A fish-light complementary photovoltaic power station optimization method based on power generation efficiency influencing factor analysis

Abstract

The application provides a fish-light complementary photovoltaic power station optimization method based on power generation efficiency influencing factor analysis, influencing factor data collection, input of influencing factor data into a trained neural network model, prediction of power generation efficiency of the fish-light complementary photovoltaic power station, judgment of whether the power generation efficiency meets a preset threshold value, optimization and adjustment of the influencing factors when the power generation efficiency does not meet the preset threshold value, return to step S1 for new round of data collection after the above optimization operation is completed, and sequential execution of step S2 and step S3, continuous monitoring of the power generation efficiency until the preset threshold value is met, so that the fish-light complementary photovoltaic power station optimization is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fish-light complementary photovoltaic power station, in particular to a fish-light complementary photovoltaic power station optimization method based on power generation efficiency influencing factor analysis. BACKGROUND

[0002] As a new energy utilization mode, fish-light complementary photovoltaic power station combines photovoltaic power generation with fish farming, realizing the comprehensive utilization of land and water. However, the power generation efficiency of fish-light complementary photovoltaic power station is affected by many factors, such as light intensity, photovoltaic component angle, water surface reflectivity, fish farming activities, weather conditions, etc. Currently, the optimization method for fish-light complementary photovoltaic power station mainly focuses on traditional empirical analysis and simple mathematical models, which is difficult to accurately predict and optimize the power generation efficiency, and further optimize the fish-light complementary photovoltaic power station. SUMMARY

[0003] The purpose of the present application is to solve the above-mentioned problems in the prior art, and to provide a fish-light complementary photovoltaic power station optimization method based on power generation efficiency influencing factor analysis, which comprehensively analyzes the factors affecting the power generation efficiency, accurately predicts and optimizes the power generation efficiency using a neural network model, and improves the power generation efficiency of the fish-light complementary photovoltaic power station.

[0004] The present application is realized by the following technical solutions:

[0005] A fish-light complementary photovoltaic power station optimization method based on power generation efficiency influencing factor analysis, the method comprising the following steps:

[0006] Step S1: collecting influencing factor data, including meteorological data of the fish-light complementary photovoltaic power station, latitude of the project site, photovoltaic array tracking mode, photovoltaic component angle, and orientation of the photovoltaic system array;

[0007] Step S2: inputting the meteorological data, latitude of the project site, photovoltaic array tracking mode, photovoltaic component angle, and orientation of the photovoltaic system array into the trained neural network model to predict the power generation efficiency of the fish-light complementary photovoltaic power station;

[0008] Step S3: determining whether the power generation efficiency meets the preset threshold, and when it does not, optimizing and adjusting the influencing factors, after completing the above optimization operation, returning to step S1 for new round of data collection, and sequentially executing steps S2 and S3, continuously monitoring the power generation efficiency until the preset threshold is met.

[0009] Further, the meteorological data includes light intensity, temperature, humidity, and wind speed.

[0010] Further, the photovoltaic array tracking mode includes fixed type, single-axis tracking, and double-axis tracking.

[0011] Further, step S2 before including the construction of neural network model includes using sample data on neural network model training and optimization, get trained neural network model.

[0012] Compared with the prior art, the present application has the following advantages: the present application proposes a fishlight complementary photovoltaic power station optimization method based on power generation efficiency influencing factor analysis, the influencing factor data is collected, the influencing factor data is input into the trained neural network model, and the power generation efficiency of the fishlight complementary photovoltaic power station is predicted and output; it is judged whether the power generation efficiency meets the preset threshold value, when it does not meet, the influencing factors are optimized and adjusted, after the above optimization operation, return to step S1 to collect new round of data, and execute step S2 and step S3 in turn, continuously monitor the power generation efficiency, until the preset threshold value is met, so as to realize the optimization of fishlight complementary photovoltaic power station. BRIEF DESCRIPTION OF DRAWINGS

[0013] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only one embodiment of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0014] Figure 1 It is a flow chart of a fishlight complementary photovoltaic power station optimization method based on power generation efficiency influencing factor analysis.

[0015] Figure 2 It is a structure of a neural network model. DETAILED DESCRIPTION

[0016] In order to make the technical means, creative features, purposes and effects realized by the present application easy to understand, the technical solutions in the specific embodiments of the present application will be described clearly and completely below to further illustrate the present application. Obviously, the described specific embodiments are only part of the embodiments of the present application, not all.

[0017] The present application will be described in further detail below in conjunction with the drawings:

[0018] As shown in the accompanying Figure 1 The present application discloses a fishlight complementary photovoltaic power station optimization method based on power generation efficiency influencing factor analysis, the method comprises the following steps:

[0019] Step S1: influence factor data collection, the data includes meteorological data of the fish-light complementary photovoltaic power station, latitude of the project site, photovoltaic array tracking mode, photovoltaic component angle, and azimuth of the photovoltaic system array;

[0020] Step S2: input the meteorological data, latitude of the project site, photovoltaic array tracking mode, photovoltaic component angle, and azimuth of the photovoltaic system array into the trained neural network model to predict the output power generation efficiency of the fish-light complementary photovoltaic power station;

[0021] Step S3: determine whether the power generation efficiency meets the preset threshold, when it does not meet, optimize and adjust the influence factors, after completing the above optimization operation, return to step S1 to perform new round of data collection, and sequentially execute step S2 and step S3, continuously monitor the power generation efficiency until the preset threshold is met.

[0022] Further, the meteorological data includes light intensity, temperature, humidity, and wind speed.

[0023] Further, the photovoltaic array tracking mode includes fixed type, single-axis tracking, and double-axis tracking.

[0024] Further, step S2 includes training and optimizing the neural network model by using sample data before the step S2, adopting the back propagation algorithm and the stochastic gradient descent algorithm, constantly adjusting the weight and bias of the neural network model, and obtaining the trained neural network model.

[0025] The optimization of the fish-light complementary photovoltaic power station highly depends on the prediction of the power generation efficiency, however, the accurate prediction of the power generation efficiency is extremely challenging. Therefore, in this work, a double-branch neural network model for power generation efficiency prediction is introduced. The first branch of the proposed network uses a convolutional network for feature learning, and the second branch uses a BiGRU module connected in sequence for feature learning, then integrates these features together to form a single but representative feature vector, which is used as the input of the attention mechanism for further selection for the prediction of the power generation efficiency.

[0026] Further, the neural network model adopted by the present application is a double-branch neural network model, after preprocessing the meteorological data, latitude of the project site, photovoltaic array tracking mode, photovoltaic component angle, and azimuth of the photovoltaic system array, a feature vector F0 is constructed as the input of the double-branch neural network model, the feature vector F0 is input into the first feature extraction branch and the second feature extraction branch respectively to obtain features F1 and F2, the features F1 and F2 are fused, and the fused features are input into a self-attention module, and a prediction output is obtained through a Dense layer.

[0027] Further, the first feature extraction branch comprises sequentially connected convolutional layers, maximum pooling layers, convolutional layers, maximum pooling layers and a flattening layer.

[0028] Further, the second feature extraction branch comprises sequentially connected first BiGRU, second BiGRU and a flattening layer to obtain a feature Fb, the Fb is input into a weight processing module to obtain a feature F2, the F2 is obtained by the following formula:

[0029]

[0030] Wherein, w = sigmoid(Fb), w represents a weight, Relu() represents a Relu operation, and linear() represents a linear transformation.

[0031] By inputting the Fb into the weight processing module, the extracted features are further optimized, and gradient disappearance can be prevented.

[0032] Further, the features F1 and F2 are fused based on a Concat layer.

[0033] The above describes the main technical features and basic principles of the present application and related advantages, for those skilled in the art, it is obvious that the present application is not limited to the details of the above exemplary embodiments, and the present application can be realized in other specific forms without departing from the concept or basic features of the present application. Therefore, no matter from which point of view, the above-mentioned specific embodiments should be regarded as exemplary and non-limiting, the scope of the present application is defined by the appended claims rather than the above description, therefore all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present application.

[0034] In addition, it should be understood that although the present specification is described according to each embodiment, not every embodiment contains only one independent technical solution, the description manner of the specification is only for the sake of clarity, those skilled in the art should consider the specification as a whole, the technical solutions in each embodiment can also be combined appropriately to form other embodiments which can be understood by those skilled in the art.

Claims

1. A method for optimizing a fishery-solar complementary photovoltaic power station based on analysis of factors affecting power generation efficiency, characterized in that: The method comprises the following steps: Step S1: collecting influencing factor data, including meteorological data of the fishery-solar complementary photovoltaic power station, the latitude of the project site, the photovoltaic array tracking mode, the photovoltaic module angle, and the orientation of the photovoltaic system array; Step S2: inputting the meteorological data, the latitude of the project site, the photovoltaic array tracking mode, the photovoltaic module angle, and the orientation of the photovoltaic system array into the trained neural network model to predict and output the power generation efficiency of the fish-solar hybrid photovoltaic power station; Step S3: Determine whether the power generation efficiency meets the preset threshold. If not, optimize and adjust the influencing factors. After completing the above optimization operation, return to step S1 for a new round of data collection, and execute steps S2 and S3 in sequence to continuously monitor the power generation efficiency until the preset threshold is met. The neural network model is a dual-branch neural network model. After preprocessing the meteorological data, the latitude of the project site, the photovoltaic array tracking mode, the photovoltaic module angle, and the orientation of the photovoltaic system array, a feature vector F0 is constructed as the input of the dual-branch neural network model. The feature vector F0 is input into the first feature extraction branch and the second feature extraction branch respectively to obtain features F1 and F2. The features F1 and F2 are feature fused, and the fused features are input into the self-attention module, and the predicted output is obtained through Dense layer prediction.

2. The method for optimizing a fishery-solar complementary photovoltaic power station based on analysis of factors affecting power generation efficiency according to claim 1, characterized in that: The meteorological data includes light intensity, temperature, humidity, and wind speed.

3. The method for optimizing a fishery-solar complementary photovoltaic power station based on analysis of factors affecting power generation efficiency according to claim 1, characterized in that: The photovoltaic array tracking modes include fixed, single-axis tracking, and dual-axis tracking.

4. The method for optimizing a fishery-solar complementary photovoltaic power station based on analysis of factors affecting power generation efficiency according to claim 1, characterized in that: The steps before step S2 include constructing a neural network model, using sample data to train and optimize the neural network model, using back propagation algorithm and stochastic gradient descent algorithm, and continuously adjusting the weights and biases of the neural network model to obtain a trained neural network model.

5. The method for optimizing a fishery-solar complementary photovoltaic power station based on analysis of factors affecting power generation efficiency according to claim 1, characterized in that: The first feature extraction branch includes a convolution layer, a maximum pooling layer, a convolution layer, a maximum pooling layer, and a flattening layer connected in sequence.

6. The method for optimizing a fishery-solar complementary photovoltaic power station based on analysis of factors affecting power generation efficiency according to claim 1, characterized in that: The second feature extraction branch includes a first BiGRU, a second BiGRU, and a flattening layer connected in sequence to obtain a feature Fb, and the Fb is input into a weight processing module to obtain a feature F2. The weight processing module is: ； Among them, w=sigmoid(Fb), w represents weight, Relu() represents Relu operation, and linear() represents linear transformation.

7. The method for optimizing a fishery-solar complementary photovoltaic power station based on analysis of factors affecting power generation efficiency according to claim 1, characterized in that: The feature fusion of the features F1 and F2 is specifically performed based on the Concat layer.

Images