Elastic electricity transaction decision-making method and system

Abstract

The invention discloses an elastic electricity transaction decision-making method and system, relates to the technical field of data processing, and solves the technical problems that historical electricity transaction data are not sufficiently analyzed and utilized, and the transaction decision-making method is not flexible enough in dealing with market fluctuation. According to the method, the fluctuation trend and seasonal change of the price are accurately analyzed, effective transaction rule information is mined, an accurate generating capacity prediction model is established, the short-term generating capacity can be effectively predicted according to the transaction rule information and historical power transaction data, corresponding transaction parameter intervals are determined according to fluctuation influences of different conditions, and the generating capacity prediction efficiency is improved. According to the invention, a reasonable transaction decision is generated to realize balance between power supply and demand, the generated transaction decision is periodically monitored, judgment and adjustment are carried out in time according to the demand quantity change condition in the monitoring period, and the accuracy and adaptability of the transaction decision are improved.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a flexible power trading decision-making method and system. Background Art

[0002] With the development of economy and the progress of society, electricity plays an increasingly important role in people's life and production. The electricity trading market is also becoming increasingly complex and diversified, with frequent price fluctuations and affected by many factors, such as seasonal changes, weather conditions, economic indicators, electricity consumption, etc.

[0003] According to publication number CN116452327A, a power trading decision-making method and related equipment are provided, including: obtaining initial weather forecast values, inputting the initial weather forecast values ​​into a pre-constructed multi-source ensemble forecast model to obtain weather forecast values; wherein the multi-source ensemble forecast model is obtained by training a convolutional neural network with weather-related sample data, and the initial weather forecast values ​​are weather forecast observation values; forecasting the weather forecast values ​​through an optimal forecast algorithm model to obtain electric power forecast values ​​corresponding to the weather forecast values; obtaining a historical electricity price sequence, forecasting the historical electricity price sequence through a cosine combination forecast model to obtain a real-time electricity price forecast value; forecasting the electric power forecast value and the real-time electricity price forecast value through a time period combination forecast model to obtain auxiliary decision data for power trading.

[0004] At present, in power trading, the analysis and utilization of historical power trading data is not sufficient, and there is a lack of effective methods to explore the trading rules in historical power trading data, making it difficult to accurately predict the changes in short-term power generation and user demand. At the same time, the existing trading decision-making methods are not flexible enough in dealing with market fluctuations, and cannot adjust the trading parameter range in time according to market changes, resulting in inaccurate and inefficient trading decisions. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a flexible electricity trading decision-making method and system, which solves the problem that the analysis and utilization of historical electricity trading data is not sufficient and the trading decision-making method is not flexible enough in dealing with market fluctuations.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a flexible power trading decision-making method, comprising the following steps: S101: Acquire historical power transaction data, analyze price fluctuation trends and seasonal changes based on the historical power transaction data, and generate transaction rule information; S102: establishing a power generation prediction model based on the transaction rule information and predicting short-term power generation, and comparing short-term power generation with user demand based on historical power transaction data to generate a comparison result, wherein the comparison result includes: a large supply signal and a large demand signal; S103: Analyze the obtained comparison results, determine the corresponding trading parameter range according to the fluctuation impact in different situations, generate trading parameter range information, and then generate a trading decision according to the trading parameter range information; S014: Perform periodic monitoring based on the generated transaction decision, and make judgments based on changes in demand during the monitoring period, while generating judgment information, and the judgment information includes normal monitoring information and interval adjustment information.

[0007] As a further solution of the present invention: the specific method of generating the transaction rule information in step S101 is: Obtaining power transaction data before time t, and analyzing the corresponding price fluctuation trend in the power transaction data, the analysis methods include: correlation analysis and cluster analysis; The method of correlation analysis is to calculate the correlation between prices and other factors at different time points; The cluster analysis method is to use cluster analysis method for a large amount of time node data to cluster similar time nodes into one category, based on the results of correlation analysis and cluster analysis; Comprehensive correlation analysis and cluster analysis generate transaction pattern information.

[0008] As a further solution of the present invention: the specific method of establishing the power generation prediction model according to the transaction rule information in step S102 is: The first step is to collect historical power generation data, and the time span covers the impact of different seasons, weather conditions and economic activity levels on power generation; The second step is to check the completeness and accuracy of the data and deal with missing values ​​and outliers; The third step is to test the stability of the power generation time series. If the time series is not stable, it will be processed by difference to make it a stable series. The fourth step is to calculate the autocorrelation function and partial autocorrelation function of the stable power generation time series, observe its decay mode, and preliminarily determine the order range of autoregression (AR) and moving average (MA) according to the characteristics of ACF and PACF; The fifth step is to estimate the parameters of the ARMA model using the maximum likelihood estimation method, and generate the corresponding power generation forecast model. The mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE) are used to evaluate the prediction effect of the model, and the power generation forecast model is established. The corresponding power generation forecast is carried out according to the established power generation forecast model, the historical power transaction data within the forecast time t1 is obtained, and the historical power transaction data is substituted into the power generation forecast model to obtain the short-term power generation.

[0009] As a further solution of the present invention: the specific method of generating the comparison result in step S012 is: Obtain the historical electricity trading data within the prediction time t1, and at the same time obtain the corresponding electricity demand, and compare the electricity demand with the short-term power generation. If the short-term power generation is greater than the electricity demand, a large supply signal is generated; if the short-term power generation is less than the electricity demand, a large demand signal is generated.

[0010] As a further solution of the present invention: the specific method of analyzing the comparison result in step S013 is: Obtaining a comparison result, wherein the comparison result includes a large supply signal and a large demand signal, and taking the large supply signal as an example for analysis, obtaining a prediction time t1, and obtaining historical power trading data corresponding to the prediction time t1, and obtaining corresponding trading parameters according to the historical power trading data, and integrating the trading parameters to obtain a historical trading parameter range; Then, taking the analysis situation corresponding to the large supply signal as the standard, the data corresponding to the same situation in the historical power trading data is obtained, and then the trading parameters corresponding to the obtained data of the same situation are integrated to obtain the secondary integration interval. At the same time, the secondary integration interval is adjusted based on the impact of price fluctuations within the predicted time t1.

[0011] As a further solution of the present invention: the specific method of adjusting the secondary integration interval in step S103 is: Obtain the energy transaction change and energy demand, and calculate the corresponding total energy consumption change cost, then obtain the total cost corresponding to the same energy demand in the historical power transaction data, and calculate the ratio of the total energy consumption change cost to the total cost as the energy fluctuation ratio; Then, the energy fluctuation ratio is multiplied by the secondary integration interval to obtain the product value interval, and the secondary integration interval and the product value interval are summed to obtain the trading parameter interval, and a trading decision is generated according to the trading parameter interval.

[0012] As a further solution of the present invention: the specific method of generating the judgment information in step S104 is: Conduct periodic monitoring of trading decisions, obtain trading demand within time T, and obtain changes in trading demand. If the trading demand increases, generate normal monitoring information; if the trading demand decreases, generate interval adjustment information.

[0013] A flexible power trading decision system, comprising: a trading data collection unit, a trading data analysis unit, a trading decision generation unit and a trading decision output unit; A transaction data collection unit, which is used to transmit the acquired power transaction data to the transaction data analysis unit; A transaction data analysis unit, which is used to analyze the fluctuation trend and seasonal changes of the acquired power transaction data and generate transaction law information, and at the same time establish a power generation prediction model and predict short-term power generation based on the transaction law information, and then analyze the relationship between short-term power generation and user demand based on historical power transaction data to generate a comparison result, wherein the comparison result includes a large supply signal and a large demand signal, and transmit the generated comparison result to the transaction decision generation unit; A trading decision generating unit, which is used to analyze the obtained comparison results, determine the trading parameter interval by calculating the fluctuation influence under different situations, integrate the trading parameter interval based on the same situation corresponding to the historical power trading data to obtain a secondary integration interval, and generate the trading parameter interval in combination with the fluctuation influence, and generate a trading decision, and then transmit the trading decision to the trading decision output unit; The transaction decision output unit is used to periodically monitor the acquired transaction decision information, and to generate judgment information based on the change of the corresponding demand within the monitoring period, and to display the judgment information to the corresponding operator.

[0014] The present invention provides a flexible power trading decision-making method and system. Compared with the prior art, it has the following beneficial effects: The present invention utilizes historical electricity trading data to accurately analyze price fluctuation trends and seasonal changes, dig out effective trading rule information, and establish an accurate power generation prediction model. It can effectively predict short-term power generation based on trading rule information and historical electricity trading data, determine the corresponding trading parameter range according to the fluctuation impact of different situations, generate reasonable trading decisions to achieve a balance between electricity supply and demand, periodically monitor the generated trading decisions, make timely judgments and adjustments based on changes in demand within the monitoring period, and improve the accuracy and adaptability of trading decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flow chart of the steps of the present invention; Figure 2 This is the principle framework of the system of the present invention. DETAILED DESCRIPTION

[0016] 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.

[0017] For example, see Figure 1 The present application provides a flexible power trading decision-making method, which includes the following steps: In step S101, historical power transaction data is acquired, and price fluctuation trends and seasonal changes are analyzed based on the historical power transaction data, and transaction rule information is generated.

[0018] First, the electricity trading data before time t is obtained, and the specific value of time t here is set by the operator, which may be a quarter or a month. At the same time, the corresponding price fluctuation trend in the electricity trading data is analyzed, and the relationship between the price fluctuation and the different time nodes is judged to generate trading rule information. The judgment method here is: through correlation analysis and cluster analysis. The correlation analysis method is to calculate the correlation between the price of different time nodes and other factors (such as electricity consumption, weather conditions, economic indicators, etc.). The cluster analysis method is that for a large amount of time node data, a cluster analysis method can be used to cluster similar time nodes into one category, and some trading rules are extracted according to the results of correlation analysis and cluster analysis.

[0019] In a specific example, by observing long-term price trends, we can discover the cyclical pattern of prices. For example, prices may rise during the peak electricity consumption period in the summer, and may fall during the low period in the winter. For another example, when analyzing based on quarterly time nodes, we observed that in the second quarter (April-June) and the third quarter (July-September), due to frequent business activities and the combined influence of temperature factors, electricity price fluctuations showed a cyclical upward pattern; while in the first quarter (January-March), due to the fact that some companies were in the off-season and the temperature was relatively mild, the fluctuations in electricity trading parameters were relatively gentle and at a low level.

[0020] Next, in step S102, a power generation prediction model is established based on the transaction rule information and short-term power generation is predicted. At the same time, short-term power generation is compared with user demand based on historical power transaction data to generate a comparison result, wherein the comparison result includes: a large supply signal and a large demand signal.

[0021] Obtain price influencing factors in the transaction rule information, where the price influencing factors include power generation and supply and demand relationship, and perform forecast analysis on the power generation, and the specific forecast analysis method is to establish a corresponding power generation forecast model for forecasting, as follows: First, the prediction time scale is clarified, and the prediction time scale here includes: short-term prediction (usually a few hours to a few days), medium-term prediction (usually a few weeks to a few months) and long-term prediction (several months to a few years). In this application, it is a short-term prediction, and the power generation data in the past period of time is collected, including data of different time scales, so as to analyze the changing trend and seasonal characteristics of power generation. Then, the time series analysis method is selected as the standard for corresponding prediction, and the corresponding model is selected for prediction analysis. The model selected here is the autoregressive moving average model (ARMA), and the corresponding prediction model is established at the same time; The first step is to collect historical power generation data. The time span should be long enough to cover the impact of different seasons, weather conditions, and economic activity levels on power generation. The frequency of the data can be determined based on the forecast time scale, such as daily, weekly, or monthly power generation data. The second step is to check the completeness and accuracy of the data and deal with missing values ​​and outliers. Missing values ​​can be handled by interpolation, mean filling, etc., and outliers can be removed or corrected. First, clean the data and check whether there are missing values. If there are missing values, the average value of adjacent months can be used to fill them.

[0022] The third step is to test the stationarity of the power generation time series. Commonly used methods include unit root test (such as ADF test). If the time series is not stationary, it needs to be differentiated to make it a stationary series. Specifically, by drawing the power generation time series graph and performing ADF test, it is found that the time series is not stationary. After first-order difference processing, it is tested for stationarity again. At this time, the time series becomes a stationary series.

[0023] The fourth step is to calculate the autocorrelation function and partial autocorrelation function of the stable power generation time series, observe its decay mode, and preliminarily determine the order range of autoregression (AR) and moving average (MA) based on the characteristics of ACF and PACF; it is observed that ACF is tailing and PACF is truncated after a certain order. It is preliminarily judged that a low-order ARMA model may be suitable, such as ARMA (1,1) or ARMA (2,1). Then use the AIC and BIC information criteria for evaluation. After calculation, it is found that the AIC and BIC values ​​of the ARMA (2,1) model are relatively small, and the model is determined to be a better choice.

[0024] The fifth step is to estimate the parameters of the ARMA model using the maximum likelihood estimation method. Specifically, the autoregressive coefficient and moving average coefficient in the model are determined by maximizing the likelihood function. Through the optimization algorithm, the autoregressive coefficients are 0.5 and -0.2, and the moving average coefficient is 0.4. At the same time, the corresponding power generation forecast model is generated, and the mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE) are used to evaluate the forecast effect of the model; assuming that the actual power generation is 990 MWh in the 25th month, 1060 MWh in the 26th month, and 1110 MWh in the 27th month. The calculated MAE is 10, the RMSE is 15, and the MAPE is 1%. The evaluation results show that the model has a good forecast effect.

[0025] According to the established power generation forecasting model, the corresponding power generation forecast is carried out, the historical power trading data within time t1 is obtained, and the historical power trading data is substituted into the power generation forecasting model to obtain the short-term power generation. Assuming that the current month is the 24th month, the model predicts that the power generation in the 25th month will be 1000 MWh, the power generation in the 26th month will be 1050 MWh, and the power generation in the 27th month will be 1100 MWh. At the same time, the 95% confidence interval of the predicted value is calculated, assuming that the confidence interval of the 25th month is [980, 1020], the confidence interval of the 26th month is [1020, 1080], and the confidence interval of the 27th month is [1080, 1120].

[0026] Then, based on the historical electricity trading data within the prediction time t1, the corresponding electricity demand is obtained, and the electricity demand is compared with the short-term power generation. If the short-term power generation is greater than the electricity demand, a large supply signal is generated; if the short-term power generation is less than the electricity demand, a large demand signal is generated.

[0027] Next, in step S103, the obtained comparison result is analyzed, and the corresponding trading parameter range is determined according to the fluctuation impact in different situations, and trading parameter range information is generated at the same time, and then a trading decision is generated according to the trading parameter range information.

[0028] Obtaining comparison results, wherein the comparison results include a large supply signal and a large demand signal, and taking the large supply signal as an example for analysis (the large demand signal is analyzed in the same way), obtaining the forecast time t1, and obtaining the historical power trading data corresponding to the forecast time t1, and the historical power trading data obtained here is in the same time period as the forecast time t1, and obtaining the corresponding trading parameters according to the historical power trading data, and integrating the trading parameters to obtain the historical trading parameter range; Then, taking the analysis situation corresponding to the large supply signal as the standard, the data corresponding to the same situation in the historical power trading data is obtained. For example, the average daily power generation at that time reached 50 million kWh, while the daily power consumption was only 45 million kWh. At this time, the corresponding signal is a large supply signal. Further, the data of the situation in which the power generation is greater than the power consumption in the historical power trading data is obtained. Then, the transaction parameters corresponding to the obtained data of the same situation are integrated to obtain the secondary integration interval. At the same time, the secondary integration interval is adjusted based on the impact of price fluctuations within the forecast time t1, and the specific conditions are as follows: Obtain the energy transaction change and energy demand, where the energy transaction change represents the unit energy transaction change. The transaction change is an interval value, which is determined based on the historical power transaction data, and the average of the sum of the minimum and maximum values ​​is taken as the standard for calculation, including increases and decreases. The energy demand represents the energy demand corresponding to the short-term power generation, and the corresponding total energy consumption change cost is calculated. Then, the total cost corresponding to the same energy demand in the historical power transaction data is obtained, and the ratio of the total energy consumption change cost to the total cost is calculated and recorded as the energy fluctuation ratio; Then, the energy fluctuation ratio is multiplied by the secondary integration interval to obtain the product value interval, and the secondary integration interval and the product value interval are summed. Specifically, the minimum value and the maximum value of the secondary integration interval are respectively multiplied by the energy fluctuation ratio to obtain the trading parameter interval, and a trading decision is generated based on the trading parameter interval.

[0029] Next, in step S014, periodic monitoring is performed based on the generated transaction decision, and a judgment is made according to the change in demand during the monitoring period, and judgment information is generated.

[0030] Conduct periodic monitoring of trading decisions, obtain trading demand within time T, and obtain changes in trading demand. If the trading demand increases, generate normal monitoring information; if the trading demand decreases, generate interval adjustment information.

[0031] In a specific example, a monitoring cycle T is one month. During this cycle, the demand for electricity trading is obtained by collecting and analyzing market data. If the demand for electricity trading continues to increase from the beginning of the month to the middle of the month, it means that the market demand for electricity is relatively strong, and normal monitoring information is generated at this time. For example, the launch of some new industrial projects has led to an increase in electricity consumption, which has increased the demand for electricity trading. However, if the demand for electricity trading begins to decrease from the middle of the month to the end of the month, it may be due to some companies adjusting their production plans and reducing electricity consumption, or the weather turning cooler and the demand for electricity from residents decreasing. At this time, interval adjustment information will be generated, prompting relevant departments and market participants to adjust their electricity trading decisions. For example, the price of electricity can be appropriately lowered to stimulate demand, or the power generation plan can be adjusted to avoid oversupply of electricity.

[0032] For example 2, please refer to Figure 2 A flexible power trading decision system includes: a trading data collection unit, a trading data analysis unit, a trading decision generation unit and a trading decision output unit, and each of the above units is unidirectionally electrically connected.

[0033] The transaction data acquisition unit is used to transmit the acquired power transaction data to the transaction data analysis unit.

[0034] A transaction data analysis unit, which is used to analyze the fluctuation trend and seasonal changes of the acquired power transaction data, and generate transaction rule information. At the same time, a power generation prediction model is established based on the transaction rule information and short-term power generation is predicted. Then, the relationship between short-term power generation and user demand is analyzed based on historical power transaction data to generate a comparison result, wherein the comparison result includes a large supply signal and a large demand signal. At the same time, the generated comparison result is transmitted to the transaction decision generation unit, and the specific processing method here is the same as that of step S102 in Example 1.

[0035] A trading decision generating unit is used to analyze the obtained comparison results, determine the trading parameter range by calculating the fluctuation impact under different situations, and integrate the trading parameter range based on the same situation corresponding to the historical power trading data to obtain a secondary integrated range, and generate the trading parameter range in combination with the fluctuation impact, and generate a trading decision, and then transmit the trading decision to the trading decision output unit, and the processing method here is the same as the processing method of step S103 in the first embodiment.

[0036] The transaction decision output unit is used to periodically monitor the acquired transaction decision information, and to generate judgment information based on the change of the corresponding demand within the monitoring period, and to display the judgment information to the corresponding operator.

[0037] Some of the data in the above formulas are dimensionless and numerically calculated. Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0038] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A flexible power trading decision-making method, characterized in that: The following steps are involved: S101: Acquire historical power transaction data, analyze price fluctuation trends and seasonal changes based on the historical power transaction data, and generate transaction rule information; S102: establishing a power generation prediction model according to the transaction rule information and predicting the short-term power generation, and comparing the short-term power generation with the user demand based on the historical power transaction data to generate a comparison result, wherein the comparison result includes: a large supply signal and a large demand signal; S103: Analyze the obtained comparison results, determine the corresponding trading parameter range according to the fluctuation impact in different situations, generate trading parameter range information, and then generate a trading decision according to the trading parameter range information; S014: Perform periodic monitoring based on the generated transaction decision, make judgments based on changes in demand during the monitoring period, and generate judgment information, which includes normal monitoring information and interval adjustment information.

2. A flexible power trading decision-making method according to claim 1, characterized in that: The specific method of generating the transaction rule information in step S101 is: Obtaining power transaction data before time t, and analyzing the corresponding price fluctuation trend in the power transaction data, the analysis methods include: correlation analysis and cluster analysis; The method of correlation analysis is to calculate the correlation between prices and other factors at different time points; The cluster analysis method is as follows: for a large amount of time node data, cluster analysis method is used to cluster similar time nodes into one category; According to the results of correlation analysis and cluster analysis, transaction pattern information is generated by integrating correlation analysis and cluster analysis.

3. A flexible power trading decision-making method according to claim 1, characterized in that: The specific method of establishing the power generation prediction model according to the transaction rule information in step S102 is: The first step is to collect historical power generation data, and the time span covers the impact of different seasons, weather conditions and economic activity levels on power generation; The second step is to check the completeness and accuracy of the data and deal with missing values ​​and outliers; The third step is to test the stability of the power generation time series. If the power generation time series is not stable, differential processing is performed to obtain a stable power generation time series. The fourth step is to calculate the autocorrelation function and partial autocorrelation function of the stable power generation time series, observe its decay mode, and preliminarily determine the order range of autoregression and moving average according to the characteristics of ACF and PACF; The fifth step is to estimate the parameters of the ARMA model using the maximum likelihood estimation method, and use the mean absolute error, root mean square error and mean absolute percentage error to evaluate the prediction effect of the model, and establish a power generation prediction model; The corresponding power generation forecast is carried out according to the established power generation forecast model, the historical power transaction data within the forecast time t1 is obtained, and the historical power transaction data is substituted into the power generation forecast model to obtain the short-term power generation information.

4. A flexible power trading decision-making method according to claim 1, characterized in that: The specific method of generating the comparison result in step S012 is: Obtain the historical electricity trading data within the prediction time t1, and at the same time obtain the corresponding electricity demand, and compare the electricity demand with the short-term power generation. If the short-term power generation is greater than the electricity demand, a large supply signal is generated; if the short-term power generation is less than the electricity demand, a large demand signal is generated.

5. A flexible power trading decision-making method according to claim 1, characterized in that: The specific method of analyzing the comparison result in step S013 is: Obtaining a comparison result, wherein the comparison result includes a large supply signal and a large demand signal, and taking the large supply signal as an example for analysis, obtaining a prediction time t1, and obtaining historical power trading data corresponding to the prediction time t1, and obtaining corresponding trading parameters according to the historical power trading data, and integrating the trading parameters to obtain a historical trading parameter range; Then, taking the analysis situation corresponding to the large supply signal as the standard, the data corresponding to the same situation in the historical power trading data is obtained, and then the trading parameters corresponding to the obtained data of the same situation are integrated to obtain the secondary integration interval. At the same time, the secondary integration interval is adjusted based on the impact of price fluctuations within the predicted time t1.

6. A flexible power trading decision-making method according to claim 5, characterized in that: The specific method of adjusting the secondary integration interval in step S103 is: Obtain the energy transaction change and energy demand, and calculate the corresponding total energy consumption change cost, then obtain the total cost corresponding to the same energy demand in the historical power transaction data, and calculate the ratio of the total energy consumption change cost to the total cost as the energy fluctuation ratio; Then, the energy fluctuation ratio is multiplied by the secondary integration interval to obtain the product value interval, and the secondary integration interval and the product value interval are summed to obtain the trading parameter interval, and a trading decision is generated according to the trading parameter interval.

7. A flexible power trading decision-making method according to claim 1, characterized in that: The specific method of generating the judgment information in step S104 is: Conduct periodic monitoring of trading decisions, obtain trading demand within time T, and obtain changes in trading demand. If the trading demand increases, generate normal monitoring information; if the trading demand decreases, generate interval adjustment information.

8. A flexible power trading decision system, used to execute a flexible power trading decision method according to any one of claims 1 to 7, characterized in that: include: Transaction data collection unit, transaction data analysis unit, transaction decision generation unit and transaction decision output unit; A transaction data collection unit, which is used to transmit the acquired power transaction data to the transaction data analysis unit; A transaction data analysis unit, which is used to analyze the fluctuation trend and seasonal changes of the acquired power transaction data and generate transaction law information, and at the same time establish a power generation prediction model and predict short-term power generation based on the transaction law information, and then analyze the relationship between short-term power generation and user demand based on historical power transaction data to generate a comparison result, wherein the comparison result includes a large supply signal and a large demand signal, and transmit the generated comparison result to the transaction decision generation unit; A trading decision generating unit, which is used to analyze the obtained comparison results, determine the trading parameter interval by calculating the fluctuation influence under different situations, integrate the trading parameter interval based on the same situation corresponding to the historical power trading data to obtain a secondary integration interval, and generate the trading parameter interval in combination with the fluctuation influence, and generate a trading decision, and then transmit the trading decision to the trading decision output unit; The transaction decision output unit is used to periodically monitor the acquired transaction decision information, and to generate judgment information based on the change of the corresponding demand within the monitoring period, and to display the judgment information to the corresponding operator.

Images