Method for correlation analysis of multiple groups of same-frequency and same-dimension data in three-dimensional space

Abstract

The application discloses a method for correlation analysis of time-varying data in a three-dimensional space, which comprises the following steps: step one, collecting and processing data of the same type and different types in the same frequency and the same dimension; step two, color rendering of the data based on daily understanding of business attributes; step three, generating a three-dimensional space cube according to time and date dimensions; step four, stacking and constructing a new data cube based on the three-dimensional space cube; step five, mathematical analysis of the three-dimensional stacked data of the new data cube; and step six, combined analysis of time domain characteristics and internal relations of the data of the same type and different types. The application can not only display a plurality of massive continuous data in a limited space by taking into account data sampling accuracy, but also perfectly retain characteristic data to present time domain characteristics of the data, so that the data as a whole and in part can be observed, measured and visualized.

Description

Technical Field

[0001] This invention relates to a correlation analysis method for time-varying data in a power system, and more particularly to a correlation analysis method for multiple sets of time-varying data with the same frequency and dimension in a three-dimensional space. Background Technology

[0002] With the widespread application of big data technology across various industries, creating real-time data streams has become a popular trend. Data constantly records the numerical changes carried by user-operated devices. In the era of high-speed networks, data growth is extremely rapid, with most changes occurring on a second-by-second basis. Over time, if a single data object is counted every minute, the number of data points in a year will exceed 520,000. When analyzing the characteristics of changes in operational data over time, time is typically used as the horizontal axis and numerical values ​​as the vertical axis, plotting daily operational curves over 24 hours. The granularity of the horizontal time axis can be accurate to 1 minute, 5 minutes, or 15 minutes, even down to the second, provided the sampling precision is sufficient. However, as businesses expand, based on their business models, sampling and analysis of only single data objects is clearly no longer sufficient to meet their needs.

[0003] Familiar two-dimensional pie charts, histograms, scatter plots, and bar charts are traditional statistical charts and the most common and basic applications of data visualization. However, traditional statistical charts can only present basic information, reveal the structure in the data, and visualize quantitative data results. Furthermore, when multiple sets of data exist simultaneously, the traditional approach is to create traditional BI statistical charts from these multiple sets and compare them with each other. With full datasets, there is a great deal of redundant data, allowing only general data trends or point-to-point comparisons, resulting in complex analysis that is not clear, intuitive, or convenient to use.

[0004] With the continuous development of power systems, the amount of data in power systems is increasing. Chinese invention patent publication number "CN108268201A" provides a method for processing power data charts based on drag-and-drop recalculation. This method can transform a complex power data chart into multiple neatly arranged and concise power charts by dragging and dropping, further solving the problem of complex charts often appearing as a jumbled mess when displayed. However, the drag-and-drop process still requires manual operation, and it is inconvenient for staff to view densely packed charts, making it difficult to intuitively understand the specific data. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, this invention provides a method for correlation analysis of multiple sets of time-varying data with the same frequency and dimension in three-dimensional space.

[0006] This invention solves the above-mentioned technical problems through the following technical solution: a method for correlation analysis of multiple sets of time-varying data with the same frequency and dimension in three-dimensional space, characterized by comprising the following steps: Step 1: Collect multiple sets of data of the same type and different types and process them with the same frequency and dimension; Step two: Render the data with different colors based on daily understanding of business attributes; Step 3: Generate a cube in three-dimensional space according to the time and date dimensions; Step 4: Stack new data cubes on top of existing cubes in three-dimensional space; Step 5: Perform mathematical analysis of the three-dimensional stacked data on the new data cube; Step six: Combine and analyze the temporal characteristics and intrinsic relationships of multiple sets of data of the same type and different types.

[0007] Preferably, step one specifically includes the following sub-steps: The time dimension of the time-varying data with time-domain characteristics is truncated to form a daily curve over 24 hours; the curves of different days are superimposed in an orderly manner according to date and aligned by time, thereby expanding the date dimension on the basis of the traditional daily curve. Each data point carries corresponding three-dimensional attributes and is displayed in a three-dimensional space in the form of bars. The second group of data is truncated according to the above pattern and stacked on the bars of the first group of data according to date and time, and so on; after stacking, a new three-dimensional data cube of values, time, and date is formed; based on the three-dimensional engine, the expansion in dimensional, spatial, and numerical directions is realized, and massive amounts of data are fully displayed in a limited three-dimensional space.

[0008] Preferably, the time-varying data with time-domain characteristics refers to data obtained by intermittently counting values ​​with the same sampling precision within a certain time period. The values ​​are directly dependent on the time points, and the values ​​change with time. When the time-varying data uses the current time point as the coordinate, it records the changes in the values ​​of the data object in real time. During the continuous operation of the data object, a massive amount of continuous running values ​​containing time points are continuously generated over time. There is a continuous correspondence between the values ​​and the time points. The set of these running values ​​is the time-varying data with time-domain characteristics.

[0009] Preferably, step two creates a significant layering of multiple objects within a three-dimensional cube.

[0010] Preferably, the mathematical analysis in step five includes filtering and segmentation functions based on input conditions.

[0011] Preferably, the correlation analysis method for multiple sets of time-varying data with the same frequency and dimension in three-dimensional space further includes the following steps: Step 7: Based on the analysis results of Step 6, predict the data development at subsequent time points; Step 8: Make advance plans based on the prediction results from Step 7.

[0012] The correlation analysis method for multiple sets of time-varying data with the same frequency and dimension in three-dimensional space adopts the following modules: The data acquisition module collects multiple sets of data of the same type but different types. The processing module, connected to the acquisition module, performs same-frequency and same-dimensional processing on the data acquired by the acquisition module; The rendering module connects to the processing module and performs color rendering on the data processed by the processing module based on the daily understanding of the business attributes. The generation module, connected to the rendering module, generates a cube in three-dimensional space according to the time and date dimensions. The building module connects to the generation module and stacks new data cubes on the basis of the cubes generated by the generation module in three-dimensional space. The analysis module, connected to the construction module, performs mathematical analysis of the three-dimensional stacked data of the new data cube. The combination module connects with the analysis module to perform combined analysis of the temporal characteristics and intrinsic relationships of multiple sets of data of the same or different types.

[0013] Preferably, the correlation analysis method for multiple sets of time-varying data with the same frequency and dimension in three-dimensional space further employs a prediction module and a planning module. The prediction module is connected to the analysis module, and the planning module is connected to the prediction module. The prediction module predicts the data development at subsequent time nodes based on the analysis results of the analysis module, and the planning module makes advance plans based on the prediction results of the prediction module.

[0014] Preferably, the analysis module is connected to a feedback module, which is used to provide the analysis results to the user.

[0015] The significant advantages of this invention are: it can simultaneously ensure data sampling accuracy by displaying multiple sets of massive continuous data within a limited space, and perfectly preserve the temporal characteristics of the data, achieving observable and measurable visualization of both the overall and local data. Based on a 3D engine, this invention processes multiple sets of data of the same or different types at the same frequency and dimension, completely displaying massive amounts of data of the same or different types within a limited 3D space. This enables global observation of massive amounts of data within a limited display space, and ensures observability and measurability of both the overall and local data under the same sampling frequency and data length, fully preserving the overall characteristics of the data. This provides users with auxiliary decision-making tools to find potential patterns and modes from massive amounts of data. Attached Figure Description

[0016] Figure 1 This is a flowchart of the correlation analysis method for multiple sets of time-varying data with the same frequency and dimension in three-dimensional space according to the present invention.

[0017] Figure 2 This is a schematic diagram illustrating the cutting process of this invention.

[0018] Figure 3 This is a schematic diagram of the cut part of the present invention.

[0019] Figure 4 This is a block diagram illustrating the principle of the modules used in this invention. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0021] In the description of this invention, it should be understood that the terms "upper", "lower", "front", "rear", "left", "right", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0022] like Figure 1 As shown, the correlation analysis method for multiple sets of time-varying data with the same frequency and dimension in three-dimensional space according to the present invention includes the following steps: Step 1: Collect multiple sets of data of the same type and different types and process them with the same frequency and dimension; Step two: Render the data with different colors based on daily understanding of business attributes; Step 3: Generate a cube in three-dimensional space according to the time and date dimensions; Step 4: Stack new data cubes on top of existing cubes in three-dimensional space; Step 5: Perform mathematical analysis of the three-dimensional stacked data on the new data cube; Step six: Combine and analyze the temporal characteristics and intrinsic relationships of multiple sets of data of the same type and different types.

[0023] Step one specifically includes the following sub-steps: When the data starts from the initial time point, the data object continuously generates massive amounts of constantly changing values ​​over time during its operation. The data obtained through this method is continuous time-varying data, where time and data are directly dependent, and the values ​​change with time. The time dimension of the time-varying data with time-domain characteristics is extracted in 24-hour intervals to form a daily curve. The curves of different days are stacked in an orderly manner according to date and aligned by time, thus expanding the date dimension on the basis of the traditional daily curve. Each data point carries three corresponding attributes (value, time, and date), which are displayed in the form of bars in three-dimensional space. The second group of data is extracted in the same way and stacked on the bars of the first group of data according to date and time alignment, and so on. After stacking, a new data cube of value, time, and date is formed. Based on the three-dimensional engine, the expansion in dimensions, space, and value directions is realized, and massive amounts of data are fully displayed in a limited three-dimensional space. Time-varying data with time-domain characteristics refers to data obtained by intermittently counting values ​​with the same sampling precision within a certain time period. The values ​​are directly dependent on the time points, and the values ​​change over time. Time-varying data uses the current time point as the coordinate to record the numerical changes of the data object in real time. During the continuous operation of the data object, a massive amount of continuous running values ​​containing time points are continuously generated over time. There is a continuous correspondence between these values ​​and the time points. The set of these running values ​​is time-varying data with time-domain characteristics, which can realize the time-domain characteristic analysis of the data.

[0024] Step two specifically includes the following: When multiple sets of massive amounts of data are displayed in the same three-dimensional space, clear differentiation becomes essential. To achieve a more intuitive and layered representation of data fluctuations and distribution, different colors are used to render the data based on a daily understanding of business attributes, creating significant layering of multiple objects within the three-dimensional cube. In this invention, multiple sets of data of different types coexist, with values ​​for the same date and time simply superimposed numerically. To ensure data observability and facilitate analysis, different colors are used to render the data based on business attributes. This method of color differentiation based on a daily understanding of business attributes enhances the horizontal comparison between data. Layering the data with different colors makes it easier to distinguish and understand the data. Specifically, each unit of data is displayed in a three-dimensional space as a column, with the height of the column representing the magnitude of the value. By positively correlating the height of all the columns representing values ​​with their magnitudes, the values ​​are further processed by increasing their dimensionality. Different colors are used to render different value ranges. The unit data object column is assigned different colors within different value ranges, forming different color distinctions for different value ranges of a single data object. This increases the horizontal comparison between data objects and presents the overall numerical distribution and fluctuation trends macroscopically. Based on the full data cube, the range of any of the three dimensions can be adjusted as needed, achieving all permutations and combinations of adjustments to any dimensional range of the three dimensions. This forms a local spatial data cube under any specific condition, enabling local data cube analysis that satisfies any time domain and value range conditions.

[0025] In the new data cube of step four, the product formed by superimposing multiple sets of data constitutes combined data. There are clear color gradations between the multiple sets of data, allowing analysis of the influence relationship between time and values ​​using date and time. For example, analyzing electricity consumption in East China by date reveals that the total electricity consumption in East China is significantly higher in summer and winter than in spring and winter; by time, it shows that electricity consumption in East China is significantly higher during the day than at night in summer and winter. This is reflected in the three-dimensional data cube as the overall volume of summer and winter being significantly larger than that of spring and autumn; and the overall volume during the day being larger than that at night. Different data components can also be filtered to analyze the inherent correlation between data. For instance, in East China's annual electricity consumption, both summer and winter are peak seasons, but thermal power accounts for a larger proportion in winter. This is reflected in the three-dimensional data cube as, when red data is filtered out and only blue and green data are retained, the overall volume of summer values ​​is significantly larger than that of winter values.

[0026] Step five, mathematical analysis, involves the new data cube. The three dimensions of this data cube possess business significance in three-dimensional space, representing individual elements, ranges, and total quantities. By incorporating geometric algorithms, it's possible to calculate and display the extreme values, mean, and range of all values ​​within the data cube. Furthermore, it's possible to calculate and mine hidden key indicators such as the inherent correlations, numerical growth capabilities, and durations of data within intervals. Simultaneously, based on the calculation depth of the spatial cube's cross-section and volume, information such as the number of data points and the total number of values ​​within that time-domain interval can be extracted. Step five's mathematical analysis includes filtering and segmentation functions based on input conditions. This invention achieves a larger capacity display of massive, continuous, time-varying data across multiple groups within a limited three-dimensional space. The data cube, rendered using a 3D engine, can be filtered and segmented based on conditions. Geometric mathematical calculations, such as the area of ​​a specific cross-section or the volume of a data cube in a specific time domain, assist in in-depth data analysis. Mathematical algorithms applied to specific cross-sections or data cubes generated from data cubes formed in three-dimensional space after arbitrary dimensional filtering and segmentation enable comparative analysis of data such as cross-sectional area, volume, mean, and extreme values. Cutting, based on a data cube, allows for customized cutting along three data dimensions to create local spatial cubes under specific conditions. This provides a basis for analyzing the intrinsic relationships between multiple sets of local data. Specifically, it uses three dimensions: numerical, time, and date. The beginning and end of each dimension represent the data interval for that dimension. Dragging either end allows for cropping and retaining the data within the new endpoint, forming a local 3D data cube. Furthermore, it allows for filtering different data sets and combining them for display. For example, assuming three sets of data A, B, and C exist, the full-data 3D data cube displays the combined effect of these three sets. Set A can be filtered out, retaining sets B and C; alternatively, set B can be filtered out, retaining sets A and C. The results of these different data combinations allow for analysis of the intrinsic relationships between the data. The 3D data cube, based on a 3D rendering engine, supports arbitrary combinations and dragging cutting along the three data dimensions. During dragging, the data is dynamically "cut" and rendered using a 2D plane passing through the "drag point" and perpendicular to its axis. Figure 2 and Figure 3 As shown, when the left end of the X-axis, the "drag point," moves from position X1 to position X2 along the X-axis, the result of this dynamic process is the cutting off of the data within the three-dimensional space formed by the movement from surface A1 to surface A2, i.e., as... Figure 3The data within space V shown is cut off. Arbitrary dimension combination analysis involves adjusting the range of any three dimensions based on the full data cube. This creates local data cubes with arbitrary combinations of date / seasonal characteristics, time / moment characteristics, and numerical / different numerical object characteristics. Mathematical analysis is used to uncover explicit and implicit patterns and characteristics within the data cube. Customized dimension selection enables temporal characteristic analysis of the feature data cube, providing data support for user decision-making. In three-dimensional space, any cross-sectional view formed by combining any two dimensions has unique data value and business significance. Specifically: the date-time dimension combination, through the complete fluctuation curve of the three-dimensional data cube from 0:00 to 24:00, plays an important role in analyzing the changes in the time-domain characteristics of extreme value distribution of data, providing strong help and support for precise command and dispatch, early warning of emergencies, and scientific and efficient formulation of diversion strategies; the value-time dimension combination, through the different heights of the front and back data "pillars" between the three-dimensional data cube, generates a cross-sectional view of the maximum value for each minute of the year. On the time dimension, each coordinate point shows the fluctuation of the maximum value at that moment throughout the year, providing an analysis of the changes in the time-domain characteristics of the data maximum value affected by day and night throughout the year; the combination of values / different value objects, through the stacking and layering effect of the three-dimensional data cube, shows the impact of different data combinations on the whole under the condition that the time is constant, providing an analysis of the internal correlation between data. Any dimension axis can be moved and cut from either end of the axis to retain a three-dimensional data cube that satisfies both logical filtering conditions, thus forming a local data cube that satisfies specific conditions in any dimension, providing conditions for performing local data temporal characteristic analysis.

[0027] like Figure 4 As shown, the correlation analysis method for multiple sets of time-varying data with the same frequency and dimension in three-dimensional space according to the present invention further includes the following steps: Step 7: Based on the analysis results of Step 6, predict the data development at subsequent time points, such as the electricity consumption in the next few days. Step 8: Make advance plans based on the prediction results of Step 7, such as planning electricity usage in advance.

[0028] The correlation analysis method for multiple sets of time-varying data with the same frequency and dimension in three-dimensional space in this invention employs the following modules: The data acquisition module collects multiple sets of data of the same type but different types. The processing module, connected to the acquisition module, performs same-frequency and same-dimensional processing on the data acquired by the acquisition module; The rendering module connects to the processing module and performs color rendering on the data processed by the processing module based on the daily understanding of the business attributes. The generation module, connected to the rendering module, generates a cube in three-dimensional space according to the time and date dimensions. The building module connects to the generation module and stacks new data cubes on the basis of the cubes generated by the generation module in three-dimensional space. The analysis module, connected to the construction module, performs mathematical analysis of the three-dimensional stacked data of the new data cube. The combination module connects with the analysis module to perform combined analysis of the temporal characteristics and intrinsic relationships of multiple sets of data of the same or different types.

[0029] The correlation analysis method for multiple sets of time-varying data with the same frequency and dimension in three-dimensional space in this invention also employs a prediction module and a planning module. The prediction module is connected to the analysis module, and the planning module is connected to the prediction module. The prediction module predicts the data development at subsequent time nodes based on the analysis results of the analysis module, such as the electricity consumption in the next few days. The planning module can make advance plans based on the prediction results of the prediction module, such as planning electricity consumption arrangements in advance.

[0030] The analysis module is connected to a feedback module, which is used to provide the analysis results to users (such as local electricity management systems) for reference and judgment.

[0031] This invention extends two-dimensional time-varying data ("time-value") to a three-dimensional spatial display of values, moments, and dates. It combines and filters multiple dimensions within this three-dimensional space to generate specific cross-sectional diagrams or filtered data cubes. By treating these data cubes or cross-sections as single objects in three-dimensional space, it enables geometric and mathematical calculations to assist in in-depth data analysis, such as determining the specific cross-sectional area of ​​the data cube or the volume of a data cube in a specific time domain. Based on the specific cross-sectional diagrams or filtered data cubes generated through the combination and filtering of multiple dimensions in three-dimensional space, mathematical algorithms are applied to each data object "column" within the generated data cube or cross-section to achieve comparative analysis of extreme values, mean values, and other data from all data object "columns."

[0032] In summary, this invention can simultaneously display multiple sets of massive datasets, improving the readability of such data and uncovering the inherent relationships between them. When analyzing the interrelationships between multiple datasets, it departs from the traditional method of creating BI statistical charts and comparing charts directly. Instead, it stacks the data in a three-dimensional space according to time periods and uses color to create layers, making the distribution of data values ​​across all time periods readily apparent and facilitating the observation of overall data trends. This invention is used to analyze the inherent relationships between multiple stacked datasets, clearly and intuitively displaying detailed data, data trends, and time-varying characteristics in a three-dimensional space.

[0033] This invention not only enables the full display of multiple sets of data within the same space, but also allows users to filter and segment the 3D data cube of the full data according to their own needs, forming local data cubes for temporal characteristic analysis under specific conditions. It achieves the preservation of feature data while ensuring sampling accuracy, taking into account both overall and local data analysis, which is more conducive to the quantitative analysis of characteristics and the identification of patterns. Simultaneously, based on the mathematical analysis of the data cube, it fully explores the potential value of the data, thereby transforming complex time-varying data with temporal characteristics into clear, intuitive, and usable data that facilitates the exploration of potential value and the identification of patterns.

[0034] The data objects supported by this invention can span multiple time-varying data with time-domain characteristics, and any loading of multiple groups of time-varying data with time-domain characteristics can achieve a clear and intuitive data cube display. This invention not only achieves a comprehensive and intuitive display of the entire data set but also enables cross-day comparisons of time-varying data. This makes the time-domain characteristics of data trend changes at the same point in time on different dates throughout the year clearer and more intuitive. The data cube, rendered using a 3D engine, can also be rotated and scaled, allowing users to achieve a globally observable and measurable view of the entire running data (minutes / seconds) from different sides and angles within a limited page.

[0035] Addressing the temporal characteristics of massive, continuously running data, this invention enables a larger-capacity display of such data within a limited visual space. Traditional two-dimensional curves for displaying annual data at the minute and second level require continuously extending or compressing the coordinate axes, increasing visual unavailability and loss of data accuracy. This invention, without sacrificing sampling accuracy, eliminates the need for lengthy and monotonous two-dimensional curves, making the temporal characteristics of time-varying data, such as its size and distribution, readily apparent in three-dimensional space. This facilitates the analysis of data fluctuation patterns across different seasons and specific dates.

[0036] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0037] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for correlation analysis of multiple sets of time-varying data with the same frequency and dimension in three-dimensional space, characterized in that, The method for correlation analysis of multiple sets of time-varying data with the same frequency and dimension in three-dimensional space includes the following steps: Step 1: Collect multiple sets of data of the same type and different types and process them with the same frequency and dimension; Step two: Render the data with different colors based on daily understanding of business attributes; Step 3: Generate a cube in three-dimensional space according to the time and date dimensions; Step 4: Stack new data cubes on top of existing cubes in three-dimensional space; Step 5: Perform mathematical analysis of the three-dimensional stacked data on the new data cube; Step 6: Combine and analyze the temporal characteristics and internal relationships of multiple sets of data of the same type and different types; Step one specifically includes the following sub-steps: The time dimension of time-varying data with time-domain characteristics is extracted in 24-hour intervals to form a daily curve. Curves from different days are then stacked in order by date and aligned by time, thus expanding the date dimension beyond the traditional daily curve. Each data point carries three attributes: numerical value, time, and date, displayed in a bar-like format in three-dimensional space. The second group of data is extracted using the same pattern: 'the time dimension of time-varying data with time-domain characteristics is extracted in 24-hour intervals to form a daily curve,' and stacked on top of the bars of the first group of data, aligned by date and time. This process is repeated until a new three-dimensional data cube of numerical value, time, and date is formed. Based on a three-dimensional engine, the expansion in dimensions, space, and numerical direction is achieved, allowing massive amounts of data to be fully displayed in a limited three-dimensional space. The time-varying data with time-domain characteristics refers to data obtained by intermittently counting values ​​with the same sampling precision within a certain time period. The values ​​are directly dependent on the time points, and the values ​​change with time. When the time-varying data uses the current time point as the coordinate, it records the changes in the values ​​of the data object in real time. During the continuous operation of the data object, a massive amount of continuous running values ​​containing time points are continuously generated over time. There is a continuous correspondence between the values ​​and the time points. The set of these running values ​​is the time-varying data with time-domain characteristics.

2. The method for correlation analysis of multiple sets of time-varying data with the same frequency and dimension in three-dimensional space as described in claim 1, characterized in that, Step two creates layers of multiple objects within a three-dimensional cube.

3. The method for correlation analysis of multiple sets of time-varying data with the same frequency and dimension in three-dimensional space as described in claim 1, characterized in that, The mathematical analysis in step five includes filtering and segmentation functions based on input conditions.

4. The method for correlation analysis of multiple sets of time-varying data with the same frequency and dimension in three-dimensional space as described in claim 1, characterized in that, The correlation analysis method for multiple sets of time-varying data with the same frequency and dimension in three-dimensional space also includes the following steps: Step 7: Based on the analysis results of Step 6, predict the data development at subsequent time points; Step 8: Make advance plans based on the prediction results from Step 7.

5. The method for correlation analysis of multiple sets of time-varying data with the same frequency and dimension in three-dimensional space as described in claim 1, characterized in that, The correlation analysis method for multiple sets of time-varying data with the same frequency and dimension in three-dimensional space adopts the following modules: The data acquisition module collects multiple sets of data of the same type but different types. The processing module is connected to the acquisition module and performs same-frequency, same-dimensional processing on the data acquired by the acquisition module. The rendering module connects to the processing module and performs color rendering on the data processed by the processing module based on the daily understanding of the business attributes. The generation module, connected to the rendering module, generates a cube in three-dimensional space according to the time and date dimensions. The building module connects to the generation module and stacks new data cubes on the basis of the cubes generated by the generation module in three-dimensional space. The analysis module, connected to the construction module, performs mathematical analysis of the three-dimensional stacked data of the new data cube. The combination module connects with the analysis module to perform combined analysis of the temporal characteristics and intrinsic relationships of multiple sets of data of the same or different types.

6. The method for correlation analysis of multiple sets of time-varying data with the same frequency and dimension in three-dimensional space as described in claim 5, characterized in that, The correlation analysis method for multiple sets of time-varying data with the same frequency and dimension in three-dimensional space also employs a prediction module and a planning module. The prediction module is connected to the analysis module, and the planning module is connected to the prediction module. The prediction module predicts the data development at subsequent time nodes based on the analysis results of the analysis module, and the planning module makes advance plans based on the prediction results of the prediction module.

7. The method for correlation analysis of multiple sets of time-varying data with the same frequency and dimension in three-dimensional space as described in claim 5, characterized in that, The analysis module is connected to a feedback module, which is used to provide the analysis results to the user.

Images