Fishing light system photovoltaic power prediction method considering environmental change and photovoltaic panel aging

Abstract

The invention relates to a fishing light system photovoltaic power prediction method considering environment change and photovoltaic panel aging, and the method comprises the steps: obtaining fishing light system data, carrying out the power index analysis, classifying weather types, and obtaining a data set of each weather type; analyzing the correlation between each meteorological factor influencing the photovoltaic power generation and the photovoltaic power generation power, and screening out main influence factors; based on a photovoltaic array mechanism model, calculating theoretical power values of parameters in the data set, and identifying and optimizing power abnormal points; the relative aging degree of the photovoltaic module is calculated, so that the photovoltaic output power in the data set is corrected; searching historical data of similar days from the data set according to measured data of a prediction day, and constructing a sample set to perform iterative training on the neural network; and inputting the actually measured data of the prediction day into the trained neural network to obtain a photovoltaic power prediction result. Compared with the prior art, the method effectively improves the accuracy of power prediction of the fishing light system and the effectiveness of power supply.

Description

Technical Field

[0001] The invention relates to the technical field of photovoltaic power prediction, and in particular to a photovoltaic power prediction method for a fishing light system taking environmental changes and photovoltaic panel aging into consideration. Background Art

[0002] The fishery photovoltaic storage system includes solar photovoltaic panels, energy storage equipment (such as battery packs), intelligent control systems, etc. The load includes fishing boats, fishing gear, breeding equipment and other types of fishery-specific electrical equipment. The photovoltaic output power of the fishery photovoltaic system is significantly affected by weather conditions (such as light intensity, temperature, humidity, wind speed, etc.) and environmental factors (such as wetlands, rivers, lakes, etc.). On the one hand, changes in weather conditions, such as fluctuations in light intensity, temperature, humidity and wind speed, will directly affect the power generation of the photovoltaic system. In special environments such as wetlands, rivers, lakes, etc., factors such as high humidity and water reflection also further aggravate the fluctuations in the performance of photovoltaic modules. On the other hand, when photovoltaic modules are exposed to changing environmental conditions such as sunlight, temperature, humidity, wind, etc. for a long time, the material degradation of the modules, the increase in contact resistance and the aging of the packaging materials will occur, which are specifically manifested as light attenuation, mechanical damage and thermal stress, which will lead to a significant loss of power generation.

[0003] The accuracy of short-term photovoltaic power forecasting is directly affected by meteorological factors. Deeply exploring the impact of meteorological factors on short-term photovoltaic output and finding historical similar day data for the day to be predicted will help improve the accuracy of power forecasting. Some research has been conducted at home and abroad on finding historical similar day data for the day to be predicted. Song Weiqiong et al. selected similar days by calculating the prediction accuracy of the grey correlation rate. Liu Hui et al. divided the characteristic data into linear and nonlinear, and used the Pearson correlation co-efficient (PCC) and the maximum information coefficient to calculate the correlation between meteorological factors and power load. Such methods all attempt to find similar day data through a single algorithm, and the model prediction accuracy is low. Zhang Dahai et al. determined the meteorological influencing factors through PCC, decomposed the historical data set into intrinsic mode function components at different frequencies, and used long-term and short-term time series networks and extreme learning machine models to predict the high-frequency and low-frequency IMF components respectively. Li Bin et al. combined grey correlation, correlation weighting method and maximum information coefficient to determine the similar day curve to improve the prediction accuracy.

[0004] In terms of the aging characteristics of photovoltaic modules, Ma Mingyao et al. used the battery filling factor to determine the degree of aging failure and analyzed its current-voltage output characteristics. The existing photovoltaic output prediction technology lacks consideration of the component aging phenomenon caused by uncertain factors in the installation environment and changes in the meteorological environment. Aging data is often filtered out as abnormal data to reduce the impact on output data. In order to improve the accuracy and reliability of photovoltaic power prediction, it is necessary to propose a prediction method that comprehensively considers different weather scenarios and aging and degradation environments. By collecting and analyzing photovoltaic power generation data under different weather and environmental conditions, a high-precision power prediction model is established using a combination of data physics modeling.

[0005] Based on the above background, in order to improve the management level of fishery-photovoltaic systems in special environments such as wetlands, rivers, lakes, etc., it is urgent to study the photovoltaic power prediction method of fishery-photovoltaic systems that takes into account the influence of environmental changes and the aging characteristics of photovoltaic panel components, so as to achieve all-weather and reliable power supply and ensure the normal operation of fisheries, which has reference significance for optimizing energy management and energy storage scheduling. Summary of the invention

[0006] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a photovoltaic power prediction method for a fishery light system that takes into account environmental changes and photovoltaic panel aging, analyzes weather conditions such as light intensity, temperature, humidity, wind speed, and photovoltaic component aging characteristics and power-related characteristics, and uses a BP neural network to accurately predict photovoltaic power generation.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] A photovoltaic power prediction method for a fishery photovoltaic system considering environmental changes and photovoltaic panel aging includes the following steps:

[0009] Obtain data for photovoltaic power prediction of fishery-photovoltaic systems, perform power index analysis, and classify weather types to obtain data sets for various weather types;

[0010] According to the data set, the correlation between various meteorological factors affecting photovoltaic power generation and photovoltaic power generation power is analyzed, so as to screen out the main factors that have a greater impact on power;

[0011] Based on the photovoltaic array mechanism model, the theoretical power value is calculated for the irradiance and temperature parameters in the data set, and compared with the corresponding measured power value to identify and optimize power anomalies;

[0012] The IV curve of the photovoltaic module is tested by an IV tester, and the relative aging degree of the photovoltaic module is calculated, so as to correct the photovoltaic output power in the data set;

[0013] Analyze the power index based on the measured data of the forecast day, classify the weather type, and find the historical data of similar days from the data set to construct a sample set;

[0014] Iteratively train the neural network using the sample set until a trained neural network is obtained;

[0015] The features corresponding to the main influencing factors are extracted according to the measured data on the prediction day and input into the trained neural network to obtain the photovoltaic power prediction results of the fishery-photovoltaic system.

[0016] Furthermore, the process of classifying the weather types is specifically as follows:

[0017] Instantaneous power output and daily cumulative power generation are used as power indicators, and the similarity between different weather type data is calculated through Euclidean distance, so as to classify the weather types;

[0018] The calculation expression of the Euclidean distance is:

[0019]

[0020] In the formula, and are the average power generation of two different weather types at time i; E p and E q are the daily cumulative power generation of the two weather types respectively; n is the number of time periods divided by the selected day.

[0021] Furthermore, the method adopts a photovoltaic power generation power influencing factor analysis method based on grey correlation to analyze the correlation between various meteorological factors and photovoltaic power generation;

[0022] The photovoltaic power generation power influencing factor analysis method based on grey correlation includes:

[0023] Normalize the data in the dataset;

[0024] For the normalized data, construct the feature vector with each meteorological factor as the similarity factor;

[0025] Based on the feature vector, a reference sequence and a comparison sequence are constructed, and differences are made to obtain a difference sequence;

[0026] Define the minimum value of each difference sequence as the minimum range, and the maximum value of each difference sequence as the maximum range; thereby calculate the correlation coefficient between the comparison sequence and the reference sequence to reflect the shape difference of each output timing curve;

[0027] The correlation coefficients of each comparison sequence are normalized to obtain the corresponding factor weight coefficients to describe the relative effect of each meteorological factor on the system, thereby screening out the main influencing factors with a greater impact on power.

[0028] Furthermore, the calculation expression of the normalization process is:

[0029]

[0030] In the formula, x i is the i-th sequence; x i (k) is the kth data point in sequence i; x i ′(k) is the normalized value of the kth data point; are the largest and smallest data points in the ith sequence, respectively; m is the total number of sequences; n is the total number of data points in the sequence;

[0031] The calculation expression of the difference sequence is:

[0032] Δ i (k)=|x′0(k)-x′ i (k)|(i=0,1,2,...,m; k=1,2,3,...,n)

[0033] In the formula, Δ i =(Δ i (1),Δ i (2),…,Δ i (n)); x0=(x0(1),x0(2),…,x0(n)) is the reference sequence. When i>0, x i =(x i (1),x i (2),…,x i (n)) is a comparison sequence; and x′0 and x′ i All are normalized sequences;

[0034] The calculation expression of the correlation coefficient is:

[0035]

[0036] In the formula, ξ 0i is the correlation coefficient; ρ is the resolution coefficient; is the minimum range; is the maximum range;

[0037] The calculation expression of the factor weight coefficient is:

[0038]

[0039] In the formula, ξi is the factor weight coefficient of the i-th comparison sequence.

[0040] Further, the photovoltaic array mechanism model includes a current source, a diode, an equivalent parallel resistance reflecting the non-ideal characteristics of the PN junction and the influence of impurities near it, a contact resistance formed by the electrode and the semiconductor, and a series resistance formed by the semiconductor and the electrode resistance;

[0041] The diode and the equivalent parallel resistor are respectively connected in parallel at both ends of the current source, and the contact resistor and the series resistor are connected in series with each other and then connected in parallel at both ends of the current source.

[0042] Furthermore, based on the IV relationship expression of the photovoltaic array mechanism model, the theoretical power value is calculated, and the IV relationship expression is:

[0043]

[0044] Where, I is the load current; I p is the current flowing through the parallel resistor; I ph is the photocurrent; I d is the dark current; I sd is the reverse saturation current of the diode; U is the load voltage; R s and R p are the series resistance and equivalent parallel resistance respectively; a is the ideal factor of the diode; T emp is the photovoltaic cell temperature; k0 is the Boltzmann constant, k0=1.381×10 -23 J / K; q is the electron charge, q = 1.6021 × 10 -19 C;

[0045] in,

[0046]

[0047] Where G is the actual irradiance; G ref is the reference irradiance; T ref is the reference temperature; I sc_ref is the short-circuit current at reference irradiance and reference temperature; α i is the short-circuit current temperature coefficient;

[0048]

[0049] In the formula, E g is the band gap energy; I os is the reverse saturation current of the diode under reference temperature conditions;

[0050]

[0051] In the formula, I sc and U oc are the short-circuit current and open-circuit voltage of the photovoltaic cell at the reference temperature.

[0052] Furthermore, the calculation process of the relative aging degree of the photovoltaic module includes:

[0053] According to the IV curve of the tested photovoltaic module, the measured open circuit voltage, short circuit current and maximum power point are obtained to calculate the fill factor. The calculation expression of the fill factor is:

[0054]

[0055] In the formula, σ FF is the filling factor; U ocm is the open circuit voltage; I scm is the short-circuit current; U max is the voltage corresponding to the maximum power point; I max is the current corresponding to the maximum power point;

[0056] According to the characteristic that the filling factor of photovoltaic modules is basically greater than 0.7 when they are working normally, the relative aging degree is calculated, and the corresponding calculation expression is:

[0057]

[0058] Where, d e is the relative degree of aging;

[0059] The calculation expression for correcting the photovoltaic output power in the data set is:

[0060] P s =P t (1-K loss )

[0061] Where P t P is the photovoltaic output power without correction and without considering the aging effect; s is the photovoltaic output power after considering the aging effect; K loss is the power loss coefficient.

[0062] Furthermore, the process of searching for historical data of similar days from the data set is specifically as follows:

[0063] First, we search for similar days to the forecast day by using the highest temperature, lowest temperature and weather conditions. Then we select the day with the closest characteristics as the similar day, and use the comprehensive similarity index as the evaluation criterion. The calculation expression of the comprehensive similarity index is:

[0064]

[0065] In the formula, is the calculated value of the comprehensive similarity index; ξ 0i (k) is the correlation coefficient between the current sequence i and the reference sequence; N0 is the number of data in the sequence.

[0066] Furthermore, the neural network is a BP neural network, which predicts the photovoltaic power of the fishery light system after normalizing the collected data, and uses the normalized absolute mean error and the normalized root mean square error to evaluate the normalized error.

[0067] Furthermore, the evaluation indicators used in the BP neural network training process include:

[0068]

[0069] Wherein, MAPE is the mean absolute percentage error; MAE is the mean absolute error; MBE is the mean error; RMSE is the root mean square error; x′ i is the predicted output value of the BP neural network; x i is the actual value; N is the number of data in the sequence.

[0070] Compared with the prior art, the present invention has the following advantages:

[0071] (1) The present invention first classifies the weather type of the data according to the power index, and analyzes the correlation between photovoltaic power generation and the meteorological factors affecting photovoltaic power generation, extracts the main influencing factors, and selects similar day data of the corresponding weather type for the predicted day data under different weather conditions, and further selects similar day data according to the main influencing factors and the comprehensive similarity index for network training;

[0072] On the other hand, by testing the IV curve of the photovoltaic module, the relative aging degree of the photovoltaic module is calculated, so as to correct the photovoltaic output power in the data set. When the aging degree of the photovoltaic module is serious, the measured power data is used as a separate category for power prediction to improve the accuracy of the prediction results.

[0073] Overall, the present invention effectively improves the accuracy of power prediction and the effectiveness of power supply, helps to optimize the scheduling and power distribution of energy storage equipment in complex environments, improves the overall efficiency of the system, ensures a stable power supply to fisheries under various weather conditions, and improves fishery production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 A schematic flow chart of a photovoltaic power prediction method for a fishing-photovoltaic system taking into account environmental changes and photovoltaic panel aging provided in an embodiment of the present invention;

[0075] Figure 2 An equivalent circuit diagram of a single diode model of a photovoltaic cell provided in an embodiment of the present invention;

[0076] Figure 3 A schematic diagram of a grey relational analysis method provided in an embodiment of the present invention;

[0077] Figure 4 Schematic diagram of an IV curve of a photovoltaic module before and after aging provided in an embodiment of the present invention, wherein the solid line is a normal IV curve, and the dotted line is a measured IV curve after aging;

[0078] Figure 5 The present invention provides a schematic diagram of a photovoltaic power prediction process based on a BP neural network. DETAILED DESCRIPTION

[0079] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0080] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0081] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0082] Example 1

[0083] like Figure 1 As shown, this embodiment provides a photovoltaic power prediction method for a fishing light system taking into account environmental changes and photovoltaic panel aging, including the following steps:

[0084] S1: Obtain data for photovoltaic power prediction of fishery-photovoltaic system, perform power index analysis, and classify weather types to obtain data sets of various weather types;

[0085] S2: Based on the data set, the correlation between various meteorological factors affecting photovoltaic power generation and photovoltaic power generation is analyzed to screen out the main factors that have a greater impact on power;

[0086] S3: Based on the photovoltaic array mechanism model, the theoretical power value is calculated for the irradiance and temperature parameters in the data set, and compared with the corresponding measured power value to identify and optimize the power anomaly points;

[0087] S4: Testing the IV curve of the photovoltaic module by an IV tester, calculating the relative aging degree of the photovoltaic module, and thus correcting the photovoltaic output power in the data set;

[0088] S5: Analyze the power index based on the measured data of the forecast day, classify the weather type, and find the historical data of similar days from the data set to construct a sample set;

[0089] S6: Iteratively train the neural network using the sample set until a trained neural network is obtained;

[0090] S7: Extract the features corresponding to the main influencing factors based on the measured data on the prediction day, and input them into the trained neural network to obtain the photovoltaic power prediction results of the fish-to-light system.

[0091] The following is a detailed description of each step:

[0092] 1. Weather type classification considering environmental differences in step S1

[0093] The process of classifying weather types is as follows:

[0094] Instantaneous power output and daily cumulative power generation are used as power indicators, and the similarity between different weather type data is calculated through Euclidean distance, so as to classify the weather types;

[0095] The calculation expression of Euclidean distance is:

[0096]

[0097] In the formula, is the average power generation of the standard weather type in the i-th period; is the average value of the power generation of the weather type to be classified in the i-th period; E p and E q are the daily cumulative power generation of the two weather types respectively; n is the number of time periods divided by the selected day.

[0098] Specifically, on sunny days, solar radiation is the strongest, and photovoltaic modules can achieve the highest power output; on cloudy days, clouds partially block sunlight, resulting in unstable radiation intensity and large fluctuations in power output; on overcast days, clouds completely cover the sky, significantly reducing radiation intensity and lower power output; on rainy days, not only are there clouds blocking the sky, but raindrops will further reduce light, increase humidity and cooling effects, causing a significant drop in power output; on foggy days, water droplets suspended in the air scatter and absorb sunlight, which weakens radiation intensity and reduces the power output of photovoltaic modules. Weather conditions include a variety of meteorological factors, such as sky cloud cover, ambient temperature, wind speed, air pressure, etc.

[0099] By monitoring and calculating power indicators, we can fully understand the operating status and efficiency of the system. The main power indicators include instantaneous power output, daily cumulative power generation, capacity factor, etc. Instantaneous power output reflects the power generation capacity of the photovoltaic system at a specific moment and is an important parameter for real-time monitoring; daily cumulative power generation is a measure of the total power generation throughout the day and can evaluate the daily performance of the system.

[0100] Combining the main power indicators with weather quality, weather conditions, etc., the weather types are divided into sunny, cloudy, overcast, rainy and foggy, so as to make power classification prediction.

[0101] Instantaneous power output (such as deriving a power value every 15 minutes) and daily cumulative power generation are used as power indicators. The similarity between different types is calculated through Euclidean distance, so as to classify weather types.

[0102] 2. Correlation analysis between multiple factors and photovoltaic power generation in step S2

[0103] Methods The photovoltaic power generation power influencing factor analysis method based on grey correlation was used to analyze the correlation between various meteorological factors and photovoltaic power generation power.

[0104] The analysis methods of factors affecting photovoltaic power generation based on grey correlation include:

[0105] Normalize the data in the dataset;

[0106] For the normalized data, construct the feature vector with each meteorological factor as the similarity factor;

[0107] Based on the feature vector, a reference sequence and a comparison sequence are constructed, and the difference is made to obtain a difference sequence;

[0108] Define the minimum value of each difference sequence as the minimum range, and the maximum value of each difference sequence as the maximum range; thereby calculate the correlation coefficient between the comparison sequence and the reference sequence to reflect the shape difference of each output timing curve;

[0109] The correlation coefficients of each comparison sequence are normalized to obtain the corresponding factor weight coefficients to describe the relative effect of each meteorological factor on the system, thereby screening out the main influencing factors with a greater impact on power.

[0110] Specifically, the relationship between irradiance, ambient temperature, wind speed, relative humidity, geographical location and photovoltaic output power is analyzed, and a method for analyzing factors affecting photovoltaic power generation based on grey correlation is established. The data is normalized to the interval of 0 to 1, which is:

[0111]

[0112] In the formula, x i is the i-th sequence of the grey system factor set; x i (k) is the kth data point in sequence i; x i ′(k) is the normalized value of the kth data point; are the largest and smallest data points in the ith sequence, respectively.

[0113] Grey correlation analysis is used to determine the degree to which each influencing factor interferes with system changes based on the degree of correlation. The steps of grey correlation analysis are as follows: Figure 3 shown.

[0114] 2.1. Feature vector selection.

[0115] Construct a feature vector X=[E,T m ,…]. Where E is the solar irradiance; T m is the panel temperature.

[0116] 2.2. Calculate the correlation coefficient between the feature vector and each time period in the historical data set, that is, the similarity index.

[0117] Define the difference Δ between the feature vector and the historical data to be compared i (k), as shown in Eq.

[0118] Δ i (k) = |x′0(k)-x′ i (k)|(i=0,1,2,...,m; k=1,2,3,...,n) (3)

[0119] Where Δ i =(Δ i (1),Δ i (2),…,Δ i (n)); x0=(x0(1),x0(2),…,x0(n)) is the reference sequence. When i>0, x i =(xi (1),x i (2),…,x i (n)) is a comparison sequence; and x′0 and x′ i is the normalized sequence.

[0120] Define the minimum value of each difference sequence as the minimum range The maximum value in each difference sequence is the maximum range Calculate the correlation coefficient ξ between the i-th comparison sequence and the reference sequence 0i (k) is used to reflect the shape differences of multiple output timing curves of photovoltaic panels, which can be expressed as:

[0121]

[0122] Where ρ is the resolution coefficient, which ranges from 0 to 1 and is usually 0.5.

[0123] 2.3. Grey correlation calculation.

[0124] Normalize the grey relational degree to obtain the factor weight coefficient ξ of the i-th comparison sequence i for:

[0125]

[0126] The factor weight coefficient describes the relative effect of each influencing factor on the system.

[0127] 3. Abnormal point analysis based on mechanism model and power comparison in step S3

[0128] The photovoltaic array mechanism model includes a current source, a diode, an equivalent parallel resistance reflecting the non-ideal characteristics of the PN junction and the influence of impurities nearby, a contact resistance formed by the electrode and the semiconductor, and a series resistance formed by the semiconductor and electrode resistances;

[0129] The diode and the equivalent parallel resistor are respectively connected in parallel at both ends of the current source, and the contact resistor and the series resistor are connected in series with each other and then connected in parallel at both ends of the current source.

[0130] In terms of mechanism model, the equivalent circuit of a single diode model of a photovoltaic cell is as follows: Figure 2 As shown, the IV relationship expression is shown in formula (2). It includes a current source connected in parallel with the diode, an equivalent parallel resistor that reflects the non-ideal characteristics of the PN junction and the influence of impurities nearby, and a series resistor composed of the contact resistance between the electrode and the semiconductor and the resistance of the semiconductor and the electrode.

[0131]

[0132] Where, I is the load current; Ip is the current flowing through the parallel resistor; I ph is the photocurrent; I d is the dark current; I sd is the reverse saturation current of the diode; U is the load voltage, which is the open circuit voltage of the photovoltaic cell when no load is connected; R s and R p are the equivalent series and parallel resistances respectively; a is the diode ideal factor; T emp is the photovoltaic cell temperature; k0 is the Boltzmann constant, k0=1.381×10 -23 J / K; q is the electron charge, q = 1.6021 × 10 -19 C.

[0133] The photocurrent I in formula (6) ph It will change with the change of irradiance and temperature, and its relationship with irradiance and temperature is expressed as:

[0134]

[0135] Where G is the actual irradiance; G ref is the reference irradiance, usually G ref =1000W / m 2 ; T ref is the reference temperature, usually T ref =25℃;I sc_ref is the short-circuit current at reference irradiance and reference temperature; α i is the short-circuit current temperature coefficient.

[0136] The diode reverse saturation current I in formula (6) sd The expression is:

[0137]

[0138] In the formula, E g is the band gap energy, which is about 1.12 eV for silicon at standard temperature 25°C; I os is the reverse saturation current of the diode under reference temperature conditions.

[0139] The reverse saturation current I of the diode at the reference temperature os for:

[0140]

[0141] In the formula, I sc and U oc are the short-circuit current and open-circuit voltage of the photovoltaic cell at the reference temperature.

[0142] From the relationship shown in equations (6)-(9), it can be seen that light intensity and temperature have a great influence on photovoltaic power generation and are direct factors affecting photovoltaic power. Salt fog in the environment mainly changes the temperature of solar cells, and seawater mainly changes the received radiation intensity to affect the output characteristics, thereby affecting the output power.

[0143] 4. Photovoltaic output power correction taking into account component aging characteristics in step S4

[0144] The calculation process of the relative aging degree of photovoltaic modules includes:

[0145] According to the IV curve of the tested photovoltaic module, the measured open circuit voltage, short circuit current and maximum power point are obtained to calculate the fill factor;

[0146] According to the characteristic that the filling factor of photovoltaic modules is basically greater than 0.7 when they are working normally, the relative aging degree is calculated;

[0147] The PV output power in the dataset is corrected based on the relative aging degree.

[0148] Specifically: Schematic diagram of the IV curve of photovoltaic modules before and after aging is as follows Figure 4 As shown. The IV curve of the component is tested by an IV tester, based on the measured open circuit voltage V ocm , short circuit current I scm and the maximum power point P max Calculate the fill factor using the expression:

[0149]

[0150] When the components are working normally, the fill factor is generally greater than 0.7. The relative aging degree d is calculated. e , the expression is:

[0151]

[0152] Considering the aging characteristics of components, the attenuation rate is introduced into the photovoltaic output power for power correction:

[0153] P s =P t (1-K loss ) (12)

[0154] Where P t P is the photovoltaic output power without correction and without considering the aging effect; s K is the photovoltaic output power after considering the aging effect; loss is the power loss coefficient.

[0155] As the use time and environmental factors affect the components, when it drops below 0.5, it is considered that the components have been seriously aged. When it is judged that the degree of component aging is serious, the power data measured under severe aging conditions is taken as a separate category for power prediction.

[0156] 5. Similar day selection based on grey correlation analysis in step S5

[0157] The process of finding historical data of similar days from the data set is as follows:

[0158] First, similar days to the predicted day are found through the maximum temperature, minimum temperature and weather conditions. Secondly, the day with the closest characteristics is selected as the similar day, and the comprehensive similarity index is used as the evaluation criterion.

[0159] Specifically, the power generation of photovoltaic panels is mainly determined by factors such as irradiance and atmospheric visibility. However, for the fishery-photovoltaic complementary project in a specific area, its photovoltaic output sequence has certain similarities with a certain period of historical data in the same period. Considering the weather conditions at different time periods during the day, first find similar days through the highest temperature, lowest temperature and weather conditions; then select the similar day with the closest characteristics for power prediction.

[0160] Since the input data has more dimensions and more correlation coefficients, it is difficult to obtain the overall similarity between the curves. Therefore, a comprehensive similarity index is defined. It can be expressed as:

[0161]

[0162] In the formula, N0 is the number of data in the sequence.

[0163] Then, the similarity index and comprehensive similarity are used to sort the data, and the historical curves with high similarity to the input data at the time to be predicted are selected as the similar day data set to train the predictor and verify the prediction accuracy.

[0164] 6. Power prediction method based on BP neural network in steps S6 and S7

[0165] The neural network is a BP neural network. The BP neural network predicts the photovoltaic power of the fishery light system after normalizing the collected data, and uses the normalized absolute mean error and normalized root mean square error to evaluate the normalized error.

[0166] Specifically: Figure 5 As shown in the figure, the BP neural network is a type of multi-layer feedforward neural network with error back propagation correction capability, strong nonlinear mapping capability, capable of processing complex mapping relationships between input and output, self-learning and adaptive capabilities, and able to automatically adjust the weights and restrictions of the network to improve analysis accuracy. It is used to analyze the impact of complex environments on photovoltaic output power.

[0167] Step 1: Form a sample set for BP neural network analysis. Form data sets for each type of weather and perform preprocessing. Perform correlation analysis on multi-factor meteorological indicators to screen out correlation parameters that have a greater impact on power and obtain a meteorological feature set. Combine the correlation factors and power input and output vectors to form training samples.

[0168] Step 2: BP neural network training based on the sample set. Normalize the collected data and normalize the absolute mean error e NMAE and the normalized root mean square error e NRMSE The expression is:

[0169]

[0170] Where x′(i) is the predicted output value of the neural network; x(i) is the corresponding measured power value after normalization; and N is the number of predicted samples.

[0171] The processed data is divided into training set, verification set and test set in proportion. The first two are used for BP neural network training, and the test set is used to test the generalization ability of the trained network. The neural network structure such as the number of layers, the number of neurons in each layer and the type of activation function, training method and parameters (such as learning rate) are further determined, and the neural network training is completed based on the selected training method and samples. Finally, the test set is used for testing, and the network output and actual output are compared to evaluate the analysis accuracy. When the accuracy does not meet the requirements, the structure, learning method or parameters of the BP neural network are changed and retrained to meet the requirements.

[0172] By comparing the network output after verification with the actual output, the accuracy of power prediction can be evaluated and analyzed. The evaluation indicators include:

[0173]

[0174] Wherein, MAPE is the mean absolute percentage error; MAE is the mean absolute error; MBE is the mean error; RMSE is the root mean square error.

[0175] Step 3: Use the trained BP neural network to predict photovoltaic power. Take the meteorological indicators of the target day as input, input the trained BP neural network to obtain the corresponding photovoltaic power data, and perform denormalization to obtain the actual predicted value.

[0176] When the BP neural network is trained using historical data, the meteorological information of the predicted day is input into the network to obtain the output power at each moment of the predicted day. After denormalization, it is compared with the measured photovoltaic power. The prediction error of the network can be calculated based on different error assessment methods. If it meets the requirements (generally the prediction error is 15%-35%), it means that the network can be used to predict the output power of the photovoltaic power generation component.

[0177] Establish a regression model between power prediction and measured power. Calculate the goodness of fit R 2 The indicator is calculated as follows:

[0178]

[0179] In the formula, SSR is the regression sum of squares; SSE is the residual sum of squares; SST is the total deviation sum of squares.

[0180] According to the model test results, the goodness of fit R 2 It is close to 1, indicating that the actual PV power and the predicted PV output have a good fit.

[0181] Example 2

[0182] This embodiment provides a photovoltaic power prediction device for a fishing-photovoltaic system taking into account environmental changes and aging of photovoltaic panels, characterized in that it includes a memory and a processor, the memory stores a computer program, and the processor calls the computer program to execute the steps of the photovoltaic power prediction method for a fishing-photovoltaic system taking into account environmental changes and aging of photovoltaic panels as in Example 1.

[0183] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A photovoltaic power prediction method for a fishery photovoltaic system considering environmental changes and photovoltaic panel aging, characterized in that: The following steps are involved: Obtain data for photovoltaic power prediction of fishery-photovoltaic systems, perform power index analysis, and classify weather types to obtain data sets for various weather types; According to the data set, the correlation between various meteorological factors affecting photovoltaic power generation and photovoltaic power generation power is analyzed, so as to screen out the main factors that have a greater impact on power; Based on the photovoltaic array mechanism model, the theoretical power value is calculated for the irradiance and temperature parameters in the data set, and compared with the corresponding measured power value to identify and optimize power anomalies; The IV curve of the photovoltaic module is tested by an IV tester, and the relative aging degree of the photovoltaic module is calculated, so as to correct the photovoltaic output power in the data set; Analyze the power index based on the measured data of the forecast day, classify the weather type, and find the historical data of similar days from the data set to construct a sample set; Iteratively train the neural network using the sample set until a trained neural network is obtained; The features corresponding to the main influencing factors are extracted according to the measured data on the prediction day and input into the trained neural network to obtain the photovoltaic power prediction results of the fishery-photovoltaic system.

2. A photovoltaic power prediction method for a fishing-photovoltaic system considering environmental changes and photovoltaic panel aging according to claim 1, characterized in that: The process of classifying weather types is specifically as follows: Instantaneous power output and daily cumulative power generation are used as power indicators, and the similarity between different weather type data is calculated through Euclidean distance, so as to classify the weather types; The calculation expression of the Euclidean distance is: In the formula, and are the average power generation of two different weather types at time i; E p and E q are the daily cumulative power generation of the two weather types respectively; n is the number of time periods divided by the selected day.

3. The photovoltaic power prediction method of a fishing light system considering environmental changes and photovoltaic panel aging according to claim 1 is characterized in that: The method adopts a photovoltaic power generation power influencing factor analysis method based on grey correlation to analyze the correlation between various meteorological factors and photovoltaic power generation; The photovoltaic power generation power influencing factor analysis method based on grey correlation includes: Normalize the data in the dataset; For the normalized data, construct the feature vector with each meteorological factor as the similarity factor; Based on the feature vector, a reference sequence and a comparison sequence are constructed, and differences are made to obtain a difference sequence; Define the minimum value of each difference sequence as the minimum range, and the maximum value of each difference sequence as the maximum range; thereby calculate the correlation coefficient between the comparison sequence and the reference sequence to reflect the shape difference of each output timing curve; The correlation coefficients of each comparison sequence are normalized to obtain the corresponding factor weight coefficients to describe the relative effect of each meteorological factor on the system, thereby screening out the main influencing factors with a greater impact on power.

4. The photovoltaic power prediction method of a fishing light system considering environmental changes and photovoltaic panel aging according to claim 3 is characterized in that: The calculation expression of the normalization process is: In the formula, x i is the i-th sequence; x i (k) is the kth data point in sequence i; x i ′(k) is the normalized value of the kth data point; are the largest and smallest data points in the ith sequence, respectively; m is the total number of sequences; n is the total number of data points in the sequence; The calculation expression of the difference sequence is: Δ i (k)=|x′0(k)-x′ i (k)|(i=0,1,2,...,m;k=1,2,3,...,n) In the formula, Δ i =(Δ i (1),Δ i (2),…,Δ i (n)); x0=(x0(1),x0(2),…,x0(n)) is the reference sequence. When i>0, x i =(x i (1),x i (2),…,x i (n)) is a comparison sequence; and x′0 and x′ i All are normalized sequences; The calculation expression of the correlation coefficient is: In the formula, ξ 0i is the correlation coefficient; ρ is the resolution coefficient; is the minimum range; is the maximum range; The calculation expression of the factor weight coefficient is: In the formula, ξ i is the factor weight coefficient of the i-th comparison sequence.

5. The photovoltaic power prediction method of a fishing light system considering environmental changes and photovoltaic panel aging according to claim 1 is characterized in that: The photovoltaic array mechanism model includes a current source, a diode, an equivalent parallel resistance reflecting the non-ideal characteristics of the PN junction and the influence of impurities near it, a contact resistance formed by the electrode and the semiconductor, and a series resistance formed by the semiconductor and the electrode resistance; The diode and the equivalent parallel resistor are respectively connected in parallel at both ends of the current source, and the contact resistor and the series resistor are connected in series with each other and then connected in parallel at both ends of the current source.

6. A photovoltaic power prediction method for a fishing-photovoltaic system considering environmental changes and photovoltaic panel aging according to claim 5, characterized in that: Based on the IV relationship expression of the photovoltaic array mechanism model, the theoretical power value is calculated, and the IV relationship expression is: Where, I is the load current; I p is the current flowing through the parallel resistor; I ph is the photocurrent; I d is the dark current; I sd is the reverse saturation current of the diode; U is the load voltage; R s and R p are the series resistance and equivalent parallel resistance respectively; a is the ideal factor of the diode; T emp is the photovoltaic cell temperature; k0 is the Boltzmann constant, k0=1.381×10 -23 J / K; q is the electron charge, q = 1.6021 × 10 -19 C; in, Where G is the actual irradiance; G ref is the reference irradiance; T ref is the reference temperature; I sc_ref For reference Short-circuit current at illumination and reference temperature; α i is the short-circuit current temperature coefficient; In the formula, E g is the band gap energy; I os is the reverse saturation current of the diode under reference temperature conditions; In the formula, I sc and U oc are the short-circuit current and open-circuit voltage of the photovoltaic cell at the reference temperature.

7. The photovoltaic power prediction method of a fishing light system considering environmental changes and photovoltaic panel aging according to claim 1 is characterized in that: The calculation process of the relative aging degree of the photovoltaic module includes: According to the IV curve of the tested photovoltaic module, the measured open circuit voltage, short circuit current and maximum power point are obtained to calculate the fill factor. The calculation expression of the fill factor is: In the formula, σ FF is the filling factor; U ocm is the open circuit voltage; I scm is the short-circuit current; U max is the voltage corresponding to the maximum power point; I max is the current corresponding to the maximum power point; According to the characteristic that the filling factor of photovoltaic modules is basically greater than 0.7 when they are working normally, the relative aging degree is calculated, and the corresponding calculation expression is: Where, d e is the relative degree of aging; The calculation expression for correcting the photovoltaic output power in the data set is: P s =P t (1-K loss ) Where P t P is the photovoltaic output power without correction and without considering the aging effect; s is the photovoltaic output power after considering the aging effect; K loss is the power loss coefficient.

8. The photovoltaic power prediction method of a fishing light system considering environmental changes and photovoltaic panel aging according to claim 1 is characterized in that: The process of finding historical data of similar days from the data set is specifically as follows: First, we search for similar days to the forecast day by using the highest temperature, lowest temperature and weather conditions. Then we select the day with the closest characteristics as the similar day, and use the comprehensive similarity index as the evaluation criterion. The calculation expression of the comprehensive similarity index is: In the formula, is the calculated value of the comprehensive similarity index; ξ 0i (k) is the correlation coefficient between the current sequence i and the reference sequence; N0 is the number of data in the sequence.

9. The photovoltaic power prediction method of a fishing light system considering environmental changes and photovoltaic panel aging according to claim 1, characterized in that: The neural network is a BP neural network, which predicts the photovoltaic power of the fishery light system after normalizing the collected data, and uses the normalized absolute mean error and the normalized root mean square error to evaluate the normalized error.

10. A photovoltaic power prediction method for a fishing-photovoltaic system taking into account environmental changes and photovoltaic panel aging according to claim 9, characterized in that: The evaluation indicators used in the BP neural network training process include: Wherein, MAPE is the mean absolute percentage error; MAE is the mean absolute error; MBE is the mean error; RMSE is the root mean square error; x′ i is the predicted output value of the BP neural network; x i is the actual value; N is the number of data in the sequence.

Images