Intelligent analysis method for multivariate data fusion mining of electric power energy system

Abstract

The invention discloses an intelligent analysis method for multivariate data fusion mining of an electric power energy system, and the method comprises the steps: collecting and obtaining environment data parameters of the electric power energy system, and carrying out the analysis to obtain renewable energy power generation parameters; acquiring influence factor parameters of the electric quantity load demand to obtain electric quantity load parameters; analyzing the renewable energy power generation parameters and the power load parameters to obtain an analysis result; setting an expected value, and judging whether the analysis result meets the expected value or not; if the analysis result meets the expected value, the analysis result is output, and if the analysis result does not meet the expected value, the analysis result is analyzed based on the expected value, and an optimization report is generated. According to the output analysis result and the optimization report, the overall operation efficiency of the power system can be improved, the adaptability and flexibility of the system to high-proportion renewable energy access are enhanced, the user experience is optimized, and the power supply service quality is improved.

Description

Technical Field

[0001] This invention relates to the field of power energy systems, specifically to an intelligent analysis method for multi-source data fusion and mining in power energy systems. Background Technology

[0002] The power energy system, as the cornerstone of modern society, is undergoing unprecedented and profound changes. The traditional centralized, unidirectional power supply model is gradually evolving into a distributed, bidirectional, interactive, and multi-energy-coordinated smart grid. This transformation has not only brought higher efficiency, stronger reliability, and better environmental performance, but has also made the power system unprecedentedly complex. Supporting this complex system is massive, heterogeneous, and highly dynamic multi-dimensional data. How to effectively integrate, understand, and utilize this data to extract valuable information to guide system optimization, fault warning, and decision-making has become a core challenge for the power industry. Consequently, intelligent analysis methods for multi-dimensional data fusion and mining in power energy systems have emerged, driven by the collaborative development and integration of multiple key technologies. These methods primarily address the challenges of massive data volume, high heterogeneity, and high real-time requirements in power systems. The core idea is to integrate equipment operation data, environmental data, and user electricity consumption data, using intelligent algorithms to uncover potential patterns within the data, providing support for optimized power system scheduling, fault warning, and energy management.

[0003] Power systems typically consist of multiple stages, including generation, transmission, distribution, and consumption, each involving numerous physical devices and complex operating mechanisms. Traditional power systems primarily focus on generation, transmission, distribution, and consumption, with relatively limited data sources, mainly from SCADA (Supervisory Control and Data Acquisition) and EMS (Energy Management System). Traditional power systems often rely on manual operation and accumulated experience for dispatch and management. However, with increasing load demand volatility, the integration of renewable energy, and continuous changes in environmental and policy conditions, power system dispatch and management are becoming increasingly complex. Therefore, methods relying on data analysis and intelligent decision-making are particularly important. Especially in the face of rapidly changing electricity markets and climate environments, the effective integration and analysis of massive amounts of data is crucial for achieving smart grids and sustainable development goals.

[0004] With the integration of renewable energy sources, the balance and stability of the power system are challenged due to their volatility and unpredictability. In particular, renewable energy output and user load exhibit strong randomness and volatility, making improving the intelligence level of the power system and addressing energy challenges a primary objective of power system optimization. As technology continues to advance, people's electricity demands are becoming increasingly diversified, placing higher demands on the intelligent management of the power system. However, many power systems still employ traditional management models, making it impossible to monitor and regulate electricity supply and demand in real time, and difficult to cope with emergencies. The level of intelligence in the power system needs to be improved. Summary of the Invention

[0005] This application provides an intelligent analysis method for multi-source data fusion and mining in power energy systems, which is used to address the technical problem that the level of intelligence in power systems needs to be improved in the existing technology.

[0006] In view of the above problems, this application provides an intelligent analysis method for multi-source data fusion and mining of power energy systems.

[0007] The first aspect of this application provides an intelligent analysis method for multi-source data fusion and mining of a power energy system. The method comprises: collecting environmental data parameters of the power energy system and analyzing them to obtain renewable energy generation parameters; collecting parameters of influencing factors of electricity load demand and obtaining electricity load parameters; analyzing the renewable energy generation parameters and the electricity load parameters to obtain analysis results; setting an expected value and determining whether the analysis results meet the expected value; if the analysis results meet the expected value, outputting the analysis results; if the analysis results do not meet the expected value, analyzing the analysis results based on the expected value and generating an optimization report; and outputting the analysis results and optimization report accordingly.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] This application embodiment employs an intelligent analysis method for multi-source data fusion and mining in a power energy system. It collects environmental data parameters of the power energy system and analyzes them to obtain renewable energy generation parameters; it also collects parameters influencing electricity load demand to obtain electricity load parameters; it analyzes the renewable energy generation parameters and electricity load parameters to obtain analysis results; it sets an expected value and determines whether the analysis results meet the expected value; if the analysis results meet the expected value, the analysis results are output; if the analysis results do not meet the expected value, the analysis results are analyzed based on the expected value to generate an optimization report; based on the output analysis results and optimization report, the overall operating efficiency of the power system can be improved, the system's adaptability and flexibility to high-proportion renewable energy access can be enhanced, user experience can be optimized, and the quality of power supply services can be improved.

[0010] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 The system flowchart provided for this application. Detailed Implementation

[0013] This application provides an intelligent analysis method for multi-source data fusion and mining in power energy systems, which is intended to address the technical problem that the level of intelligence in existing power systems needs to be improved.

[0014] After introducing the basic principles of this application, the technical solutions in this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0015] Example 1

[0016] like Figure 1 As shown, this application provides an intelligent analysis method for multi-source data fusion and mining in power energy systems, the method comprising:

[0017] S100: Collects and obtains environmental data parameters of the power energy system, and analyzes and obtains renewable energy power generation parameters;

[0018] In this embodiment, the acquisition of environmental data parameters for the power energy system refers to the acquisition of meteorological forecast data, historical output data, real-time monitoring data, and power grid operation status data. Specifically, meteorological forecast data refers to future weather conditions. This data typically includes the following elements: weather conditions, temperature, humidity, wind force, precipitation, air pressure, ultraviolet radiation, air quality, etc., and can be obtained by accessing the China Meteorological Data Network. Historical output data refers to renewable energy power generation parameters over a past period, including power generation capacity, power generation efficiency, and power generation, which can be obtained from local equipment in wind farms or cloud servers. Real-time monitoring data refers to data parameters obtained by continuously and dynamically acquiring and analyzing data from target objects through sensors, monitoring equipment, or systems to obtain the latest status information. This data can be obtained by real-time monitoring of environmental parameters using high-precision sensors. Power grid operation status data refers to various data generated during the operation of the power system, including equipment status data, load data, power grid topology data, environmental data, etc., and can be acquired using various sensors through wireless sensor networks and Internet of Things (IoT) technology.

[0019] Step S100 in the method provided in this application embodiment includes:

[0020] S110. Collect and acquire meteorological forecast data to obtain the first environmental data parameters;

[0021] S120. Collect and acquire historical power output data to obtain the second environmental data parameters;

[0022] S130. Collect and acquire real-time monitoring data to obtain third-environmental data parameters;

[0023] S140. Collect and acquire power grid operation status data to obtain the fourth environmental data parameters;

[0024] S150, The first environmental data parameter, the second environmental data parameter, the third environmental data parameter and the fourth environmental data parameter are used as the environmental data set;

[0025] S160. Substitute the environmental data set into the regression formula to obtain the power generation parameters. The regression formula is:

[0026] y=β0+β1x1+β2x2+β3x3+β4x4+∈;

[0027] Where y is the predicted power generation, x1 is the first environmental data parameter, x2 is the second environmental data parameter, x3 is the third environmental data parameter, x4 is the fourth environmental data parameter, β0 is the intercept term, β1, β2, β3, and β4 are the regression coefficients to be learned, and ∈ is the error term.

[0028] In this embodiment, the data is first preprocessed, including data cleaning to remove missing values, outliers, and duplicate data. If the environmental parameters vary significantly, data normalization or standardization may be necessary. Then, historical data is used to train the model, and the regression coefficient β can be estimated using the least squares method. By collecting the above environmental data parameters, constructing a linear regression model, training and evaluating the model, future renewable energy generation can be predicted. For example, a project, through the deep integration of photovoltaic, energy storage systems, and intelligent monitoring technologies, configured a 1.06MW distributed photovoltaic power generation system and a 215kWh energy storage system. Using sensors, the Internet of Things (IoT), and intelligent monitoring technologies, this project achieved comprehensive collection and analysis of data on photovoltaics, energy storage, and load, providing an intelligent and efficient solution for future energy management.

[0029] S200: Collect and obtain parameters of factors affecting electricity load demand to obtain electricity load parameters;

[0030] In this embodiment, the parameters influencing electricity load demand collected refer to historical electricity consumption data, weather forecast data, seasonal fluctuation data, and social activity data. Historical electricity consumption data refers to actual electricity consumption records over a past period, reflecting electricity consumption habits, periodic patterns, and load change trends. This data is typically obtained through power monitoring platforms or power management systems. Weather forecast data refers to data predicting future weather conditions, including temperature, humidity, wind speed, and precipitation. Real-time and future weather forecast data can be obtained from meteorological stations. Seasonal fluctuation data refers to fluctuations in electricity demand caused by seasonal changes. Statistical software or data analysis platforms can be used to perform seasonal analysis of historical data. Social activity data refers to social events or activities that affect electricity demand, such as holidays, large-scale events, or social incidents. These activities often lead to abnormal fluctuations in electricity demand. Relevant information can be obtained through social event calendars, news platforms, and public event databases. Combining this data with electricity demand fluctuation data allows for the identification of the impact of specific activities on the electricity load. By integrating this data, changes in electricity load demand can be more accurately predicted and analyzed, helping to optimize power supply and dispatch.

[0031] Step S200 in the method provided in this application embodiment includes:

[0032] S210 collects historical electricity consumption data to obtain the first object data parameters;

[0033] S220 collects weather forecast data to obtain the second object data parameters;

[0034] S230 collects seasonal fluctuation data to obtain the data parameters of the third object;

[0035] S240 collects social activity data to obtain the fourth object data parameters;

[0036] S250 uses the first object data parameter, the second object data parameter, the third object data parameter, and the fourth object data parameter as an object data set;

[0037] S260 constructs a SARIMAX (Seasonal Autoregressive Integral Moving Average) model, inputs the object data parameter set for analysis, and obtains the electricity load parameters;

[0038] S270 uses the first object data parameter as the target variable, the second and fourth object data parameters as exogenous variables, and the third object data parameter as a seasonality term.

[0039] The formula for the SARIMAX model is:

[0040] Y t =φ1Y t-1 +φ2Y t-2 +θ1∈ t-1 +θ2∈ t-2 +∈ t +m+n;

[0041] Where Y represents the electrical load at time t, and φ1, φ2, θ1, θ2 are the coefficients of the AR and MA processes; ∈ t is the error term, AR is the autoregressive term, MA is the moving average term, m is the seasonal term, and n is the exogenous variable.

[0042] In this application example, SARIMAX is an advanced statistical model used in time series analysis and forecasting; its full name is the Seasonal Autoregressive Integral Moving Average model. It is an extension of the ARIMA model and is mainly used to process time series data with seasonal trends and external influencing factors.

[0043] Using the SARIMAX model to predict real-time electricity load, combined with historical electricity consumption data, weather forecasts, seasonal fluctuations, and social activities, can effectively improve forecast accuracy. The SARIMAX model is suitable for time series data, especially for electricity load forecasting with seasonality and external factors. Preprocessing of the above-mentioned data parameters is necessary. All data should be timestamped, meteorological data should be completed using interpolation, and social activity data should be filled with data similar to historical dates. Weather data, seasonal fluctuations, and social activities should be transformed into numerical forms or categorical variables. Accurate load forecasting can provide a reference for power system dispatching, helping the grid to make load adjustments in advance, optimize generation plans, and allocate power resources. Electricity load forecasting can help electricity market participants make reasonable trading decisions, reduce the risks caused by fluctuations in electricity demand, and help power companies predict equipment load fluctuations and arrange maintenance, repair, and upgrade plans in advance.

[0044] S300: Analyze the renewable energy power generation parameters and power load parameters to obtain the analysis results;

[0045] In this embodiment, the renewable energy generation parameters and power load parameters are analyzed. First, real-time or near-real-time power load data of the power grid is continuously received and analyzed. Based on the changing trend and fluctuation range of the power load parameters over time, a judgment is made. When the load data is continuously at a relatively low level, the identified time period is clearly marked as a low-valley period. When the load data is continuously at a relatively high level, the identified time period is clearly marked as a peak period. The generated time period classification results are used for subsequent electricity price strategy formulation, demand-side response scheduling, or power grid optimization operation.

[0046] Step S300 in the method provided in this application embodiment includes:

[0047] S310 generates time series data based on the change of the power load parameter over time, and obtains an intermediate threshold based on the time series data;

[0048] The S320 acquires real-time power load parameters as comparison parameters and compares them with an intermediate threshold. If the detected load parameter is lower than the intermediate threshold, the identified time period is clearly marked as an off-peak period. If the detected load parameter is higher than the intermediate threshold, the identified time period is clearly marked as a peak period.

[0049] The S330 generates an energy storage scheme based on power generation parameters and power load parameters for different time periods, and uses the energy storage charge during off-peak and peak periods as the analysis results.

[0050] In this embodiment, during off-peak hours, power companies can prioritize the use of low-cost power generation methods. During peak hours, the system can dispatch efficient power generation resources while reducing reliance on heavily polluting coal-fired power plants. Dynamically adjusting load and power generation ensures stable operation of the power system at all times. Combined with advanced energy storage technology, peak-valley load regulation is achieved. The energy storage system stores excess electricity during off-peak hours and releases it during peak hours, reducing reliance on traditional thermal power units.

[0051] S400: Analyze the analysis results based on the expected value;

[0052] In this embodiment, the expected value is set based on a comprehensive consideration of multiple objectives, including power system operation, planning, economics, and policy. The power system needs to maintain sufficient reserve capacity to cope with sudden fluctuations or failures in generation or load, and the system must have sufficient flexibility resources to smooth out rapid fluctuations in renewable energy and load. The system should have a certain degree of anti-interference capability, capable of maintaining power supply to critical loads or rapidly restoring them in the event of a fault or extreme event. Thus, the real-time balance between generation and load must be within allowable deviations. Setting different expected values ​​based on different situations allows for flexible and efficient control of the power system, improving operational efficiency and enhancing the adaptability of renewable energy. After obtaining the above-mentioned expected values ​​and analysis results, different solutions are proposed based on the comparison between different analysis results and expected values, generating an optimization report.

[0053] Step S400 in the method provided in this application embodiment includes:

[0054] S410 If the analysis result meets the expected value, the analysis result is output; if the analysis result does not meet the expected value, the power consumption data is collected and uploaded in real time.

[0055] The S420 analyzes electricity consumption data to establish a personalized electricity consumption behavior model for users;

[0056] S430 obtains the currently executed electricity price time table, matches and analyzes the user's electricity consumption behavior model with the electricity price time table, and generates an optimization report based on the results of behavior modeling and electricity price correlation analysis.

[0057] By analyzing users' electricity consumption behavior through big data, an electricity consumption behavior model is constructed based on a regression model. The electricity consumption data is substituted into the regression formula, which is: y=β0+β1x1+β2x2+β3x3+β4x4+∈; where y is the electricity consumption, x1 is the temperature, x2 is the number of people in the household, x3 is the type of equipment and the frequency of use, x4 is the time characteristic, β0, β1, β2, β3, β4 are the coefficients of the model, which can be estimated through training data, and ∈ is the error term.

[0058] After obtaining a user's personalized electricity consumption behavior model, the electricity consumption behavior is matched and analyzed with the electricity price schedule. By analyzing the user's electricity consumption in different time periods, it is determined whether electricity expenses can be reduced by adjusting the electricity consumption behavior. Based on the user's historical electricity consumption data, this data is mapped to the electricity price schedule, and the electricity consumption in different price periods is calculated. If the user's electricity consumption is high during peak hours when the electricity price is high, then electricity costs can be reduced by optimizing the user's electricity consumption behavior, adjusting peak-hour electricity consumption, and shifting it to off-peak hours. By adjusting the user's electricity consumption behavior, the user's electricity expenses are minimized without affecting normal living needs. Based on the user's daily electricity consumption patterns, the time periods for equipment use are intelligently scheduled to balance the load in different time periods and reduce dependence on peak hours. Weight coefficients for electricity price periods can be introduced into the user's electricity consumption model, and the model can be retrained based on these weights to optimize the user's electricity consumption behavior.

[0059] Personalized electricity usage suggestions are pushed based on user electricity consumption data to help users adjust their electricity consumption habits appropriately during different electricity price periods and reduce energy consumption. These personalized suggestions can be delivered to users promptly and clearly through methods such as push notifications, message reminders, or on-screen displays.

[0060] S500: Outputs analysis results and optimization reports;

[0061] In this embodiment, the obtained analysis results are first interpreted and summarized in detail, comprehensively presenting the performance and trends of various data indicators. Then, based on these analysis results, corresponding optimization strategies are formulated, and improvement suggestions are proposed for potential problems or deficiencies, forming a detailed optimization report. The report includes not only the analysis process and results summary, but also the specific implementation steps of the optimization plan, expected effects, and possible risk assessments.

[0062] Step S500 in the method provided in this application embodiment includes:

[0063] S510 identifies and analyzes major electricity-related issues based on the analysis report content;

[0064] S520 identifies problems such as poor electricity usage habits and behavioral patterns in high-energy-consuming areas in the report, and proposes suggestions for adjusting electricity usage behavior.

[0065] S530 analyzes the output results and optimizes the report.

[0066] In this embodiment, personalized visualization interfaces and reports can be provided according to different user roles and needs, but are not limited to, allowing users to monitor power usage in real time, automatically adjust equipment operating modes, and improve energy efficiency. This provides users with a platform offering personalized recommendations, energy efficiency diagnostics, online services, and other functions, enhancing user satisfaction and energy-saving awareness. For example, power dispatchers focus on the real-time operating status of the power grid and power flow distribution, while equipment maintenance personnel focus on equipment status monitoring and fault warnings, improving the effectiveness of visualization in supporting user decision-making.

[0067] The steps of the methods or algorithms described in this application can be directly embedded in hardware, a software unit executed by a processor, or a combination of both. The software unit can be stored in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM, or any other storage medium of any form in the art. Exemplarily, the storage medium can be connected to the processor so that the processor can read information from the storage medium and write information to the storage medium. Optionally, the storage medium can also be integrated into the processor. The processor and storage medium can be disposed in an ASIC, which can be disposed in a terminal. Optionally, the processor and storage medium can also be disposed in different components within the terminal. These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0068] Although this application has been described in conjunction with specific features and embodiments, it is obvious that various modifications and combinations can be made thereto without departing from the spirit and scope of this application. Accordingly, this specification and drawings are merely illustrative examples of this application and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Thus, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. An intelligent analysis method for multi-source data fusion and mining in power energy systems, characterized in that, The method includes: Collect and acquire environmental data parameters of the power energy system, and analyze them to obtain renewable energy power generation parameters; Collect and obtain parameters of factors affecting electricity load demand to obtain electricity load parameters; The parameters of renewable energy power generation and power load are analyzed to obtain the analysis results; Set expected values ​​and determine whether the analysis results meet the expected values; If the analysis results meet the expected value, the analysis results are output; if the analysis results do not meet the expected value, the analysis results are analyzed based on the expected value, and an optimization report is generated. Based on the output analysis results and optimization report.

2. The method according to claim 1, characterized in that, The process of collecting and acquiring environmental data parameters of the power energy system and analyzing them to obtain renewable energy power generation parameters includes: Collect and acquire meteorological forecast data to obtain the first environmental data parameters; Historical power output data is collected to obtain the second environmental data parameters; Collect and acquire real-time monitoring data to obtain third-party environmental data parameters; Collect and acquire power grid operation status data to obtain the fourth environmental data parameters; The first environmental data parameter, the second environmental data parameter, the third environmental data parameter, and the fourth environmental data parameter are used as the environmental data set; Substituting the environmental data set into the regression formula, the power generation parameters are obtained. The regression formula is as follows: y=β0+β1x1+β2x2+β3x3+β4x4+∈; Where y is the predicted power generation, x1 is the first environmental data parameter, x2 is the second environmental data parameter, x3 is the third environmental data parameter, x4 is the fourth environmental data parameter, β0 is the intercept term, β1, β2, β3, and β4 are the regression coefficients to be learned, and ∈ is the error term.

3. The method according to claim 1, characterized in that, The process of collecting and acquiring parameters influencing electricity load demand to obtain electricity load parameters includes: Collect historical electricity consumption data to obtain the first object's data parameters; Collect weather forecast data to obtain the data parameters of the second object; Collect seasonal fluctuation data to obtain the data parameters of the third object; Collect social activity data to obtain the fourth object's data parameters; The first object data parameter, the second object data parameter, the third object data parameter, and the fourth object data parameter are used as the object data set; A SARIMAX model is constructed, and the set of object data parameters is used for analysis to obtain the power load parameters. The first object data parameter is used as the target variable, the second and fourth object data parameters are used as exogenous variables, and the third object data parameter is used as a seasonality term. The formula for the SARIMAX model is: Y t =φ1Y t-1 +φ2Y t-2 +θ1∈ t-1 +θ2∈ t-2 +∈ t +m+n; Where Y represents the electrical load at time t, and φ1, φ2, θ1, θ2 are the coefficients of the AR and MA processes; ∈ t is the error term, AR is the autoregressive term, MA is the moving average term, m is the seasonal term, and n is the exogenous variable.

4. The method according to claim 1, characterized in that, The analysis of the renewable energy power generation parameters and power load parameters to obtain analysis results includes: Time series data is generated based on the change of the power load parameter over time, and an intermediate threshold is obtained based on the time series data. Real-time power load parameters are acquired as comparison parameters. The comparison parameters are compared with the intermediate threshold. If the load parameter is detected to be lower than the intermediate threshold, the identified time period is clearly marked as the off-peak period. If the load parameter is detected to be higher than the intermediate threshold, the identified time period is clearly marked as the peak period. Power storage schemes are generated based on power generation parameters and power load parameters for different time periods, and the energy storage charge during off-peak and peak periods is used as the analysis result.

5. The method according to claim 1, characterized in that, The analysis of the results based on the expected value includes: If the analysis results meet the expected value, the analysis results are output; if the analysis results do not meet the expected value, electricity consumption data is collected and uploaded in real time. Based on the analysis of electricity consumption data, a personalized electricity consumption behavior model for users is established; Obtain the currently executed electricity price time period table, match and analyze the user electricity consumption behavior model with the electricity price time period table, and generate an optimization report based on the results of behavior modeling and electricity price correlation analysis.

6. The method according to claim 1, characterized in that, The output analysis results and optimization report include: Based on the analysis report, the main electricity-related issues were identified and analyzed. Based on the identified problems, suggestions for adjusting electricity consumption behavior are proposed, including identifying poor electricity usage habits and behavioral patterns in high-energy-consuming areas. Based on the output analysis results and optimization report.

Images