22 results about "Market simulation" patented technology

The invention discloses a sports market demand prediction method and device based on deep learning and a medium, and relates to the technical field of market demand prediction, and the method comprises the steps: collecting sports demand data, and carrying out the preprocessing; performing relation mining and graph structure learning on the preprocessed sports demand data through a graph attention space-time network to generate a macroscopic demand potential energy graph; performing potential area identification on the macroscopic demand potential energy diagram by adopting a pre-trained sports market simulation model, and outputting local demand prediction data; performing weighted fusion and error correction on the macroscopic demand potential energy map and the local demand prediction data, and outputting a sports demand prediction score; and making a sports market demand strategy according to the sports demand prediction score and the multi-granularity demand prediction report, and transmitting the sports market demand strategy to an enterprise manager through an enterprise decision support interface. According to the method, multi-level accurate prediction and decision support of sports market demands are realized through dual-mechanism cooperation of the graph attention space-time network and the space-time convolution.

The embodiment of the invention relates to an analog simulation method based on electric power spot transaction. The analog simulation method comprises the steps of obtaining static structure data such as power grid topology and unit parameters and dynamic time sequence data such as load and new energy prediction; performing preprocessing and multi-time scale division on the dynamic data; constructing a multi-agent model with differentiated characteristics based on the processed data; each agent generates and reports energy block information such as a time-sharing block, a continuous block, a curve block or a variable block according to a perceived market environment; constructing a market clearing model considering system power balance, network power flow, unit operation and energy block characteristic constraint based on all declarations, and solving to obtain clearing electricity price and electric quantity; a result is fed back to the intelligent agent, and the strategy is updated through deep reinforcement learning; and performing rolling co-simulation between the day-ahead market and the real-time market, wherein a day-ahead clearing result is used as a boundary condition of the real-time market. The authenticity, accuracy and practicability of market simulation are improved.

The invention relates to an energy storage electric power spot market simulation clearing method based on hierarchical model control prediction, and the method comprises the following steps: S1, building an electric power spot market global simplified clearing model containing energy storage, and taking the model as an upper prediction control model; s2, establishing an electric power spot market detailed clearing model containing energy storage as a lower execution model; s3, dividing long-time simulation into M solving periods; in each solving period, global simplification clearing optimization is firstly utilized to obtain an energy storage final state SOC constraint value of each remaining period; on the basis of the energy storage final state SOC constraint value of the current period, market simulation clearing of the current period is completed by using a detailed clearing model; and finally, updating the system state according to the current clearing result, and rolling to advance to the next cycle until clearing of all cycles is completed. According to the method, dynamic global optimization of the energy storage SOC constraint is realized through a hierarchical model prediction control mechanism, and the calculation efficiency and the market simulation clearing effect are effectively balanced.

The present application relates to the technical field of power spot market transaction and thermal power unit operation optimization, in particular to a double-layer deduction method for thermal power unit group spot bidding strategy, which adopts a double-layer optimization model for solution, the upper layer is a similar layer of price unit value, the target is to minimize the weighted relative deviation of market simulation price and historical real price, and the upper and lower limits of segmented bidding and bidding quantity are set according to the capacity scale of the unit to set differentiated constraints; the lower layer is a market clearing layer, the target is to minimize the total system cost, and the system power balance, unit operation, new energy consumption, power transmission safety and other system operation and market clearing full-dimensional constraints are met, the optimal bidding strategy combination is obtained through the iterative interaction of bidding and price signals, the method adopts DE-PSO hybrid algorithm for solution, can accurately fit the real market price law, provides scientific bidding decision basis for power suppliers, and is suitable for different market operation scenes such as heating period and non-heating period, etc.

The invention discloses a method for a friendly power station of a new energy electrothermal coupling system to participate in electric power market simulation and benefit evaluation, and belongs to the technical field of electrical engineering. And a simulation model of the friendly power station of the new energy electrothermal coupling system is constructed, the simulation model is comprehensively incorporated into electricity market electricity price signals, system peak power supply function requirements and detailed operation constraints of all equipment, and accurate simulation of the operation process of the power station is achieved. On the basis, the benefit evaluation method is provided, the investment and operation and maintenance cost of various devices of the power station, the income of participating in the spot market and the auxiliary service market and the peak capacity compensation income are comprehensively considered, and an evaluation system covering the whole life cycle of the power station is constructed. The method provides key technical support for planning, operation strategy making and economic benefit evaluation of the new energy electrothermal coupling system friendly power station in a power market environment.

The invention discloses an insurance policy pricing method under a market simulation scene and a related device, and belongs to the technical field of data analysis. The method comprises the following steps: configuring economic scene environment parameters, obtaining insurance policy information of an insurance policy to be priced, and analyzing the insurance policy information to obtain risk factor information related to a product; based on the insurance policy information and the risk factor information, performing matching in a data source to obtain related scene data; constructing an economic scene generation model based on the related scene data, and generating a plurality of economic scene paths representing future market conditions through the economic scene generation model; adopting a customer behavior model to simulate and generate customer behavior data of each time node under each economic scene path according to the economic scene paths; and according to the customer behavior data, calculating a cash flow under each economic scene path, and summarizing to obtain an expected price of an insurance policy. According to the invention, more accurate insurance policy pricing is realized, and the risk loss is effectively reduced.

The invention belongs to the technical field of power system simulation, and discloses a distributed resource aggregation control strategy verification system and method combined with hardware-in-loop simulation. In the distributed resource aggregation control strategy verification system, a physical simulation layer receives an equipment control instruction of an aggregation layer, simulates an electromagnetic transient process based on a high-fidelity power grid and a distributed resource model, and outputs an equipment state and electrical quantity data; the aggregation layer performs protocol adaptation on the physical simulation data, and splits a macroscopic instruction of the aggregation scene optimization control layer into an equipment control instruction; the optimization control layer combines physical data and market feedback to execute rolling optimization, update instructions and transaction declaration; the market simulation layer generates a price sequence and outputs deviation assessment; and the output unit summarizes the data and compares a target output strategy verification evaluation result. According to the method, the problem of difficulty in high-fidelity experiment simulation verification in distributed resource aggregation control is solved.

The invention discloses an automatic electricity market modeling simulation method and system based on a large language model agent. The method comprises the following steps: (1) obtaining and preprocessing an unstructured electricity market description document; (2) based on the preprocessed document content, automatically constructing a highly structured power market mathematical model through a multi-level reasoning and extraction mechanism by adopting a hierarchical thought chain information extraction method; (3) on the basis of the constructed structured power market mathematical model, generating a simulation result by adopting an inverse coding and market simulation method, utilizing the code generation capability of a large language model and combining execution feedback and mathematical model logic; the method provides support for intelligent decision-making of the electricity market.

The invention discloses a multi-agent simulation system for electricity market bidding decision making. The multi-agent simulation system comprises an environment simulation module and a multi-agent module. The environment simulation module comprises a market output unit and a market settlement unit and is used for executing a market output settlement method, performing financial settlement and generating a reward signal and a market environment state; the multi-agent module comprises a plurality of agents, and each agent is composed of an agent strategy network and an agent action decoder; the multi-agent module is used for receiving the reward signal and the market environment state information fed back by the environment simulation module, and carrying out autonomous strategy learning and bidding decision making; the agent action decoder comprises an original action receiving and analyzing unit, a normalization and scale mapping unit, a constraint projection and structure integration unit and a compliance quotation output unit. According to the method, the convergence speed of multi-agent reinforcement learning in power market simulation and the economical efficiency of the strategy are remarkably improved.

Embodiments of the present application relate to a kind of simulation method based on power spot trading, comprising: obtaining grid topology, unit parameter and other static structure data and dynamic time series data such as load, new energy prediction;Dynamic data is preprocessed and divided into multiple time scales;Based on the data processed, construct a multi-agent model with different characteristics;Each agent generates and reports energy block information such as time block, continuous block, curve block or variable block according to the perceived market environment;Based on all reports, construct a market clearing model considering system power balance, network power flow, unit operation and energy block characteristic constraints and solve, obtain clearing price and power;The results are fed back to the agent, and the strategy is updated by deep reinforcement learning;Rolling collaborative simulation is carried out between day-ahead and real-time market, wherein the day-ahead clearing result is used as the boundary condition of real-time market.The present application improves the authenticity, accuracy and practicality of market simulation.

This invention discloses a multi-agent-based electricity-carbon coupling market simulation system. The technical solution involves constructing three main types of agents: a supervisory agent simulating the carbon trading authority, a grid agent simulating power grid companies, and a power company agent simulating heterogeneous power companies. The system includes the following modules: a power production module for determining the planned power generation and grid-connected electricity price bidding for each power company agent; a power trading module for simulating the grid agent's power dispatching and market clearing processes; a low-carbon technology transformation module for simulating the power company agents' decisions on energy-saving and carbon-reducing technology transformation under carbon constraints; a carbon quota allocation module for simulating the supervisory agent's allocation of carbon quotas and the operation of the primary carbon market; a carbon quota trading module for simulating the operation of the secondary carbon market; and a unit investment and decommissioning module for simulating the power company agents' decisions on long-term asset structure adjustments based on future expectations.

We provide the system. [Solution] A means for receiving user attribute information and calculating an optimized asset allocation based on it, A means for acquiring market data in real time from an external data source and analyzing said market data, A means for automatically adjusting the user's asset group based on the analysis results, A means of learning from past investment history and providing investment strategies suitable for the user, A means of providing learning content to support users in improving their financial knowledge, A means of visualizing market simulations using augmented reality technology, A system that includes this.

The application discloses a kind of enterprise image planning digitization intelligent customization matching method and system, specifically related to artificial intelligence and enterprise brand design technical field, the present application is by collecting enterprise multi-modal source data, constructs enterprise brand heterogeneous information network and adopts graph attention network to extract enterprise DNA embedding vector;Multi-modal conditional generative adversarial network is constructed, color matching scheme, copy style parameter and IP image sketch are generated simultaneously, and cross-modal consistency score is introduced to ensure the coordination of scheme;Further, the generator is used as reinforcement learning intelligent agent, and the prediction index of the pre-trained market simulation environment model is used as the reward, and the near-end strategy optimization algorithm is used for multiple iterations of fine-tuning, to output the optimal planning scheme.The present application can significantly improve the differentiation degree and cross-modal consistency of enterprise image planning scheme, shorten the planning cycle, and realize intelligent, personalized automatic matching.

The application discloses a power spot market operation evaluation method based on a multi-factor evaluation model, relates to the technical field of power spot market operation evaluation, and comprises the following steps: performing initial market simulation based on back testing of historical data and obtaining a first state set; adopting a clustering algorithm to perform feature division on the first state set, taking the division result as a preset Bayesian network input signal, and constructing an operation evaluation decision model through the Bayesian network; deciding on a preset first action set according to the operation evaluation decision model to obtain a simulation alternation sequence; adopting a preset adversarial network to perform feature mapping on the simulation alternation sequence to obtain a projection offset feature; and supplementing the state transition probability of the preset Bayesian network according to the projection offset feature to optimize the operation evaluation decision model, thereby solving the problem of mode switching decision distortion caused by feature space mismatch in the traditional evaluation method.

The application discloses a power spot market simulation clearing method, which combines electrical distance and dynamic boundary output to perform spatial division; adopts a dynamic programming algorithm integrated with a multi-target cost function and unit constraint to perform time division; establishes a mixed integer programming model of a time-space block and a master-slave iteration framework with the mixed integer programming model as a sub-problem; the specific steps are as follows: S1, data acquisition and pretreatment; S2, spatial division; S3, time division; S4, generation of a time-space block and a mixed integer programming model of the time-space block; S5, establishment of a master-slave iteration framework; and S6, iteration solving process. The application further discloses a power spot market simulation clearing system.

The invention discloses a personalized credit product intelligent design system and method based on generative AI, and relates to the technical field of financial credit product design. The system comprises a data acquisition and processing module, a generative AI product design module, a market response simulation module, a default probability prediction module and a product scheme output module. Intelligent integration of multi-source data is realized through a dynamic weight fusion algorithm based on a Bayesian network; generating a personalized credit product scheme by adopting a conditional generative adversarial network and a variational automatic encoder; constructing a three-layer coupling market simulation model to predict market response; accurately predicting the default probability by using a time-varying t-Copula model and Monte Carlo simulation; and outputting an optimal scheme through a multi-objective optimization algorithm. According to the method, the data accuracy is improved to 98.7%, the product design efficiency is improved by 99.9%, the default prediction KS value reaches 0.75, the transformation of credit products from template design to intelligent and personalized design is realized, and a complete technical solution is provided for digital transformation of financial institutions.

The invention discloses a text travel consumption product popularity prediction and evaluation method based on artificial intelligence, and relates to the technical field of intelligent decision making, and the method comprises the steps: collecting text travel consumption product attribute data, user interaction data and environment event data, carrying out the data cleaning, standardization and correlation integration, and generating a market data set; defining a state space, an action space and an interaction rule according to the market data set, and creating a corresponding dynamic agent for each text travel consumption product; constructing a multi-agent reinforcement learning framework, inputting the dynamic agent into the multi-agent reinforcement learning framework, and generating a virtual market simulation environment; and driving the dynamic agents to select marketing actions according to the state attribute values in the current state space, and generating a market state evolution sequence according to the marketing actions executed by all the dynamic agents and the interaction rules. The market dynamic state is simulated through the multi-agent reinforcement learning framework, and the accuracy and reliability of popularity prediction are improved.

To provide a transaction behavior evaluation device capable of evaluating a transaction behavior not limited to a transaction actually performed in a market.SOLUTION: In a transaction action evaluation device 100, a strategic action model setting processing part 102 prepares strategic action data 151 for which a strategic action model being the model of the strategic action of a market participant is determined, and a peculiar action model setting processing part 103 prepares peculiar action data 152 for which a peculiar action model being the model of peculiar action being action different from the strategic action model is determined. A market simulation processing part 105 simulates a market according to the bidding action of a market participant determined from strategy action data 151 and peculiar action data 152, and action evaluation data 155 evaluates whether or not the market participant acts according to a strategy action model in the simulated market and the influence of the action taken by the market participant on the market.SELECTED DRAWING: Figure 1

A computer-implemented system simulates financial markets using blockchain-based assets to support trading, portfolio management, and market-dynamic gameplay. The system is implemented digitally and may optionally support physical or hybrid gameplay formats that integrate physical game components with digital portfolio interfaces. Players interact with blockchain-based assets, including cryptocurrencies, digital tokens, and non-fungible tokens, whose values dynamically adjust in response to simulated economic factors. Program modules apply rule-based logic to stored asset representations to update asset values and portfolio state during gameplay, synchronizing interactions across digital and physical formats. The system further supports reward mechanisms that grant digital assets or redeemable points and provides adaptive visualizations that reflect changing market conditions, enabling an interactive simulation of financial asset behavior.

The invention relates to a multi-dimensional electric power spot market simulation scene hybrid and reduction system and method, and the method comprises the steps: extracting time sequence characteristics and related structures of loads, photovoltaic power and wind power from historical data based on SOM clustering, a Markov chain and a Copula function; then introducing a reference output curve and constraint parameters such as a load level, wind power output, photovoltaic output and a supply-demand ratio, performing sequence reconstruction on the candidate day scene, generating a target scene time sequence power curve consistent with a planning scene, and finally realizing typical scene reduction through FCMC. While the number of scenes is remarkably reduced and the spot clearing and production simulation calculation amount is reduced, good approximation of the reduced scene set to the original multi-dimensional random characteristic is kept, so that the calculation scale of the electric power spot market clearing and production simulation is effectively reduced, the simulation efficiency and engineering availability are improved, and the economic benefit is increased. Therefore, standardized scene input is directly provided for simulation modules such as security constraint unit combination, spot clearing and flexibility evaluation.

The system according to this embodiment aims to propose optimal strategies and measures to respond to highly immediate market fluctuations. [Solution] The system according to the embodiment comprises a data collection unit, an analysis unit, a simulation unit, and a proposal unit. The data collection unit collects data. The analysis unit analyzes the data collected by the data collection unit. The simulation unit performs market simulations based on the analysis results obtained by the analysis unit. The proposal unit proposes optimal strategies and measures based on the simulation results obtained by the simulation unit.

The application discloses an enterprise rent pricing optimization decision system based on reinforcement learning, relates to the technical field of enterprise rent, and comprises a data extraction module, a market simulation module, a reinforcement learning module, a hierarchical evaluation module and a constraint control module.The data extraction module is used for collecting multi-modal leasing data of a target leasing unit and constructing each competitive leasing unit as a topological graph node.In the application, the reinforcement learning module relies on the market simulation module, performs multi-step forward deduction on a current pricing action and multiple sets of counterfactual pricing actions respectively, calculates expected cumulative income, and embeds the difference between the maximum counterfactual income and the current income as a regret value into a policy network loss function, so that the model not only pursues immediate rewards in training, but also actively avoids opportunity losses, effectively inhibits the short-sighted greedy behavior of traditional reinforcement learning, significantly reduces vacancy losses and income decay caused by short-sighted decision-making, and optimizes long-term income of assets.