Multi-market linkage carbon transaction strategy analysis method

Abstract

The invention discloses a multi-market linkage carbon trading strategy analysis method, relates to the technical field of market analysis, and solves the technical problems that a carbon emission trading market is affected by different markets, analysis and prediction are performed on a single carbon emission trading market, an analysis result is not accurate enough, and carbon trading of an enterprise is not facilitated. The trend of the electricity market is analyzed according to the electricity influence data; analyzing the trend of the energy market according to the energy market data; analyzing the electricity price based on the trend of the electricity market and the trend of the energy market to obtain an electricity price fluctuation curve; predicting the carbon price in the carbon transaction according to the electricity price fluctuation curve to obtain a predicted carbon price; predicting a carbon transaction quota according to the historical carbon transaction data to obtain a predicted quota; adjusting the predicted carbon price based on the carbon transaction quota, and analyzing a transaction strategy; performing carbon transaction according to the transaction strategy; and the accuracy of carbon transaction analysis can be improved.

Description

Technical Field

[0001] The present invention belongs to the field of market analysis, and relates to carbon trading strategy analysis technology, and specifically is a multi-market linked carbon trading strategy analysis method. Background Art

[0002] Carbon trading refers to the trading of carbon emission rights, which treats carbon dioxide emission rights as a commodity. The buyer obtains a certain amount of carbon dioxide emission rights by paying a certain amount to the seller. Carbon trading, as an environmental protection policy aimed at reducing greenhouse gas emissions, can force companies to adopt more environmentally friendly technologies and measures to reduce carbon emissions by establishing carbon emission quotas and carbon trading markets, thereby achieving the goal of energy conservation and emission reduction. Strategic analysis helps companies better understand the operating mechanism of the carbon trading market and formulate reasonable emission reduction plans, thereby achieving economic benefits while protecting the environment. The establishment of the carbon trading market has made carbon emission rights a scarce resource, and its price is determined by market supply and demand. Through strategic analysis, companies can grasp market dynamics more accurately and reasonably arrange the purchase and sale of carbon emission rights according to their own circumstances, thereby achieving optimal allocation of resources, reducing corporate operating costs and improving resource utilization in society as a whole.

[0003] The prior art (invention patent application with publication number CN115330081A) discloses a carbon trading market intelligent analysis and prediction method and system based on big data, the method includes: step S101, collecting multi-dimensional enterprise carbon emission data of multiple emission-controlled enterprises, obtaining the actual emission reduction value of each emission-controlled enterprise based on a preset emission reduction reference value calculation formula, and obtaining a basic big data set; S102, based on the obtained basic big data set, conducting a detailed analysis of the market price trend of the emission-controlled enterprises to obtain an analysis result set, the analysis result set is the regional support degree of the emission-controlled enterprises; S103, based on the analysis result set and the preset carbon trading market expected condition set, the preset carbon trading market expected condition set is the comprehensive regional support degree QZ, judging whether the analysis result set meets the expected condition set; S104, if the analysis result set meets the expected condition result, then outputting the analysis result and predicting the carbon emission trading market.

[0004] The existing technology analyzes the multi-dimensional carbon emission data and actual emission reduction values ​​of comprehensive enterprises to obtain a set of analysis results, and analyzes the set of analysis results with the expected condition results to obtain a prediction of the carbon emission trading market; however, the carbon emission trading market is affected by different markets. Analyzing and predicting a single carbon emission trading market will result in inaccurate analysis results, which is not conducive to carbon trading for enterprises.

[0005] The present invention provides a multi-market linked carbon trading strategy analysis method to solve the above technical problems. Summary of the invention

[0006] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a multi-market linked carbon trading strategy analysis method, which is used to solve the technical problem that the prior art analyzes the multi-dimensional carbon emission data and actual emission reduction values ​​of a comprehensive enterprise to obtain an analysis result set, and analyzes the analysis result set with the expected condition results to obtain a prediction of the carbon emission trading market; however, the carbon emission trading market is affected by different markets, and analyzing and predicting a single carbon emission trading market will result in inaccurate analysis results, which is not conducive to the company's carbon trading.

[0007] To achieve the above objectives, the first aspect of the present invention provides a multi-market linkage carbon trading strategy analysis method, comprising:

[0008] Step S1: Obtain power impact data, and analyze the trend of the power market based on the power impact data;

[0009] Step S2: Obtain energy market data, and analyze the trend of the energy market based on the energy market data; analyze the electricity price based on the trend of the electricity market and the trend of the energy market to obtain an electricity price fluctuation curve;

[0010] Step S3: predicting the carbon price in carbon trading according to the electricity price fluctuation curve to obtain a predicted carbon price; obtaining historical carbon trading data and carbon trading quotas; predicting the carbon trading quota according to the historical carbon trading data to obtain a predicted quota; and adjusting the predicted carbon price based on the carbon trading quota;

[0011] Step S4: Analyze the trading strategy based on the adjusted predicted carbon price and carbon trading quota; and conduct carbon trading based on the trading strategy.

[0012] Preferably, the method further includes: step S5: randomly extracting an encryption algorithm from an encryption algorithm library, and using the encryption algorithm to encrypt the carbon trading data; constructing a carbon trading blockchain, and storing the encrypted carbon trading data in a node corresponding to the blockchain.

[0013] The present invention randomly extracts an encryption algorithm from an encryption algorithm library to encrypt carbon trading data, which can prevent others from maliciously cracking the encryption algorithm and causing leakage of carbon trading data, and stores the encrypted carbon trading data in the blockchain; it can ensure the security and non-tamperability of carbon trading data, and verified users can view the stored data in the blockchain, which is conducive to improving the security and transparency of carbon trading data.

[0014] Preferably, the analyzing the trend of the power market according to the power impact data includes:

[0015] Retrieve power impact data, including historical electricity prices, environmental factors, number of electricity users, and fuel prices;

[0016] The wavelet transform is used to predict the electricity price in the power market, which includes the following steps:

[0017] Step S11: preprocessing the power impact data, completing the missing values ​​in the power impact data, and removing the abnormal values ​​in the power impact data;

[0018] Step S12: Integrate the power impact data into an original power sequence in chronological order; select a suitable wavelet basis function according to the original power sequence;

[0019] Step S13: Analyze the original power sequence using wavelet basis function to obtain a low-frequency subsequence and a high-frequency subsequence;

[0020] Step S14: constructing a low-frequency prediction model and a high-frequency prediction model respectively; inputting the low-frequency subsequence and the high-frequency subsequence into the low-frequency prediction model and the high-frequency prediction model respectively, and obtaining prediction values ​​of several subsequences;

[0021] Step S15: Recombining the predicted values ​​of several subsequences through inverse wavelet transform to restore them to the original time scale to obtain the electricity price prediction value.

[0022] It should be noted that common wavelet basis functions include Haar, Daubechies, Coiflets, etc.; different types of basis functions are suitable for different types of data features; it is also necessary to determine the appropriate decomposition level, which is usually determined based on the characteristics of the data and the prediction objectives.

[0023] The present invention predicts the electricity price in the electricity market by wavelet transform according to the power influencing data, comprehensively considers the historical data and the factors affecting the electricity market, predicts the electricity price by wavelet transform, and obtains the predicted electricity price value; it can better understand the data structure and remove unnecessary interference, which is conducive to improving the accuracy of the prediction results.

[0024] Preferably, analyzing the trend of the energy market based on energy market data includes:

[0025] Retrieving energy market data; dividing the energy market data into certain periods to obtain several periods of energy market data; wherein the energy market data includes: historical energy consumption and historical power generation;

[0026] Calculate the average value of energy market data for the previous n periods and mark the average value as the initial value; set appropriate smoothing coefficient α and trend component smoothing coefficient β; calculate the first exponential smoothing value and the second exponential smoothing value through the formula;

[0027] The difference between the first exponential smoothing value and the second exponential smoothing value is calculated to obtain the trend component; a prediction model is constructed based on the trend component; the prediction model is used to analyze the trend of the energy market to obtain the energy consumption forecast and the power generation forecast.

[0028] It should be noted that the trend of the energy market is analyzed, and the energy consumption forecast and power generation forecast are analyzed separately, and both types of data are predicted using exponential smoothing.

[0029] The present invention uses the exponential smoothing method to predict the trend of energy market data, and predicts the power generation and energy consumption respectively. The prediction method of the exponential smoothing method is real-time. When new observation values ​​appear, the new observation values ​​can be used to predict the data, which is conducive to timely updating of the prediction results.

[0030] Preferably, the analysis of electricity prices based on the trends of the electricity market and the energy market to obtain an electricity price fluctuation curve includes:

[0031] Retrieving electricity price forecast values, energy consumption forecast amounts and power generation forecast amounts; integrating electricity price forecast values, energy consumption forecast amounts and power generation forecast amounts into an electricity price adjustment sequence;

[0032] Calling the electricity price prediction model; inputting the electricity price adjustment sequence into the electricity price prediction model to obtain the corresponding predicted electricity price; fitting the predicted electricity price within the set time period into an electricity price fluctuation curve; wherein, the electricity price prediction model is constructed based on the artificial intelligence model.

[0033] The present invention integrates the electricity price forecast value, energy consumption forecast and power generation forecast into an electricity price adjustment sequence, and uses an electricity price forecast model to analyze the forecasted electricity price; after comprehensively considering the factors of the energy market and the electricity market, the forecasted electricity price is obtained; the market can be linked, which is conducive to improving the accuracy of the forecast results.

[0034] Preferably, the electricity price prediction model is constructed based on an artificial intelligence model, including:

[0035] Acquire a standard data set; wherein the standard data set includes a standard input sequence consistent with the content attributes of the electricity price adjustment sequence, and a standard output sequence consistent with the content attributes of the predicted electricity price;

[0036] Divide the standard data set into a training set, a validation set, and a test set; use the training set to train the artificial intelligence model; use the validation set to adjust the internal parameters of the artificial intelligence model; use the test set to test the trained artificial intelligence model to obtain the test indicators of the artificial intelligence model;

[0037] Obtain the indicator threshold of the artificial intelligence model, and determine whether the test indicators of the artificial intelligence model are all greater than the indicator threshold; if yes, mark the artificial intelligence model as an electricity price prediction model; if not, retrain the artificial intelligence model.

[0038] It should be noted that the test indicators include accuracy, recall, F1 score and stability; the indicator thresholds are set by experts based on actual conditions.

[0039] Preferably, the predicting of the carbon price in carbon trading according to the electricity price fluctuation curve to obtain the predicted carbon price includes:

[0040] Obtain historical carbon trading data; divide the historical carbon trading data by time to obtain a number of periods i; wherein the carbon trading data includes historical electricity prices and historical carbon prices; i = 1, 2, ..., n, where n is a positive integer;

[0041] By formula Calculate the correlation coefficient r between electricity price and carbon price; extract the predicted electricity price for the forecast period from the electricity price fluctuation curve, and multiply the predicted electricity price by the correlation coefficient r to obtain the predicted carbon price; where xi represents the historical electricity price of the i-th period, yi represents the historical carbon price of the i-th period; x represents the average value of the historical electricity price, and y represents the average value of the historical carbon price.

[0042] The present invention calculates the correlation coefficient between electricity price and carbon price based on historical carbon trading data, extracts the predicted electricity price for the period requiring prediction from the electricity price fluctuation curve, and calculates the predicted carbon price based on the predicted electricity price. It can preliminarily predict the carbon price for different periods, laying the foundation for the subsequent adjustment of the carbon price.

[0043] Preferably, the step of predicting carbon trading quotas based on historical carbon trading data to obtain predicted quotas includes:

[0044] Retrieve historical carbon trading data; divide the historical trading data according to set periods to obtain carbon trading data for several periods; add up the carbon trading data for several periods to obtain carbon trading quotas;

[0045] The carbon trading quota is fitted into a quota curve according to several periods, and the carbon trading quota in the forecast period is forecasted according to the quota curve to obtain the forecast quota.

[0046] Preferably, the adjusting the predicted carbon price based on the carbon trading quota includes:

[0047] Retrieve carbon trading quotas and forecast quotas, and calculate the difference between forecast quotas and carbon trading quotas;

[0048] Determine whether the difference is a positive number; if so, keep the original predicted carbon price; if not, multiply the difference by the adjustment coefficient and add it to the original predicted carbon price to obtain a new predicted carbon price.

[0049] The present invention adjusts the predicted carbon price according to the carbon trading quota and the predicted quota, can adjust the predicted carbon price according to the actual situation of the carbon trading market, can make the predicted carbon price more in line with reality, and is conducive to improving the accuracy of the predicted carbon price.

[0050] Preferably, analyzing the trading strategy based on the adjusted predicted carbon price and carbon trading quota includes:

[0051] Retrieve the predicted carbon price and carbon trading quota; when the predicted carbon price is lower than the price threshold, generate a purchase signal to notify the power generation company's managers to purchase carbon emission rights;

[0052] When the carbon trading quota is lower than the quota threshold and the carbon price is higher than the set threshold, a selling signal is generated to notify managers of other companies to sell carbon emission rights.

[0053] It should be noted that other enterprises refer to those that have a portion of the carbon emission rights quota, but when the carbon emission rights quota is greater than the quota required by the enterprise, the carbon emission rights quota can be sold.

[0054] The present invention generates trading strategies for different enterprises according to the predicted carbon price and carbon trading quota, which can ensure that the enterprises obtain greater profits in the carbon trading process.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] 1. The present invention predicts the electricity price in the electricity market by wavelet transform according to the power influencing data, comprehensively considers the historical data and the factors affecting the electricity market, predicts the electricity price by wavelet transform, and obtains the electricity price prediction value; can better understand the data structure and remove unnecessary interference, which is beneficial to improve the accuracy of the prediction result; uses the exponential smoothing method to predict the trend of energy market data, and predicts the power generation and energy consumption respectively. The prediction method of the exponential smoothing method is real-time. When new observation values ​​appear, the new observation values ​​can be used to predict the data, which is beneficial to timely update the prediction results; the electricity price prediction value, the energy consumption prediction amount and the power generation prediction amount are integrated into the electricity price adjustment sequence, and the predicted electricity price is analyzed by the electricity price prediction model; after comprehensively considering the factors of the energy market and the electricity market, the predicted electricity price is obtained; the market can be linked, which is beneficial to improve the accuracy of the prediction result.

[0057] 2. The present invention calculates the correlation coefficient between electricity price and carbon price based on historical carbon trading data, extracts the predicted electricity price for the period requiring prediction from the electricity price fluctuation curve, and calculates the predicted carbon price based on the predicted electricity price; can make preliminary predictions on carbon prices in different periods, laying a foundation for subsequent adjustment of carbon prices; can adjust the predicted carbon price based on carbon trading quotas and predicted quotas, can adjust the predicted carbon price based on the actual situation of the carbon trading market, can make the predicted carbon price more in line with reality, and is conducive to improving the accuracy of the predicted carbon price; generates trading strategies for different enterprises based on the predicted carbon price and carbon trading quota, can ensure that enterprises obtain greater profits in the process of carbon trading; randomly extracts encryption algorithms from the encryption algorithm library to encrypt carbon trading data, can prevent others from maliciously cracking the encryption algorithm and causing leakage of carbon trading data, and stores the encrypted carbon trading data in the blockchain; can ensure the security and non-tamperability of carbon trading data, and verified users can view the stored data in the blockchain, which is conducive to improving the security and transparency of carbon trading data. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0059] Figure 1 It is a schematic diagram of the overall steps of the present invention;

[0060] Figure 2 It is a schematic diagram of the steps of predicting electricity prices in the electricity market according to the present invention;

[0061] Figure 3 A schematic diagram of the steps of analyzing the energy market and adjusting the predicted electricity price according to the present invention;

[0062] Figure 4 It is a schematic diagram of the carbon price prediction steps of the carbon trading market of the present invention. DETAILED DESCRIPTION

[0063] The technical scheme of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0064] See also Figure 1 The first embodiment of the present invention provides a multi-market linkage carbon trading strategy analysis method, including:

[0065] Step S1: Obtain power impact data, and analyze the trend of the power market based on the power impact data;

[0066] Step S2: Obtain energy market data, and analyze the trend of the energy market based on the energy market data; analyze the electricity price based on the trend of the electricity market and the trend of the energy market to obtain an electricity price fluctuation curve;

[0067] Step S3: predicting the carbon price in carbon trading according to the electricity price fluctuation curve to obtain a predicted carbon price; obtaining historical carbon trading data and carbon trading quotas; predicting the carbon trading quota according to the historical carbon trading data to obtain a predicted quota; and adjusting the predicted carbon price based on the carbon trading quota;

[0068] Step S4: analyzing the trading strategy according to the adjusted predicted carbon price and carbon trading quota; and conducting carbon trading according to the trading strategy;

[0069] Step S5: randomly extract an encryption algorithm from the encryption algorithm library, and use the encryption algorithm to encrypt the carbon trading data; build a carbon trading blockchain, and store the encrypted carbon trading data in the corresponding node of the blockchain.

[0070] See also Figure 2-Figure 3 , obtain power impact data, where the power impact data includes historical electricity prices, environmental factors, number of electricity users, and fuel prices;

[0071] The wavelet transform is used to predict the electricity price in the power market, which includes the following steps:

[0072] Step S11: preprocessing the power impact data, completing the missing values ​​in the power impact data, and removing the abnormal values ​​in the power impact data;

[0073] Step S12: Integrate the power impact data into an original power sequence in chronological order; select a suitable wavelet basis function according to the original power sequence;

[0074] Step S13: Analyze the original power sequence using wavelet basis function to obtain a low-frequency subsequence and a high-frequency subsequence;

[0075] Step S14: constructing a low-frequency prediction model and a high-frequency prediction model respectively; inputting the low-frequency subsequence and the high-frequency subsequence into the low-frequency prediction model and the high-frequency prediction model respectively, and obtaining prediction values ​​of several subsequences;

[0076] Step S15: Recombining the predicted values ​​of several subsequences through inverse wavelet transform to restore them to the original time scale to obtain the electricity price prediction value.

[0077] It should be noted that the low-frequency prediction model adopts traditional statistical methods, such as regression models, etc.; the high-frequency prediction model adopts machine learning methods, such as support vector machines, long short-term memory networks, etc.

[0078] Obtain energy market data, divide the energy market data into certain periods, and obtain several periods of energy market data; wherein the energy market data includes: historical energy consumption and historical power generation;

[0079] Calculate the average value of energy market data for the previous n periods and mark the average value as the initial value; set appropriate smoothing coefficient α and trend component smoothing coefficient β; calculate the first exponential smoothing value and the second exponential smoothing value through the formula;

[0080] The difference between the first exponential smoothing value and the second exponential smoothing value is calculated to obtain the trend component; a prediction model is constructed based on the trend component; the prediction model is used to analyze the trend of the energy market to obtain the energy consumption forecast and the power generation forecast.

[0081] It is worth noting that the calculation formula of the first exponential smoothing value is St = αyt + (1-α) St-1; where yt represents the actual observation value of the tth period; St-1 is the first exponential smoothing value of the previous period;

[0082] The calculation formula of the quadratic exponential smoothing value is St(2)=αSt+(1-α)St-1(2); where St-1(2) is the quadratic exponential smoothing value of the previous period;

[0083] The trend component bt = [α / (1-α)][St-St(2)], and the forecast for period m is Ft+m = at+mbt; where at = 2St-St(2);

[0084] The value of the trend component needs to be updated regularly, bt = β[St(2)-St-1(2)]+(1-β)bt-1; the larger the β value, the faster the model responds to trend changes in the data and can adapt to new trend directions more quickly; conversely, a smaller β value will make the model more conservative and tend to keep the existing trend estimate unchanged unless there is enough new evidence to indicate that the trend has changed.

[0085] Retrieving electricity price forecast values, energy consumption forecast amounts and power generation forecast amounts; integrating electricity price forecast values, energy consumption forecast amounts and power generation forecast amounts into an electricity price adjustment sequence;

[0086] Calling the electricity price prediction model; inputting the electricity price adjustment sequence into the electricity price prediction model to obtain the corresponding predicted electricity price; fitting the predicted electricity price within the set time period into an electricity price fluctuation curve; wherein, the electricity price prediction model is constructed based on the artificial intelligence model.

[0087] For example: suppose that we analyze the carbon trading data for the first quarter of next year, obtain the electricity impact data for the past three years, process the electricity impact data using wavelet transform, and obtain the electricity price forecast value; obtain the energy market data, divide the energy market data by quarter, and use exponential smoothing to predict the energy consumption and power generation for each month in the first quarter of next year to obtain the energy consumption forecast and power generation forecast; integrate the electricity price forecast, energy consumption forecast, power generation and measurement into an electricity price adjustment sequence; input the electricity price adjustment sequence into the electricity price prediction model to obtain the predicted electricity price for each month in the first quarter of next year; fit the predicted electricity price for each month to obtain the electricity price fluctuation curve for the first quarter of next year.

[0088] It is worth noting that the electricity price prediction model is built based on an artificial intelligence model, including:

[0089] Acquire a standard data set; wherein the standard data set includes a standard input sequence consistent with the content attributes of the electricity price adjustment sequence, and a standard output sequence consistent with the content attributes of the predicted electricity price;

[0090] Divide the standard data set into a training set, a validation set, and a test set; use the training set to train the artificial intelligence model; use the validation set to adjust the internal parameters of the artificial intelligence model; use the test set to test the trained artificial intelligence model to obtain the test indicators of the artificial intelligence model;

[0091] Obtain the indicator threshold of the artificial intelligence model, and determine whether the test indicators of the artificial intelligence model are all greater than the indicator threshold; if yes, mark the artificial intelligence model as an electricity price prediction model; if not, retrain the artificial intelligence model.

[0092] See also Figure 4 , obtain historical carbon trading data; divide the historical carbon trading data according to time to obtain a number of periods i; wherein the carbon trading data includes historical electricity prices and historical carbon prices; i = 1, 2, ..., n, where n is a positive integer;

[0093] By formula Calculate the correlation coefficient r between electricity price and carbon price; extract the predicted electricity price for the forecast period from the electricity price fluctuation curve, and multiply the predicted electricity price by the correlation coefficient r to obtain the predicted carbon price; where xi represents the historical electricity price of the i-th period, yi represents the historical carbon price of the i-th period; x represents the average value of the historical electricity price, and y represents the average value of the historical carbon price.

[0094] It is worth noting that when the correlation between electricity price and carbon price is not a linear relationship, the Spearman rank correlation coefficient can be used to calculate the relationship between electricity price and carbon price; the specific steps are as follows:

[0095] Sort electricity prices and carbon prices in ascending order; assign a rank to each observation, with the smallest observation being assigned rank 1 and the largest observation being assigned the highest rank; calculate the rank difference d between each pair of observations i =R(x i )-R(y i ); find the sum of squares of all differences Calculate the correlation coefficient according to the formula n is the number of observations; the value range of rs is -1 to 1. When rs is greater than 0, it means that the two variables are positively correlated; when rs is less than 0, it is negatively correlated. The closer rs is to ±1, the stronger the monotonic correlation between the two variables.

[0096] Obtain historical carbon trading data; divide the historical trading data according to set periods to obtain carbon trading data for several periods; add up the carbon trading data for several periods to obtain carbon trading quotas;

[0097] Fitting carbon trading quotas into quota curves according to several periods, predicting the carbon trading quotas in the forecast period according to the quota curves, and obtaining forecast quotas;

[0098] Obtain carbon trading quotas and forecast quotas, and calculate the difference between forecast quotas and carbon trading quotas;

[0099] Determine whether the difference is a positive number; if so, keep the original predicted carbon price; if not, multiply the difference by the adjustment coefficient and add it to the original predicted carbon price to obtain a new predicted carbon price.

[0100] It should be noted that the adjustment formula for the predicted carbon price is YT=ST+θ×PC; among them, ST is the original predicted carbon price, PC is the quota difference, θ is the adjustment coefficient, and the adjustment coefficient is set according to actual experience.

[0101] For example: based on the correlation coefficient between the calculated electricity price and the carbon price of historical carbon trading, the predicted electricity price for the forecast period is extracted from the electricity price fluctuation curve, and the predicted electricity price is multiplied by the correlation coefficient r to obtain a predicted carbon price of 90 yuan / ton; according to the quota curve, the predicted quota for the first quarter of next year is 800 tons, and the quota for the first quarter of next year is 700 tons, so the calculated difference is 100; then the adjustment coefficient is set to 0.1, and the adjusted predicted carbon price is calculated to be 100 yuan / ton according to the formula.

[0102] Retrieve the predicted carbon price and carbon trading quota; when the predicted carbon price is lower than the price threshold, generate a purchase signal to notify the power generation company's managers to purchase carbon emission rights;

[0103] When the carbon trading quota is lower than the quota threshold and the carbon price is higher than the set threshold, a selling signal is generated to notify managers of other companies to sell carbon emission rights.

[0104] It should be noted that the price threshold and setting threshold are set based on actual experience.

[0105] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0106] Working principle of the present invention: The present invention obtains power impact data, and analyzes the trend of the power market according to the power impact data; obtains energy market data, and analyzes the trend of the energy market according to the energy market data; analyzes the power price based on the trend of the power market and the trend of the energy market to obtain the power price fluctuation curve; predicts the carbon price in carbon trading according to the power price fluctuation curve to obtain the predicted carbon price;

[0107] Obtain historical carbon trading data and carbon trading quotas; predict carbon trading quotas based on historical carbon trading data to obtain predicted quotas; adjust predicted carbon prices based on carbon trading quotas; analyze trading strategies based on adjusted predicted carbon prices and carbon trading quotas; conduct carbon trading based on trading strategies; randomly extract encryption algorithms from the encryption algorithm library, and use the encryption algorithms to encrypt carbon trading data; build a carbon trading blockchain, and store the encrypted carbon trading data in the corresponding nodes of the blockchain.

[0108] 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 multi-market linkage carbon trading strategy analysis method, characterized in that: include: Step S1: Obtain power impact data, and analyze the trend of the power market based on the power impact data; Step S2: Obtain energy market data, and analyze the trend of the energy market based on the energy market data; analyze the electricity price based on the trend of the electricity market and the trend of the energy market to obtain an electricity price fluctuation curve; Step S3: predicting the carbon price in carbon trading according to the electricity price fluctuation curve to obtain a predicted carbon price; obtaining historical carbon trading data and carbon trading quotas; predicting the carbon trading quotas according to the historical carbon trading data to obtain a predicted quota; Adjust the forecast carbon price based on carbon trading quotas; Step S4: Analyze the trading strategy based on the adjusted predicted carbon price and carbon trading quota; and conduct carbon trading based on the trading strategy.

2. The multi-market linkage carbon trading strategy analysis method according to claim 1 is characterized in that: The method also includes: step S5: randomly extracting an encryption algorithm from an encryption algorithm library, and using the encryption algorithm to encrypt the carbon trading data; constructing a carbon trading blockchain, and storing the encrypted carbon trading data in a node corresponding to the blockchain.

3. The multi-market linkage carbon trading strategy analysis method according to claim 1 is characterized in that: The analysis of the trend of the electricity market based on the electricity impact data includes: Retrieve power impact data, including historical electricity prices, environmental factors, number of electricity users, and fuel prices; The wavelet transform is used to predict the electricity price in the power market, which includes the following steps: Step S11: preprocessing the power impact data, completing the missing values ​​in the power impact data, and removing the abnormal values ​​in the power impact data; Step S12: Integrate the power impact data into an original power sequence in chronological order; select a suitable wavelet basis function according to the original power sequence; Step S13: Analyze the original power sequence using wavelet basis function to obtain a low-frequency subsequence and a high-frequency subsequence; Step S14: constructing a low-frequency prediction model and a high-frequency prediction model respectively; inputting the low-frequency subsequence and the high-frequency subsequence into the low-frequency prediction model and the high-frequency prediction model respectively, and obtaining prediction values ​​of several subsequences; Step S15: Recombining the predicted values ​​of several subsequences through inverse wavelet transform to restore them to the original time scale to obtain the electricity price prediction value.

4. The multi-market linkage carbon trading strategy analysis method according to claim 1 is characterized in that: The analysis of energy market trends based on energy market data includes: Retrieving energy market data; dividing the energy market data into certain periods to obtain several periods of energy market data; wherein the energy market data includes: historical energy consumption and historical power generation; Calculate the average value of energy market data for the previous n periods and mark the average value as the initial value; set appropriate smoothing coefficient α and trend component smoothing coefficient β; calculate the first exponential smoothing value and the second exponential smoothing value through the formula; The difference between the first exponential smoothing value and the second exponential smoothing value is calculated to obtain the trend component; a prediction model is constructed based on the trend component; the prediction model is used to analyze the trend of the energy market to obtain the energy consumption forecast and the power generation forecast.

5. The multi-market linkage carbon trading strategy analysis method according to claim 1 is characterized in that: The electricity price is analyzed based on the trend of the electricity market and the trend of the energy market to obtain the electricity price fluctuation curve, including: Retrieving electricity price forecast values, energy consumption forecast amounts and power generation forecast amounts; integrating electricity price forecast values, energy consumption forecast amounts and power generation forecast amounts into an electricity price adjustment sequence; Calling the electricity price prediction model; inputting the electricity price adjustment sequence into the electricity price prediction model to obtain the corresponding predicted electricity price; fitting the predicted electricity price within the set time period into an electricity price fluctuation curve; wherein, the electricity price prediction model is constructed based on the artificial intelligence model.

6. A multi-market linkage carbon trading strategy analysis method according to claim 5, characterized in that: The electricity price prediction model is constructed based on an artificial intelligence model and includes: Acquire a standard data set; wherein the standard data set includes a standard input sequence consistent with the content attributes of the electricity price adjustment sequence, and a standard output sequence consistent with the content attributes of the predicted electricity price; Divide the standard data set into a training set, a validation set, and a test set; use the training set to train the artificial intelligence model; use the validation set to adjust the internal parameters of the artificial intelligence model; use the test set to test the trained artificial intelligence model to obtain the test indicators of the artificial intelligence model; Obtain the indicator threshold of the artificial intelligence model, and determine whether the test indicators of the artificial intelligence model are all greater than the indicator threshold; if yes, mark the artificial intelligence model as an electricity price prediction model; if not, retrain the artificial intelligence model.

7. The multi-market linkage carbon trading strategy analysis method according to claim 1 is characterized in that: The method of predicting the carbon price in carbon trading according to the electricity price fluctuation curve to obtain the predicted carbon price includes: Obtain historical carbon trading data; divide the historical carbon trading data by time to obtain a number of periods i; wherein the carbon trading data includes historical electricity prices and historical carbon prices; i = 1, 2, ..., n, where n is a positive integer; By formula Calculate the correlation coefficient r between electricity price and carbon price; extract the predicted electricity price for the forecast period from the electricity price fluctuation curve, and multiply the predicted electricity price by the correlation coefficient r to obtain the predicted carbon price; where xi represents the historical electricity price of the i-th period, yi represents the historical carbon price of the i-th period; x represents the average value of the historical electricity price, and y represents the average value of the historical carbon price.

8. The multi-market linkage carbon trading strategy analysis method according to claim 1 is characterized in that: The method of predicting carbon trading quotas based on historical carbon trading data to obtain predicted quotas includes: Retrieve historical carbon trading data; divide the historical trading data according to set periods to obtain carbon trading data for several periods; add up the carbon trading data for several periods to obtain carbon trading quotas; The carbon trading quota is fitted into a quota curve according to several periods, and the carbon trading quota in the forecast period is forecasted according to the quota curve to obtain the forecast quota.

9. A multi-market linkage carbon trading strategy analysis method according to claim 1 or 8, characterized in that: The adjustment of the predicted carbon price based on the carbon trading quota includes: Retrieve carbon trading quotas and forecast quotas, and calculate the difference between forecast quotas and carbon trading quotas; Determine whether the difference is a positive number; if so, keep the original predicted carbon price; if not, multiply the difference by the adjustment coefficient and add it to the original predicted carbon price to obtain a new predicted carbon price.

10. The multi-market linkage carbon trading strategy analysis method according to claim 1, characterized in that: The analysis of trading strategies based on the adjusted predicted carbon price and carbon trading quotas includes: Retrieve the predicted carbon price and carbon trading quota; when the predicted carbon price is lower than the price threshold, generate a purchase signal to notify the power generation company's managers to purchase carbon emission rights; When the carbon trading quota is lower than the quota threshold and the carbon price is higher than the set threshold, a selling signal is generated to notify managers of other companies to sell carbon emission rights.

Images