherein the second extraction frequency is the interval time for collecting data from the temperature sensor determined according to the reliability coefficient.
[0071] Specifically, the preset extraction frequency is a preset interval time set by a person skilled in the art to collect temperature data from a temperature sensor in a target wind turbine. The N position influence coefficients are divided by the total value of the N position influence coefficients, and the ratio is multiplied by the preset extraction frequency to obtain N first extraction frequencies. Similarly, the N reliability coefficients are divided by the total value of the N reliability coefficients, and the ratio is multiplied by the preset extraction frequency to obtain N second extraction frequencies.
[0072] Step S700: collecting data information of N temperature sensors within a preset time window to obtain a data set of N temperature sensors;
[0073] Step S800: inputting the N first extraction frequencies, the N second extraction frequencies and the N temperature sensor data sets into a temperature perception model to obtain a temperature perception result;
[0074] Furthermore, step S800 of the embodiment of the present application further includes:
[0075] Step S810: constructing the temperature perception model, wherein the temperature perception model includes a first perception branch, a second perception branch and a temperature perception analysis layer;
[0076] Step S820: inputting the N first extraction frequencies and the N temperature sensor data sets into the first perception branch to obtain a first perception data extraction result set;
[0077] Step S830: inputting the N second extraction frequencies and the N temperature sensor data sets into the second perception branch to obtain a second perception data extraction result set;
[0078] Step S840: Input the first perception data extraction result and the second perception data extraction result into the temperature perception analysis layer to obtain the temperature perception result.
[0079] Furthermore, step S800 of the embodiment of the present application further includes:
[0080] Step S850: Acquire multiple sample first extraction frequencies, multiple sample temperature sensor data sets, and multiple sample first perception data extraction result sets as a first training set;
[0081] Step S860: Use the first training set to perform supervised training on the framework constructed based on the feedforward neural network until the output reaches convergence to obtain the first perception branch.
[0082] In a possible embodiment, the preset time window is a preset time period for collecting temperature sensor data. Within the preset time window, the data information collected by N temperature sensors is acquired to obtain N temperature sensor data sets. The N temperature sensor data sets reflect the temperature change of the target wind turbine set within the preset time window.
[0083] In one embodiment, the temperature perception model is an intelligent model for analyzing the temperature of a target wind turbine, the input data is N first extraction frequencies, N second extraction frequencies and N temperature sensor data sets, and the output data is a temperature perception result. The temperature perception result reflects the temperature change of the target wind turbine.
[0084] Specifically, the temperature perception model includes a first perception branch, a second perception branch and a temperature perception analysis layer. The N first extraction frequencies and the N temperature sensor data sets are input into the first perception branch to obtain a first perception data extraction result set, the N second extraction frequencies and the N temperature sensor data sets are input into the second perception branch to obtain a second perception data extraction result set, and the first perception data extraction result and the second perception data extraction result are input into the temperature perception analysis layer to obtain a temperature perception result.
[0085] Specifically, by obtaining multiple sample first extraction frequencies, multiple sample temperature sensor data sets, and multiple sample first perception data extraction result sets as the first training set, the first training set is used to perform supervised training on the framework built based on the feedforward neural network until the output reaches convergence, and the first perception branch is obtained. The technical effect of improving the intelligence level of temperature perception analysis is achieved.
[0086] Step S900: using the temperature sensing result as an index, searching based on the temperature-unit control scheme mapping relationship, obtaining the wind turbine unit optimization control scheme, and optimizing the control of the target wind turbine unit according to the wind turbine unit optimization control scheme.
[0087] Specifically, the temperature sensing result is used as an index, and a search is performed based on the temperature-unit control scheme mapping relationship to obtain a wind turbine unit optimization control scheme, and the target wind turbine unit is optimized and controlled according to the wind turbine unit optimization control scheme. The wind turbine unit optimization control scheme is used to optimize the control of the wind turbine unit, including unit control parameters, unit output power, etc.
[0088] Specifically, the temperature-unit control scheme mapping relationship is used to describe the correspondence between the temperature sensing result and the wind turbine unit optimization control scheme. Preferably, the temperature-unit control scheme mapping relationship is generated by obtaining multiple historical temperature sensing results and multiple wind turbine unit optimization control schemes according to the correspondence between the two.
[0089] In summary, the embodiments of the present application have at least the following technical effects:
[0090] The present application determines N position influence coefficients and N reliability coefficients by analyzing the position, life and fault of the temperature sensor in the target wind turbine, and then obtains N temperature sensor data sets in a preset time window, inputs N first extraction frequencies, N second extraction frequencies and N temperature sensor data sets into the temperature perception model, obtains the temperature perception result, and then uses the temperature perception result as an index to retrieve based on the temperature-unit control scheme mapping relationship to obtain the wind turbine optimization control scheme, and optimizes the target wind turbine according to the wind turbine optimization control scheme. The technical effect of improving the accuracy of wind turbine optimization control and improving control efficiency is achieved.
[0091] Embodiment 2
[0092] Based on the same inventive concept as the wind turbine optimization control method based on temperature perception in the aforementioned embodiment, Figure 4 As shown, the present application provides a wind turbine optimization control system based on temperature perception, and the system and method embodiments in the present application are based on the same inventive concept. The system includes:
[0093] A layout information acquisition module 11, wherein the layout information acquisition module 11 is used to acquire layout information of N temperature sensors of a target wind turbine group, wherein the layout information of the N temperature sensors includes location information of N temperature sensors and configuration information of N temperature sensors;
[0094] A position influence coefficient determination module 12, wherein the position influence coefficient determination module 12 is used to determine N position influence coefficients based on the position information of the N temperature sensors;
[0095] A preset life acquisition module 13, the preset life acquisition module 13 is used to extract sensor service life and preset life information from the N temperature sensor configuration information, and obtain N sensor service life information and N sensor preset life information;
[0096] A fault information acquisition module 14, the fault information acquisition module 14 is used to collect fault information of the N temperature sensors within a preset historical time period to obtain N fault information;
[0097] A reliability coefficient obtaining module 15, wherein the reliability coefficient obtaining module 15 is used to perform reliability analysis based on the N fault information, the N sensor service life information and the N sensor preset life information to obtain N reliability coefficients;
[0098] An extraction frequency determination module 16, the extraction frequency determination module 16 is used to determine N first extraction frequencies according to the N position influence coefficients, and determine N second extraction frequencies according to the N reliability coefficients;
[0099] A sensor data acquisition module 17, wherein the sensor data acquisition module 17 is used to collect data information of N temperature sensors within a preset time window to obtain N temperature sensor data sets;
[0100] A temperature sensing result obtaining module 18, wherein the temperature sensing result obtaining module 18 is used to input the N first extraction frequencies, the N second extraction frequencies and the N temperature sensor data sets into a temperature sensing model to obtain a temperature sensing result;
[0101] The optimization control module 19 is used to use the temperature sensing result as an index, search based on the temperature-unit control scheme mapping relationship, obtain the wind turbine unit optimization control scheme, and optimize the control of the target wind turbine unit according to the wind turbine unit optimization control scheme.
[0102] Furthermore, the position influence coefficient determination module 12 is used to perform the following method:
[0103] Obtaining a unit area boundary of the target wind turbine unit;
[0104] Traversing the position information of the N temperature sensors, determining the distance values between the N temperature sensors and the boundary of the unit area, and obtaining N longitudinal distance sets and N transverse distance sets;
[0105] Determine N first position influence coefficients based on the N longitudinal distance sets;
[0106] Determining N second position influence coefficients based on the N lateral distance sets;
[0107] The N first position influence coefficients and the N second position influence coefficients are weightedly calculated according to preset weight ratios to obtain N position influence coefficients.
[0108] Furthermore, the reliability coefficient obtaining module 15 is used to perform the following method:
[0109] The reliability formula is:
[0110]
[0111] Among them, G i is the reliability coefficient of the ith temperature sensor, x1 is the proportion of the temperature sensor service life affecting the reliability, x2 is the proportion of the temperature sensor failure affecting the reliability, x1+x2=1, t 0i is the preset life of the ith temperature sensor, t 1i is the service life of the ith temperature sensor, i is the number of temperature sensors, n is the total number of failures of the ith temperature sensor, j is the failure sequence value of the ith temperature sensor in the preset historical time period, i is an integer greater than or equal to 1, n is an integer greater than or equal to 1, j is an integer greater than or equal to 1, y ij is the damage of the i-th temperature sensor when the j-th failure occurs, Y ij is the performance value of the sensor when the i-th temperature sensor has the j-th fault, and α is the empirical coefficient for fault impact analysis;
[0112] The N fault information, the N sensor service life information and the N sensor preset life information are respectively input into the reliability analysis formula to perform reliability analysis to obtain the N reliability coefficients.
[0113] Furthermore, the reliability coefficient obtaining module 15 is used to perform the following method:
[0114] By collecting performance data of the i-th temperature sensor before the j-th failure occurs, a j-th performance data set is obtained, wherein the j-th performance data set includes a data transmission speed and a temperature data collection lag time;
[0115] The expert survey method is used to evaluate the j-th performance data set and generate the sensor performance value when the i-th temperature sensor has the j-th failure.
[0116] Furthermore, the extraction frequency determination module 16 is used to perform the following method:
[0117] Get the preset extraction frequency;
[0118] Divide the N position influence coefficients by the total value of the N position influence coefficients respectively, and multiply the ratio by the preset extraction frequency to obtain N first extraction frequencies;
[0119] The N reliability coefficients are respectively divided by the total value of the N reliability coefficients, and the ratio is multiplied by the preset extraction frequency to obtain N second extraction frequencies.
[0120] Furthermore, the temperature sensing result obtaining module 18 is used to perform the following method:
[0121] Constructing the temperature perception model, wherein the temperature perception model includes a first perception branch, a second perception branch and a temperature perception analysis layer;
[0122] Inputting the N first extraction frequencies and N temperature sensor data sets into a first perception branch to obtain a first perception data extraction result set;
[0123] Inputting the N second extraction frequencies and the N temperature sensor data sets into the second perception branch to obtain a second perception data extraction result set;
[0124] The first perception data extraction result and the second perception data extraction result are input into the temperature perception analysis layer to obtain the temperature perception result.
[0125] Furthermore, the temperature sensing result obtaining module 18 is used to perform the following method:
[0126] Acquire a plurality of sample first extraction frequencies, a plurality of sample temperature sensor data sets, and a plurality of sample first perception data extraction result sets as a first training set;
[0127] The first training set is used to perform supervised training on the framework built based on the feedforward neural network until the output reaches convergence, thereby obtaining the first perception branch.
[0128] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. Other embodiments are within the scope of the attached claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0129] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.
[0130] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.
Claims
1. A wind turbine optimization control method based on temperature perception, characterized in that: The method comprises: Acquire N temperature sensor layout information of the target wind turbine group, wherein the N temperature sensor layout information includes N temperature sensor position information and N temperature sensor configuration information; Determine N position influence coefficients based on the position information of the N temperature sensors; Extracting the sensor service life and the preset service life information from the N temperature sensor configuration information respectively, to obtain the N sensor service life information and the N sensor preset service life information; Collecting fault information of the N temperature sensors occurring within a preset historical time period to obtain N fault information; Performing reliability analysis based on the N fault information, the N sensor service life information and the N sensor preset life information to obtain N reliability coefficients; Determine N first extraction frequencies according to the N position influence coefficients, and determine N second extraction frequencies according to the N reliability coefficients; Collect data information of N temperature sensors within a preset time window to obtain a data set of N temperature sensors; Inputting the N first extraction frequencies, the N second extraction frequencies and the N temperature sensor data sets into a temperature perception model to obtain a temperature perception result, wherein the temperature perception model is used to analyze the temperature of a target wind turbine; Using the temperature sensing result as an index, searching based on the temperature-unit control scheme mapping relationship, obtaining the wind turbine unit optimization control scheme, and optimizing the control of the target wind turbine unit according to the wind turbine unit optimization control scheme; Wherein, the method further comprises: Constructing the temperature perception model, wherein the temperature perception model includes a first perception branch, a second perception branch and a temperature perception analysis layer; Inputting the N first extraction frequencies and N temperature sensor data sets into the first perception branch to obtain a first perception data extraction result set; Inputting the N second extraction frequencies and the N temperature sensor data sets into the second perception branch to obtain a second perception data extraction result set; The first perception data extraction result and the second perception data extraction result are input into the temperature perception analysis layer to obtain the temperature perception result.
2. The method according to claim 1, characterized in that The method comprises: Obtaining a unit area boundary of the target wind turbine unit; Traversing the position information of the N temperature sensors, determining the distance values between the N temperature sensors and the boundary of the unit area, and obtaining N longitudinal distance sets and N transverse distance sets; Determining N first position influence coefficients based on the N longitudinal distance sets; Determining N second position influence coefficients based on the N lateral distance sets; The N first position influence coefficients and the N second position influence coefficients are weightedly calculated according to preset weight ratios to obtain N position influence coefficients.
3. The method according to claim 1, characterized in that The method comprises: The reliability formula is: Among them, G i is the reliability coefficient of the ith temperature sensor, x1 is the proportion of the temperature sensor service life affecting the reliability, x2 is the proportion of the temperature sensor failure affecting the reliability, x1+x2=1, t 0i is the preset life of the ith temperature sensor, t 1i is the service life of the ith temperature sensor, i is the number of temperature sensors, n is the total number of failures of the ith temperature sensor, j is the failure sequence value of the ith temperature sensor in the preset historical time period, i is an integer greater than or equal to 1, n is an integer greater than or equal to 1, j is an integer greater than or equal to 1, y ij is the damage of the i-th temperature sensor when the j-th failure occurs, Y ij is the performance value of the sensor when the i-th temperature sensor has the j-th fault, and α is the empirical coefficient for fault impact analysis; The N fault information, the N sensor service life information and the N sensor preset life information are respectively input into the reliability analysis formula to perform reliability analysis to obtain the N reliability coefficients.
4. The method according to claim 3, characterized in that The method comprises: By collecting performance data of the i-th temperature sensor before the j-th failure occurs, a j-th performance data set is obtained, wherein the j-th performance data set includes a data transmission speed and a temperature data collection lag time; The expert survey method is used to evaluate the j-th performance data set and generate the sensor performance value when the i-th temperature sensor has the j-th failure.
5. The method according to claim 1, characterized in that The method comprises: Get the preset extraction frequency; Divide the N position influence coefficients by the total value of the N position influence coefficients respectively, and multiply the ratio by the preset extraction frequency to obtain N first extraction frequencies; The N reliability coefficients are respectively divided by the total value of the N reliability coefficients, and the ratio is multiplied by the preset extraction frequency to obtain N second extraction frequencies.
6. The method according to claim 1, characterized in that The method comprises: Acquire a plurality of sample first extraction frequencies, a plurality of sample temperature sensor data sets, and a plurality of sample first perception data extraction result sets as a first training set; The first training set is used to perform supervised training on the framework built based on the feedforward neural network until the output reaches convergence, thereby obtaining the first perception branch.
7. The wind turbine optimization control system based on temperature perception is characterized by: The system comprises: A layout information acquisition module, wherein the layout information acquisition module is used to acquire layout information of N temperature sensors of a target wind turbine group, wherein the layout information of the N temperature sensors includes location information of the N temperature sensors and configuration information of the N temperature sensors; A position influence coefficient determination module, the position influence coefficient determination module is used to determine N position influence coefficients based on the position information of the N temperature sensors; A preset life acquisition module, the preset life acquisition module is used to extract the sensor service life and preset life information from the N temperature sensor configuration information, and obtain the N sensor service life information and the N sensor preset life information; A fault information acquisition module, the fault information acquisition module is used to collect fault information of the N temperature sensors occurring within a preset historical time period to obtain N fault information; A reliability coefficient obtaining module, the reliability coefficient obtaining module is used to perform reliability analysis based on the N fault information, the N sensor service life information and the N sensor preset life information to obtain N reliability coefficients; An extraction frequency determination module, the extraction frequency determination module is used to determine N first extraction frequencies according to the N position influence coefficients, and determine N second extraction frequencies according to the N reliability coefficients; A sensor data acquisition module, wherein the sensor data acquisition module is used to collect data information of N temperature sensors within a preset time window to obtain a set of N temperature sensor data; A temperature perception result acquisition module, the temperature perception result acquisition module is used to input the N first extraction frequencies, the N second extraction frequencies and the N temperature sensor data sets into a temperature perception model to obtain a temperature perception result, wherein the temperature perception model is used to analyze the temperature of the target wind turbine; An optimization control module, the optimization control module is used to use the temperature sensing result as an index, search based on a temperature-unit control scheme mapping relationship, obtain an optimization control scheme for the wind turbine unit, and optimize the control of the target wind turbine unit according to the optimization control scheme for the wind turbine unit; The temperature sensing result obtaining module is used to perform the following steps: Constructing the temperature perception model, wherein the temperature perception model includes a first perception branch, a second perception branch and a temperature perception analysis layer; Inputting the N first extraction frequencies and N temperature sensor data sets into the first perception branch to obtain a first perception data extraction result set; Inputting the N second extraction frequencies and the N temperature sensor data sets into the second perception branch to obtain a second perception data extraction result set; The first perception data extraction result and the second perception data extraction result are input into the temperature perception analysis layer to obtain the temperature perception result.
