Day-ahead electricity price depth intelligent prediction method based on adaptive dynamic window

Abstract

The invention discloses a day-ahead electricity price depth intelligent prediction method based on an adaptive dynamic window, and belongs to the technical field of electricity price prediction, and the method comprises the steps: 1, collecting the original data of the electricity price from a database, and carrying out the preprocessing; 2, inputting the preprocessed data into a window processing subsystem to obtain a first estimation model value of the electricity price; 3, collecting related parameter information related to the electricity price, and obtaining a second estimation model value of the electricity price; and step 4, performing comprehensive analysis in combination with the first estimation model value and the second estimation model value of the electricity price to predict the day-ahead electricity price. According to the invention, comprehensive analysis can be carried out according to a day-ahead overall electricity price model and electricity price models in different time periods of the day to accurately predict the electricity price in the next time period, so that the electricity price can be predicted more accurately in real time according to factors such as the electricity consumption time period, residential electricity consumption demands and weather changes. And the adaptability of the prediction model to short-term and long-term fluctuations is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of electricity price prediction, and specifically relates to a method for deep intelligent prediction of day-ahead electricity prices based on an adaptive dynamic window. Background Art

[0002] With the rapid development of the domestic electricity spot market and the continuous improvement of the market operation mechanism, electricity price forecasting has become an indispensable tool for market participants. The accuracy of electricity price forecasting directly affects the formulation of trading strategies, the assessment of market risks, and the optimization of resource allocation.

[0003] The day-ahead electricity price in the electricity spot market is highly complex and uncertain due to the combined influence of multiple factors. For example, the relationship between electricity supply and demand will change due to the influence of multiple factors, thus affecting the fluctuation of electricity prices. In addition, the core idea of ​​the existing technology is mainly to use a large amount of historical electricity price data to train the model through traditional machine learning or neural network methods, while the domestic electricity spot market is affected by the uncertainty of inter-provincial power dispatch and new energy access, and the fluctuation of electricity prices in a shorter period is stable, which is not suitable for real-time prediction of day-ahead transaction electricity prices. Summary of the invention

[0004] The purpose of the present invention is to provide a method for deep intelligent prediction of day-ahead electricity prices based on an adaptive dynamic window, so as to solve the problems faced in the above-mentioned background technology.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A method for deep intelligent prediction of day-ahead electricity prices based on an adaptive dynamic window, the method comprising:

[0007] Step 1: Collecting raw data related to electricity prices from a database and preprocessing the collected raw data;

[0008] Step 2: input the preprocessed data into the window processing subsystem for processing and analysis to obtain a first estimated model value of the electricity price;

[0009] Step 3: Collect relevant parameter information related to the electricity price, and process and analyze it to obtain a second estimated model value of the electricity price;

[0010] Step 4: Perform a comprehensive analysis based on the first estimation model value and the second estimation model value of the electricity price to predict the day-ahead real-time electricity price.

[0011] Furthermore, the method for obtaining the first estimated model value of the electricity price in step 2 is:

[0012] S21: first, convert the preprocessed data into three-dimensional window data, where the first dimension of the three-dimensional data represents the number of samples, the second dimension represents the position of the window, and the third dimension represents the number of features;

[0013] S22: converting the preprocessed data into two-dimensional window data, where each window represents data for three consecutive days, but there is no continuity between windows, and data is obtained by random sampling;

[0014] S23: The processed window data is processed by IQR method to remove outliers. The principle is shown in formula (2):

[0015] X filtered ={x∈X|Lower Bound≤x≤Upper Bound} (2)

[0016] Among them, Lower Bound and Upper Bound are shown in formulas (3) and (4):

[0017] LowerBound=Q1-1.5*IQR (3)

[0018] UpperBound=Q3+1.5*IQR (4)

[0019] Where Q1 is the lower quartile, Q3 is the upper quartile, and the IQR formula is shown in (5):

[0020] IQR=Q3-Q1(5)

[0021] After processing the outlier data, the data is normalized as shown in formula (6):

[0022]

[0023] Among them, X is the original data, X' is the processed data, μ is the mean of the data, and σ is the standard deviation of the data;

[0024] S24: Extract features. The transformed features are shown in formula (7):

[0025]

[0026] After feature extraction, it contains:

[0027] Constant term 1;

[0028] First-order term: original feature x j ,(j=1,2,…,m);

[0029] Quadratic term: original feature product x j xk , and the square of each feature (where 1≤j≤k≤m);

[0030] Cubic terms: the product of three features in each group x j x k x l , and the cubic of each feature

[0031] (where 1≤j≤k≤l≤m);

[0032] The fourth term: the product of each group of four features x j x k x l x p , and the fourth power of each feature (where 1≤j≤k≤l≤p≤m), m is the number of original features;

[0033] S25: Perform regression training on the processed window data using the RANSAC algorithm to obtain a first estimated model value of the electricity price.

[0034] Furthermore, in S25, the RANSAC algorithm is used to perform regression training on the processed window data to obtain a first estimation model M, whose principle formula is shown in (8):

[0035] M=(θ best ,I best ) (8)

[0036] Then, through the formula Model best =p1*θ best +p2*I best Obtaining a first estimated model value of the electricity price;

[0037] Among them, p1 and p2 are preset parameters, θ best is the optimal model parameter selected by maximizing the number of inliers, as shown in formula (9):

[0038]

[0039] Among them, I best represents the maximum interior point set, and the definition of the interior point set is shown in formula (10):

[0040] I best ={x i ∈X|residual(x i ,θ best )≤∈} (10).

[0041] Furthermore, the characteristics in S24 include the interconnection line plan, new energy forecast, unified electricity load and day-ahead price of electricity prices, and the relevant parameter information in step three includes weather information in different time periods and electricity consumption information in different time periods.

[0042] Furthermore, the method for obtaining the second estimated model value of the electricity price in step 3 is:

[0043] According to the different power consumption time periods, divide the n power consumption time periods in a day into n periods and obtain the previous time period ΔT i-1 The amount of light energy converted into v And the power load Q v ;

[0044] So through the formula Get the second estimated model value Model of the electricity price est ;

[0045] Among them, ET is the environmental impact factor, a1 and a2 are the preset proportional coefficients, L0 is the preset standard light energy conversion amount in this time period, Q0 is the preset standard power load in this time period, α and β are the respective conversion coefficients, and EP is ΔT of the previous day. i The electricity price for a period of time.

[0046] Furthermore, the light energy conversion amount L v The method to obtain it is:

[0047] Get the previous time period ΔT i-1 The light intensity vs. time curve S(t) and the photovoltaic panel temperature vs. time curve T(t);

[0048] So through the formula The light energy conversion amount L v ;

[0049] Among them, σ is the light energy conversion coefficient, t a = ΔT i-1 The start time of the time period, t b = ΔT i-1 The end time of the time period, yn is ΔT i-1 The cloud cover for the time period, c1 and c2 are conversion coefficients.

[0050] Furthermore, the method for predicting the day-ahead electricity price in step 4 is:

[0051] Through the formula MO = Model best +Model est 2 Predict the day-ahead electricity price MO for each time period.

[0052] Beneficial effects of the present invention:

[0053] The present invention collects the original data of electricity prices from a historical database, pre-processes the collected original data, and then inputs the pre-processed data into a window processing subsystem for processing and analysis to obtain a first estimated model value of the electricity price. The method first calculates the empirical threshold of the data through the interquartile range method and preliminarily filters out abnormal values. In the model training stage, the RANSAC random sampling consistency algorithm is used to continuously iterate and train to obtain the optimal training set. The traditional machine learning method is combined with the IQR-RANSAC method and applied to the actual day-ahead price prediction task of electric power spot trading, so that the overall model of the electricity price of the day can be obtained more accurately, which is convenient for evaluating and predicting the day-ahead electricity price.

[0054] The present invention can also perform analysis and processing based on relevant parameter information of different time periods of the day to obtain a second estimated model value of the electricity price in each time period, and predict the electricity price situation in the current time period on the previous day based on the electricity price situation in the previous time period and the electricity price situation in the current time period on the previous day, so as to make a real-time prediction of the electricity price in the current time period, thereby more accurately understanding the real-time fluctuation of the electricity price.

[0055] The present invention can also make a comprehensive analysis based on the overall electricity price model of the day before and the electricity price models of different time periods of the day to accurately predict the electricity price for the next time period. In this way, the electricity price can be more accurately predicted in real time based on factors such as electricity consumption time period, residents' electricity demand and weather changes, so as to improve the adaptability of the prediction model to short-term and long-term fluctuations.

[0056] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0058] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0059] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. 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.

[0060] In one embodiment, a method for deep intelligent prediction of day-ahead electricity prices based on an adaptive dynamic window is disclosed, such as Figure 1 As shown, the prediction method includes:

[0061] Step 1: Collecting raw data related to electricity prices from a database and preprocessing the collected raw data;

[0062] Step 2: input the preprocessed data into the window processing subsystem for processing and analysis to obtain a first estimated model value of the electricity price;

[0063] Step 3: Collect relevant parameter information related to the electricity price, and process and analyze it to obtain a second estimated model value of the electricity price;

[0064] Step 4: Perform a comprehensive analysis based on the first estimation model value and the second estimation model value of the electricity price to predict the day-ahead real-time electricity price.

[0065] Through the above technical scheme, the present application collects raw data related to electricity price prediction from a historical database, and preprocesses the collected raw data. The preprocessing steps include but are not limited to cleaning, integrating, converting, etc. of the raw data, and then inputs the preprocessed data into the window processing subsystem for processing and analysis to obtain a first estimated model value of the electricity price. The empirical threshold of the data is first calculated by the interquartile range method (IQR) and outliers are initially filtered out. In the model training stage, the random sampling consensus algorithm (RANSAC) is continuously iterated to obtain the optimal training set, and the traditional machine learning method is combined with the IQR-RANSAC method, which is applied to the actual day-ahead price prediction task of electricity spot trading, so that the model of the electricity price of the day (history) can be obtained more accurately, which is convenient for evaluating and predicting the day-ahead electricity price. Then, the relevant parameter information of different time periods of the day is analyzed and processed to obtain the second estimated model value of the electricity price (real-time), and finally the first estimated model value of the electricity price and the second estimated model value are combined for comprehensive analysis to predict the day-ahead electricity price. In this way, we can conduct a comprehensive analysis based on the overall electricity price model of the day before and the electricity price models of different time periods on the day to accurately predict the electricity price for the next time period. This can make it possible to more accurately predict electricity prices in real time based on factors such as electricity consumption time periods, residential electricity demand, and weather changes, thereby improving the adaptability of the prediction model to short-term and long-term fluctuations.

[0066] As an implementation mode of the present invention, the method for obtaining the first estimated model value of the electricity price in step 2 is:

[0067] S21: First, the preprocessed data is converted into three-dimensional window data, where the first dimension of the three-dimensional data represents the number of samples, the second dimension represents the position of the window (i.e., the time step), and the third dimension represents the number of features;

[0068] S22: converting the preprocessed data into two-dimensional window data, where each window represents data for three consecutive days, but there is no continuity between windows, and data is obtained by random sampling;

[0069] S23: The processed window data is processed by IQR method to remove outliers and obtain the filtered data X filtered , the principle is shown in formula (2):

[0070] X filtered ={x∈X|Lower Bound≤x≤Upper Bound} (2)

[0071] Among them, the Lower Bound (lower bound threshold) and the Upper Bound (upper bound threshold) are shown in formulas (3) and (4):

[0072] LowerBound=Q1-1.5*IQR (3)

[0073] UpperBound=Q3+1.5*IQR (4)

[0074] Where Q1 is the lower quartile, Q3 is the upper quartile, and the IQR formula is shown in (5):

[0075] IQR=Q3-Q1(5)

[0076] After processing the outlier data, the data is normalized as shown in formula (6):

[0077]

[0078] Among them, X is the original data, X' is the normalized data, μ is the mean of the data, and σ is the standard deviation of the data;

[0079] S24: Extract features, including the interconnection line plan of electricity price, new energy forecast, unified electricity load and day-ahead price, etc. The transformed features are shown in formula (7):

[0080]

[0081] After feature extraction, it contains:

[0082] Constant term 1;

[0083] First-order term: original feature xj ,(j=1,2,…,m);

[0084] Quadratic term: original feature product x j x k , and the square of each feature (where 1≤j≤k≤m);

[0085] Cubic terms: the product of three features in each group x j x k x l , and the cubic of each feature

[0086] (where 1≤j≤k≤l≤m);

[0087] The fourth term: the product of each group of four features x j x k x l x p , and the fourth power of each feature (where 1≤j≤k≤l≤p≤m), m is the number of original features;

[0088] S25: Use the RANSAC algorithm to perform regression training on the processed window data to obtain the first estimation model M, whose principle formula is shown in (8):

[0089] M=(θ best ,I best ) (8)

[0090] Then, through the formula Model best =p1*θ best +p2*I best Obtaining a first estimated model value of the electricity price;

[0091] Among them, p1 and p2 are preset parameters, θ best is the optimal model parameter selected by maximizing the number of inliers, as shown in formula (9):

[0092]

[0093] Among them, I best represents the maximum interior point set, and the definition of the interior point set is shown in formula (10):

[0094] I best ={x i ∈X|residual(x i ,θ best )≤∈} (10).

[0095] Through the above technical solution, this embodiment provides a method for obtaining the first estimated model value of the electricity price, which specifically includes the following steps S21: first, convert the preprocessed data into three-dimensional window data, wherein the first dimension of the three-dimensional data represents the number of samples, the second dimension represents the position of the window (i.e., the time step), and the third dimension represents the number of features; S22: convert the preprocessed data into two-dimensional window data, wherein each window represents data for three consecutive days, but there is no continuity between windows, and the data are obtained by random sampling; S23: remove outliers from the processed window data by the IQR method to obtain filtered data X filtered , the principle is shown in formula (2):

[0096] X filtered ={x∈X|Lower Bound≤x≤Upper Bound} (2)

[0097] Among them, the Lower Bound (lower bound threshold) and the Upper Bound (upper bound threshold) are shown in formulas (3) and (4):

[0098] LowerBound=Q1-1.5*IQR (3)

[0099] UpperBound=Q3+1.5*IQR (4)

[0100] Where Q1 is the lower quartile, Q3 is the upper quartile, and the IQR formula is shown in (5):

[0101] IQR=Q3-Q1 (5)

[0102] After processing the outlier data, the data is normalized as shown in formula (6):

[0103]

[0104] Among them, X is the original data, X' is the normalized data, μ is the mean of the data, and σ is the standard deviation of the data;

[0105] S24: Extract features, including the interconnection line plan of electricity price, new energy forecast, unified electricity load and day-ahead price, etc. The transformed features are shown in formula (7):

[0106]

[0107] After feature extraction, it contains:

[0108] Constant term 1;

[0109] First-order term: original feature x j,(j=1,2,…,m);

[0110] Quadratic term: original feature product x j x k , and the square of each feature (where 1≤j≤k≤m);

[0111] Cubic terms: the product of three features in each group x j x k x l , and the cubic of each feature

[0112] (where 1≤j≤k≤l≤m);

[0113] The fourth term: the product of each group of four features x j x k x l x p , and the fourth power of each feature (where 1≤j≤k≤l≤p≤m), m is the number of original features;

[0114] S25: Use the RANSAC algorithm to perform regression training on the processed window data to obtain the first estimation model M, whose principle formula is shown in (8):

[0115] M=(θ best ,I best ) (8)

[0116] Then, through the formula Model best =p1*θ best +p2*I best Obtaining a first estimated model value of the electricity price;

[0117] Among them, p1 and p2 are preset parameters, θ best is the optimal model parameter selected by maximizing the number of inliers, as shown in formula (9):

[0118]

[0119] Among them, I best represents the maximum interior point set, and the definition of the interior point set is shown in formula (10):

[0120] I best ={x i ∈X|residual(x i ,θ best )≤∈} (10).

[0121] It can be seen that this method mainly consists of three key subsystems, among which (1) data preprocessing system: in this stage, the raw data undergoes a series of processing steps to improve the quality and efficiency of subsequent analysis, including feature extraction, that is, constructing important features related to electricity prices based on the raw data. In addition, the data will be normalized to adjust the numerical ranges of different features to the same scale to avoid excessive impact on the model due to differences in feature values. (2) Window processing subsystem: the preprocessed data is transmitted to the window processing subsystem. The system analyzes the data of the day, dynamically selects the appropriate window size and groups it, and then divides the data into training set and test set based on the grouping results. (3) Forecasting subsystem: the data after window processing finally enters the forecasting subsystem for forecasting to obtain the first estimated model value of the electricity price of the day. In this way, the size of the daily training window can be flexibly adjusted according to the market fluctuations and data characteristics of the day, so as to capture more timely price change trends, thereby more accurately deriving a model of the day-ahead electricity price, and improving the adaptability of the forecasting model to short-term and long-term fluctuations.

[0122] As an implementation mode of the present invention, the method for obtaining the second estimated model value of the electricity price in step 3 is:

[0123] According to the different power consumption time periods, divide the n power consumption time periods in a day into n periods and obtain the previous time period ΔT i-1 The amount of light energy converted into v And the power load Q v ;

[0124] So through the formula Get the second estimated model value Model of the electricity price est ;

[0125] Among them, ET is the environmental impact factor, a1 and a2 are the preset proportional coefficients, L0 is the preset standard light energy conversion amount in this time period, Q0 is the preset standard power load in this time period, α and β are the respective conversion coefficients, and EP is ΔT of the previous day. i The electricity price in the time period, the amount of light energy conversion L v The method of obtaining is: obtain the previous time period ΔT i-1 The light intensity vs. time curve S(t) and the photovoltaic panel temperature vs. time curve T(t);

[0126] So through the formula The light energy conversion amount L v ;

[0127] Among them, σ is the light energy conversion coefficient, t a = ΔT i-1 The start time of the time period, t b = ΔTi-1 The end time of the time period, yn is ΔT i-1 The cloud cover for the time period, c1 and c2 are conversion coefficients.

[0128] Through the above technical solution, this embodiment provides a method for obtaining the second estimated model value of the electricity price. Since the supply and demand in each time period are different, the electricity price will also fluctuate in the short term, and the prediction model can only predict the overall electricity price of the day, so it is impossible to make a more accurate real-time prediction of the electricity price. Therefore, this application divides n electricity consumption time periods in a day according to different electricity time periods, and obtains the previous time period ΔT i-1 The amount of light energy converted into v And the power load Q v , where the power load is obtained based on the power consumption data, and the light energy conversion amount L v By formula It is concluded that the amount of light energy conversion is mainly related to the light intensity, the amount of cloud cover and the temperature of the photovoltaic panel. Generally speaking, the greater the light intensity, the greater the light energy conversion. The higher the temperature of the photovoltaic panel or the greater the cloud cover, the lower its photovoltaic conversion efficiency. Therefore, ΔT is obtained. i-1 The light intensity changes with time curve S(t) and the photovoltaic panel temperature changes with time curve T(t), so that the formula The light energy conversion amount L v , the influence of temperature and cloud cover can be eliminated to make the result more accurate; then through the formula Get the second estimated model value Model of the electricity price est In this way, the electricity price situation in the current time period can be predicted based on the electricity price situation in the previous time period and the electricity price situation in the current time period of the previous day, so as to make a real-time prediction of the electricity price in the current time period and understand the real-time fluctuation of electricity prices more accurately.

[0129] It should be noted that the environmental impact factor ET is preset in real time according to the weather forecast for the day. The preset proportional coefficients a1 and a2, the respective conversion coefficients α and β, and the light energy conversion coefficient σ can all be determined based on historical data combined with empirical data. The preset standard light energy conversion amount L0 can be predicted based on weather forecasts for different time periods. The standard electricity load Q0 is determined based on historical electricity consumption data, and the cloud cover ∈ is obtained through cloud radar technology. It is worth mentioning again.

[0130] As an implementation mode of the present invention, the method for predicting the day-ahead electricity price in step 4 is:

[0131] Through the formula MO = Model best +Model est 2 Predict the day-ahead electricity price MO for each time period.

[0132] Through the above technical solution, the real-time final formula MO = Model best +Model est 2 The MO of the day-ahead electricity price for each time period is predicted. Based on the overall model of the day-ahead and the model of the current time period, the real-time electricity price of each time period can be understood more accurately in real time, which can improve the adaptability of the prediction model to short-term and long-term fluctuations, and better formulate trading plans, thereby minimizing the risk of electricity price trading.

[0133] The above contents are merely examples and explanations of the concept of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. A method for deep intelligent prediction of day-ahead electricity prices based on an adaptive dynamic window, characterized in that: The method comprises: Step 1: Collecting raw data related to electricity prices from a database and preprocessing the collected raw data; Step 2: input the preprocessed data into the window processing subsystem for processing and analysis to obtain a first estimated model value of the electricity price; Step 3: Collect relevant parameter information related to the electricity price, and process and analyze it to obtain a second estimated model value of the electricity price; Step 4: Perform a comprehensive analysis based on the first estimation model value and the second estimation model value of the electricity price to predict the day-ahead real-time electricity price.

2. The method for deep intelligent prediction of day-ahead electricity prices based on an adaptive dynamic window according to claim 1 is characterized in that: The method for obtaining the first estimated model value of the electricity price in step 2 is: S21: first, convert the preprocessed data into three-dimensional window data, where the first dimension of the three-dimensional data represents the number of samples, the second dimension represents the position of the window, and the third dimension represents the number of features; S22: converting the preprocessed data into two-dimensional window data, where each window represents data for three consecutive days, but there is no continuity between windows, and data is obtained by random sampling; S23: The processed window data is processed by IQR method to remove outliers. The principle is shown in formula (2): X filtered ={x∈X|Lower Bound≤x≤Upper Bound} (2) Among them, Lower Bound and Upper Bound are shown in formulas (3) and (4): LowerBound=Q1-1.5*IQR (3) UpperBound=Q3+1.5*IQR (4) Where Q1 is the lower quartile, Q3 is the upper quartile, and the IQR formula is shown in (5): IQR=Q3-Q1 (5) After processing the outlier data, the data is normalized as shown in formula (6): Among them, X is the original data, X' is the processed data, μ is the mean of the data, and σ is the standard deviation of the data; S24: Extract features. The transformed features are shown in formula (7): After feature extraction, it contains: Constant term 1; First-order term: original feature x j ,(j=1,2,…,m); Quadratic term: original eigenproduct x j x k , and the square of each feature (where 1≤j≤k≤m); Cubic terms: the product of three features in each group x j x k x l , and the cubic of each feature (where 1≤j≤k≤l≤m); The fourth term: the product of each group of four features x j x k x l x p , and the fourth power of each feature (where 1≤j≤k≤l≤p≤m), m is the number of original features; S25: Perform regression training on the processed window data using the RANSAC algorithm to obtain a first estimated model value of the electricity price.

3. The method for deep intelligent prediction of day-ahead electricity prices based on an adaptive dynamic window according to claim 2 is characterized in that: In S25, the RANSAC algorithm is used to perform regression training on the processed window data to obtain a first estimation model M, and its principle formula is shown in (8): M=(θ best ,I best ) (8) Then, through the formula Model best =p1*θ best +p2*I best Obtaining a first estimated model value of the electricity price; Among them, p1 and p2 are preset parameters, θ best is the optimal model parameter selected by maximizing the number of inliers, as shown in formula (9): Among them, I best represents the maximum interior point set, and the definition of the interior point set is shown in formula (10): I best ={x i ∈X|residual(x i ,θ best )≤∈} (10)。 4. The method for deep intelligent prediction of day-ahead electricity prices based on an adaptive dynamic window according to claim 3 is characterized in that: The features in S24 include the interconnection line plan, new energy forecast, unified electricity load and day-ahead price of electricity prices, and the relevant parameter information in step three includes weather information in different time periods and electricity consumption information in different time periods.

5. The method for deep intelligent prediction of day-ahead electricity prices based on an adaptive dynamic window according to claim 4 is characterized in that: The method for obtaining the second estimated model value of the electricity price in step 3 is: According to the different power consumption time periods, divide the n power consumption time periods in a day into n periods and obtain the previous time period ΔT i-1 The amount of light energy converted into v And the power load Q v ; So through the formula Get the second estimated model value Model of the electricity price est ; Among them, ET is the environmental impact factor, a1 and a2 are the preset proportional coefficients, L0 is the preset standard light energy conversion amount in this time period, Q0 is the preset standard power load in this time period, α and β are the respective conversion coefficients, and EP is ΔT of the previous day. i The electricity price for a period of time.

6. The method for deep intelligent prediction of day-ahead electricity prices based on an adaptive dynamic window according to claim 5, characterized in that: The light energy conversion amount L v The method to obtain it is: Get the previous time period ΔT i-1 The light intensity vs. time curve S(t) and the photovoltaic panel temperature vs. time curve T(t); So through the formula The light energy conversion amount L v ; Among them, σ is the light energy conversion coefficient, t a = ΔT i-1 The start time of the time period, t b = ΔT i-1 The end time of the time period, yn is ΔT i-1 The cloud cover for the time period, c1 and c2 are conversion coefficients.

7. The method for deep intelligent prediction of day-ahead electricity prices based on an adaptive dynamic window according to claim 6, characterized in that: The method for predicting the day-ahead electricity price in step 4 is: Through the formula MO = Model best +Model est 2 Predict the day-ahead electricity price MO for each time period.

Images