Method, device and equipment for predicting output power of tidal current energy power generation device

Abstract

The invention discloses an output power prediction method, device and equipment of a tidal current energy power generation device, and relates to the field of tidal current energy power generation, and the method comprises the steps: building a matching data set between first tidal current observation data of a test sea area and average output power data of the tidal current energy power generation device; performing data quality control processing on the matched data set to remove abnormal data in the matched data set to obtain a target data set; based on the target data set, constructing a target odd data set and a target even data set by using an odd-even extraction method of data serial numbers; based on the target odd data set, constructing an initial output power learning model of the tidal current energy power generation device; and based on the target even data set, correcting the initial output power learning model to obtain a target output power learning model, and performing output power prediction of the tidal current energy power generation device based on the target output power learning model.

Description

Technical Field

[0001] The present application relates to the technical field of tidal energy power generation, and in particular to an output power prediction method, device and equipment for a tidal energy power generation device. Background Art

[0002] With the increasing demand for energy in the global economy and society, various environmental problems such as the greenhouse effect caused by the consumption of traditional fossil energy have become increasingly prominent, and people are in urgent need of building a green and low-carbon energy system. Tidal energy is a green, safe, and pollution-free renewable energy source. Compared with other marine renewable energy sources, tidal energy has the characteristics of high energy density and good predictability. Its development and utilization have been widely valued by countries around the world, so many tidal energy power generation devices have carried out demonstration applications. In the demonstration application of tidal energy power generation devices, field testing and evaluation of their output power are carried out, and then the annual power generation index of tidal energy power generation devices is calculated, which is related to the overall economic performance of tidal energy power generation devices and the investment willingness of social capital.

[0003] However, the calculation results of the average output power of the tidal energy power generation device in the related art are not accurate enough. Summary of the invention

[0004] The purpose of the present application is to provide a method, device, equipment, medium and product for predicting the output power of a tidal energy power generation device.

[0005] To achieve the above objectives, this application provides the following solutions: In a first aspect, the present application provides a method for predicting output power of a tidal energy power generation device, comprising: Establishing a matching data set between the first tidal current observation data of the test sea area and the average output power data of the tidal current energy power generation device; Performing data quality control processing on the matching data set to remove abnormal data in the matching data set to obtain a target data set; Based on the target data set, using an odd-even extraction method of data sequence numbers, a target odd-numbered data set and a target even-numbered data set are constructed; Based on the target odd-numbered data set, constructing an initial output power learning model of the tidal energy power generation device; Based on the target even-numbered data set, the initial output power learning model is modified to obtain a target output power learning model, and the output power of the tidal energy power generation device is predicted based on the target output power learning model.

[0006] In a second aspect, the present application provides an output power prediction device for a tidal energy power generation device, comprising: A building module, configured to build a matching data set between the first tidal current observation data of a test sea area and the average output power data of a tidal current power generation device; A processing module, configured to perform data quality control processing on the matching data set to eliminate abnormal data in the matching data set, so as to obtain a target data set; A first construction module, configured to construct a target odd data set and a target even data set based on the target data set by using the odd-even extraction method of data serial numbers; A second construction module, configured to construct an initial output power learning model of the tidal current power generation device based on the target odd data set; A correction module, configured to correct the initial output power learning model based on the target even data set to obtain a target output power learning model, and perform output power prediction of the tidal current power generation device based on the target output power learning model.

[0007] In a third aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the computer program to implement the steps of the output power prediction method of the tidal current power generation device described in any one of the above.

[0008] In a fourth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the output power prediction method of the tidal current power generation device described in any one of the above are implemented.

[0009] In a fifth aspect, the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps of the output power prediction method of the tidal current power generation device described in any one of the above are implemented.

[0010] According to the specific embodiments provided by the present application, the following technical effects are disclosed in the present application: The present application provides a method, device, equipment, medium and product for predicting the output power of a tidal current energy generation device. By performing data quality control processing on the matching data set to eliminate abnormal data in the matching data set and obtain a target data set, the reliability of the input data can be improved, and the interference of outliers on model training can be avoided. By using the odd-even extraction method of data serial numbers based on the target data set to construct a target odd data set and a target even data set, the target data set is split into a training set (target odd data set) and a validation set (target even data set) according to the parity of the serial numbers, avoiding local biases that may be introduced by random partitioning, enabling the model to learn more comprehensive data distribution characteristics, maximizing the utilization rate of limited data, and reducing the risk of overfitting. Through the iterative process of training with the target odd data set and correcting with the target even data set, the prediction stability of the model for unknown data is enhanced (suitable for complex and variable marine environments). The present application improves the accuracy and reliability of predicting the output power of a tidal current energy generation device through data quality control, separate construction of odd and even data sets, and model iterative correction. Description of the Drawings

[0011] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0012] Figure 1 It is a schematic flowchart of a method for predicting the output power of a tidal current energy generation device provided by an embodiment of the present application; Figure 2 It is a schematic curve diagram of an initial output power learning model provided by an embodiment of the present application; Figure 3 It is a schematic diagram of the average output power predicted by an initial output power learning model provided by an embodiment of the present application; Figure 4 It is a schematic diagram of a correction coefficient provided by an embodiment of the present application; Figure 5 It is a schematic fitting curve diagram of a second mathematical expression provided by an embodiment of the present application; Figure 6 It is a schematic diagram of the average output power predicted by a target output power learning model provided by an embodiment of the present application; Figure 7 It is a schematic comparison diagram of the prediction results of the average output power by an initial output power learning model and a target output power learning model provided by an embodiment of the present application; Figure 8A schematic diagram of functional modules of an output power prediction device for a tidal energy power generation device provided in one embodiment of the present application; Figure 9 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0013] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions 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.

[0014] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0015] In an exemplary embodiment, Figure 1 As shown, a method for predicting the output power of a tidal power generation device is provided, which is executed by a computer device, specifically, it can be executed by a computer device such as a terminal or a server alone, or it can be executed by a terminal and a server together. In the embodiment of the present application, the following steps 102 to 110 are included. Among them: Step 102: establishing a matching data set between the first tidal current observation data of the test sea area and the average output power data of the tidal current energy power generation device; Among them, the test sea area can be an ocean area divided for scientific research, engineering, environmental assessment and other purposes, and the first tidal observation data can be data obtained by measuring and recording the motion characteristics of tidal currents (tidal currents) in the ocean at multiple times; a tidal energy power generation device is a device that uses the kinetic energy of ocean tidal flows to generate electricity, and its principle is similar to that of a wind power generation device, but the energy source is converted from air flow to seawater flow; the average output power data of the tidal energy power generation device can be calculated by measuring the actual output power of the tidal energy power generation device measured at multiple times, and a matching data set of the first tidal observation data and the average output power data can be established to determine the corresponding relationship between the first tidal observation data and the average output power data, and provide a data basis and model training basis for the subsequent prediction of the average output power data of the tidal energy power generation device based on the first tidal observation data.

[0016] Step 104: performing data quality control processing on the matching data set to remove abnormal data in the matching data set to obtain a target data set; Among them, a data quality control algorithm can be adopted to perform data quality control on the matched data set after the matching in step 102 to form a target data set after quality control; data quality control can be a process of making data meet quality standards such as accuracy, integrity, consistency, timeliness, and reliability during the processes of data collection, storage, processing, and analysis. By performing data quality control on the said matched data set, data errors, noises, and biases can be eliminated or reduced, making the data credible and suitable for subsequent analysis and decision-making.

[0017] Step 106: Based on the said target data set, use the odd-even extraction method of data serial numbers to construct a target odd data set and a target even data set; Among them, the target data set after quality control formed in step 104 can be used to construct a data set required for predicting the average output power of the tidal energy power generation device; the odd-even extraction method of data serial numbers is a data segmentation method based on data serial numbers. By separately extracting the data with odd and even serial numbers in the target data set, rapid grouping of the data is realized. The data with odd serial numbers forms the target odd data set, and the data with even serial numbers forms the target even data set. The target odd data set can be expressed as and the target even data set can be expressed as .

[0018] Step 108: Based on the said target odd data set, construct an initial output power learning model for the tidal energy power generation device; Among them, the initial output power learning model can also be called an initial mathematical model of the average output power of the tidal energy power generation device, an initial average output power prediction model, or an initial model. The target odd data set in step 106 can be used to construct an initial mathematical model of the average output power of the tidal energy power generation device, that is, the target odd data set is used as a training set to train the initial output power learning model.

[0019] Step 110: Based on the said target even data set, correct the initial output power learning model to obtain a target output power learning model, and perform output power prediction for the tidal energy power generation device based on the said target output power learning model.

[0020] Among them, the target output power learning model can also be called a final mathematical model of the average output power of the tidal energy power generation device, a final average output power prediction model, or a final model. The target even data set in step 106 can be used , correct the initial mathematical model of the average output power of the tidal energy power generation device in step 108 to form the final mathematical model of the average output power of the tidal energy power generation device, that is, the target even-numbered data set The initial output power learning model is corrected as a validation set to obtain a target output power learning model. Subsequently, new power flow observation data can be input into the target output power learning model so that the target output power learning model outputs a predicted average output power.

[0021] By implementing the above-mentioned steps 102 to 110, data quality control processing is performed on the matching data set to eliminate abnormal data in the matching data set to obtain a target data set, which can improve the reliability of input data and avoid abnormal values ​​interfering with model training; based on the target data set, a target odd data set and a target even data set are constructed by using the odd and even extraction method of the data sequence number, and the target data set is split into a training set (target odd data set) and a verification set (target even data set) according to the parity of the sequence number, avoiding local deviations that may be introduced by random division, so that the model learns more comprehensive data distribution characteristics, maximizes the utilization of limited data, reduces the risk of overfitting, and enhances the prediction stability of the model for unknown data through the iterative process of training the target odd data set and correcting the target even data set (applicable to complex and changeable marine environments); the present application improves the accuracy and reliability of the output power prediction of the tidal energy power generation device through data quality control, construction of odd and even data sets and iterative model correction.

[0022] In another exemplary embodiment of the present application, the above step 102 may be replaced by the following steps 1021 to 1023: Step 1021: acquiring original tidal current observation data of the test sea area, performing spatial conversion processing on the original tidal current observation data to obtain first tidal current observation data, wherein the original tidal current observation data is vertical profile data of the test sea area at a historical moment, and the first tidal current observation data is impeller swept section characteristic data input to the tidal current energy power generation device at a historical moment; The original tidal current observation data may be data directly acquired by the tidal current observation equipment in the test sea area, and the original tidal current observation data may include observation time data ( ), vertical profile velocity data of tidal flow velocity ( ), vertical section flow data of tidal flow direction ( ); where k is the sequence number of the tidal velocity and flow direction data in the vertical section, and its value can be expressed as , It is the total number of tidal flow velocity and flow direction data in the vertical section.

[0023] The first power flow observation data may include observation time data ( ), the characteristic velocity data of the tidal current within the swept cross-section of the impeller of the tidal current power generation device ( ), the characteristic direction data of the tidal current within the swept cross-section of the impeller of the tidal current power generation device ( ); where i is the serial number of the tidal current observation data, and its value can be expressed as , is the total number of the tidal current observation data.

[0024] By performing spatial conversion processing on the vertical profile velocity data in the original tidal current observation data, the characteristic velocity data of the tidal current can be obtained, as shown in the following formula (1). The characteristic velocity data of the tidal current input into the swept cross-section of the impeller of the tidal current power generation device can be calculated through the vertical profile velocity data of the tidal current velocity ( ): ): (1); Where A is the swept area of the impeller of the tidal current power generation device, is the swept area of the impeller of the tidal current power generation device at the k th vertical profile, a is the serial number of the vertical profile at the lowest end of the impeller of the tidal current power generation device, b is the serial number of the vertical profile at the uppermost end of the impeller of the tidal current power generation device.

[0025] By performing spatial conversion processing on the vertical profile flow direction data in the original tidal current observation data, the characteristic flow direction data of the tidal current can be obtained, as shown in the following formula (2). The characteristic direction data of the tidal current input into the swept cross-section of the impeller of the tidal current power generation device can be calculated through the vertical profile flow direction data of the tidal current flow direction ( ): ): (2); It should be noted that: when the calculation result of is negative, the final result is obtained only after adding 360°.

[0026] Step 1022: Obtain the original output power measurement data of the tidal current power generation device, and perform mean processing on the original output power measurement data to obtain the average output power data of the tidal current power generation device; Wherein, the original output power measurement data can be the data obtained by the power measurement equipment of the tidal current power generation device, and the original output power measurement data can include measurement time data ( ) and the original output power measurement data output by the tidal current power generation device ( ), where j is the serial number of the data acquired by the power measurement device, and its value can be expressed as , is the total number of data acquired by the power measurement device. The average output power data refers to the average output power data within the time interval of the tidal current observation data ( ).

[0027] Since in the on-site test and analysis of the tidal current power generation device, the data acquisition frequency of the power measurement device of the tidal current power generation device is higher than that of the tidal current observation device, therefore, within the time interval of the tidal current observation data, the power measurement device will acquire n original output power measurement data output by the tidal current power generation device ( ). It is easy to know that when the time of the tidal current observation data is synchronized with the time of the original output power measurement data output by the tidal current power generation device, that is, when, the time corresponding to the original output power measurement data output by the tidal current power generation device corresponding to is . Among them,

[0028] Calculate the average output power data of the tidal current power generation device within the time interval of the tidal current observation data, that is, within the time interval of ( ), as shown in the following formula (3), the average output power data ( ) can be calculated: (3); Among them, represents that when calculating the average output power data of the tidal current power generation device within the time interval from to , the original output power measurement data corresponding to the moment participates in the calculation, while the original output power measurement data corresponding to the moment does not participate in the calculation.

[0029] Step 1023: Match the first tidal current observation data and the average output power data to obtain a matching data set.

[0030] Among them, the matching data set may include observation time data ( ), the tidal current characteristic velocity data input into the impeller swept cross-section range of the tidal current power generation device ( ), the tidal current characteristic direction data within the impeller swept cross-section of the tidal current power generation device ( ), and the average output power data of the tidal current power generation device ( ).

[0031] In the embodiment of the present application, the calculation method of "the tidal current characteristic direction data within the impeller swept cross-section of the tidal current power generation device ( )" is formula (2). The proposal of formula (2) avoids the calculation error caused by using the "direct arithmetic mean" calculation method and improves the accuracy of calculating the tidal current characteristic direction data. For example, the arithmetic mean of 30° and 315° is 172.5°, but the tidal current characteristic direction data calculated by using the method proposed in the embodiment of the present application is 352.5°, and the calculation result is more consistent with the actual situation.

[0032] In the embodiment of the present application, by converting the original vertical profile data into the impeller swept cross-section characteristic data, the effective energy capture area can be focused, and the hydrodynamic characteristics of the actual force on the impeller can be more accurately reflected, improving the physical rationality of the prediction model; by performing mean processing on the original output power measurement data to obtain the average output power data, the instantaneous fluctuation interference of the power data can be reduced, highlighting the long-term change trend of the tidal current energy, avoiding high-frequency noise interference in model training, and improving the signal-to-noise ratio of the data; by time-aligning and matching the first tidal current observation data after spatial transformation processing and the average output power to construct a matching data set, the physical causal relationship between the input (flow velocity) and the output (power) is ensured to be clear, avoiding false associations caused by time asynchronization.

[0033] In another exemplary embodiment of the present application, the matching data set includes the observation time data, the tidal current characteristic flow velocity data within the impeller swept cross-section of the tidal current power generation device, the tidal current characteristic direction data within the impeller swept cross-section of the tidal current power generation device, and the average output power data of the tidal current power generation device. The above step 104 can be replaced by the following steps 1041 to 1043: Step 1041: Divide the matching data set according to a preset first flow velocity interval to obtain a plurality of first data groups; Among them, the first flow velocity interval can be expressed as , and the matching data set can be divided according to a fixed first flow velocity interval . Then, the number of the divided first data groups can be calculated by the following formula (4): (4); Among them, can represent the maximum tidal current velocity in the tidal current characteristic flow velocity data of the matching data set, It can represent the minimum tidal current velocity in the tidal current characteristic velocity data, denotes rounding up, represents the total number of the first data groups divided, and is denoted by m represents the serial number of the first data group after division, and its value can be expressed as , and the tidal current velocity range of the m-th first data group can be expressed as .

[0034] Specifically, the matching data set can be divided into data groups according to a fixed first flow velocity interval, and then the mutually matching observation time data in the matching data set ( ), the tidal current characteristic velocity data ( ) input into the impeller swept cross-section range of the tidal current energy generation device, the tidal current characteristic direction data ( ) input into the impeller swept cross-section range of the tidal current energy generation device, and the average output power data ( ) of the tidal current energy generation device are divided into the corresponding data groups.

[0035] Step 1042: Using the data quality control algorithm, perform data quality control processing on the average output power data in each of the first data groups to eliminate the abnormal output power data in the corresponding first data group and the observation time data, tidal current characteristic velocity data, and tidal current characteristic direction data corresponding to the abnormal output power data, and obtain the corresponding first sub-data groups; Among them, in the on-site test work of the tidal current energy generation device, due to the existence of factors such as measurement errors and drift of measurement data, there will inevitably be certain abnormal data in the on-site test data set. In order to ensure the accuracy of the prediction result of the average output power of the tidal current energy generation device, it is necessary to perform quality control on the data in the divided first data groups.

[0036] First, the Shapiro-Wilk test method can be used to perform a normality test on the average output power data ( ) of the tidal current energy generation device in each first data group m. Secondly, judge the normality test result of the average output power data ( m ) of the tidal current energy generation device in each first data group . If the discrimination test result is a normal distribution, then use the following formula (5) to calculate the distribution range m of the data to be retained in the average output power data ( ) of the tidal current energy generation device in each first data group , and for the average output power data of the tidal current energy generation device exceeding range ), eliminate them and retain the distribution range of the data .

[0037] (5); Among them, represents the number of average output power data of the tidal current energy generation devices within the m -th first data group ( ), represents the average value of average output power data, represents the standard deviation of average output power data.

[0038] If the discriminant test result is not a normal distribution, the first quartile m and the third quartile of the average output power data of the tidal current energy generation devices within each first data group ( and ) can be calculated using software such as Origin or MATLAB. Then, use the following formula (6) to calculate the distribution range m of the average output power data of the tidal current energy generation devices within each first data group ( ) that should retain the data. Eliminate the average output power data of the tidal current energy generation devices ( ) that exceed the range of , and retain the distribution range of the data. .

[0039] (6); Among them, according to the m or range within each first data group , eliminate the average output power data of the tidal current energy generation devices ( or ) that exceed the or range, as well as the corresponding observation time data ( m ) within each first data group , the tidal current characteristic velocity data ( m ) input into the impeller swept cross-section range of the tidal current energy generation devices within each first data group , and the tidal current characteristic direction data ( m ) input into the impeller swept cross-section range of the tidal current energy generation devices within each first data group to obtain the first sub-data group corresponding to each first data group m. The first sub-data group is the first data group after eliminating abnormal data.

[0040] Step 1043: Sort multiple first sub-data groups according to the observation time data to obtain a target data set.

[0041] Among them, the observation time data within each first sub-data group ( ), the tidal current characteristic velocity data input into the impeller swept cross-section range of the tidal current power generation device ( ), the tidal current characteristic direction data input into the impeller swept cross-section range of the tidal current power generation device ( ), and the average output power data of the tidal current power generation device ( ) can be combined into a new data set in the order of the observation time data, which is the target data set after quality control.

[0042] Specifically, the target data set may include the observation time data ( ), the tidal current characteristic velocity data input into the impeller swept cross-section range of the tidal current power generation device ( ), the tidal current characteristic direction data input into the impeller swept cross-section range of the tidal current power generation device ( ), and the average output power data of the tidal current power generation device ( ), where is the serial number of the data in the data set after quality control, and its value can be expressed as , is the total number of data in the target data set after quality control.

[0043] In the embodiment of the present application, by proposing a "data quality control algorithm" in the average output power prediction work of the tidal current power generation device, for the average output power data of the tidal current power generation device within each divided data group ( On the basis of performing a normality test, the characteristics of the quartile method that are insensitive to extreme outliers and the excellent identification characteristics of the 3σ method for abnormal data in a normal distribution dataset are fully utilized. At the same time, this application also considers the false alarm and missed alarm risks in data quality control work, and then proposes a "data quality control algorithm" in the average output power prediction of tidal current power generation devices, providing reliable data support for constructing a mathematical model of the average output power of tidal current power generation devices; by grouping and cleaning abnormal data according to the first flow velocity interval and finally reconstructing a target dataset with consistent time series, the refinement degree of data quality control can be improved, avoiding the deviation of global outlier detection. Traditional methods may lead to misjudgment due to uneven flow velocity distribution (such as less high-flow velocity data), while grouped processing can more accurately identify local outliers. Retain the true physical characteristics of different flow velocity intervals (such as large power fluctuations at low flow velocities and different noise sources at high flow velocities); when removing abnormal data, synchronously remove the associated observation time, flow velocity, and direction data to ensure the integrity of data records, avoid implicating valid tidal current data due to a single abnormal power value, and maintain the physical association between the input (flow velocity / direction) and output (power); the first sub-dataset after cleaning is re-sorted according to the observation time to generate the final target dataset, restoring the time series characteristics of the data, preventing time breaks caused by data cleaning, and facilitating subsequent model training.

[0044] In another exemplary embodiment of this application, the matching dataset includes observation time data, tidal current characteristic flow velocity data within the impeller swept cross-section range input to the tidal current power generation device, tidal current characteristic direction data within the impeller swept cross-section range input to the tidal current power generation device, and average output power data of the tidal current power generation device. The above step 106 can be replaced by the following steps 1061 to 1066: Step 1061: Use the odd-even extraction method of data serial numbers to split the target dataset into a first odd dataset and a first even dataset; Among them, for the observation time data ( ), tidal current characteristic flow velocity data ( ), tidal current characteristic direction data ( ), and average output power data ( ) in the target dataset, the data when is odd can be extracted to form a first odd dataset , and the data when is even can be extracted to form a first even dataset .

[0045] The first odd dataset may include: observation time data ( ), the characteristic velocity data of the tidal current within the impeller swept cross-section of the tidal current power generation device ( ), the characteristic direction data of the tidal current within the impeller swept cross-section of the tidal current power generation device ( ), the average output power data of the tidal current power generation device ( ), represents the data sequence number in the odd dataset, and its value can be expressed as , is the total number of data in the first odd dataset.

[0046] The first even dataset may include: observation time data ( ), the characteristic velocity data of the tidal current within the impeller swept cross-section of the tidal current power generation device ( ), the characteristic direction data of the tidal current within the impeller swept cross-section of the tidal current power generation device ( ), the average output power data of the tidal current power generation device ( ). represents the data sequence number in the even dataset, and its value can be expressed as , is the total number of data in the first even dataset.

[0047] Step 1062: Divide the first odd dataset according to a preset second flow velocity interval to obtain multiple second data groups; divide the first even dataset according to the second flow velocity interval to obtain multiple third data groups; Among them, the second flow velocity interval can be expressed as , and the first odd dataset and the first even dataset can be divided respectively according to a fixed second flow velocity interval . Then, the number of the divided second data groups can be calculated by the following formula (7), and the number of the divided third data groups can be calculated by the following formula (8): (7); (8); Among them, represents the total number of the second data groups that the first odd dataset can be divided into, represents the serial number of the odd number array, and its value can be expressed as . represents the total number of the third data groups that the first even dataset can be divided into, represents the serial number of the even data group, and its value is , . Represents the maximum tidal current velocity in the first odd dataset in Represents the first odd dataset the minimum tidal current velocity in Represents the first even dataset the maximum tidal current velocity in Represents the first even dataset the minimum tidal current velocity in the tidal current velocity range of the th second data group can be expressed as . The tidal current velocity range of the th third data group can be expressed as . The tidal current characteristic velocity data ( ) within the impeller swept cross-section range of the tidal energy power generation device input in the first odd dataset can be used to divide the corresponding tidal current characteristic direction data (

[0048] ) within the impeller swept cross-section range of the tidal energy power generation device input in the first even dataset ), and the average output power data (

[0049] ) of the tidal energy power generation device into the corresponding second data group.

[0049] It should be noted that to distinguish the tidal current characteristic velocity data, tidal current characteristic direction data, and average output power data in the second data group obtained by dividing the first odd dataset and the third data group obtained by dividing the first even dataset , the tidal current characteristic velocity data in the second data group can be named the first tidal current characteristic velocity data, the tidal current characteristic direction data in the second data group can be named the first tidal current characteristic direction data, and the average output power data in the second data group can be named the first average output power data; similarly, the tidal current characteristic velocity data in the third data group can be named the second tidal current characteristic velocity data, the tidal current characteristic direction data in the third data group can be named the second tidal current characteristic direction data, and the average output power data in the third data group can be named the second average output power data.

[0050] Step 1063: Based on the first tidal current characteristic velocity data in each of the second data groups, determine the first velocity mean value in the corresponding second data group; based on the first tidal current characteristic direction data in each of the second data groups, determine the first coefficient mean value of the utilization coefficient of the tidal current energy generation device for tidal current energy resources in the corresponding second data group; based on the first average output power data in each of the second data groups, determine the first power mean value in the corresponding second data group. First, as shown in the following formula (9), the mean value of the first tidal current characteristic velocity data in each second data group can be calculated. Among them, ): (9); Where, represents the first velocity mean value of the first tidal current characteristic velocity data of the th second data group, represents the number of the first tidal current characteristic velocity data in the th second data group, represents the first tidal current characteristic velocity data located in the th second data group.

[0051] Secondly, as shown in the following formula (10), the mean value of the utilization coefficient of the tidal current energy generation device for tidal current energy resources in each second data group can be calculated. Among them, (10); Where, represents the first coefficient mean value of the utilization coefficient of the tidal current energy generation device for tidal current energy resources in the th second data group, represents the number of the first tidal current characteristic velocity data in the th second data group, that is, the number of the utilization coefficients of the tidal current energy generation device for tidal current energy resources in the th second data group, is the utilization coefficient of the tidal current energy generation device for tidal current energy resources in the th second data group, is the first odd data set the acute angle between the first tidal current characteristic direction data in and the perpendicular line of the impeller swept cross-section of the tidal current energy generation device; the utilization coefficient is calculated based on the angular relationship between the tidal current characteristic direction data input to the impeller swept cross-section of the tidal current energy generation device and the perpendicular line of the impeller swept cross-section of the tidal current energy generation device, and is crucial for analyzing in detail the energy index of the tidal current energy input to the impeller swept cross-section of the tidal current energy generation device.

[0052] Finally, as shown in the following formula (11), calculate the mean value of the first average output power data in each second data group (): ): (11) Wherein,[[]] represents the first power mean value of the first average output power data in the th second data group, represents the th second data group, that is, the number of the first average output power data in the th second data group ( ), represents the first average output power data located in the th second data group.

[0053] Step 1064: Based on the second tidal current characteristic velocity data in each of the third data groups, determine the second velocity mean value in the corresponding third data group; based on the second tidal current characteristic direction data in each of the third data groups, determine the second coefficient mean value of the utilization coefficient of the tidal energy power generation device for tidal energy resources in the corresponding third data group; based on the second average output power data in each of the third data groups, determine the second power mean value in the corresponding third data group; First, as shown in the following formula (12), the mean value of the second tidal current characteristic velocity data ( ) in each third data group can be calculated: ): (12); Wherein,[[]] represents the second velocity mean value of the second tidal current characteristic velocity data of the th third data group, represents the number of the second tidal current characteristic velocity data in the th third data group, represents the second tidal current characteristic velocity data located in the th third data group.

[0054] Secondly, as shown in the following formula (13), the mean value of the utilization coefficient of the tidal energy power generation device for tidal energy resources in each third data group can be calculated: ): (13); Wherein,[[]] represents the second coefficient mean value of the utilization coefficient of the tidal energy power generation device for tidal energy resources in the th third data group, represents the quantity of the second tidal current characteristic velocity data in the th third data group, that is, the quantity of the utilization coefficient of the tidal current energy generation device for the tidal current energy resource in the th third data group, is the utilization coefficient of the tidal current energy generation device for the tidal current energy resource in the th second data group, is the acute angle between the second tidal current characteristic direction ( ) in the first even data set and the perpendicular line of the impeller swept cross-section of the tidal current energy generation device.

[0055] Finally, as shown in the following formula (14), calculate the mean value of the second average output power ( ) in each third data group: (14); wherein, represents the second power mean value of the second average output power data in the th third data group. represents the quantity of the tidal current characteristic velocity data in the th third data group, that is, the quantity of the second average output power data ( ) in the th third data group, represents the second average output power data located in the th third data group.

[0056] Step 1065: Based on the first flow velocity mean value, the first coefficient mean value, and the first power mean value of the multiple second data groups, construct a target odd data set; wherein, the first flow velocity mean value ( ) of the first tidal current characteristic velocity data, the first coefficient mean value ( ) of the utilization coefficient of the tidal current energy generation device for the tidal current energy resource, and the first power mean value ( ) of the first average output power data of the tidal current energy generation device in the th second data group can be sorted in ascending order according to the data group serial number divided by the first odd data set to construct the target odd data set required for predicting the average output power of the tidal current energy generation device .

[0057] Step 1066: Based on the second flow velocity mean value, the second coefficient mean value, and the second power mean value of the multiple third data groups, construct a target even data set.

[0058] Among them, the second flow velocity mean value of the second tidal current characteristic flow velocity data in the th third data group ( ), the second coefficient mean value of the utilization coefficient of the tidal current energy generation device for tidal current energy resources ( ), and the second power mean value of the second average output power data of the tidal current energy generation device ( ), according to the first even data set The data group serial numbers divided are Sorted in ascending order, the target even data set required to predict the average output power of the tidal current energy generation device is constructed .

[0059] Based on analyzing the angular relationship between the tidal current characteristic direction data input into the impeller swept cross-section range of the tidal current energy generation device and the perpendicular line of the impeller swept cross-section of the tidal current energy generation device, the present application embodiment proposes the first coefficient mean value ( ) and the second coefficient mean value ( ) of the utilization coefficient of the tidal current energy generation device for tidal current energy resources. By introducing the coefficient mean value into the established mathematical model of the average output power of the tidal current energy generation device, the influence of the characteristic direction of the tidal current on the mathematical model of the average output power of the tidal current energy generation device is corrected, making the mathematical model of the average output power of the tidal current energy generation device constructed in the present application more scientific and reasonable; first, the data set is split according to odd and even serial numbers, and then the mean values (flow velocity, utilization coefficient, power) are calculated by grouping according to the flow velocity interval. Through the grouped mean value processing, the measurement error and instantaneous fluctuation can be effectively smoothed; the mean values in different flow velocity intervals can accurately reflect the energy capture characteristic curve of the impeller; by independently performing the mean value processing on the odd and even data sets, the training set (odd) and the verification set (even) are respectively constructed, avoiding data leakage, ensuring that the verification set is completely independent of the training set, and realizing more accurate model evaluation. The verification result reflects the true generalization ability.

[0060] In another exemplary embodiment of the present application, the above step 108 may be replaced by the following steps 1081 to 1083: Step 1081: Set the first power mean value as the first dependent variable, and set the first flow velocity mean value and the first coefficient mean value as the first independent variable and the second independent variable respectively; Among them, the first power mean value ( ) of the first average output power data of the tidal current energy generation device in the target odd data set can be set as the dependent variable ( ), and the first flow velocity mean value ( ) of the first tidal current characteristic flow velocity data can be set as the first independent variable ( ), and the first coefficient mean value ( ), set the first coefficient mean value of the utilization coefficient of the tidal current energy generation device for tidal current energy resources ( ) as the second independent variable ( ).

[0061] Step 1082: Set the first mathematical expression, and determine the first parameter, the second parameter, and the third parameter in the first mathematical expression based on the first power mean value, the first flow velocity mean value, the first coefficient mean value, and the first preset parameter of the tidal current energy generation device; Among them, the first preset parameter of the tidal current energy generation device may include the mean value of the tidal current characteristic flow velocity when the tidal current energy generation device starts to generate electricity (i.e., the cut-in flow velocity ) and the rated flow velocity value of the tidal current energy generation device ( ), , can be provided by the R & D unit of the tidal current energy generation device, or can be obtained according to the on-site test results of the power characteristics of the tidal current energy generation device; the first mathematical expression can be set as , and in the first mathematical expression are the first parameter, the second parameter, and the third parameter to be determined respectively, and the first independent variable in the first mathematical expression has a domain of definition: .

[0062] Based on multiple first flow velocity mean values ( ) in the target odd data set, the first coefficient mean value ( ) corresponding to each first flow velocity mean value, and the first power mean value ( ) of the first average output power data, the Levenberg-Marquardt iterative optimization algorithm (the maximum number of iterations of the Levenberg-Marquardt iterative optimization algorithm is set to 400, and the tolerance is set to ) can be used to determine the parameters in the non-linear mathematical model . parameters.

[0063] It should be noted that setting the maximum number of iterations of the Levenberg-Marquardt iterative optimization algorithm to 400 and the tolerance to is mainly to prevent infinite loops, balance accuracy and computational efficiency, and handle special cases where the algorithm cannot converge.

[0064] Step 1083: Based on the second preset parameter of the tidal current energy generation device and the first mathematical expression, construct the initial output power learning model of the tidal current energy generation device.

[0065] Among them, the second preset parameter of the tidal current energy generation device may include the cut-out flow velocity of the tidal current energy generation device ( ), and the rated output power of the tidal current energy generation device under the rated flow velocity condition ( ). Similarly, and can be provided by the R & D unit of the tidal current energy generation device, or obtained according to the on-site test results of the power characteristics of the tidal current energy generation device.

[0066] According to the first mathematical expression determined in step 1082 and the parameters determined in step 1082, and at the same time according to the second preset parameter of the tidal current energy generation device, the initial average output power mathematical model of the tidal current energy generation device can be determined, as shown in the following formula (15): (15); As Figure 2 shown, the initial output power learning model can be represented as curve 21. Under the condition that the parameters are determined and the mean value of the first coefficient remains unchanged, curve 21 reflects the corresponding relationship between the first independent variable (i.e., the mean value of the first flow velocity) and the first dependent variable (i.e., the mean value of the first power). R square is an index of the fitting degree, which is used to reflect the ability of curve 21 to explain the data.

[0067] In the embodiment of the present application, the mean value of the first flow velocity and the mean value of the first coefficient are used as independent variables, and the first preset parameter is introduced to constrain the first mathematical expression. Based on the first mathematical expression and the second preset parameter, the initial output power learning model of the tidal current energy generation device is constructed, thereby enhancing the credibility of the model.

[0068] In another exemplary embodiment of the present application, the above step 110 may be replaced by the following steps 1101 to 1105: Step 1101: Input the mean value of the second flow velocity and the mean value of the second coefficient into the initial output power learning model, and obtain the mean value of the third power output by the initial output power learning model; Among them, the mean value of the second flow velocity ( ) of the second tidal current characteristic flow velocity in each third data group in the target even data set and the mean value of the second coefficient ( ) of the utilization coefficient of the tidal current energy generation device for tidal current energy resources can be respectively substituted into and in formula (15) to obtain the mean value of the third power ( ) calculated by applying the initial output power learning model in each third data group; As Figure 3As shown, it is a schematic diagram of the average output power predicted by the initial output power learning model. The second power mean is calculated based on the actual output power of the tidal current energy generation device. To distinguish it from the second power mean, the average output power predicted by the initial output power learning model can be named the third power mean. The initial output power learning model results in some predicted average output power data points being below the horizontal line where the average output power is zero, which obviously requires correction of the initial output power learning model.

[0069] Step 1102: Determine the correction coefficient of the initial output power learning model based on the second power mean and the third power mean; Using the calculated third power mean and the target even data set the second power mean of the second average output power of the tidal current energy generation device in each third data group in ( ), calculate the correction coefficient of the initial output power learning model in each third data group ( ), and its calculation method is shown in the following formula (16): (16); As Figure 4 shown, according to the correction coefficient calculation method proposed in this application, the correction coefficient of the initial output power learning model can be calculated. Since the second flow velocity mean, the second coefficient mean, and the second power mean in different third data groups may be different, each of the third data groups corresponds to a correction coefficient.

[0070] Step 1103: Set the correction coefficient as the second dependent variable and the second flow velocity mean as the third independent variable; Among them, the correction coefficient obtained in step 1102 ( ) can be set as the second dependent variable ( ), and the second flow velocity mean of the second tidal current characteristic flow velocity in each third data group ( ) can be set as the third independent variable ( ).

[0071] Step 1104: Set the second mathematical expression and determine the fourth parameter, the fifth parameter, and the sixth parameter in the second mathematical expression based on the correction coefficient and the second flow velocity mean; As shown in the following formula (17), a non-linear mathematical relationship (i.e., the second mathematical expression) between the correction coefficient and the second flow velocity mean of the second tidal current characteristic flow velocity can be constructed: (17); Among them, They are the fourth parameter, the fifth parameter, and the sixth parameter to be determined respectively. According to the correction coefficient ( ), and the second flow velocity mean value of the second tidal current characteristic flow velocity in each third data group ( ), by using the mathematical method of polynomial fitting, the parameter in the second mathematical expression is determined.

[0072] As Figure 5 shown, according to the correction coefficient fitting method proposed in this application, the non-linear mathematical relationship between the correction coefficient and the second flow velocity mean value (i.e., the second mathematical expression) 31 can be obtained.

[0073] Step 1105: Based on the second mathematical expression, correct the initial output power learning model to obtain the target output power learning model.

[0074] Among them, the target output power learning model can also be called the final mathematical model of the average output power of the tidal current energy generation device. The second power mean value of the second average output power of the tidal current energy generation device predicted in each third data group can be set as the first dependent variable , the second flow velocity mean value of the second tidal current characteristic flow velocity in each third data group can be set as the first independent variable , and the second coefficient mean value of the utilization coefficient of the tidal current energy generation device for the tidal current energy resource in each third data group can be set as the second independent variable .

[0075] Substitute the set , , into formula (15) respectively, and at the same time use the second mathematical expression to correct the initial output power learning model, the final mathematical model of the average output power of the tidal current energy generation device can be obtained, and its mathematical expression is as shown in the following formula (18): (18); As Figure 6 shown, the average output power predicted by the target output power learning model can be obtained. The average output power predicted by the target output power learning model can be named the fourth power mean value. As Figure 7 shown, the average output power predicted by the target output power learning model can be compared with the average output power predicted by the initial output power learning model. As shown in Table 1 below, it is the comparison table of the average output power prediction results of the initial output power learning model and the target output power learning model.

[0076] Table 1 Comparison table of the average output power prediction results of the initial mathematical model and the final mathematical model

[0077] As can be seen from Table 1 above, the average output power predicted by the target output power learning model is closer to the actual average output power of the tidal current energy generation device, and the prediction result is more accurate.

[0078] In the embodiments of the present application, the limitations of the prediction result of the average output power of the initial tidal current energy generation device average output power mathematical model constructed for the average output power of the tidal current energy generation device are studied, and a calculation method for the "correction coefficient of the initial tidal current energy generation device average output power mathematical model" is proposed, which improves the accuracy of the finally constructed tidal current energy generation device average output power mathematical model. The calculation results in Table 1 also prove that the accuracy of the finally constructed tidal current energy generation device average output power mathematical model has been improved.

[0079] The embodiments of the present application propose a fitting method for the non-linear mathematical expression between the correction coefficient and the second flow velocity mean value of the second tidal current characteristic flow velocity data, and then determine the mathematical expression between the correction coefficient and the mean value of the tidal current characteristic flow velocity, making it possible to determine the correction coefficient of the tidal current energy generation device within any flow velocity data group. Further, according to the determined mathematical expression between the correction coefficient and the mean value of the tidal current characteristic flow velocity, the initial tidal current energy generation device average output power mathematical model is corrected, and the construction of the finally constructed tidal current energy generation device average output power mathematical model is completed, forming a prediction method for the average output power of the tidal current energy generation device; the accuracy and robustness of the output power prediction model of the tidal current energy generation device are significantly improved by establishing an error compensation mechanism.

[0080] In another exemplary embodiment of the present application, the output power prediction method of the tidal current energy generation device further includes: Step 111: Obtain the second tidal current observation data of the test sea area, where the second tidal current observation data is the impeller swept cross-section characteristic data input to the tidal current energy generation device at the current moment; Among them, the second tidal current observation data may be the observation time data obtained at the current moment ( ), the tidal current characteristic flow velocity data within the impeller swept cross-section range input to the tidal current energy generation device ( ), and the tidal current characteristic direction data within the impeller swept cross-section range input to the tidal current energy generation device ( ); where i is the serial number of the tidal current observation data, and its value can be expressed as , and O is the total number of tidal current observation data.

[0081] Step 112: Input the second tidal current observation data into the target output power learning model to obtain the average output power of the tidal current energy generation device output by the target output power learning model.

[0082] Among them, the newly obtained second tidal current observation data can be input into the trained target output power learning model, and the predicted average output power can be output by the target output power learning model. It should be noted that the target output power learning model can also be continuously updated through the second tidal current observation data and the actual average output power to achieve dynamic update of the target output power learning model.

[0083] Based on the same inventive concept, an embodiment of the present application also provides an output power prediction device for a tidal current power generation device for implementing the output power prediction method of the tidal current power generation device involved above. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the output power prediction device for the tidal current power generation device provided below can refer to the limitations on the optimization scheduling method in the above text and will not be repeated here.

[0084] In an exemplary embodiment, as Figure 8 shown, an output power prediction device 400 for a tidal current power generation device is provided, including: A building module 401, configured to establish a matching data set between the first tidal current observation data of the test sea area and the average output power data of the tidal current power generation device; A processing module 402, configured to perform data quality control processing on the matching data set to eliminate abnormal data in the matching data set and obtain a target data set; A first construction module 403, configured to construct a target odd data set and a target even data set based on the target data set by using the odd-even extraction method of data serial numbers; A second construction module 404, configured to construct an initial output power learning model of the tidal current power generation device based on the target odd data set; A correction module 405, configured to correct the initial output power learning model based on the target even data set to obtain a target output power learning model, and perform output power prediction of the tidal current power generation device based on the target output power learning model.

[0085] Exemplarily, the output power prediction device 400 of the tidal current power generation device may be a component of the tidal current power generation device itself. In some embodiments, the output power prediction device 400 of the tidal current power generation device may also be a device independent of the tidal current power generation device.

[0086] As an alternative implementation, the establishing module 401 includes: a first obtaining sub-module, configured to obtain the original tidal current observation data of the test sea area, perform spatial transformation processing on the original tidal current observation data to obtain first tidal current observation data, where the original tidal current observation data is the vertical profile data of the test sea area at a historical moment, and the first tidal current observation data is the impeller swept cross-section characteristic data input to the tidal current power generation device at a historical moment; a second obtaining sub-module, configured to obtain the original output power measurement data of the tidal current power generation device, perform mean processing on the original output power measurement data to obtain the average output power data of the tidal current power generation device; and a matching sub-module, configured to match the first tidal current observation data and the average output power data to obtain a matching data set.

[0087] As an alternative implementation, the matching data set includes observation time data, tidal current characteristic velocity data within the impeller swept cross-section range input to the tidal current power generation device, tidal current characteristic direction data within the impeller swept cross-section range input to the tidal current power generation device, and the average output power data of the tidal current power generation device. The processing module 402 includes: a first dividing sub-module, configured to divide the matching data set according to a preset first flow velocity interval to obtain a plurality of first data groups; a processing sub-module, configured to use a data quality control algorithm to perform data quality control processing on the average output power data in each first data group to remove the abnormal output power data in the corresponding first data group and the observation time data, tidal current characteristic velocity data, and tidal current characteristic direction data corresponding to the abnormal output power data, to obtain corresponding first sub-data groups; and a sorting sub-module, configured to sort the plurality of first sub-data groups according to the observation time data to obtain a target data set.

[0088] As an alternative implementation, the matching data set includes observation time data, tidal current characteristic velocity data input within the impeller swept cross-section of the tidal current power generation device, tidal current characteristic direction data input within the impeller swept cross-section of the tidal current power generation device, and average output power data of the tidal current power generation device. The first construction module 403 includes: a splitting sub-module, configured to split the target data set into a first odd data set and a first even data set by using the odd-even extraction method of data serial numbers; a second partitioning sub-module, configured to partition the first odd data set according to a preset second flow velocity interval to obtain a plurality of second data groups; partition the first even data set according to the second flow velocity interval to obtain a plurality of third data groups; a first determination sub-module, configured to determine a first flow velocity mean value corresponding to each second data group based on the first tidal current characteristic velocity data in each second data group; determine a first coefficient mean value of the utilization coefficient of the tidal current power generation device for tidal current energy resources corresponding to each second data group based on the first tidal current characteristic direction data in each second data group; determine a first power mean value corresponding to each second data group based on the first average output power data in each second data group; a second determination sub-module, configured to determine a second flow velocity mean value corresponding to each third data group based on the second tidal current characteristic velocity data in each third data group; determine a second coefficient mean value of the utilization coefficient of the tidal current power generation device for tidal current energy resources corresponding to each third data group based on the second tidal current characteristic direction data in each third data group; determine a second power mean value corresponding to each third data group based on the second average output power data in each third data group; a first construction sub-module, configured to construct a target odd data set based on the first flow velocity mean value, the first coefficient mean value, and the first power mean value of the plurality of second data groups; a second construction sub-module, configured to construct a target even data set based on the second flow velocity mean value, the second coefficient mean value, and the second power mean value of the plurality of third data groups.

[0089] As an alternative implementation, the second construction module 404 includes: a first setting sub-module, configured to set the first power mean value as a first dependent variable, and set the first flow velocity mean value and the first coefficient mean value as a first independent variable and a second independent variable respectively; a second setting sub-module, configured to set a first mathematical expression, and determine a first parameter, a second parameter, and a third parameter in the first mathematical expression based on the first power mean value, the first flow velocity mean value, the first coefficient mean value, and a first preset parameter of the tidal current power generation device; a third construction sub-module, configured to construct an initial output power learning model of the tidal current power generation device based on a second preset parameter of the tidal current power generation device and the first mathematical expression.

[0090] As an alternative implementation, the correction module 405 includes: a third acquisition sub-module, configured to input the second flow velocity mean value and the second coefficient mean value into the initial output power learning model, and acquire a third power mean value output by the initial output power learning model; a third determination sub-module, configured to determine a correction coefficient of the initial output power learning model based on the second power mean value and the third power mean value; a third setting sub-module, configured to set the correction coefficient as a second dependent variable, and set the second flow velocity mean value as a third independent variable; a fourth setting sub-module, configured to set a second mathematical expression, and determine a fourth parameter, a fifth parameter, and a sixth parameter in the second mathematical expression based on the correction coefficient and the second flow velocity mean value; and a correction sub-module, configured to correct the initial output power learning model based on the second mathematical expression to obtain a target output power learning model.

[0091] As an alternative implementation, the device further includes: a first acquisition module, configured to acquire second tidal current observation data of the test sea area, where the second tidal current observation data is impeller swept cross-section characteristic data of the tidal current power generation device input at the current moment; and a second acquisition module, configured to input the second tidal current observation data into the target output power learning model, and acquire an average output power of the tidal current power generation device output by the target output power learning model.

[0092] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal, and its internal structure diagram may be as Figure 9 shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is configured to store data. The input / output interface of the computer device is configured to exchange information between the processor and external devices. The communication interface of the computer device is configured to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a method for predicting the output power of a tidal current power generation device.

[0093] Those skilled in the art can understand, Figure 9The structure shown is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0094] In an exemplary embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0095] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0096] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0097] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0098] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memories can include read-only memory (ROM), magnetic tapes, floppy disks, flash memories, optical memories, high-density embedded non-volatile memories, resistive random access memories (ReRAM), magnetoresistive random access memories (MRAM), ferroelectric random access memories (FRAM), phase change memories (PCM), graphene memories, etc. Volatile memories can include random access memory (RAM) or external cache memories, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0099] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.

Claims

1. A method for predicting the output power of a tidal current energy generation device, characterized in that, The output power prediction method of the tidal current energy generation device includes: Establishing a matching data set between the first tidal current observation data of the test sea area and the average output power data of the tidal current energy generation device; Performing data quality control processing on the matching data set to eliminate abnormal data in the matching data set and obtain a target data set; Based on the target data set, using the odd-even extraction method of data serial numbers, constructing a target odd data set and a target even data set; Based on the target odd data set, constructing an initial output power learning model of the tidal current energy generation device; Based on the target even data set, correcting the initial output power learning model to obtain a target output power learning model, and predicting the output power of the tidal current energy generation device based on the target output power learning model.

2. The output power prediction method of the tidal current energy generation device according to claim 1, characterized in that The establishment of the matching data set between the first tidal current observation data of the test sea area and the average output power data of the tidal current energy generation device includes: Obtaining the original tidal current observation data of the test sea area, performing spatial conversion processing on the original tidal current observation data to obtain the first tidal current observation data, where the original tidal current observation data is the vertical profile data of the test sea area at a historical moment, and the first tidal current observation data is the impeller swept cross-section characteristic data input into the tidal current energy generation device at a historical moment; Obtaining the original output power measurement data of the tidal current energy generation device, performing mean value processing on the original output power measurement data to obtain the average output power data of the tidal current energy generation device; Matching the first tidal current observation data and the average output power data to obtain a matching data set.

3. The output power prediction method of the tidal current energy generation device according to claim 1, characterized in that, The matching data set includes observation time data, tidal current characteristic velocity data within the impeller swept cross-section range input into the tidal current energy generation device, tidal current characteristic direction data within the impeller swept cross-section range input into the tidal current energy generation device, and the average output power data of the tidal current energy generation device. The performing data quality control processing on the matching data set to eliminate abnormal data in the matching data set and obtain a target data set includes: Dividing the matching data set according to a preset first flow velocity interval to obtain a plurality of first data groups; Using a data quality control algorithm to perform data quality control processing on the average output power data in each first data group to eliminate abnormal output power data in the corresponding first data group and the observation time data, tidal current characteristic velocity data, and tidal current characteristic direction data corresponding to the abnormal output power data, and obtaining a corresponding first sub-data group; Sorting the plurality of first sub-data groups according to the observation time data to obtain a target data set.

4. The output power prediction method of the tidal current energy generation device according to claim 1, characterized in that The matching data set includes observation time data, tidal current characteristic velocity data within the impeller swept cross-section range input into the tidal current energy generation device, tidal current characteristic direction data within the impeller swept cross-section range input into the tidal current energy generation device, and the average output power data of the tidal current energy generation device. The constructing a target odd data set and a target even data set based on the target data set using the odd-even extraction method of data serial numbers includes: Using the odd-even extraction method of data serial numbers, the target data set is split into a first odd data set and a first even data set; According to a preset second flow rate interval, the first odd data set is divided to obtain a plurality of second data groups; according to the second flow rate interval, the first even data set is divided to obtain a plurality of third data groups; Based on the first tidal current characteristic flow rate data in each of the second data groups, the first flow rate mean value of the corresponding second data group is determined; based on the first tidal current characteristic direction data in each of the second data groups, the first coefficient mean value of the utilization coefficient of the tidal current energy generation device for tidal current energy resources in the corresponding second data group is determined; based on the first average output power data in each of the second data groups, the first power mean value of the corresponding second data group is determined; Based on the second tidal current characteristic flow rate data in each of the third data groups, the second flow rate mean value of the corresponding third data group is determined; based on the second tidal current characteristic direction data in each of the third data groups, the second coefficient mean value of the utilization coefficient of the tidal current energy generation device for tidal current energy resources in the corresponding third data group is determined; based on the second average output power data in each of the third data groups, the second power mean value of the corresponding third data group is determined; Based on the first flow rate mean value, the first coefficient mean value, and the first power mean value of the plurality of second data groups, a target odd data set is constructed; Based on the second flow rate mean value, the second coefficient mean value, and the second power mean value of the plurality of third data groups, a target even data set is constructed.

5. The output power prediction method of the tidal current energy generation device according to claim 4, characterized in that Constructing the initial output power learning model of the tidal current energy generation device based on the target odd data set includes: Setting the first power mean value as the first dependent variable, and setting the first flow rate mean value and the first coefficient mean value as the first independent variable and the second independent variable respectively; Setting a first mathematical expression, and determining the first parameter, the second parameter, and the third parameter in the first mathematical expression based on the first power mean value, the first flow rate mean value, the first coefficient mean value, and the first preset parameter of the tidal current energy generation device; Based on the second preset parameter of the tidal current energy generation device and the first mathematical expression, the initial output power learning model of the tidal current energy generation device is constructed.

6. The output power prediction method of the tidal current energy generation device according to claim 4, characterized in that, Correcting the initial output power learning model based on the target even data set to obtain a target output power learning model includes: Inputting the second flow rate mean value and the second coefficient mean value into the initial output power learning model to obtain the third power mean value output by the initial output power learning model; Based on the second power mean value and the third power mean value, determining the correction coefficient of the initial output power learning model; Setting the correction coefficient as the second dependent variable, and setting the second flow rate mean value as the third independent variable; Setting a second mathematical expression, and determining the fourth parameter, the fifth parameter, and the sixth parameter in the second mathematical expression based on the correction coefficient and the second flow rate mean value; Based on the second mathematical expression, the initial output power learning model is modified to obtain a target output power learning model.

7. The output power prediction method of the tidal current energy generation device according to any one of claims 1 to 6, characterized in that The output power prediction method of the tidal energy power generation device also includes: Acquiring second tidal current observation data of the test sea area, where the second tidal current observation data is impeller swept cross-section characteristic data input to the tidal current energy power generation device at a current moment; The second power flow observation data is input into the target output power learning model to obtain the average output power of the power generation device output by the target output power learning model.

8. An output power prediction device for a tidal current energy generation device, characterized in that, The output power prediction device of the tidal energy power generation device comprises: An establishment module is used to establish a matching data set between the first tidal current observation data of the test sea area and the average output power data of the tidal current energy power generation device; A processing module, used for performing data quality control processing on the matching data set to remove abnormal data in the matching data set to obtain a target data set; A first construction module is used to construct a target odd-numbered data set and a target even-numbered data set based on the target data set by using an odd-numbered and even-numbered extraction method of data sequence numbers; A second construction module is used to construct an initial output power learning model of the tidal energy power generation device based on the target odd-numbered data set; A correction module is used to correct the initial output power learning model based on the target even-numbered data set to obtain a target output power learning model, and to predict the output power of the tidal energy power generation device based on the target output power learning model.

9. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method for predicting the output power of a tidal energy power generation device described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, the steps of the method for predicting the output power of a tidal energy power generation device described in any one of claims 1 to 7 are implemented.

Images