Multi-agent electricity price prediction intelligent computing cloud platform system based on autonomous feature evaluation
The multi-agent electricity price forecasting system, which uses autonomous feature evaluation, dynamically optimizes the selection of feature variables and data collection, solving the problems of low accuracy and high cost in traditional electricity price forecasting, and achieving efficient and real-time electricity price forecasting.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Filing Date
- 2026-03-15
- Publication Date
- 2026-07-10
AI Technical Summary
Traditional electricity price forecasting methods struggle to fully capture the combined effects of complex, multi-dimensional factors, resulting in limited forecast accuracy and generalization ability. Furthermore, the selection of feature variables relies on manual collaborative work, which is costly, inefficient, and makes it difficult to achieve real-time dynamic updates.
A multi-agent intelligent computing cloud platform system for electricity price prediction based on autonomous feature evaluation is adopted. By mining candidate feature variables through feature evaluation agents, and combining data acquisition, preprocessing and model evaluation agents, a closed-loop processing flow is formed to dynamically optimize feature variable selection and data collection, thereby realizing the automated screening and updating of feature variables.
It improves the accuracy and efficiency of electricity price forecasting, reduces interference from redundant features, ensures the continuous improvement and real-time performance of the forecasting model, reduces the subjectivity and cost of feature selection, and adapts to the rapid changes in the electricity market.
Smart Images
Figure CN122367530A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of cloud computing technology, and in particular to a multi-agent intelligent computing cloud platform system for electricity price prediction based on autonomous feature evaluation. Background Technology
[0002] As the core regulatory signal for the operation of the electricity market, accurate forecasting of electricity prices is crucial for power companies to formulate production strategies. Electricity prices are influenced by complex, multi-dimensional factors, exhibiting nonlinear, time-varying, and multi-scale coupling characteristics. Traditional, single-method electricity price forecasting approaches struggle to fully capture the combined effects of various factors, resulting in significant limitations in forecast accuracy and generalization ability.
[0003] In related technologies, electricity price forecasting mainly relies on manually selecting feature variables for prediction. The lack of scientific quantitative evaluation of the correlation between candidate features and electricity prices easily introduces redundant features or omits key influencing factors, resulting in insufficient input information for the prediction model and consequently hindering the continuous improvement of prediction accuracy. Summary of the Invention
[0004] This application aims to at least partially address one of the technical problems in the related art.
[0005] Therefore, this application proposes a system, apparatus, electronic device, and storage medium.
[0006] One embodiment of this application proposes a multi-agent intelligent computing cloud platform system for electricity price prediction based on autonomous feature evaluation, including: Feature evaluation agent, data acquisition agent, data preprocessing agent, electricity price prediction agent, model evaluation agent; The feature evaluation agent is used to mine candidate feature variables, determine target feature variables based on the correlation between the candidate feature variables and electricity prices, and send the target feature variables to the data acquisition agent. The data acquisition agent is used to receive the target feature variable, collect feature data based on the target feature variable and preset basic feature variables, and send the feature data to the data preprocessing agent; wherein, the basic variable and the candidate feature variable are feature variables related to power production; The data preprocessing agent is used to preprocess the feature data and send the preprocessed feature data to the electricity price prediction agent. The electricity price prediction agent is used to call the prediction model, make a prediction based on the preprocessed feature data, and output the predicted value of the electricity price. The model evaluation agent is used to evaluate the prediction accuracy based on the actual value and the predicted value of the electricity price, and to instruct the feature evaluation agent to mine candidate feature variables based on the prediction accuracy.
[0007] Optionally, the feature evaluation agent is used for: Retrieve data related to power production and extract feature variables related to power production as candidate feature variables; Based on the predicted value of the electricity price output by the electricity price prediction agent, a first correlation coefficient between each candidate feature variable and the predicted value is determined; Calculate the second correlation coefficient among each of the candidate feature variables; The confidence level of the candidate feature variables is determined based on the timeliness of their acquisition and the frequency of their update. The target feature variable is determined from the candidate feature variables based on the first correlation coefficient, the second correlation coefficient, and the confidence level.
[0008] Optionally, determining the target feature variable from the candidate feature variables based on the first correlation coefficient, the second correlation coefficient, and the confidence score includes: The candidate feature variables are filtered based on the first correlation coefficient, and candidate feature variables with a first correlation coefficient higher than a preset first threshold are retained. The candidate feature variables are screened based on the second correlation coefficient. In response to the existence of multiple candidate feature variables whose second correlation coefficient is higher than a preset second threshold, one of the multiple candidate features is retained. The candidate feature variables are filtered based on the confidence level, and candidate feature variables with a confidence level higher than a preset third threshold are retained; The candidate feature variables that remain after the screening are determined as the target feature variables.
[0009] Optionally, the feature evaluation agent is also used for: The target feature variables are input into the model evaluation agent to train the prediction model; Determine the accuracy of the trained prediction model, and determine whether to retain the target feature variable based on the improvement in accuracy.
[0010] Optionally, the data acquisition agent is used for: Based on the target feature variables and basic feature variables, the corresponding data collection channels are determined; According to the preset acquisition strategy, the feature data corresponding to the target feature variable and the basic feature variable are acquired from the acquisition channels.
[0011] Optionally, the data preprocessing agent is used for: The feature data is filtered to identify outliers and missing values, and corrections are made for the outliers and missing values. The corrected feature data is then standardized. The standardized feature data is converted into a preset format and sent to the electricity price prediction agent.
[0012] Optionally, the correction for the outliers and missing values includes at least one of the following: Interpolation is performed based on the adjacent values of the outlier to obtain a corrected value, and the outlier is replaced by the corrected value. The feature data is averaged to obtain a first fill value, and the missing values are filled according to the first fill value. The data adjacent to the missing value are averaged to obtain a second filler value, and the missing value is filled according to the second filler value.
[0013] Optionally, the standardization process for the corrected feature data includes: Determine the maximum and minimum values of the corresponding feature data for each target feature variable and basic feature variable, and calculate the difference between the maximum and minimum values as a reference value; Calculate the first difference between each feature data corresponding to the target feature variable and the minimum value; Divide the first difference by the reference value to obtain the standardized feature data.
[0014] Optionally, the model evaluates the agent for: A second difference is calculated based on the actual value of the electricity price and the predicted value, and the prediction accuracy is determined based on the second difference; If the prediction accuracy is lower than a preset fourth threshold, the feature variable to be replaced is determined, and the feature evaluation agent is instructed to re-mine the candidate feature variable to replace the feature variable to be replaced.
[0015] Optionally, determining the feature variable to be replaced includes: Calculate the third correlation coefficient between the target feature variable, the basic feature variable, and the predicted value; The target feature variable or basic feature variable whose third correlation coefficient is lower than the preset fifth threshold is determined as the feature variable to be replaced.
[0016] Optionally, the data acquisition agent is also used for: Delete the data collection channel corresponding to the feature variable to be replaced.
[0017] Another embodiment of this application proposes an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the system as described in the foregoing aspect.
[0018] Another embodiment of this application proposes a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the system as described in the foregoing aspect.
[0019] Another embodiment of this application proposes a chip including processing circuitry configured to execute the system as described in the foregoing aspect.
[0020] Another embodiment of this application proposes a computer program product that, when executed by a processor, implements the system as described in the foregoing aspect.
[0021] The multi-agent intelligent computing cloud platform system, device, electronic equipment, chip, and storage medium for electricity price prediction based on autonomous feature evaluation proposed in this application can achieve the following beneficial effects: The feature evaluation agent actively mines candidate feature variables and selects target feature variables based on their correlation with electricity prices. This optimizes the prediction input dimension from the source, reduces redundant and irrelevant feature interference, lowers data processing complexity, and lays the foundation for subsequent high-precision prediction.
[0022] The model evaluation agent quantifies the prediction accuracy based on actual and predicted values, and uses the evaluation results to instruct the feature evaluation agent to re-mine candidate features. This enables continuous updates to the feature variables used to predict electricity prices, ensuring continuous improvement in the system's prediction accuracy over long-term operation, and enhancing prediction efficiency and real-time performance.
[0023] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description
[0024] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A schematic diagram of the structure of a multi-agent intelligent computing cloud platform system for electricity price prediction based on autonomous feature evaluation is provided in this application embodiment; Figure 2 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application; Figure 3This is a schematic diagram of the structure of a chip proposed in an embodiment of this application. Detailed Implementation
[0025] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.
[0026] Electricity price forecasting is a core part of power plant production decisions. Electricity prices are affected by a combination of factors, and the selection and optimization of characteristic variables are key prerequisites for improving the accuracy of electricity price forecasting.
[0027] In the existing electricity price forecasting process, the discovery and selection of feature variables heavily relies on the collaborative work of experts from multiple fields. Experts need to manually read literature, analyze patterns to discover potential features, then rely on experience to select effective features that are highly correlated with electricity prices, and finally guide staff to expand the scope of data collection. This approach has significant drawbacks: 1) High expert costs, with significant time and economic costs associated with collaboration among experts from multiple fields; 2) Low efficiency in feature mining and selection, as manual operation cannot achieve real-time, dynamic feature updates, making it difficult to adapt to the rapidly changing needs of the electricity market; 3) Strong subjectivity in feature selection, with differences in experience among different experts easily leading to biases in feature selection results, affecting the accuracy of subsequent forecasts; 4) In existing automated forecasting schemes, the feature extraction stage does not form a closed-loop linkage with other stages, and feature optimization cannot drive the iterative upgrade of data collection and model prediction.
[0028] The following describes an embodiment of the intelligent computing cloud platform system for multi-agent electricity price prediction based on autonomous feature evaluation, with reference to the accompanying drawings.
[0029] This application proposes a multi-agent intelligent computing cloud platform device for electricity price prediction based on autonomous feature evaluation.
[0030] Figure 1 This is a schematic diagram of the structure of a multi-agent intelligent computing cloud platform system for predicting electricity prices based on autonomous feature evaluation, provided in an embodiment of this application.
[0031] like Figure 1 As shown, the system includes: a feature evaluation agent 10, a data acquisition agent 20, a data preprocessing agent 30, an electricity price prediction agent 40, and a model evaluation agent 50. The feature evaluation agent is used to mine candidate feature variables, determine target feature variables based on the correlation between the candidate feature variables and electricity prices, and send the target feature variables to the data acquisition agent. The data acquisition agent is used to receive the target feature variable, collect feature data based on the target feature variable and preset basic feature variables, and send the feature data to the data preprocessing agent; wherein, the basic variable and the candidate feature variable are feature variables related to power production; The data preprocessing agent is used to preprocess the feature data and send the preprocessed feature data to the electricity price prediction agent. The electricity price prediction agent is used to call the prediction model, make a prediction based on the preprocessed feature data, and output the predicted value of the electricity price. The model evaluation agent is used to evaluate the prediction accuracy based on the actual value and the predicted value of the electricity price, and to instruct the feature evaluation agent to mine candidate feature variables based on the prediction accuracy.
[0032] In this embodiment, multiple intelligent agents work together to form a closed-loop processing flow of "feature evaluation - data expansion - model optimization - accuracy feedback - feature re-optimization". The core is to realize the dynamic feature evaluation and selection of feature variables through the feature evaluation intelligent agent.
[0033] The time series of electricity price forecasts is ,in , The current moment; The target forecast value includes the day-to-day price. and intraday real-time prices ; for The set of characteristic variables at time 1, , The total number of feature variables includes target feature variables and preset basic feature variables. Basic feature variables consist of fundamental features such as power plant-related data and meteorological data. These are subsequently dynamically expanded into a feature set containing multi-dimensional related factors through a feature evaluation agent. The core objective of the system is to optimize the feature variable set through dynamic feature evaluation and selection by the feature evaluation agent. And through the collaboration of various intelligent agents, accurate prediction of electricity prices can be achieved.
[0034] Optionally, the predicted electricity price can help power plants adjust their production strategies. Lower electricity prices indicate lower demand, allowing them to reduce power generation and conserve resources to avoid waste. Conversely, higher prices indicate higher demand, allowing them to increase power generation to ensure a sufficient supply. This ensures the efficient use of power production resources (such as power plant equipment and fuel) and prevents waste.
[0035] Optionally, the feature evaluation agent is used for: Retrieve data related to power production and extract feature variables related to power production as candidate feature variables; Based on the predicted value of the electricity price output by the electricity price prediction agent, a first correlation coefficient between each candidate feature variable and the predicted value is determined; Calculate the second correlation coefficient among each of the candidate feature variables; The confidence level of the candidate feature variables is determined based on the timeliness of their acquisition and the frequency of their update. The target feature variable is determined from the candidate feature variables based on the first correlation coefficient, the second correlation coefficient, and the confidence level.
[0036] Optionally, determining the target feature variable from the candidate feature variables based on the first correlation coefficient, the second correlation coefficient, and the confidence score includes: The candidate feature variables are filtered based on the first correlation coefficient, and candidate feature variables with a first correlation coefficient higher than a preset first threshold are retained. The candidate feature variables are screened based on the second correlation coefficient. In response to the existence of multiple candidate feature variables whose second correlation coefficient is higher than a preset second threshold, one of the multiple candidate features is retained. The candidate feature variables are filtered based on the confidence level, and candidate feature variables with a confidence level higher than a preset third threshold are retained; The candidate feature variables that remain after the screening are determined as the target feature variables.
[0037] In this embodiment, the core execution unit of the system is responsible for the automatic discovery, dynamic screening, validity verification, and feedback update of electricity price-related features, thereby forming a dynamic optimization of the feature variables throughout the entire process.
[0038] During the retrieval of power production-related data, multi-channel automated literature and data retrieval was initiated. The search scope included core journals in the power industry, power plant data, and other resources. Quantitative data on physical attributes directly related to power system operation were mined within the power industry. Natural Language Processing (NLP) technology was used to extract keywords, perform semantic analysis, and mine patterns in the search results to uncover connections with electricity prices. Candidate feature variables that have potential correlations form a set of potential features. , The number of potential features includes, but is not limited to: power production-related feature variables. (I) Physical characteristics of power plants: These characteristics reflect the physical state of the power production process, directly affecting power supply capacity and serving as a core physical factor influencing electricity prices. Specifically, they include: Thermal power units: number of units in operation, unit output power, unit utilization hours, boiler combustion efficiency, unit start-up / shutdown status (quantified as 0 / 1, start-up / shutdown as 1, operation as 0); New energy units (photovoltaic / wind power / hydropower): photovoltaic irradiance, wind turbine wind speed, hydropower inflow, actual power generation of new energy units, curtailment rate of solar / wind power, and prediction deviation of new energy output; Energy storage units: State of charge (SOC), energy storage charging and discharging power, and energy storage charging and discharging duration.
[0039] (II) Physical characteristics of the power grid side: This primarily reflects the physical constraints of power transmission / dispatch. Physical bottlenecks in power transmission directly affect regional electricity prices, specifically including: Power grid operation: regional power flow value, transmission and transformation line load rate, substation bus voltage, transmission capacity of transmission channels, and degree of grid congestion (quantified as a congestion coefficient, 0 for no congestion and 1 for complete congestion). Power grid dispatching: cross-regional power transmission volume, power grid reserve capacity, peak-shaving unit response speed, and load adjustment amount of dispatching instructions.
[0040] (III) Load-side physical characteristics: This core reflects the physical demand in the electricity consumption process. Changes in the physical load are a direct cause of electricity price fluctuations, specifically including: Regional load: total system power load, peak-valley load difference, load factor, real-time load change rate, and industry-specific load (specific values for industrial / residential / commercial loads); Load forecasting: short-term load forecast, load forecast deviation rate, load response capability (quantification of interruptible load).
[0041] (iv) Physical characteristics of electricity fuels: physical indicators of fuels directly related to electricity production, specifically including: Thermal power fuel: calorific value of coal fed into the furnace, gas supply pressure, and fuel consumption (standard coal consumption / gas consumption). Fuel transportation: power plant coal inventory balance and fuel transportation time to the plant (quantified in hours / day, reflecting the physical timeliness of fuel supply).
[0042] (v) Physical characteristics of the power system operating environment: Natural physical environment data strongly related to power production / operation, specifically including: Extreme weather: Regional extreme temperatures (highest / lowest), meteorological level quantification of gale / rainstorm / snowstorm (e.g., gale level quantified as 0-12), duration of high temperature / cold wave; Equipment operating environment: ice thickness of power transmission and transformation lines, dust accumulation rate of photovoltaic panels, and ambient temperature at the wind turbine hub.
[0043] Then, a multi-dimensional quantitative analysis method was used to analyze the candidate feature set. Dynamic filtering is performed to remove invalid candidate feature variables and retain valid feature variables, forming a new feature set that includes the target feature variable. The specific process is as follows: Correlation analysis: Calculating candidate feature variables With electricity prices Pearson correlation coefficient (That is, the first correlation coefficient), set the first threshold. (like ), retain candidate feature variables whose correlation coefficients meet the threshold; Collinearity test: Calculate the variance inflation factor (i.e., the second correlation coefficient) among the candidate feature variables after screening, and set a second threshold. (like Candidate feature variables with severe multicollinearity are eliminated to avoid feature redundancy; Timeliness Verification: Verify the timeliness and update frequency of candidate feature variables. Calculate the confidence level based on the timeliness and update frequency. The confidence level reflects the reliability and real-time nature of the candidate feature variables. A third threshold is set, and candidate feature variables with confidence levels higher than the third threshold are retained. These candidate feature variables can be crawled on a scheduled / real-time basis, and their data update frequency matches the electricity price prediction frequency, ensuring the reliability and real-time nature of the feature data.
[0044] Optionally, the feature evaluation agent is also used for: The target feature variables are input into the model evaluation agent to train the prediction model; Determine the accuracy of the trained prediction model, and determine whether to retain the target feature variable based on the improvement in accuracy.
[0045] In this embodiment, a new feature set containing the target feature variables will be added. , and the set containing the original basic characteristic variables Fusion, forming a set of fusion features ,Will The data is fed back to the electricity price prediction agent, which retrains the model based on the fused features and calculates the prediction index. The prediction accuracy of the model under the fused feature set is improved by a percentage greater than or equal to a preset value. (like If so, then determine For valid target feature variables; if the preset value is not reached. If so, then re-examine candidate feature variables and adjust the screening threshold.
[0046] Optionally, the feature evaluation agent is also used for: The validated set of newly added features The system generates feature expansion instructions, which are fed back to the data crawling agent to guide it in adding corresponding data crawling channels and rules; simultaneously, it integrates the feature set. Feedback is sent to the electricity price prediction agent as the basis for subsequent model training and prediction. In addition, the feature evaluation agent continuously receives information from the model evaluation agent. If the prediction accuracy does not meet the preset target, the feature re-evaluation process is automatically triggered to adjust the screening threshold and mining range, thereby realizing dynamic iterative optimization of feature variables.
[0047] Optionally, the data acquisition agent is used for: Based on the target feature variables and basic feature variables, the corresponding data collection channels are determined; According to the preset acquisition strategy, the feature data corresponding to the target feature variable and the basic feature variable are acquired from the acquisition channels.
[0048] In this embodiment, the core function of the data acquisition agent is to receive feedback instructions from the feature evaluation agent, complete the initial feature data crawling and dynamic feature data expansion, and output the original time series data. In the initial state, publicly available data related to power production is crawled according to preset rules as basic feature variables. Upon receiving a new feature instruction from the feature evaluation agent, the system automatically matches the corresponding data crawling channel, completes the timed and real-time crawling of the target feature variables, and updates the feature variable set. .
[0049] Optionally, the crawling rules may include the following: The specific time period for scheduled triggering (e.g., 96 times / day for power data crawling, 24 times / day for meteorological data crawling) will be adaptively adjusted for different regions or scenarios. The trigger conditions for real-time monitoring are quantitatively defined (such as data platform update prompts), and re-crawl will be triggered when the deviation between crawled data and historical data exceeds a certain threshold to ensure data reliability. The data fusion rules for multi-channel crawling will select or discard feature data when the feature data corresponding to the same feature variable is crawled from different platforms with inconsistent results, or fuse feature data from multiple platforms through a weighted algorithm.
[0050] Optionally, the data preprocessing agent is used for: The feature data is filtered to identify outliers and missing values, and corrections are made for the outliers and missing values. The corrected feature data is then standardized. The standardized feature data is converted into a preset format and sent to the electricity price prediction agent.
[0051] Optionally, the correction for the outliers and missing values includes at least one of the following: Interpolation is performed based on the adjacent values of the outlier to obtain a corrected value, and the outlier is replaced by the corrected value. The feature data is averaged to obtain a first fill value, and the missing values are filled according to the first fill value. The data adjacent to the missing value are averaged to obtain a second filler value, and the missing value is filled according to the second filler value.
[0052] In this embodiment, the outlier handling process uses the 3σ criterion to identify outliers in the feature data (calculating the sample mean and variance, then calculating the interval consisting of the mean ± 3 times the standard deviation), and performs interpolation repair (linear interpolation or polynomial interpolation) on the outliers. The formula is as follows: (linear interpolation, This is an outlier. , (This refers to normal data before and after the outlier). During missing value handling, missing data in the feature data are filled using either mean imputation or K-nearest neighbor imputation. The formula for mean imputation is as follows: ( (The effective amount of data for variables with the same characteristics). Optionally, the standardization process for the corrected feature data includes: Determine the maximum and minimum values of the corresponding feature data for each target feature variable and basic feature variable, and calculate the difference between the maximum and minimum values as a reference value; Calculate the first difference between each feature data corresponding to the target feature variable and the minimum value; Divide the first difference by the reference value to obtain the standardized feature data.
[0053] In this embodiment, during the standardization process, all feature data are normalized to the [0,1] interval to eliminate the influence of dimensions. The formula is as follows: ( , The first (Minimum and maximum values of each characteristic variable); During the formatted output process, the processed data is organized into a CSV two-dimensional table or a JSON file and output to the electricity price prediction agent.
[0054] Optionally, the model evaluates the agent for: A second difference is calculated based on the actual value of the electricity price and the predicted value, and the prediction accuracy is determined based on the second difference; If the prediction accuracy is lower than a preset fourth threshold, the feature variable to be replaced is determined, and the feature evaluation agent is instructed to re-mine the candidate feature variable to replace the feature variable to be replaced.
[0055] In this embodiment, the model evaluation agent is used to receive feedback from the electricity price prediction agent. Predicted value at time And at the next moment, that is At any given moment, collect the true value of electricity prices. The prediction accuracy is determined based on a second difference between the actual value and the predicted value; the smaller the second difference, the higher the prediction accuracy.
[0056] If the prediction accuracy is lower than the preset fourth threshold, it indicates that the prediction accuracy is low, meaning that the current feature variables used for prediction have a low correlation with electricity prices, and feature variables need to be retrieved again for prediction.
[0057] Optionally, determining the feature variable to be replaced includes: Calculate the third correlation coefficient between the target feature variable, the basic feature variable, and the predicted value; The target feature variable or basic feature variable whose third correlation coefficient is lower than the preset fifth threshold is determined as the feature variable to be replaced.
[0058] Optionally, the data acquisition agent is also used for: Delete the data collection channel corresponding to the feature variable to be replaced.
[0059] In this embodiment, the model evaluation agent performs feature importance analysis on the prediction model to obtain the third correlation coefficient (calculated based on the feature gain value of XGBoost, the attention weight of LSTM, the feature embedding weight of Transformer, etc.) of the target feature variable and the basic feature variable on the electricity price prediction result. The agent then sets the third correlation coefficient below a preset fifth threshold. (Can be set according to actual needs, such as) The basic features of the data are marked as "low-contribution features" (i.e., features to be replaced), and their weights will be increased in subsequent iterations. The basic features are labeled as "high contribution features" and used as a reference for feature mining.
[0060] The data acquisition agent will no longer collect feature data corresponding to the feature variable to be replaced.
[0061] To implement the above embodiments, this application also proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the system as described in the foregoing system embodiments.
[0062] To implement the above embodiments, this application also proposes a computer program product having a computer program stored thereon, which, when executed by a processor, implements the system as described in the foregoing system embodiments.
[0063] To implement the above embodiments, this application also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the system as described in the foregoing system embodiments.
[0064] Figure 2 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. For example, the electronic device 800 may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.
[0065] Reference Figure 2 The electronic device 800 may include one or more of the following components: processing component 802, memory 804, power component 806, multimedia component 808, audio component 810, input / output (I / O) interface 812, sensor component 814, and communication component 816.
[0066] Processing component 802 typically controls the overall operation of electronic device 800, such as operations associated with display, telephone calls, data communication, camera operation, and recording operations. Processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the system described above. Furthermore, processing component 802 may include one or more modules to facilitate interaction between processing component 802 and other components. For example, processing component 802 may include a multimedia module to facilitate interaction between multimedia component 808 and processing component 802.
[0067] Memory 804 is configured to store various types of data to support the operation of electronic device 800. Examples of this data include instructions for any application or system operating on electronic device 800, contact data, phonebook data, messages, pictures, videos, etc. Memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.
[0068] Power component 806 provides power to various components of electronic device 800. Power component 806 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 800.
[0069] Multimedia component 808 includes a screen that provides an output interface between the electronic device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 808 includes a front-facing camera and / or a rear-facing camera. When the electronic device 800 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.
[0070] Audio component 810 is configured to output and / or input audio signals. For example, audio component 810 includes a microphone (MIC) configured to receive external audio signals when electronic device 800 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 804 or transmitted via communication component 816. In some embodiments, audio component 810 also includes a speaker for outputting audio signals.
[0071] I / O interface 812 provides an interface between processing component 802 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.
[0072] Sensor assembly 814 includes one or more sensors for providing state assessments of various aspects of electronic device 800. For example, sensor assembly 814 can detect the on / off state of electronic device 800, the relative positioning of components such as the display and keypad of electronic device 800, changes in position of electronic device 800 or a component of electronic device 800, the presence or absence of user contact with electronic device 800, orientation or acceleration / deceleration of electronic device 800, and temperature changes of electronic device 800. Sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 814 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 814 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.
[0073] Communication component 816 is configured to facilitate wired or wireless communication between electronic device 800 and other devices. Electronic device 800 can access wireless networks based on communication standards, such as WiFi, 4G, or 5G, or combinations thereof. In one exemplary embodiment, communication component 816 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 816 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
[0074] In an exemplary embodiment, the electronic device 800 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the system described above.
[0075] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, which can be executed by a processor 820 of an electronic device 800 to complete the system described above. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.
[0076] To implement the above embodiments, this application also proposes a chip, including: the chip includes processing circuitry configured to execute the system provided in the foregoing embodiments.
[0077] Figure 3 This is a schematic diagram of the structure of a chip according to an embodiment of this application. See also... Figure 3 The diagram shown is a schematic representation of the structure of chip 1100, but it is not limited to this.
[0078] Chip 1100 includes processing circuitry 1101, which is configured to execute any of the above systems.
[0079] In some embodiments, chip 1100 further includes one or more interface circuits 1102. Optionally, the interface circuit 1102 is connected to memory 1103, and the interface circuit 1102 can be used to receive signals from memory 1103 or other devices, and the interface circuit 1102 can be used to send signals to memory 1103 or other devices. For example, the interface circuit 1102 can read instructions stored in memory 1103 and send the instructions to processing circuit 1101.
[0080] In some embodiments, the interface circuit 1102 performs at least one of the communication steps such as sending and / or receiving in the above system, while the processing circuit 1101 performs other steps.
[0081] In some embodiments, the terms interface circuit, interface, transceiver pin, transceiver, etc., can be used interchangeably.
[0082] In some embodiments, chip 1100 further includes one or more memories 1103 for storing instructions. Optionally, all or part of the memories 1103 may be located outside of chip 1100.
[0083] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.
[0084] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.
[0085] Any process or system description in the flowchart or otherwise described herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.
[0086] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.
[0087] It should be understood that various parts of this application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or systems can be implemented using software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.
[0088] Those skilled in the art will understand that all or part of the steps of the system implementing the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the system embodiments.
[0089] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.
[0090] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of this application.
Claims
1. A multi-agent intelligent computing cloud platform system for electricity price prediction based on autonomous feature evaluation, characterized in that, include: Feature evaluation agent, data acquisition agent, data preprocessing agent, electricity price prediction agent, model evaluation agent; The feature evaluation agent is used to mine candidate feature variables, determine target feature variables based on the correlation between the candidate feature variables and electricity prices, and send the target feature variables to the data acquisition agent. The data acquisition agent is used to receive the target feature variable, collect feature data based on the target feature variable and preset basic feature variables, and send the feature data to the data preprocessing agent; wherein, the basic variable and the candidate feature variable are feature variables related to power production; The data preprocessing agent is used to preprocess the feature data and send the preprocessed feature data to the electricity price prediction agent. The electricity price prediction agent is used to call the prediction model, make a prediction based on the preprocessed feature data, and output the predicted value of the electricity price. The model evaluation agent is used to evaluate the prediction accuracy based on the actual value and the predicted value of the electricity price, and to instruct the feature evaluation agent to mine candidate feature variables based on the prediction accuracy.
2. The system according to claim 1, characterized in that, The feature evaluation agent is used for: Retrieve data related to power production and extract feature variables related to power production as candidate feature variables; Based on the predicted value of the electricity price output by the electricity price prediction agent, a first correlation coefficient between each candidate feature variable and the predicted value is determined; Calculate the second correlation coefficient among each of the candidate feature variables; The confidence level of the candidate feature variables is determined based on the timeliness of their acquisition and the frequency of their update. The target feature variable is determined from the candidate feature variables based on the first correlation coefficient, the second correlation coefficient, and the confidence level.
3. The system according to claim 2, characterized in that, The step of determining the target feature variable from the candidate feature variables based on the first correlation coefficient, the second correlation coefficient, and the confidence score includes: The candidate feature variables are filtered based on the first correlation coefficient, and candidate feature variables with a first correlation coefficient higher than a preset first threshold are retained. The candidate feature variables are screened based on the second correlation coefficient. In response to the existence of multiple candidate feature variables whose second correlation coefficient is higher than a preset second threshold, one of the multiple candidate features is retained. The candidate feature variables are filtered based on the confidence level, and candidate feature variables with a confidence level higher than a preset third threshold are retained. The candidate feature variables that remain after the screening are determined as the target feature variables.
4. The system according to claim 3, characterized in that, The feature evaluation agent is also used for: The target feature variables are input into the model evaluation agent to train the prediction model; Determine the accuracy of the trained prediction model, and determine whether to retain the target feature variable based on the improvement in accuracy.
5. The system according to claim 3, characterized in that, The data acquisition agent is used for: Based on the target feature variables and basic feature variables, the corresponding data collection channels are determined; According to the preset acquisition strategy, the feature data corresponding to the target feature variable and the basic feature variable are acquired from the acquisition channels.
6. The system according to claim 5, characterized in that, The data preprocessing agent is used for: The feature data is filtered to identify outliers and missing values, and corrections are made for the outliers and missing values. The corrected feature data is then standardized. The standardized feature data is converted into a preset format and sent to the electricity price prediction agent.
7. The system according to claim 6, characterized in that, The correction of the outliers and missing values includes at least one of the following: Interpolation is performed based on the adjacent values of the outlier to obtain a corrected value, and the outlier is replaced by the corrected value. The feature data is averaged to obtain a first fill value, and the missing values are filled according to the first fill value. The data adjacent to the missing value are averaged to obtain a second filler value, and the missing value is filled according to the second filler value.
8. The system according to claim 6, characterized in that, The standardization process for the corrected feature data includes: Determine the maximum and minimum values of the corresponding feature data for each target feature variable and basic feature variable, and calculate the difference between the maximum and minimum values as a reference value; Calculate the first difference between each feature data corresponding to the target feature variable and the minimum value; Divide the first difference by the reference value to obtain the standardized feature data.
9. The system according to claim 1, characterized in that, The model evaluation agent is used for: A second difference is calculated based on the actual value of the electricity price and the predicted value, and the prediction accuracy is determined based on the second difference; If the prediction accuracy is lower than a preset fourth threshold, the feature variable to be replaced is determined, and the feature evaluation agent is instructed to re-mine the candidate feature variable to replace the feature variable to be replaced.
10. The system according to claim 9, characterized in that, The process of determining the feature variable to be replaced includes: Calculate the third correlation coefficient between the target feature variable, the basic feature variable, and the predicted value; The target feature variable or basic feature variable whose third correlation coefficient is lower than the preset fifth threshold is determined as the feature variable to be replaced.
11. The system according to claim 10, characterized in that, The data acquisition agent is also used for: Delete the data collection channel corresponding to the feature variable to be replaced.
12. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements the system as described in any one of claims 1-11.
13. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the system as described in any one of the preceding claims 1-11.
14. A chip, characterized in that, The chip includes processing circuitry configured to execute the system as described in any one of claims 1-11.
15. A computer program product, characterized in that, It includes a computer program, which, when executed by a processor, implements the system as described in any one of claims 1-11.