Multi-target dynamic pricing method for cross-regional electricity transaction market

Abstract

The invention discloses a multi-target dynamic pricing method for a cross-regional electricity transaction market, and the method comprises the following steps: S1, carrying out data collection and modeling, and constructing a dynamic pricing data set; s2, based on the dynamic pricing data set, a multi-objective optimization model is established, and multiple objectives comprise power transaction economy maximization, carbon emission minimization and power grid stability optimization; the invention belongs to the technical field of electricity markets, and aims to solve the problems that in the prior art, it is difficult to adapt to cross-regional complex scenes, and carbon emission constraints are not considered; according to the technical scheme of the invention, the method achieves the technical effects of covering the whole link of data-model-execution-verification through deep fusion of space-time modeling, game theory, block chain, reinforcement learning and digital twinning, achieving the multi-scene self-adaption through a weight factor, game balance and risk hedging, improving the cross-regional power resource distribution efficiency, and achieving the multi-scene self-adaption of the whole link of data-model-execution-verification of the whole link of time-space modeling, game theory, block chain, reinforcement learning and digital twinning. Carbon emission and market risks are reduced, and construction of a novel electric power system is facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of power markets, and in particular to a multi-objective dynamic pricing method for a cross-regional power trading market. Background Art

[0002] Traditional electricity trading pricing often uses fixed rates or single-objective optimization models, which are difficult to adapt to the complex scenarios of dynamic supply and demand fluctuations and multi-objective conflicts (such as economic efficiency, environmental protection, and stability) in cross-regional electricity markets. They do not consider carbon emission constraints and ignore grid stability risks. In addition, cross-regional data transmission delays and low pricing transparency further hinder transaction efficiency.

[0003] Therefore, it does not meet the existing needs. We propose a multi-objective dynamic pricing method for the cross-regional electricity trading market. Summary of the Invention

[0004] To this end, the present invention provides a multi-objective dynamic pricing method for a cross-regional power trading market to solve the above-mentioned problems in the prior art.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] According to a first aspect of the present invention, a multi-objective dynamic pricing method for a cross-regional power trading market includes the following steps:

[0007] S1, data collection and modeling, building a dynamic pricing dataset;

[0008] S2. Establishing a multi-objective optimization model based on the dynamic pricing dataset, wherein the multi-objectives include maximizing the economic efficiency of power trading, minimizing carbon emissions, and optimizing grid stability;

[0009] S3. Solve the multi-objective optimization model by cooperating with the distributed optimization algorithm and the game equilibrium model, find the optimal pricing range, and generate a dynamic pricing plan;

[0010] S4. According to the dynamic pricing scheme, the electricity transaction price of each region is adjusted in real time, and the pricing information is synchronized to the trading platform through the blockchain network.

[0011] Furthermore, the step S1 further includes:

[0012] S101. Construct a spatiotemporal correlation model of cross-regional power supply and demand, and identify the spatiotemporal coupling relationship of inter-regional power transmission;

[0013] S102, collecting power supply and demand data, power generation cost data, and transmission constraint data from the cross-regional power trading market in real time to construct a dynamic pricing data set;

[0014] S103. Based on historical transaction data and user behavior analysis, a user demand response model is constructed to predict the electricity price elasticity in each region.

[0015] Furthermore, the distributed optimization algorithm in step S3 is an improved NSGA-II algorithm, which also includes:

[0016] S301, initializing the population to screen non-dominated solutions based on the Pareto frontier;

[0017] S302, using adaptive crossover and mutation operators to optimize population diversity;

[0018] S303. Introduce regional weight factors to dynamically adjust the pricing priority of each region.

[0019] Furthermore, in step S303, the weight factor of each region is dynamically calculated based on the power shortage rate, renewable energy penetration rate and user load elasticity of each region.

[0020] Furthermore, the step S3 also includes: S32, a dynamic pricing scheme, including a time-based pricing strategy, wherein the peak period pricing weight is biased towards grid stability, and the off-peak period pricing weight is biased towards economy.

[0021] Furthermore, the method further includes step S5, dynamically correcting the parameters of the multi-objective optimization model through a reinforcement learning model based on user feedback data and actual transaction results.

[0022] Furthermore, the smart contract in step S4 also includes a heating fluctuation warning mechanism, which triggers manual intervention and review process when the real-time price exceeds a preset threshold.

[0023] Furthermore, the reinforcement learning model in step S5 adopts a deep deterministic policy gradient algorithm, and the reward function is the inverse square of the deviation between the actual return and the predicted return.

[0024] The present invention has the following advantages:

[0025] This multi-objective dynamic pricing method for the cross-regional electricity trading market deeply integrates spatiotemporal modeling, game theory, blockchain, reinforcement learning and digital twins, covering the entire "data-model-execution-verification" chain, and achieving multi-scenario adaptation through weight factors, game equilibrium and risk hedging, thereby improving the efficiency of cross-regional electricity resource allocation, reducing carbon emissions and market risks, and assisting in the construction of new power systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a flowchart of the pricing method for the multi-objective dynamic pricing method for the cross-regional power trading market proposed by the present invention;

[0027] Figure 2This is a flow chart of step S1 of the multi-objective dynamic pricing method for the cross-regional power trading market proposed by the present invention;

[0028] Figure 3 This is a flow chart of step S2 of the multi-objective dynamic pricing method for the cross-regional power trading market proposed by the present invention;

[0029] Figure 4 The multi-objective dynamic pricing method for the cross-regional power trading market proposed by this invention Figure 3 Flow diagram of step S31;

[0030] Figure 5 This is a schematic diagram of the pricing system of the multi-objective dynamic pricing method for the cross-regional power trading market proposed in the present invention. DETAILED DESCRIPTION

[0031] The following describes the implementation of the present invention using specific embodiments. Those skilled in the art will readily understand the other advantages and benefits of the present invention from the disclosure herein. Obviously, the embodiments described are only a portion of the present invention, not all of it. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are intended to fall within the scope of protection of the present invention.

[0032] Reference Figure 1-5 A multi-objective dynamic pricing method for a cross-regional power trading market, according to a first aspect of the present invention, comprises the following steps:

[0033] S1, data collection and modeling, building a dynamic pricing dataset;

[0034] S2. Establishing a multi-objective optimization model based on the dynamic pricing dataset, wherein the multi-objectives include maximizing the economic efficiency of power trading, minimizing carbon emissions, and optimizing grid stability;

[0035] S3. Solve the multi-objective optimization model by cooperating with the distributed optimization algorithm and the game equilibrium model, find the optimal pricing range, and generate a dynamic pricing plan;

[0036] S4. According to the dynamic pricing plan, the electricity transaction price of each region is adjusted in real time, and the pricing information is synchronized to the trading platform through the blockchain network.

[0037] Step S1 further includes:

[0038] S101. Construct a spatiotemporal correlation model of cross-regional power supply and demand. Use a spatiotemporal graph neural network to model the spatiotemporal coupling relationship of inter-regional power transmission. Quantify the capacity of transmission corridors, renewable energy output fluctuations, and spatiotemporal load migration characteristics, thereby identifying the spatiotemporal coupling relationship of inter-regional power transmission.

[0039] S102. Real-time collection of power supply and demand data, power generation cost data, and transmission constraint data from cross-regional power trading markets, simultaneous integration of meteorological data such as wind speed and sunlight; forecasting of renewable energy output and building a dynamic pricing data set;

[0040] S103. Based on historical transaction data and user behavior analysis, a user demand response model is constructed to predict the elasticity of electricity prices in each region, divide rigid loads into adjustable loads, and quantify the demand-side response potential.

[0041] The distributed optimization algorithm in step S3 is an improved NSGA-II algorithm, which also includes:

[0042] S301, initialize the population and filter the non-dominated solutions based on the Pareto frontier, introduce the crowding distance and the preservation

[0043] Retention strategy to improve solution set diversity;

[0044] S302, using adaptive crossover and mutation operators to optimize population diversity;

[0045] S303. Introduce a regional weight factor to dynamically adjust the pricing priority of each region. The weight factor = 0.4 × power gap rate + 0.3 × renewable energy penetration rate + 0.3 × user load elasticity. The weight coefficient is dynamically corrected through fuzzy logic.

[0046] In step S303, the weight factor of each region is dynamically calculated based on the power shortage rate, renewable energy penetration rate and user load elasticity of each region.

[0047] Step S3 also includes: S32, a dynamic pricing scheme, including a time-based pricing strategy, where the peak-time pricing weight is biased towards grid stability, and the off-peak-time pricing weight is biased towards economy.

[0048] It also includes step S5, dynamically correcting the parameters of the multi-objective optimization model through the reinforcement learning model based on user feedback data and actual transaction results, specifically: S501, deep deterministic policy gradient model: reward function = 1 / (predicted benefit - actual benefit)2 + 0.5×carbon emission reduction rate, dynamically adjusting the model weight through the Actor-Critic network; S502, digital twin verification platform: building a cross-regional power market digital twin, simulating extreme scenarios, such as typhoon-induced power transmission outages, and verifying pricing strategies.

[0049] The smart contract in step S4 also includes a heating fluctuation warning mechanism, which triggers manual intervention in the review process when the real-time price exceeds the preset threshold.

[0050] The reinforcement learning model in step S5 adopts a deep deterministic policy gradient algorithm, and the reward function is the inverse square of the deviation between the actual return and the predicted return.

[0051] Working principle:

[0052] The spatiotemporal graph neural network (STGNN) captures the spatiotemporal dependencies of inter-regional power transmission;

[0053] Game models quantify user response behavior;

[0054] The NSGA-II algorithm generates a Pareto solution set, and the game equilibrium model selects the optimal solution that satisfies the interests of the power generation side and the user side;

[0055] The regional weight factor dynamically adjusts the optimization direction;

[0056] Sharded blockchains reduce data storage pressure, and smart contracts automatically execute transactions and trigger hedging tools;

[0057] Price fluctuation early warning mechanism prevents market manipulation, and manual review ensures compliance of abnormal transactions;

[0058] By correcting and optimizing model parameters based on historical deviations, the digital twin platform simulates scenarios of "extreme weather + surge in demand" to verify pricing strategies;

[0059] By deeply integrating spatiotemporal modeling, game theory, blockchain, reinforcement learning and digital twins, covering the entire "data-model-execution-verification" chain, and achieving multi-scenario adaptation through weight factors, game equilibrium and risk hedging, the efficiency of cross-regional power resource allocation is improved, carbon emissions and market risks are reduced, and the construction of new power systems is assisted.

Claims

1. A multi-objective dynamic pricing method for cross-regional power trading markets, characterized by: The steps include: S1, data collection and modeling, building a dynamic pricing dataset; S2. Establishing a multi-objective optimization model based on the dynamic pricing dataset, wherein the multi-objectives include maximizing the economic efficiency of power trading, minimizing carbon emissions, and optimizing grid stability; S3. Solve the multi-objective optimization model by cooperating with the distributed optimization algorithm and the game equilibrium model, find the optimal pricing range, and generate a dynamic pricing plan; S4. Adjust the electricity transaction price in each region in real time according to the dynamic pricing scheme, and synchronize the pricing information to the trading platform via the blockchain network; The method further includes step S5 of dynamically modifying the parameters of the multi-objective optimization model through a reinforcement learning model based on user feedback data and actual transaction results; The smart contract in step S4 also includes a heating fluctuation early warning mechanism, which triggers manual intervention in the review process when the real-time price exceeds a preset threshold.

2. The multi-objective dynamic pricing method for the cross-regional power trading market according to claim 1 is characterized in that: Said S1 further comprises: S101. Construct a spatiotemporal correlation model of cross-regional power supply and demand, and identify the spatiotemporal coupling relationship of inter-regional power transmission; S102, collecting power supply and demand data, power generation cost data, and transmission constraint data from the cross-regional power trading market in real time to construct a dynamic pricing data set; S103. Based on historical transaction data and user behavior analysis, a user demand response model is constructed to predict the electricity price elasticity in each region.

3. The multi-objective dynamic pricing method for the cross-regional power trading market according to claim 1 is characterized in that: The distributed optimization algorithm in S3 is an improved NSGA-II algorithm, which also includes: S301, initializing the population to screen non-dominated solutions based on the Pareto frontier; S302, using adaptive crossover and mutation operators to optimize population diversity; S303. Introduce regional weight factors to dynamically adjust the pricing priority of each region.

4. The multi-objective dynamic pricing method for the cross-regional power trading market according to claim 3 is characterized in that: The weight factor of each region in S303 is dynamically calculated based on the power shortage rate, renewable energy penetration rate and user load elasticity of each region.

5. The multi-objective dynamic pricing method for the cross-regional power trading market according to claim 1 is characterized in that: The S3 also includes: S32, a dynamic pricing scheme, including a time-based pricing strategy, where the peak-time pricing weight is biased towards grid stability, and the off-peak-time pricing weight is biased towards economy.

6. The multi-objective dynamic pricing method for the cross-regional power trading market according to claim 5 is characterized in that: The reinforcement learning model in step S5 adopts a deep deterministic policy gradient algorithm, and the reward function is the inverse square of the deviation between the actual return and the predicted return.

Images