304 results about "Price prediction" patented technology

The invention relates to an internet big data extraction method and device, equipment and a storage medium, and the method comprises the following steps: carrying out distributed crawler collection on an internet data source, obtaining original network data, and converting the original network data into a structured data matrix; and performing multi-level semantic analysis on the matrix, constructing a semantic feature map, and performing topic segmentation and classification to form a topic domain knowledge tree. Association rules in the knowledge tree are further mined, and an implicit knowledge network is constructed. Performing semantic decomposition and expansion on query conditions based on the network to generate expanded query data, and performing similarity matching with the knowledge network to obtain a candidate data set; and finally, performing multi-factor sorting and extraction on the candidate data, and outputting target data, thereby solving the technical problem that in the second-hand car market, due to wide data sources and fuzzy semantics, an existing system has relatively large deviation when performing price prediction and maintenance cost analysis.

The invention relates to the technical field of electric power market price prediction, in particular to an electric power spot price prediction method and system based on similar day probability correction, and the method comprises the following steps: S1, electric power data collection: obtaining historical market information of a day-ahead spot market and day-ahead electric power spot price data; s2, feature engineering construction: performing feature engineering construction on the collected power data; s3, data preprocessing: forming a data set; s4, performing model training: performing training on the LightGBM model; s5, similar day matching: screening k historical days with the highest similarity with the prediction day; s6, performing interval probability distribution modeling of the bidding space and the day-ahead electricity price: establishing a probability mapping matrix of the bidding space interval and the day-ahead electricity price interval; and S7, prediction result correction and final electricity price prediction: generating a final day-ahead spot electricity price prediction sequence. According to the invention, the method is closer to the actual market situation when the prediction logic is constructed, and the stability and adaptability of electricity price trend modeling are enhanced.

The invention belongs to the technical field of intelligent heat supply, and provides a green power consumption heat storage and release matching control method and system based on a water heat storage boiler, and the method comprises the steps: firstly obtaining a heat load prediction sequence and an electric power spot market price prediction sequence of a heat supply system; secondly, regarding the heat supply network as a graph structure composed of nodes and pipelines, and establishing a pipe network dynamic model for describing the hydraulic and thermal characteristics of the heat supply pipe network; and finally, with the purpose of minimizing the comprehensive operation cost of the heat supply system in the whole scheduling period, and with the whole network heat balance, heat source operation, centralized heat storage device operation, heat supply network safety and user comfort as constraint conditions, solving to obtain a source network load storage cooperative control strategy. The purpose of integrated collaborative optimization of the heat source, the energy storage, the pipe network and the load is achieved, a scheduling strategy with better economical efficiency can be generated, and the problems that the absorption capacity is insufficient, and the overall operation economical efficiency of the system is not high are solved.

The invention relates to the field of financial stock price markets, in particular to a stock price prediction method based on a Transform-KAN model, which comprises the following steps: S1, selecting a proper data set on a Kaggle platform, processing missing values and abnormal values of the stock price data set, dividing the data set into a training set and a test set, carrying out maximum-minimum normalization processing on the training set and the test set; s2, generating a time window feature by using a time sequence feature extraction method, and converting the time window feature into an input format suitable for a Transform structure; s3, a Transform-KAN model is constructed for stock price prediction, a long-term dependency relationship of a time sequence is extracted by using a Transform, and an MLP layer of the Transform is replaced by using KAN (Kolmogorov-Arnold Networks), so that the nonlinear fitting capability of the model is enhanced; s4, defining a loss function and an optimizer, and setting hyper-parameters such as a learning rate and a batch size; s5, inputting the data into the network, training the model by using the training set, and storing the trained model; and S6, loading the stored model, predicting the test set, and evaluating by using various evaluation indexes.

The invention provides a multi-mode hybrid expert model electricity transaction data processing method and system, a storage medium and electronic equipment. The method comprises the steps of collecting multi-source heterogeneous data in an electricity market environment; extracting multi-modal features of the multi-source heterogeneous data, and performing dynamic weighted fusion to obtain fusion features, wherein the multi-modal features comprise time sequence features, meteorological features, image features, text features and market features; different expert models including a load prediction model, a price prediction model, a risk assessment model and a strategy optimization model are allocated for processing and analysis results are obtained according to the fusion features; and generating output data in combination with the analysis result, wherein the output data comprises a service instruction and an analysis report. According to the method, the power transaction prediction precision is remarkably improved, the reasoning delay is greatly reduced, the dynamic adaptive capacity and a closed-loop self-evolution mechanism are achieved, the decision interpretability and the system robustness are enhanced, and the strict requirements of the power market for high precision, real-time performance and stability can be met.

The invention relates to the technical field of electric power spot market real-time price prediction, and discloses an electric power spot market real-time price prediction method based on low-sample high-dimension data, and the method comprises the steps: extracting historical data from a database; carrying out feature extraction from historical data, and carrying out standardization processing on the extracted features; constructing a real-time electricity price prediction model by adopting a generative neural network model, and inputting the extracted features of the historical data into the real-time electricity price prediction model to obtain a first predicted electricity price; inputting the first predicted electricity price into a heuristic predictor for adjustment to obtain a second predicted electricity price; adjusting factors for adjusting the first predicted electricity price by the heuristic predictor comprise similarity, a trend factor, a seasonal factor and a special event adjusting factor, and fusing the adjusting factors to comprehensively adjust the first predicted electricity price; existing electricity price prediction cannot be reasonably quantified due to external factors in the early stage of development of an electric power spot market, so that the prediction result is inaccurate.

The invention relates to the technical field of bulk commodities, in particular to a bulk commodity transaction price prediction method, which comprises the following steps: acquiring multi-modal data of bulk commodities in the current period; the multi-modal data comprises meteorological data, macroeconomic index quantized values and news event analysis parameters processed by natural languages; the news event analysis parameters comprise market emotion indexes and policy change influence factors; performing data preprocessing on the multi-modal data of the bulk commodities in the current period to obtain preprocessed multi-modal data; inputting the preprocessed multi-modal data into a trained price prediction model to obtain a final prediction price; the price prediction model is a hybrid neural network model fused with an LSTM-Transformer (LSTM-Transformer). According to the method, the accuracy of bulk commodity price prediction is improved, and the calculation efficiency is high.

The invention provides a real-time commodity price prediction system based on multi-source data fusion, and the system comprises a data integration module which is used for obtaining multi-source heterogeneous data, carrying out the alignment processing of the multi-source heterogeneous data according to the commodity types and timestamps, and generating a standardized fusion data set; the dynamic weight distribution module is used for dynamically calculating the weight coefficient of each data source according to the relevance between each data source in the standardized fusion data set and the commodity category; the model integration module is used for performing weighted fusion on the price trend prediction result output by the time sequence prediction model and the game correction output by the market game model based on the weight coefficient to generate initial price prediction; and the optimization output module is used for performing periodic rule correction and strategy collaborative analysis on the initial price prediction according to the real-time market dynamic data to obtain a predicted commodity price trend. By adopting the system, the commodity price trend can be dynamically predicted based on multi-source big data, an enterprise is helped to optimize a purchasing strategy, and the inventory overstock risk is reduced.

The invention discloses an intelligent energy-saving scheduling system and method based on an industrial computing power data center, and the method comprises the steps: collecting all kinds of operation data in the data center and external environment data in real time, and carrying out the preprocessing and cleaning of the data; using a machine learning algorithm to predict future load demand and energy price fluctuation; in combination with load prediction and energy price prediction, a multi-objective optimization algorithm is adopted to generate a resource scheduling plan; dynamically adjusting the working state of the cooling system according to the load change and the environment temperature and humidity data; executing a task according to the scheduling plan, feeding back according to the real-time monitoring data and an execution result, and adjusting a scheduling strategy; according to the intelligent energy-saving scheduling system and method based on the industrial computing power data center, by introducing an intelligent algorithm and multi-source data analysis, the energy utilization efficiency of the data center is improved, the operation cost is reduced, and optimal scheduling is achieved under the condition of dynamic load and energy price fluctuation.

The invention relates to an optimal scheduling method and system for participation of actual energy storage and virtual energy storage in a power market, and a medium. The method comprises the steps of collecting user historical load, historical spot transaction price and local medium and long term transaction time-of-use electricity price data; predicting the user total load, the virtual energy storage capacity and the local day-ahead electric power spot transaction price in the next day; constructing a hybrid energy storage model based on the actual energy storage physical property parameters; the minimum user operation cost is set as an optimization target, and constraint conditions are set according to the actual energy storage capacity, the virtual energy storage capacity, the battery charging and discharging efficiency, the transformer capacity and the like; and a genetic algorithm is adopted to obtain a 24-hour actual energy storage and virtual energy storage collaborative scheduling strategy in the next day. According to the method, through accurate load, schedulable capacity and price prediction, the optimization algorithm is adopted to carry out collaborative optimization of actual energy storage and virtual energy storage, and the economic benefit and flexibility of the whole system are remarkably improved.

The invention specifically discloses a lithium carbonate price prediction method and device based on a large language model, and a medium, and relates to the technical field of financial prediction and intelligent decision support. The method comprises the following steps: firstly, collecting and processing numerical type and text type data, constructing a causal graph network by the numerical type data, and extracting factual abstracts by the text type data to construct a news event library; then designing a three-layer progressive prompt template, and driving the large language model to generate a prediction result with explanation; then, implementing a dual-verification self-optimization reflection mechanism, and triggering an reflection generation improvement strategy when misjudgment is carried out or an error exceeds a threshold value; using a KTO algorithm to construct asymmetric loss function fine tuning model parameters; and finally, deploying the optimized model to a production system. According to the method, the dynamic adaptability and robustness of the model are improved through a'prediction-reflection-optimization 'closed-loop optimization architecture and a KTO algorithm driven by a foreground theory, prediction precision and interpretability are considered, and multi-dimensional decision support is provided for lithium carbonate price prediction.

The invention relates to the technical field of real estate, and discloses an ant colony algorithm and neural network combined house price data prediction method, which comprises the following steps: S1, multi-source data acquisition, S2, data preprocessing, S3, ant colony algorithm parameter optimization, S4, neural network dual-stage training and S5, adaptive optimization feedback. According to the house price data prediction technology combining the ant colony algorithm and the neural network, neural network model parameters are optimized by adopting the ant colony algorithm; in the house price prediction model, the parameters are combined to form a path, and ants traverse different parameter combination paths; along with iteration, pheromones on a better parameter combination path are continuously accumulated and enhanced, and ants are guided to explore more; in this way, the parameter space can be fully traversed, local optimum caused by parameter random initialization of a traditional neural network is avoided, and therefore the model can obtain a better parameter combination, and the prediction precision is remarkably improved.

The invention belongs to the field of financial science and technology and data analysis, and particularly relates to a financial transaction emotion analysis and price trend prediction method based on data fusion, which is mainly used for solving the problems of complexity and real-time demand of stock price prediction in a financial market. According to the method, a fundamental analysis module, an emotion evaluation module and a price prediction module are integrated, multiple data sources are integrated, an advanced machine learning algorithm is adopted, and the real dynamic state of the market is more comprehensively and accurately captured. The fundamental analysis module is used for systematically analyzing financial data, announcement information and other related indexes of the stock and providing deep understanding of a target stock; the emotion evaluation module analyzes text data in social media and financial forums and combines various emotion analysis methods to generate dynamic indexes reflecting market emotions; the price prediction module carries out modeling based on data of different time scales such as minute, hour and day lines, and the prediction precision is improved through a layer-by-layer feedback mode.

The invention discloses an electric power price prediction method and device based on time dimension, equipment and a storage medium, and relates to the technical field of electric power systems, and the method comprises the steps: carrying out the preprocessing of a multi-source data set comprising historical electric power transaction data, a load demand curve, energy production data and fuel cost, performing feature extraction on the obtained processed data set to obtain each data feature, and determining a training set and a verification set based on the data features by using Bernoulli distribution; constructing an initial power price prediction model based on a multi-head attention mechanism, a convolutional layer, a preset activation function, a bidirectional long-short-term memory network and integration of a low-rank adaptation technology, and training and verifying the initial power price prediction model by using the training set and the verification set respectively; and performing multi-scale prediction by using the obtained target power price prediction model to obtain prediction results of the power price in different time dimensions. And the real-time, day-ahead and weekly prediction accuracy of the electric power price is improved.

The invention relates to the technical field of intelligent pricing, in particular to an intelligent pricing management system based on machine learning and data analysis, and aims to improve the purity and consistency of data and lay an accurate data foundation for subsequent analysis through cleaning and format unified processing of market original data and elimination of redundant information. Through a non-dominated sorting genetic algorithm, the influence weight of each data point is analyzed, comprehensive quantitative modeling of price influence factors is realized, a model basis with better interpretation and adaptability is provided for price prediction, and market changes and competition patterns are dynamically captured by combining a linear regression model, so that the price prediction accuracy is improved. And the price is ensured to be dynamically and accurately adjusted along with the market through a fine adjustment strategy, the dynamic matching capability between the price and the market fluctuation is realized, the adaptability of the strategy under different market conditions is verified and optimized through a multi-scene simulation test and market feedback monitoring, and the high efficiency and stability of the price adjustment strategy in practical application are ensured.

The invention provides a small sample electricity market electricity price prediction method and system based on meta-learning, and belongs to the technical field of electricity market analysis based on artificial intelligence, and the method comprises the steps: obtaining the operation state data of an electric power system of a to-be-predicted region; and processing the acquired operation state data of the power system by using a pre-trained electricity price prediction model to obtain an electricity price prediction result. By constructing the cross-domain common feature extraction network and the small sample meta-learning framework, the problem of prediction model failure caused by data scarcity and frequent rule change of a new electricity market is solved, and the accuracy, generalization and adaptability of electricity price prediction in a data limited scene are remarkably improved.

The invention discloses a financial time sequence stock price prediction method and system based on deep learning. The method comprises the following steps: data preprocessing; constructing an improved generative adversarial network (GAN) model; training and optimizing the model; and outputting a prediction result. According to the financial time sequence stock price prediction method and system based on deep learning, the CNN, the LSTM, the attention mechanism and the improved GAN model are combined, the long-term dependence problem of stock data can be effectively solved, complex modes and trends in the data are captured, the prediction accuracy is improved, and the prediction accuracy is improved from experimental data. The improved GAN-LSTM-CNN-Attention model is excellent in performance on indexes such as MAPE, RMSE, MAE and R2, compared with models such as LSTM, CNN and LSTM-CNN, the error is lower, the fitting effect is better, more reliable price prediction reference can be provided for investors, and the generalization ability of the model is enhanced by adopting multiple strategies to reduce model overfitting such as data packet training, network structure simplification and use of proper activation functions.

The invention provides a power transaction decision-making method, device and equipment and a storage medium, and the method comprises the steps: obtaining time sequence operation data of a power market; inputting the time sequence operation data into a pre-trained transaction decision model, and obtaining predicted transaction data output by the transaction decision model; wherein the transaction decision-making model comprises a price prediction sub-model and a transaction decision-making sub-model, the price prediction sub-model obtains the capability of predicting and obtaining price prediction data based on time sequence operation data through pre-training, and the transaction decision-making sub-model obtains the capability of generating a transaction decision based on the price prediction data through pre-training; the activation function of the transaction decision model is composed of a plurality of nonlinear activation functions, and the transaction decision model is obtained by training with the prediction error of the power price and the prediction error of the decision income as loss functions. According to the technical scheme, support can be provided for transaction decision-making of the electricity market.

The invention discloses an electricity price prediction method and system based on fluctuation characteristic decoupling, and the method comprises the steps: obtaining electricity market data including historical electricity prices and exogenous variable characteristics, and carrying out the data preprocessing; decoupling the preprocessed historical electricity price data into an input sequence for regression prediction and a label sequence for classification prediction; predicting a regression electricity price in a future time period and predicting a time period when a low electricity price event occurs; and integrating the regression electricity price of the future time period and the time period in which the low electricity price event occurs, carrying out cascade connection to obtain a final future electricity price prediction sequence, and outputting the final future electricity price prediction sequence. According to the scheme of the invention, a conventional single and difficult electricity price prediction problem is decomposed into a conventional regression prediction task and a special low-electricity-price event classification task, so that the prediction accuracy of extremely low-price events such as zero electricity price and negative electricity price can be remarkably improved.

The invention discloses an aviation equipment intelligent pricing method and system and a medium, and the method comprises the steps: carrying out the fusion processing and feature extraction of multi-source heterogeneous data, such as aviation material internal transaction data, external market information data, supply chain state data, and regulation event text data, and generating a standardized feature vector; utilizing a natural language model to extract structured event features from the law and regulation event text data; inputting the standardized feature vector into a pre-training price prediction model adopting a mixed architecture of an integrated learning algorithm and a time sequence deep learning algorithm, and outputting a reference price prediction value and a price prediction interval; according to a pricing strategy instruction selected by a user, a final pricing result is generated through a strategy mapping module, and an interpretable price cause analysis report is output, so that the accuracy and adaptability of aerial material pricing are effectively improved, the dynamic response of various pricing strategies is supported, the transparency and interpretability of a decision-making process are enhanced, and the user experience is improved. And scientific and reliable intelligent decision support is provided for aviation equipment pricing.

The invention discloses a bulk commodity transaction market stabilization method based on a BSL-CPFS deep learning model. The method comprises the following steps: preprocessing a voice data set into a training set and a test set; introducing a DeepSpeech2 model as a speech transcription model, and training the speech transcription model by using a training set and a text set; constructing a BSL model based on a BERT model and a TextCNN, and inputting the speech transcription text data training set for training to obtain a BSL language processing model; an NER model is introduced, the data set is labeled, and the NER model is input for training to form structured data; constructing a CPFS model based on a BiLSTM network, and inputting structured data for training to obtain a price prediction model; a prediction result and historical price fluctuation are dynamically compared to identify an abnormal fluctuation signal and push the abnormal fluctuation signal in real time, the asymmetry of information is reduced, a supervision department continuously monitors the deviation between the market price and the prediction result and formulates a targeted intervention strategy, then the price correction is realized, the stability of a bulk commodity transaction market is improved, and the market competitiveness is improved. And the market pricing transparency is improved.

The invention discloses an artificial intelligence-based economic crop planting management method and system. The method comprises the steps of obtaining planting management original data, optimizing the original data, selecting target economic crops, predicting market prices of the economic crops, optimizing economic crop planting strategies and intelligently managing economic crop planting. The invention relates to the technical field of economic crop planting data processing, in particular to an artificial intelligence-based economic crop planting management method and system, which innovatively takes market price prediction as a guide and integrates a method system of target crop intelligent screening and full-cycle planting strategy optimization. The planting income stability of commercial crops is improved, price data signal decomposition processing is introduced, and the quality of commercial crop market price prediction input data is improved; a market price prediction model is innovatively designed, and the accuracy and stability of a model prediction result are improved; and an improved particle swarm optimization algorithm is introduced to realize efficient global optimization of the economic crop planting strategy combination.

The invention relates to the technical field of electricity price prediction, and discloses an electricity price prediction method based on a random forest algorithm. Comprising the steps of generating a first feature data set based on collected whole network load prediction data, new energy load prediction data, tie line plan data, necessary stop unit data, necessary start unit data, holiday sign data, total maintenance capacity data and historical day-ahead electricity price data; constructing a power price preliminary prediction model group based on the first feature data set; correcting the electric power price preliminary prediction model group in combination with thermal power bidding space data to generate an electric power price final prediction model; and generating an electricity price prediction curve of the to-be-predicted day in combination with the obtained feature data of the to-be-predicted day and the final electricity price prediction model. According to the method, prediction is carried out by constructing a plurality of decision trees and synthesizing the results of the decision trees, so that the model can better adapt to different data distributions, and the prediction accuracy is improved.

The invention discloses an electric power intra-day price prediction method based on dynamic holiday weight and multi-source fusion, and the method comprises the steps: obtaining multi-source historical data, and carrying out the time synchronization processing; holiday and festival time information is acquired, and a dynamic weight is generated based on the influence of holidays and festivals and upstream and downstream dates on the power load and the electricity price; constructing a multi-dimensional predictive factor matrix based on the multi-source historical data after time synchronization processing; and according to the multi-dimensional predictive factor matrix, on the basis of a collaborative optimization multi-model combination comprising a feedforward neural network model and a bagged regression tree model, intra-day joint prediction is executed, the intra-day joint prediction refers to a process of predicting the power load and the electricity price hourly, and an hourly prediction result of the power load and the electricity price is output. A dynamic holiday weight mechanism is introduced, a multi-source fused high-dimensional predictive factor matrix is constructed, and a multi-model combined predictive strategy of collaborative optimization of a feedforward neural network and a bagged regression tree is adopted, so that the precision and stability of intra-day electricity price and load prediction of the electricity market are improved.

A computer-implemented system for short-term stock price prediction, consisting of: • a data acquisition module (101) configured to retrieve historical intraday stock data including closing prices, trading volumes and technical indicators; • a data preprocessing module (102) configured to normalize the data and generate temporally consecutive samples using a sliding window approach; • a neural network module (103) comprising a Gated Recurrent Unit (GRU) architecture configured to receive the temporally consecutive samples and output intermediate stock price predictions; • a gradient boosting module (104) comprising an Extreme Gradient Boosting (XGBoost) regressor configured to receive the output of the GRU module along with associated residual features and to refine the predictions by modeling nonlinear patterns and residual errors; • and an output module (105) configured to present the final refined stock price prediction, where the GRU module is trained to minimize a loss function for the mean squared error, and the XGBoost module is trained to minimize an objective function for the squared error, thereby providing improved predictive accuracy for financial time series forecasts.

The invention belongs to the technical field of virtual power plants, and particularly relates to a virtual power plant day-ahead strategy optimization method in a spot mode, which comprises the following steps of: performing correlation analysis and feature extraction on Spearman time-phased medium and long-term markets, spot prices and market boundary information; a time-phased spot market price prediction model is formed, and time-phased spot market prices are predicted; adopting a bidding space linear regression method to carry out time-phased medium-and-long-term market transaction price prediction; a difference algorithm is adopted to establish a virtual power plant electric power spot market price declaration strategy optimization model; and generating a virtual power plant spot market declaration strategy. According to the method, the correlation between the market boundary factor-based time-sharing spot price and the medium and long term transaction price is analyzed, the market boundary data is utilized to predict the future time-sharing thermal power load rate mean value, and the future time-sharing thermal power load rate mean value is input into the pre-trained model, so that the accurate prediction of the time-sharing average price of the electric power spot market is realized.

The invention discloses an apple industry chain multi-source heterogeneous data coupling prediction and early warning method and system, and belongs to the field of agricultural information technology and intelligent prediction.The method comprises the steps that price data, weather data, Internet of Things sensor data and inventory data of an apple industry chain are collected, and vectorization is conducted after cleaning and dynamic time warping alignment; calculating a disaster influence factor alpha and a price conduction time lag tau based on the disaster event; inputting the multi-source data vector into a prediction model fused with bidirectional LSTM, multi-head cross attention and disaster sensing gating modules, and outputting a price prediction sequence; the invention aims to solve the problems of single data, lagging disaster response and the like in the prior art.

The invention provides an urban road transportation price prediction method and device, electronic equipment and a storage medium. The method comprises the following steps: collecting vehicle data, cargo data, route data, text data and original time sequence data; performing first preprocessing on the vehicle data, the cargo data and the first part of route data to obtain structural non-time-sequence features, performing second preprocessing on the second part of route data and the text data to obtain category features, and performing third preprocessing on the original time sequence data to obtain structural time sequence features; converting the structural non-time-sequence features, the category features and the structural time-sequence features into numeric features; inputting the numerical features into a freight rate regression prediction model, and outputting a freight rate prediction result; according to the method, feature extraction is carried out on the text data through the large language model, and the text data are converted into structured features, so that organic fusion of the deep time sequence model and the large language model is realized, high-precision freight rate dynamic prediction is realized, and the real-time response capability to emergencies is improved.

The invention belongs to the technical field of computer application, and discloses an anticipated price rolling matchmaking transaction aided decision-making method based on a big data algorithm, which integrates microscopic transaction data, mesoscopic market dynamics and macroscopic environment variables through a system, enables price prediction to more comprehensively fit the actual operation logic of the market, and improves the efficiency of price prediction. The deep forest model has a strong cross-modal feature learning ability, and can deeply mine hidden associations between different types of data, such as indirect influences between international situation changes and bulk commodity prices, conduction effects of upstream and downstream price linkage of an industrial chain on transaction varieties, and the like. A more three-dimensional and more reliable expected basis is provided for transaction decision making, so that a prediction result can better reflect the complex full view of the market, and prediction deviation caused by one-sided information is reduced; through a prediction framework combining a dynamic sliding window and incremental learning, subtle changes of the market can be tracked in real time, and it is ensured that the model is always kept synchronous with the market rhythm.

The invention relates to the technical field of electric power, provides an electric power transaction strategy generation method and system based on Monte Carlo simulation and deep learning, and solves the problem of poor scientificity and adaptability of an optimal transaction strategy of a multi-region electric power market in the prior art. The method comprises the following steps: acquiring historical price data of a multi-region power market, decomposing and extracting trends, periods and random components, and constructing a multi-dimensional feature set; the method comprises the following steps: receiving unstructured data by using a block link, packaging the unstructured data into a structured data packet through Hash verification and a smart contract, and converting the structured data packet into a multi-dimensional semantic feature vector based on a preset mapping table; associating the two types of features through a deep learning multi-modal model, and outputting price trend prediction probability distribution data with confidence; and generating a plurality of contract power splitting schemes based on Monte Carlo simulation, combining price prediction, searching a global optimal solution by using a particle swarm optimization algorithm, and generating an optimal power transaction strategy. According to the invention, the scientificity and adaptability of the optimal transaction strategy of the multi-region power market are improved.