Intelligent power transaction management platform integrating medium and long term spot green certificate

Abstract

The application discloses a medium and long term spot green certificate fusion intelligent power transaction management platform, relates to the technical field of power transaction management, and synchronously collects real-time data, adopts linear optimization to generate a dynamic reference vector, uses the vector to calculate a cross fluctuation coupling index and a certificate power synchronous deviation index, and generates a synergistic risk adjustment coefficient through a Gaussian-Laplacian hybrid risk model to guide risk management; assembles a hedging package based on a smart contract, atomizes and locks power and corresponding green certificates, and ensures compliance and traceability by means of a blockchain; real-time tracking of the profit and loss and margin state of the hedging package is performed, and an adaptive threshold is used to trigger additional or released margin; through a prediction model, settlement profit and loss data are used to iteratively correct the reference vector; and the application significantly improves price discovery, renewable energy integration and market stability, and provides double benefits of economy and environment.

Description

Technical Field

[0001] This invention relates to the field of power trading management technology, specifically to an intelligent power trading management platform that integrates medium- and long-term spot green certificates. Background Technology

[0002] In recent years, the power industry has undergone significant transformation, driven by the rapid integration of renewable energy sources such as wind and solar power. While these renewable energy sources are environmentally beneficial, they introduce intermittency and unpredictability to power supply, increasing the complexity of grid management. Simultaneously, the rise of electric vehicles, smart technologies, and industrial automation has led to constantly changing consumption patterns, requiring more flexible and responsive power trading mechanisms. Against this backdrop, medium- and long-term contracts, spot markets, and green certificates (green certificates) play roles in providing price stability, enabling real-time adjustments, and promoting sustainability, respectively. However, these trading components typically operate independently under different platforms and regulatory frameworks, limiting their overall effectiveness. This fragmentation raises a key technical issue: the lack of an integrated platform that can unify medium- and long-term contracts, spot markets, and green certificates. This deficiency is particularly pronounced in practical applications. For example, when wind power output fluctuates drastically due to sudden weather changes, spot market prices may decline rapidly, while medium- and long-term contract prices, based on static benchmarks, cannot adjust in a timely manner, leading to market participants facing the risks of price misalignment and misallocation of resources. At the same time, the independent trading of green certificates makes it difficult for companies to match electricity consumption and renewable energy certificates in real time, especially when they need to meet strict environmental compliance requirements. They may face penalties for violations due to untimely verification.

[0003] Furthermore, this problem is exacerbated during periods of surging demand (such as peak summer electricity consumption) or supply shortages (such as insufficient wind power). The lack of an integrated platform hinders the effective coordination of different market types, potentially leading to real-time supply and demand imbalances in the grid and threatening system stability. Without addressing this issue, market participants will face higher transaction costs and risks, diminishing incentives for investment in renewable energy.

[0004] To this end, the present invention provides an intelligent power trading management platform that integrates medium- and long-term spot green certificates. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] To address the shortcomings of existing technologies, this invention provides an intelligent power trading management platform integrating medium- and long-term spot green certificates. It synchronously collects real-time data and uses linear optimization to generate a dynamic reference vector, serving as a unified benchmark price sequence for cross-timescale trading. This vector is used to calculate a cross-volatility coupling index and a certificate-power synchronization deviation index. A Gaussian-Laplace hybrid risk model is then used to generate a collaborative risk adjustment coefficient to guide risk management. The platform assembles smart contract-based hedging packages, atomically locking power and its corresponding green certificates, and leverages blockchain to ensure compliance and traceability. It tracks the profit and loss of hedging packages and margin status in real time, using adaptive thresholds to trigger margin additions or releases. A predictive model iteratively corrects the reference vector using settlement profit and loss data. This significantly improves price discovery, renewable energy integration, and market stability, solving the technical problems described in the background section.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solution: a smart power trading management platform integrating medium- and long-term spot green certificates, including a data optimization module that collects unit operation data, meteorological forecast data and matching instructions, and generates a dynamic reference vector through linear optimization as a benchmark price sequence for cross-year, monthly, day-ahead and real-time trading;

[0009] The risk modeling module generates a cross-volatility coupling index and a securities-electronic synchronization deviation index based on the dynamic reference vector. These are input into a Gaussian-Laplace hybrid risk model, and the output is a collaborative risk adjustment coefficient.

[0010] The contract assembly module assembles a smart contract hedging package with electricity and green certificate atomic lock according to the collaborative risk adjustment coefficient, and ensures the immutability and traceability of the hedging package through blockchain records.

[0011] The real-time adjustment module tracks the profit and loss and margin status of the hedging package in real time, triggers the addition or release of margin based on the adaptive threshold, and writes the updated results into the risk pool.

[0012] The price correction module uses the settlement profit and loss trajectory, the cross-fluctuation coupling index, and the securities-electronic synchronization deviation index to iteratively correct the dynamic reference vector and write back the benchmark price to the next round of trading.

[0013] Furthermore, the data collected include unit operation data, meteorological forecast data, and matching instructions:

[0014] The unit's operating status is monitored in real time by sensors to obtain unit operating data; meteorological forecast data is obtained through meteorological stations or satellite data; matching instructions are obtained through the trading platform; and the unit operating data, meteorological forecast data and matching instructions are synchronized to a unified time-series event bus.

[0015] Furthermore, the step of generating the dynamic reference vector through linear optimization includes: taking the unit operation data, meteorological forecast data, and matching instructions as inputs, using a linear programming algorithm to solve the objective function to minimize price fluctuations and satisfy supply and demand balance constraints, thereby generating the dynamic reference vector.

[0016] Furthermore, the generation of the cross-fluctuation coupling index includes: performing rolling cross-correlation calculation on the spot node electricity price and the high-pass filtered renewable power prediction error, applying an exponential decay kernel, and generating the cross-fluctuation coupling index to reflect the coupling relationship between electricity price and power fluctuation.

[0017] Furthermore, the generation of the certificate-electricity synchronization deviation index includes: calculating the standardized difference between the green certificate equivalent price and the nodal electricity price, and then multiplying it by the local renewable energy penetration rate to generate the certificate-electricity synchronization deviation index, so as to quantify the synchronization deviation between green certificates and electricity prices.

[0018] Furthermore, the Gaussian-Laplace hybrid risk model uses the cross-volatility coupling index and the securities-electronic synchronization deviation index as inputs, employs a hybrid model of Gaussian and Laplace distributions to calculate the risk probability, and outputs the collaborative risk adjustment coefficient to adjust trading risk.

[0019] Furthermore, the smart contract hedging package for atomically locking electricity and green certificates includes: extracting electricity contracts and corresponding green certificates from the trading pool according to the collaborative risk adjustment coefficient, and atomically locking the electricity contracts and green certificates through a smart contract to form an indivisible hedging package.

[0020] Furthermore, ensuring the immutability and traceability of the hedging package through blockchain records includes: applying a hash function to the data of the hedging package to generate a unique identifier, recording the hedging package and its unique identifier in the distributed ledger of the blockchain, and reaching consensus among all network nodes through a consensus mechanism.

[0021] Furthermore, the adaptive threshold is dynamically adjusted based on market volatility and the profit and loss of the hedging package, including: calculating market volatility, setting a margin call threshold and a release threshold, wherein the margin call threshold is the minimum margin requirement plus the product of the risk coefficient, market volatility, and risk exposure, and the release threshold is the margin call threshold minus the adjustment parameter.

[0022] Furthermore, the iterative correction of the dynamic reference vector includes: using the gradient descent method, based on the correlation between the settlement profit and loss trajectory and the cross-volatility coupling index and the securities-electronic synchronization deviation index, calculating the adjustment amount, and updating the benchmark price of the dynamic reference vector to optimize the accuracy of the next round of trading.

[0023] (III) Beneficial Effects

[0024] This invention provides an intelligent power trading management platform that integrates medium- and long-term spot green certificates, which has the following beneficial effects:

[0025] The cross-volatility coupling index and the securities-electronics synchronization deviation index quantify market volatility and the risk of securities-electronics decoupling in real time. A Gaussian-Laplace hybrid risk model generates a collaborative risk adjustment coefficient, precisely triggering the generation of smart contract hedging packages and significantly improving the accuracy of price discovery. The predictive model uses profit and loss trajectories and risk indices to iteratively correct the dynamic reference vector, continuously optimizing the adaptability of the benchmark price and further enhancing market efficiency.

[0026] The smart contract hedging package uses atomic locking technology to achieve a unique mapping between electricity volume and green certificates, ensuring a clear and accurate correspondence between each unit of electricity volume and green certificate. A risk pool centrally stores risk data, supporting model training and validation, further improving the accuracy of accounting.

[0027] Dynamic reference vectors provide a stable price benchmark for renewable energy trading by smoothing price fluctuations. Cross-volatility coupling indices and synergistic risk adjustment coefficients help identify and address risks arising from fluctuations in renewable energy output, reducing trading uncertainty. An adaptive threshold mechanism dynamically manages margin requirements, optimizing capital occupation and alleviating cost pressures on market participants. Predictive models continuously correct benchmark prices, further enhancing price adaptability and significantly reducing risk exposure in renewable energy trading.

[0028] Smart contract hedging packages provide market participants with efficient risk hedging tools, reducing uncertainty in trading. An adaptive threshold mechanism dynamically adjusts margin requirements, ensuring flexibility in fund management. Risk pools and predictive models further enhance market stability by continuously optimizing trading strategies.

[0029] The dynamic reference vector provides a unified price benchmark for the entire trading system. The cross-volatility coupling index and the securities-electronic synchronization deviation index quantify risk in real time. The collaborative risk adjustment coefficient accurately triggers hedging strategies. The smart contract hedging package ensures trading compliance. The adaptive threshold optimizes capital utilization efficiency. The predictive model continuously improves price accuracy. Attached Figure Description

[0030] Figure 1 This is a schematic diagram of the intelligent power trading management platform of the present invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] Please see Figure 1 This invention provides an intelligent power trading management platform integrating medium- and long-term spot green certificates, comprising:

[0033] Step 1: Utilize high-frequency data acquisition technology to acquire and timestamp-aligned unit operation data, meteorological forecast data, and market matching instructions in real time, and integrate them into a time-series event bus to form a unified time-series data stream. Based on this time-series data stream, apply a linear optimization algorithm to generate a dynamic reference vector as a benchmark price series, satisfying supply and demand balance, price smoothing, and market rule constraints. The benchmark price series is directly applied to the pricing mechanisms of annual, monthly, day-ahead, and real-time transactions, unifying the pricing standards for different trading periods.

[0034] Step 101: Data Synchronization Acquisition

[0035] High-frequency data acquisition technology is used to obtain real-time unit operation data, meteorological forecast data, and market matching orders to ensure data timeliness and consistency. Unit operation data includes power generation and load levels, reflecting the actual operating status of the generator units; meteorological forecast data includes wind speed and solar irradiance, used to determine the power generation capacity of new energy sources; market matching orders include trading volume and trading price, used to reflect dynamic changes in market transactions. During the acquisition process, a timestamp is added to each set of data to ensure that all data is recorded on the same time reference. This time alignment method eliminates data deviations caused by differences in acquisition time, providing an accurate input basis for subsequent time series processing.

[0036] By using high-frequency data acquisition technology and timestamp alignment, the real-time performance and consistency of unit operation data, meteorological forecast data, and market matching instructions are ensured.

[0037] The collected unit operation data, meteorological forecast data, and market matching instructions are integrated into a time-series event bus, forming a unified time-series data stream. The time-series event bus employs a high-concurrency processing mechanism, arranging all data in chronological order and assigning a corresponding timestamp to each data entry. This mechanism ensures the continuity and temporal consistency of the data stream. By integrating multi-source data into a structured time-series data stream, the system can efficiently process heterogeneous data from different sources, providing reliable input support for subsequent optimization calculations and thus improving overall processing efficiency.

[0038] A high-concurrency processing mechanism is employed to integrate multi-source data into a time-series data stream, ensuring its temporal order and continuity. This improves data processing efficiency and structuring, enabling the system to quickly respond to input demands from multiple data sources.

[0039] Step 102: Generation of Dynamic Reference Vectors

[0040] Based on the time-series data stream in the time-series event bus, a linear optimization algorithm is applied to generate a dynamic reference vector as a benchmark price sequence. The linear optimization algorithm achieves this by balancing the relationship between power generation and market demand and reducing drastic price fluctuations. Specifically, at each time point, the difference between the total power generation of the generating units and market demand is calculated, aiming to make the difference as close to zero as possible to achieve supply and demand balance.

[0041] Meanwhile, by minimizing the magnitude of price changes between adjacent time points, a smooth transition in the price series is ensured. Furthermore, the price range is set as a constraint according to market rules to ensure that the generated benchmark price series remains within a reasonable range. This method generates a benchmark price series that reflects both market supply and demand dynamics and maintains price stability.

[0042] Using linear optimization algorithms to generate benchmark price sequences can simultaneously satisfy the requirements of supply and demand balance, price smoothing, and market rule constraints. The generated benchmark price sequences can reduce drastic price fluctuations and provide a stable and reasonable price reference for trading at different time scales.

[0043] Step 103, Application of the Benchmark Price Series

[0044] The generated dynamic reference vector serves as a benchmark price sequence, which is directly used in the pricing mechanisms for annual, monthly, day-ahead, and real-time transactions. For annual and monthly transactions, the benchmark price sequence is used to adjust the static quotes of long-term contracts; for day-ahead and real-time transactions, the benchmark price sequence is used to correct the nodal price deviations in the spot market.

[0045] By using a unified benchmark price series, trading types across different time scales are effectively linked, reducing price discrepancies between long-term and short-term trades. This consistent application supports trading entities operating within a unified price system, improving market trading efficiency and accuracy. Applying the benchmark price series to multiple trading types and unifying pricing standards across different trading periods can reduce cross-period price deviations.

[0046] Step one involves acquiring consistent, real-time input data through synchronous data collection. This data is then used to construct a structured time-series data stream via a time-series event bus. A benchmark price series is generated through a linear optimization algorithm, and finally, this series is applied to transaction pricing across multiple time scales. This tightly integrated process design ensures that every step, from data acquisition to price application, serves a unified transaction optimization goal.

[0047] Step 2: Extract the electricity price sequence of a certain node in the spot market and the power prediction error of renewable energy using dynamic reference vectors. Extract the high-frequency fluctuation component by processing the power prediction error through high-pass filtering. Calculate the cross-correlation coefficient between the electricity price of the spot node and the high-frequency fluctuation component within the rolling time window and apply an exponential decay kernel to generate a cross-fluctuation coupling index.

[0048] Extract the spot node electricity price and the equivalent price of green certificates, calculate the standardized difference between the two and multiply it by the local renewable energy penetration rate to generate the certificate-electricity synchronization deviation index; input the cross-volatility coupling index and the certificate-electricity synchronization deviation index into the Gaussian-Laplace mixed risk model, determine the distribution parameters by fitting historical data, calculate the cumulative distribution value of the two indices and generate a synergistic risk adjustment coefficient by weighted averaging, which serves as the input to the risk engine.

[0049] Step 201: Generate the cross-fluctuation coupling index

[0050] First, using the dynamic reference vector generated in step one, the electricity price sequence at a certain node in the spot market and the power prediction error of renewable energy are extracted. Next, the difference between the predicted power output and the actual output of renewable energy is calculated to obtain the power prediction error. Then, a high-pass filter is applied to the power prediction error to extract the high-frequency fluctuation component reflecting rapid changes in renewable energy output. Based on this, within a preset rolling time window, the cross-correlation coefficient between the spot node electricity price and the high-frequency fluctuation component is calculated to quantify the dynamic correlation between the two.

[0051] The cross-correlation coefficient is calculated by summing the products of the electricity price series and the high-frequency fluctuation components, and then dividing by the product of their respective fluctuation amplitudes. Subsequently, an exponential decay kernel is applied to the cross-correlation coefficient to give greater weight to recent data within the time window, ultimately generating the cross-fluctuation coupling index. The cross-fluctuation coupling index can reflect the real-time sensitivity of spot electricity prices to fluctuations in renewable energy output, serving as a quantitative indicator of market volatility characteristics for subsequent risk assessment.

[0052] The sensitivity of spot electricity prices to fluctuations in renewable energy output is a significant factor influencing market price volatility. It is necessary to quantify the dynamic correlation between the two in real time. The cross-volatility coupling index provides an accurate measure of market volatility characteristics, can provide a reliable market for risk assessment, supports accurate input for the Gauss-Laplace mixed risk model, and enhances the accuracy of risk assessment.

[0053] Step 202: Generate the certificate-electrical synchronization deviation index

[0054] Similarly, based on the dynamic reference vector generated in step one, the spot node electricity price and the equivalent price of green certificates are extracted.

[0055] Next, the difference between the green certificate price and the spot node electricity price is calculated, and then standardized by averaging the difference over a rolling time window to obtain a standardized difference, which reflects the degree of price deviation between the two. Then, the standardized difference is multiplied by the local renewable energy penetration rate, i.e., the proportion of renewable energy generation in total electricity generation, to generate the certificate-electricity synchronization deviation index. By incorporating the renewable energy penetration rate, the certificate-electricity synchronization deviation index correlates the risk of decoupling between certificate and electricity prices with the importance of renewable energy in the power system, serving as a measure of certificate-electricity synchronization for subsequent risk analysis.

[0056] The decoupling of green certificate prices from electricity prices directly affects the compliance and accounting accuracy of enterprises. This risk characteristic needs to be quantified. The certificate-electricity synchronization deviation index provides a key measure of certificate-electricity synchronization, which can provide necessary data support for the Gauss-Laplace mixed risk model and improve the comprehensiveness of risk analysis.

[0057] Step 203: Calculation of Gaussian-Laplace Mixed Risk Model

[0058] The generated cross-volatility coupling index and securities-electronic synchronization deviation index are input into a Gaussian-Laplace hybrid risk model for processing. This model assumes that the risk distribution is a combination of a Gaussian distribution and a Laplace distribution, where the Gaussian distribution is used to capture the characteristics of normal market fluctuations, and the Laplace distribution is used to capture the characteristics of extreme market events.

[0059] Distribution parameters were determined by fitting historical data, and the cumulative distribution values ​​of the cross-volatility coupling index and the securities-electronic synchronization deviation index were calculated separately. Subsequently, a weighted average of the cumulative distribution values ​​of the two was used to generate a synergistic risk adjustment coefficient, which serves as a measure of overall risk. The Gauss-Laplace hybrid risk model can comprehensively assess the risks arising from market volatility and the decoupling of securities and electronics, providing accurate risk parameters for subsequent risk management.

[0060] Market risk encompasses both normal fluctuations and extreme events. A single distribution model cannot fully describe the risk distribution, while a hybrid model can comprehensively capture both characteristics. By generating a synergistic risk adjustment coefficient, this model can provide an accurate measure of comprehensive risk.

[0061] The synergistic risk adjustment coefficient, calculated using the Gaussian-Laplace mixed risk model, is directly passed to the risk engine in step three as output. This coefficient integrates information on market volatility and the risk of decoupling between securities and telecommunications, guiding the generation of hedging packages and the formulation of risk management strategies. This ensures that step three can adjust according to real-time market risk conditions. The synergistic risk adjustment coefficient, by integrating multi-dimensional risk information from market volatility and the decoupling of securities and telecommunications, provides guidance for hedging package generation and risk strategy formulation, ensuring the targeted and real-time nature of risk management and improving risk control capabilities.

[0062] Step two quantifies the dynamic correlation between spot electricity prices and renewable energy output fluctuations, as well as the risk of decoupling between spot electricity prices and renewable energy output, by generating a cross-variability coupling index and a securities-electricity synchronization deviation index. Subsequently, a Gaussian-Laplace mixed risk model is used to process these indices, generating a collaborative risk adjustment coefficient to provide a comprehensive risk measure for the risk engine in step three.

[0063] Step 3: The risk engine receives the collaborative risk adjustment coefficient in real time, determines whether to extract the electricity contract and corresponding green certificate from the trading pool based on the preset threshold, generates a hedging package by executing atomic locking through the smart contract, writes the hedging package data into the blockchain after being processed by the hash function, records it to the distributed ledger through the consensus mechanism, and continuously monitors the changes in the collaborative risk adjustment coefficient. Based on its value, it dynamically adjusts the locking conditions or generation frequency of the hedging package, executes the adjustment through the condition update function of the smart contract, and rewrites it into the blockchain.

[0064] Step 301: The risk engine receives the collaborative risk adjustment coefficient.

[0065] The risk engine receives the collaborative risk adjustment coefficient output by the Gaussian-Laplace hybrid risk model generated in step two in real time. The collaborative risk adjustment coefficient is a comprehensive risk indicator derived by analyzing the cross-volatility coupling index and the certificate-electricity synchronization deviation index. It is used to reflect the degree of coupling between electricity price and green certificate price fluctuations, as well as the risk level of decoupling between electricity and green certificates.

[0066] The calculation process is as follows: First, historical fluctuation data of electricity prices and green certificate prices are extracted, and the synchronicity and deviation between the two are analyzed. Then, these data are integrated into a single value to reflect the dynamic changes in market risk. The risk engine uses the collaborative risk adjustment coefficient as input to the dynamic threshold for subsequent determination of whether to generate a hedging package.

[0067] The risk engine receives the coordinated risk adjustment coefficient in real time, enabling it to promptly perceive changes in the coupling degree and decoupling risk between electricity price and green certificate price fluctuations. By integrating the cross-volatility coupling index and the certificate-electricity synchronization deviation index, the coordinated risk adjustment coefficient provides a comprehensive risk quantification basis, allowing the risk engine to dynamically adjust its strategy based on market conditions, ensuring that the generation of hedging packages matches the market risk level.

[0068] Step 302: Assembling the smart contract hedging package

[0069] The risk engine assembles smart contract hedging packages containing electricity and corresponding green certificates in real time based on the value of the collaborative risk adjustment coefficient. The specific process is as follows: a preset threshold is set based on historical market data and risk tolerance, serving as the criterion for judging the collaborative risk adjustment coefficient. When the value of the collaborative risk adjustment coefficient exceeds the preset threshold, the risk engine extracts the electricity contracts and corresponding green certificates from the trading pool, ensuring that the quantity and source of both match perfectly.

[0070] Next, the electricity contract and green certificates are atomically locked using a smart contract to form a hedging package. The structure of the hedging package includes the electricity contract, green certificates, and locking conditions. The locking conditions are determined by the value of the collaborative risk adjustment coefficient, for example, stipulating that the hedging package cannot be split while the collaborative risk adjustment coefficient remains high. The locking process is automatically completed by the pre-compiled logic of the smart contract, ensuring that the binding between the electricity and green certificates is irreversible.

[0071] By assembling hedging packages using smart contracts and leveraging their automated execution capabilities, the efficiency and consistency of binding electricity contracts and green certificates are ensured. An atomic locking mechanism prevents the separation of electricity and green certificates during trading, achieving a unique correspondence between them and reducing risks to market participants arising from price fluctuations and inconsistencies in policy requirements.

[0072] Step 303: Blockchain Writing and Traceability

[0073] The assembled hedging package is written to the blockchain to ensure the immutability and traceability of transaction data. The specific steps are as follows: A hash function is applied to the hedging package data (including the electricity contract, green certificates, and locking conditions) to generate a unique identifier. The process of generating a unique identifier involves concatenating the serialized data of the electricity contract, green certificates, and locking conditions together, and then calculating a fixed-length hash value using the SHA-256 hash function. The hedging package and its unique identifier are recorded in the blockchain's distributed ledger, and consensus is reached among all network nodes through a consensus mechanism (such as proof-of-work or proof-of-stake). A query interface is also provided, allowing market participants to retrieve the transaction history, status, and source information of the hedging package using its unique identifier.

[0074] By writing hedging packages into the blockchain, the immutability of the blockchain ensures the security of transaction data, while the distributed ledger and consensus mechanism provide transparency and consistency in transactions.

[0075] Step 304: Dynamic Management of Hedging Packages

[0076] The risk engine continuously monitors market risk and the status of hedging packages, dynamically adjusting the strategies of hedging packages based on real-time updates of the collaborative risk adjustment coefficient. Specifically, when the collaborative risk adjustment coefficient falls below a preset threshold, the risk engine releases the locking conditions of some hedging packages, freeing up liquidity in the electricity contracts or green certificates.

[0077] When the collaborative risk adjustment coefficient rises above a higher threshold, the risk engine increases the frequency of hedging package generation or strengthens the constraints of the locking conditions. The adjustment process is executed through the condition update function of the smart contract, and the updated hedging package is rewritten to the blockchain to ensure that the adjusted state is recorded and verified.

[0078] The hedging package is dynamically adjusted based on real-time changes in the collaborative risk adjustment coefficient, enabling its strategy to adapt to market risk fluctuations. Liquidity is released when risk decreases to avoid excessive resource locking; constraints are strengthened when risk increases to enhance protection, thereby improving the flexibility of risk management and the efficiency of resource utilization. This reduces the cash flow pressure on market participants due to market volatility while maintaining the effectiveness of hedging measures.

[0079] Step three involves using a risk engine to assemble smart contract hedging packages of electricity and green certificates in real time based on a collaborative risk adjustment coefficient. Blockchain technology ensures the immutability and traceability of transactions, effectively addressing the risks arising from price fluctuations and decoupling between electricity and green certificates. The atomic locking of the smart contract achieves a unique mapping between electricity and green certificates, while the blockchain's write and traceability functions provide transparency and compliance guarantees. The dynamic management mechanism optimizes the hedging strategy based on changes in market risk.

[0080] Step 4: Continuously monitor the profit and loss of the hedging package and the status of the market participants' margin accounts. Calculate the profit and loss of the hedging package by multiplying the difference between the market electricity price and the locked electricity price by the electricity volume, plus the difference between the market green certificate price and the locked green certificate price multiplied by the number of green certificates, to determine the margin account balance and the frozen amount. Dynamically adjust the margin addition and release thresholds based on market volatility and the profit and loss of the hedging package. Market volatility is calculated using the rolling average absolute deviation of the recent price change rate. The addition threshold is set as the minimum margin requirement plus the product of the risk coefficient, market volatility, and risk exposure. The release threshold is set as the addition threshold minus the adjustment parameters. When the margin account balance is lower than the addition threshold, a margin addition operation is triggered, with the addition amount being the difference between the addition threshold and the margin account balance. When the margin account balance is higher than the release threshold and the hedging package profit and loss is positive, a release operation is triggered, with the release amount being the difference between the margin account balance and the release threshold. The operation is automatically executed through a smart contract and recorded on the blockchain. Write the margin adjustment records and hedging package profit and loss data into the risk pool for use by the prediction model.

[0081] Step 401: Real-time tracking of hedging package profit / loss and margin status

[0082] The profit and loss of each hedging package and the status of market participants' margin accounts are monitored in real time. The profit and loss calculation process is based on the difference between the current market price and the locked-in prices of the electricity contracts and green certificates in the hedging package. Specifically: First, the current market electricity price is subtracted from the locked-in electricity price of the electricity contracts in the hedging package, and this difference is multiplied by the electricity volume in the hedging package to obtain the profit and loss for the electricity portion; then, the current market green certificate price is subtracted from the locked-in price of the green certificates in the hedging package, and this difference is multiplied by the number of green certificates in the hedging package to obtain the profit and loss for the green certificate portion; finally, the profit and loss for the electricity portion and the profit and loss for the green certificate portion are added together to obtain the total profit and loss of the hedging package.

[0083] The determination of margin status is achieved by examining the relationship between the market participant's margin account balance and the frozen amount. The rationale for real-time tracking of hedging portfolio profits and losses and margin status is that this process promptly reflects the market participant's risk exposure and funding pressures, providing data support for subsequent dynamic adjustments. Its benefits include ensuring that the hedging portfolio's financial performance and margin levels remain aligned with market dynamics, thereby improving the responsiveness of risk management.

[0084] Real-time tracking of hedging portfolio profits and losses and margin status can promptly reflect the risk exposure and funding pressure of market participants, ensuring that the financial performance and margin level of hedging portfolios are consistent with market dynamics and improving the responsiveness of risk management.

[0085] Step 402, Adaptive Threshold Setting

[0086] Based on market volatility and the profit and loss of hedging portfolios, the margin call and release thresholds are dynamically adjusted. Market volatility is calculated using the rolling average absolute deviation of recent price change rates; that is, by statistically analyzing the magnitude of market price changes over a period of time and averaging their absolute values, the degree of market volatility is represented. The margin call threshold is set as follows: based on the minimum margin requirement, a risk coefficient multiplied by market volatility is added, and then multiplied by the market participants' risk exposure to arrive at the threshold. The margin release threshold is set by subtracting an adjustment parameter from the margin call threshold to ensure that the release threshold is lower than the margin call threshold. Dynamic adjustments avoid the inefficiency of fixed thresholds due to market changes or insufficient risk coverage, improving the flexibility of margin management and ensuring that capital occupation matches the risk level.

[0087] Step 403: Trigger margin call or release

[0088] Based on real-time tracking of the profit and loss of the hedging portfolio and the margin status, and according to the adaptive threshold setting, margin calls or releases are executed. Specifically, when the margin account balance falls below the add-on threshold, a add-on operation is triggered, with the added amount being the difference between the add-on threshold and the current margin account balance. When the margin account balance is above the release threshold and the hedging portfolio's profit and loss are positive, a release operation is triggered, with the released amount being the difference between the current margin account balance and the release threshold. Add-on or release operations are automatically completed via smart contracts, and the operation records are stored on the blockchain.

[0089] Triggering margin calls or releases allows for a rapid response to market changes, ensuring that margin levels adjust in sync with risk exposure, reducing cash flow pressure from excessive fund freezes, and providing timely margin replenishment when risks escalate to safeguard transaction security.

[0090] Step 404: Update the risk pool and make it available for model training.

[0091] Margin adjustment records and hedging package profit and loss data are written into the risk pool to support subsequent model training. Specifically: after each margin addition or release, the operation details (including operation time, amount, and trigger reason) are recorded in the risk pool; the profit and loss data of the hedging package and market volatility data are updated to the risk pool regularly; the data in the risk pool is available for the prediction model in step five to call through a preset interface for iterative correction of the dynamic reference vector.

[0092] By centrally storing risk-related data, the predictive model can obtain market feedback, improve prediction accuracy, provide real-time data support for step five, and enhance the predictive model's adaptability and the trading platform's intelligence.

[0093] By tracking the profit and loss of hedging packages and the status of margin in real time, and combining this with an adaptive threshold setting mechanism, margin calls or releases are dynamically triggered, and relevant data is updated to the risk pool. This achieves refined and dynamic optimization of margin management, addressing the characteristics of volatile markets and dynamically changing risk exposures in new energy trading. Real-time tracking and threshold adjustments effectively reduce the reserve cost pressure on market participants while ensuring the security and compliance of transactions.

[0094] Step 5: Extract profit and loss data from the smart contract hedging package through the risk pool to generate a profit and loss trajectory; calculate the correlation coefficient between the cross volatility coupling index and the profit and loss trajectory, as well as the correlation coefficient between the securities-electronic synchronization deviation index and the profit and loss trajectory, to assess the impact of market volatility and the risk of securities-electronic decoupling on the profit and loss trajectory; correct the dynamic reference vector based on the gradient descent method, with the correction amount determined by the weighted sum of the adjustment contributions of the cross volatility coupling index and the securities-electronic synchronization deviation index; update the dynamic reference vector and write it back to the market matching logic to start a new round of trading.

[0095] Step 501: Extract the profit and loss trajectory of the hedging package.

[0096] Profit and loss data of smart contract hedging packages are extracted from the risk pool in step four to form a time-series profit and loss trajectory.

[0097] The profit and loss trajectory is calculated as follows: First, take the current market electricity price and subtract the locked electricity price of the electricity contract in the hedging package. Multiply the difference by the electricity volume in the hedging package to obtain the profit and loss of the electricity volume portion. Then, take the current market green certificate price and subtract the locked price of the green certificates in the hedging package. Multiply the difference by the number of green certificates in the hedging package to obtain the profit and loss of the green certificate portion. Finally, add the profit and loss of the electricity volume portion to the profit and loss of the green certificate portion to obtain the total profit and loss of the hedging package at that point in time.

[0098] Profit and loss data directly reflects the deviation between market price and the price locked by the hedging package. It can serve as the core basis for evaluating the accuracy of the dynamic reference vector. By quantifying the relationship between market price deviation and hedging package performance, it provides reliable data support for the correction of the dynamic reference vector based on actual trading results.

[0099] Step 502: Evaluate the correlation between the cross-variability coupling index and the securities-electronic synchronization deviation index.

[0100] Using the cross-volatility coupling index and the securities-electricity synchronization deviation index generated in step two, calculate the correlation between these two indices and the profit and loss trajectory. Wherein:

[0101] Set a rolling time window. Within this window, first calculate the covariance between the cross volatility coupling index and the profit / loss trajectory. Then, calculate the standard deviation of the cross volatility coupling index and the profit / loss trajectory within the window. Divide the covariance by the product of these two standard deviations to obtain the correlation coefficient between the cross volatility coupling index and the profit / loss trajectory. Then, in the same way, calculate the covariance between the securities-electricity synchronization deviation index and the profit / loss trajectory. Divide this covariance by the product of the standard deviation of the securities-electricity synchronization deviation index and the profit / loss trajectory within the window to obtain the correlation coefficient between the securities-electricity synchronization deviation index and the profit / loss trajectory.

[0102] The cross-volatility coupling index and the securities-electronic synchronization deviation index quantify market volatility and the risk of securities-electronic decoupling, respectively. By analyzing their correlation with profits and losses, the main driving factors of benchmark price deviations can be revealed, and the degree of influence of the cross-volatility coupling index and the securities-electronic synchronization deviation index on profits and losses can be clarified, providing a clear directional basis for the adjustment of the dynamic reference vector.

[0103] Step 503: Iteratively optimize the dynamic reference vector

[0104] Based on the correlation between the profit and loss trajectory, the cross-fluctuation coupling index, and the securities-electricity synchronization deviation index, the dynamic reference vector in step one is corrected by the gradient descent method.

[0105] The specific optimization process is as follows: with the goal of minimizing the sum of absolute values ​​of profit and loss, first calculate the correlation coefficient between the cross volatility coupling index and profit and loss, multiply it by the value of the cross volatility coupling index, and obtain the adjustment contribution of the cross volatility coupling index.

[0106] Then, calculate the correlation coefficient between the securities-electricity synchronization deviation index and profit and loss, multiply it by the value of the securities-electricity synchronization deviation index to obtain the adjustment contribution of the securities-electricity synchronization deviation index; add the two adjustment contributions together, and then multiply by a preset learning rate (used to control the adjustment step size) to obtain the adjustment amount of the dynamic reference vector.

[0107] Finally, the adjustment amount is added to the current dynamic reference vector's benchmark price to obtain the corrected benchmark price. The gradient descent method can gradually reduce profit and loss deviation through multiple adjustments, while the correlation between the cross volatility coupling index and the securities-electronic synchronization deviation index ensures that the correction direction is consistent with market risk. The dynamic reference vector can gradually approach the actual market price, thereby improving the adaptability and accuracy of the benchmark price.

[0108] Step 504: Write back the benchmark price and initiate the transaction.

[0109] The corrected dynamic reference vector is written back to the market matching logic to update the trading benchmark price sequence. The corrected benchmark price serves as the basis for the next round of trading. The market matching logic generates trading orders based on the new benchmark price and initiates a new trading cycle, repeating the complete process from steps one to five. By updating the benchmark price in a timely manner, the trading platform can quickly reflect market dynamics, improve price discovery efficiency, and enhance market participants' trading confidence and market activity through continuous optimization of the benchmark price.

[0110] By extracting the profit and loss trajectory of the hedging package in step four, and combining it with the correlation analysis of the cross-volatility coupling index and the securities-electricity synchronization deviation index generated in step two, the dynamic reference vector in step one is iteratively corrected using the gradient descent method. The corrected benchmark price is then written back to the market matching logic, achieving adaptive optimization of the trading platform. In scenarios where rapid grid connection of new energy sources leads to drastic supply and demand fluctuations, step five effectively solves the deviation problem between the benchmark price and actual market demand, improving the accuracy of price series across time scales.

[0111] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0112] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0113] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0114] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0115] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A smart power trading management platform integrating medium- and long-term spot green certificates, characterized by: include, The data optimization module collects unit operation data, meteorological forecast data, and matching instructions, and generates a dynamic reference vector through linear optimization, which serves as the benchmark price sequence for cross-year, monthly, day-ahead, and real-time transactions. The risk modeling module generates a cross-volatility coupling index and a securities-electronic synchronization deviation index based on the dynamic reference vector. These are input into a Gaussian-Laplace hybrid risk model, and the output is a collaborative risk adjustment coefficient. The contract assembly module assembles a smart contract hedging package with electricity and green certificate atomic lock according to the collaborative risk adjustment coefficient, and ensures the immutability and traceability of the hedging package through blockchain records. The real-time adjustment module tracks the profit and loss and margin status of the hedging package in real time, triggers the addition or release of margin based on the adaptive threshold, and writes the updated results into the risk pool. The price correction module uses the settlement profit and loss trajectory, the cross-fluctuation coupling index, and the securities-electronic synchronization deviation index to iteratively correct the dynamic reference vector and write back the benchmark price to the next round of trading. The generation of the cross-fluctuation coupling index includes: performing rolling cross-correlation calculation on the spot node electricity price and the high-pass filtered renewable power prediction error, applying an exponential decay kernel, and generating the cross-fluctuation coupling index to reflect the coupling relationship between electricity price and power fluctuation; The generation of the certificate-electricity synchronization deviation index includes: calculating the standardized difference between the green certificate equivalent price and the nodal electricity price, and then multiplying it by the local renewable energy penetration rate to generate the certificate-electricity synchronization deviation index, so as to quantify the synchronization deviation between green certificates and electricity prices. The Gaussian-Laplace hybrid risk model takes the cross-volatility coupling index and the securities-electronic synchronization deviation index as inputs, uses a hybrid model of Gaussian and Laplace distributions to calculate the risk probability, and outputs the collaborative risk adjustment coefficient to adjust trading risk.

2. The intelligent power trading management platform integrating medium- and long-term spot green certificates as described in claim 1, characterized in that: The data collected, including unit operation data, meteorological forecast data, and matching instructions, include: The unit's operating status is monitored in real time by sensors to obtain unit operating data; meteorological forecast data is obtained through meteorological stations or satellite data; matching instructions are obtained through the trading platform; and the unit operating data, meteorological forecast data and matching instructions are synchronized to a unified time-series event bus.

3. The intelligent power trading management platform integrating medium- and long-term spot green certificates as described in claim 2, characterized in that: The process of generating the dynamic reference vector through linear optimization includes: taking the unit operation data, meteorological forecast data, and matching instructions as inputs, using a linear programming algorithm to solve the objective function to minimize price fluctuations and satisfy supply and demand balance constraints, and generating the dynamic reference vector.

4. The intelligent power trading management platform integrating medium- and long-term spot green certificates as described in claim 3, characterized in that: The smart contract hedging package for atomically locking electricity and green certificates includes: extracting electricity contracts and corresponding green certificates from the trading pool according to the collaborative risk adjustment coefficient, and atomically locking the electricity contracts and green certificates through a smart contract to form an indivisible hedging package.

5. The intelligent power trading management platform integrating medium- and long-term spot green certificates as described in claim 4, characterized in that: The step of ensuring the immutability and traceability of the hedging package through blockchain records includes: applying a hash function to the data of the hedging package to generate a unique identifier, recording the hedging package and its unique identifier in the distributed ledger of the blockchain, and reaching consensus among all nodes in the network through a consensus mechanism.

6. The intelligent power trading management platform integrating medium- and long-term spot green certificates as described in claim 5, characterized in that: The adaptive threshold is dynamically adjusted based on market volatility and the profit and loss of the hedging package, including: calculating market volatility, setting a margin call threshold and a release threshold, wherein the margin call threshold is the minimum margin requirement plus the product of the risk coefficient, market volatility, and risk exposure, and the release threshold is the margin call threshold minus the adjustment parameter.

7. The intelligent power trading management platform integrating medium- and long-term spot green certificates as described in claim 6, characterized in that: The iterative correction of the dynamic reference vector includes: using the gradient descent method, based on the correlation between the settlement profit and loss trajectory and the cross-volatility coupling index and the securities-electronic synchronization deviation index, calculating the adjustment amount, and updating the benchmark price of the dynamic reference vector to optimize the accuracy of the next round of trading.

Images