Big data visual electric energy metering analysis and prediction platform based on end-cloud collaboration

Abstract

The invention discloses a big data visualization electric energy metering analysis and prediction platform based on end-cloud collaboration. A terminal comprises a power grid data acquisition module and a power grid data screening module, and a cloud comprises a power grid data visualization platform and a next-quarter power consumption prediction and analysis module. System power data of all voltage levels of a power grid are collected through the power grid information collection module and transmitted to the database, the data are classified through the power grid data screening module, and a visual model diagram is established through the power grid data display visualization module. And then the next-quarter electricity consumption prediction analysis module analyzes and dispatches the selected electric energy data and displays the result on the visual platform, through the end-cloud cooperation technology, comparison and fusion of new data and old data acquired in real time are facilitated, and the reliability of prediction after perception, transmission and processing of the visual platform is further ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of big data visualization, and particularly relates to a big data visualization power metering analysis and prediction platform based on end-cloud collaboration. Background Art

[0002] Big data of end-cloud collaboration refers to connecting various terminal devices, such as smart phones, electronic devices, sensors, etc., through a wireless sensor network to collect, transmit, and store various digital information in real time. These data form a huge data set after being processed, that is, big data. The cloud integrates, stores, analyzes, mines, and other processes these massive and multi-source data, and then feeds back corresponding instructions or information to the terminal devices according to the processing results, so as to achieve the efficient transfer and collaborative work of data between the terminal and the cloud, thereby giving play to the advantages of big data.

[0003] The Internet of Things visualization technology can display Internet of Things data through a graphical interface to help users quickly understand and analyze the data; for example, the patent with the publication number CN119298405A discloses a power metering device load characteristic acquisition and analysis system based on Internet of Things perception. This patent can accurately identify the load characteristics of devices by building a feature library in the meter module and combining the real-time acquisition and analysis of current data. For newly connected electrical devices, the system identifies their types by comparing features such as current waveforms, frequency distributions, and harmonic components and matches them with the feature library to achieve real-time monitoring and accurate identification of the load characteristics of electrical devices, ensuring the high efficiency and flexibility of system processing to a certain extent; however, in actual applications, due to the huge data sources generated by the Internet of Things, the existing visualization tools still have deficiencies in processing speed and efficiency and are difficult to achieve real-time dynamic visualization display. Secondly, the scale of the power grids in major cities is huge, and the user demand has increased exponentially, and the demand for power grid information visualization is expanding day by day. The current power metering visualization platform has limitations in many aspects such as incomplete information transmission, inaccurate and untimely prediction, and inactive customer service. Therefore, the existing methods in the prior art still lead to problems such as the difficulty for users to obtain real-time power grid information and low information aggregation efficiency.

[0004] Therefore, a big data visualization power metering analysis and prediction platform based on end-cloud collaboration is needed to solve the above technical problems. Summary of the Invention

[0005] To solve the above technical problems, the present invention discloses a big data visualization power metering analysis and prediction platform based on end-cloud collaboration. The big data visualization power metering analysis and prediction platform proposed by the present invention collects system power data of each voltage level of the power grid through a power grid information collection module, establishes a visualization model diagram using a power grid data display visualization module, and then analyzes and schedules the selected data through a next-quarter power consumption prediction analysis module and displays the results on the visualization platform, realizing the comparison and fusion of newly acquired data and old data in real time. Through the cooperation between the terminal device and the cloud, the terminal transmits the collected data to the cloud, and after processing, the cloud can transmit relevant instructions or analysis results back to the terminal, further ensuring the reliability of the perception, transmission, processing, and prediction of the visualization platform.

[0006] To achieve the above technical effects, the present invention designs a big data visualization power metering analysis and prediction platform based on end-cloud collaboration, including: a power grid data collection module, a power grid data screening module, a power grid data visualization platform, and a next-quarter power consumption prediction analysis module; signal connections are established between the power grid data collection module, the power grid data screening module, and the power grid data visualization platform; at the same time, the power grid data screening module is bidirectionally connected to the next-quarter power consumption prediction analysis module and is signal-connected to the power grid data collection module and the power grid data visualization platform. Through the cooperation between the terminal device and the cloud, the terminal transmits the collected data to the cloud, and after processing, the cloud can transmit relevant instructions or analysis results back to the terminal, realizing efficient data interaction and collaborative work; the power grid data collection module includes a data collection unit, a big data real-time refresh unit, an Internet of Things database, and a data transmission unit, the Internet of Things database includes a data reception and fusion unit and a replacement unit, the power grid data screening module includes a data statistics unit and a data classification unit, the power grid data visualization platform includes a data reception unit, a big data modeling unit, and a visualization display interface, and the next-quarter power consumption prediction analysis module includes a big data analysis and prediction unit and a data transmission unit.

[0007] Further, the Internet of Things database is used to store the data collected by the data collection unit, the real-time refresh unit is applied to refresh the data information on the cloud platform in real time, and the data transmission unit is applied to transmit the collected data to the data screening unit.

[0008] Further, the data reception and fusion unit receives the collected data, integrates the new data into the old data in the Internet of Things database, and simply processes and classifies the data, and the replacement unit deletes the old data and retains the content of the new data.

[0009] Further, in the grid data screening module, the grid information is statistically analyzed by the data statistics unit, and the big data is classified by the data classification unit. The decision tree algorithm is used for classification by the data classification unit, and the data of the same type is classified into one category.

[0010] Further, the data receiving unit is used to receive the information transmitted by the data screening module. The big data modeling unit is used to establish line charts, bar charts, column charts, stacked charts, pie charts, and histograms of different types of data, and transfer the results to the visual display interface. The visual display interface receives and displays the model charts established by the big data modeling unit.

[0011] Further, the next-quarter electricity consumption prediction and analysis module is based on the data transmitted by the data screening unit, and combines the electricity consumption situation of the city in the previous quarter, seasonal factors, upstream and downstream events, etc., to conduct comprehensive analysis, predict the electricity consumption situation in the next quarter, clarify the peak electricity consumption time period, generate prediction data, and transfer it to the grid data visualization platform.

[0012] Further, the big data analysis and prediction unit is used for multivariate analysis and modeling of data, and predicts the electricity consumption situation in the next quarter through historical data. The data transmission unit is used to transfer the results to the grid data visualization platform. Through the cooperation of the terminal device and the cloud, the terminal transmits the collected data to the cloud, and the cloud transmits relevant instructions or analysis results back to the terminal after processing.

[0013] Further, the big data analysis and prediction unit adopts the seasonal autoregressive integrated moving average model (ARIMA). The general expression of the seasonal autoregressive integrated moving average model is, where p is the autoregressive order, d is the differencing order, and q is the moving average order; its algorithm formula is based on the combination of three models: autoregressive (AR), integration (I), and moving average (MA).

[0014] Further, the method for combining models includes the following steps:

[0015] S1: Autoregressive part: The formula of the autoregressive model AR(p) is:

[0016]

[0017] In the formula, y t is the value of the time series at time, φ1, φ2, …, φ p are autoregressive coefficients, y t-1 , y t-2 , …, y t-p are the lagged values of the time series, ∈ t is the white noise error term, representing the random fluctuation that cannot be explained by the model at time. Usually, it is assumed that ∈t obeys a normal distribution with a mean of 0 and a variance of σ 2 ;

[0018] S2: Moving average part: The formula for the moving average model MA(q) is:

[0019] y t = ∈ t + θ1∈ t-1 + θ2∈ t-2 + … + θ q ∈ t-q (2)

[0020] In the formula, θ1, θ2, …, θ q are the moving average coefficients, and ∈ t-1 , ∈ t-2 , …, ∈ t-q are the white noise error terms at past times;

[0021] S3: Integration part: The formula for the second-order difference used in the integration part algorithm is:

[0022] Δ 2 y t = Δy t - Δy t-1 = y t - 2y t-1 + y t-2 (3)

[0023] S4: Combine the autoregressive, moving average, and integration parts to obtain the formula for the ARIMA(p, m, q) model:

[0024] Φ(B)Δ d y t = Θ(B)∈ t (4)

[0025] In the formula, B is the lag operator, By k y t = y t-k , Φ(B) = 1 - φ1B - φ2B 2 - … - φ p B p is the autoregressive polynomial, and Θ(B) = 1 + θ1B + θ2B 2 + … + θ q B q is the moving average polynomial;

[0026] Among them, the autoregressive polynomial introduces the non-stationary random signal sequence {δ n}, let the m-th difference be d m δ m = δ n(n > m), if {γ n} is a stationary ARMA(p, q) sequence, that is:

[0027] γ n - b1γ n-1 - b2γ n-2 - … - b p γ n-p = W n - a1W n-1 - a2W n-2 - … - a q W n-q (5)

[0028] Then the signal model in the form of Equation (5) obtained from the non-stationary random signal sequence {δ n} is called an m-th order summation ARMA(p, q) model and is denoted by the symbol ARIMA(p, m, q);

[0029] According to dδ n = δ n - δ n-1 = (1 - D)·δ n , the difference operator d and the time-delay operator D are introduced, and the relationship is d = (1 - D); and

[0030]

[0031] The m-th order difference sequence of {δ n} is expressed as:

[0032] d m δ n = (1 - D) m δ n = γ n (7)

[0033] Then the time-domain expression of the signal model of ARIMA(p, m, q) can be written as:

[0034] B(D)(1 - D) m δ n = A(d)W n (8)

[0035] S5: By performing stationarity processing and model parameter estimation on the time series, determining θ j 、p、q、m、d、D and other parameter values, the ARIMA model can be used to predict and analyze power data.

[0036] The beneficial effects of the present invention are:

[0037] 1. The present invention designs a big data visualization power metering analysis and prediction platform based on end-cloud collaboration, including: a power grid data collection module, a power grid data screening module, a power grid data visualization platform, and a next-quarter power consumption prediction and analysis module; signal connections are established among the power grid data collection module, the power grid data screening module, and the power grid data visualization platform; at the same time, the power grid data screening module is bidirectionally connected to the next-quarter power consumption prediction and analysis module, and is signal-connected to the power grid data collection module and the power grid data visualization platform. Through the cooperation between the terminal device and the cloud, the terminal transmits the collected data to the cloud, and after being processed by the cloud, relevant instructions or analysis results can be transmitted back to the terminal to achieve efficient data interaction and collaborative work.

[0038] 2. The present invention uses the data collection module and the data screening module, and integrates and processes the data based on the cloud collaboration technology. Among them, the data collection module has a powerful access ability and can access multi-source heterogeneous data such as various sensing front-ends and business systems, effectively breaking down the barriers between data and eliminating the phenomenon of data islands. At the same time, the big data real-time refresh unit and the replacement unit can timely integrate new data into old data, continuously expanding the data storage content inside the Internet of Things database and making the database richer and more comprehensive. With its high-efficient data processing ability, the data screening module can quickly screen and analyze a large amount of power grid data, realizing real-time and accurate monitoring of the power grid operation status and in-depth information analysis. This feature not only ensures the security of data transmission, but also greatly improves the stability of data transmission, providing a solid and reliable data foundation for subsequent data analysis and applications.

[0039] 3. Through the set next-quarter power consumption prediction and analysis module, by comprehensively analyzing various key factors such as historical power consumption data, weather conditions, and economic development trends, and using advanced technologies such as statistical methods, comparison methods, and correlation analysis methods to build models and deeply analyze the data, it can more accurately provide the prediction results of the next-quarter power consumption demand. Based on this accurate prediction, power enterprises can more reasonably allocate power resources according to the power consumption demand in different time periods. This can not only ensure the stable operation of the power grid under various conditions, avoid the situation of insufficient or excessive power supply, but also effectively reduce the operating costs of power enterprises, realize the optimal allocation of power resources, and improve the utilization efficiency of power resources.

[0040] 4. Through the set big data modeling unit, diverse data charts such as line charts, bar charts, bar graphs, stacked charts, pie charts, and histograms can be established for different types of data, and the modeling analysis results are transmitted to the visual display interface. Through the visual display interface, data such as power grid-side data and user-side data can be intuitively and dynamically displayed. This intuitive data display method is applicable to multiple different application scenarios such as production, scheduling, safety supervision, and marketing. Brief Description of the Drawings

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments.

[0042] Figure 1 is the working flow chart of the Internet of Things big data visualization platform of the present invention;

[0043] Figure 2 is the schematic diagram of the module architecture of the visualization platform of the embodiments of the present invention;

[0044] Figure 3 is the module diagram of the Internet of Things database of the present invention;

[0045] Figure 4 is the flow chart of the prediction algorithm of the present invention;

[0046] Figure 5 is the comparison chart between the ARIMA sequence prediction diagram of the prediction algorithm of the present invention and the actual prediction value. Detailed Embodiments

[0047] Embodiment 1

[0048] A big data visualization power metering analysis and prediction platform based on end-cloud collaboration, comprising: a power grid data acquisition module, a power grid data screening module, a power grid data visualization platform, and a next-quarter power consumption prediction analysis module; signal connections are established between the power grid data acquisition module, the power grid data screening module, and the power grid data visualization platform; at the same time, the power grid data screening module is bidirectionally connected to the next-quarter power consumption prediction analysis module and is signal-connected to the power grid data acquisition module and the power grid data visualization platform. Through the cooperation of the terminal device and the cloud, the terminal transmits the collected data to the cloud, and the cloud can transmit relevant instructions or analysis results back to the terminal after processing, realizing efficient data interaction and collaborative work; the power grid data acquisition module includes a data acquisition unit, a big data real-time refresh unit, an Internet of Things database, and a data transmission unit, the Internet of Things database includes a data reception and fusion unit and a replacement unit, the power grid data screening module includes a data statistics unit and a data classification unit, the power grid data visualization platform includes a data reception unit, a big data modeling unit, and a visualization display interface, and the next-quarter power consumption prediction analysis module includes a big data analysis and prediction unit and a data transmission unit.

[0049] The power grid data acquisition module accesses OMS information, distribution information, relevant equipment signals, and power-on event storage data, and is used to collect power data of power grid systems at various voltage levels. Through the big data real-time refresh unit, the Internet of Things database, and the data transmission unit, the information is simply analyzed and stored and reported.

[0050] Power grid data screening module. This power grid information screening module performs secondary screening on information based on the Internet of Things database constructed by all power flow events through voltage levels, classifies the statistical data, and stores the events in the database.

[0051] Power grid data visualization platform. This power grid data visualization platform respectively makes signal connections between the power grid data screening module and the next-quarter power consumption prediction and analysis module through a data receiving unit. Among them, the big data modeling unit transmits the established model diagram to the visualization display interface through a signal transmission module, and the model diagram is displayed through the visualization display interface. Next-quarter power consumption prediction and analysis module. This next-quarter power consumption prediction and analysis module predicts the distribution of power flow information in each region in the next quarter based on the Internet of Things database refreshed by the power grid data screening module through a big data analysis and prediction unit, and reports the events to the power grid data visualization platform through a data transmission unit. In this embodiment, through the setting of the big data real-time refresh unit in the power grid data acquisition module, it is convenient to perform real-time refresh on the data information on the cloud platform. After the refresh, the data information acquisition unit collects the real-time information on the Internet of Things platform and stores the data only in the Internet of Things database. As Figure 2 shown, the new data is fused into the old data through the data receiving and fusion unit of the Internet of Things database to achieve centralized storage of new and old data. Through the replacement unit of the Internet of Things database, the duplicate information is replaced and deleted. Finally, through the setting of the signal transmission module in the module, the collected big data is transmitted to the data screening module, enabling simple classification of the information in advance before transmission, which is convenient for the data screening module to perform secondary complex classification on the data.

[0052] Existing power grid perception visualization technologies can only perceive power flow events related to the main grid for the power grid. However, in actual applications, the existing visualization tools still have deficiencies in processing speed and efficiency, and it is difficult to achieve real-time dynamic visualization display. Secondly, currently, the power grid scales of major cities are huge, and user demands are increasing in a blowout manner. The demand for power grid information visualization is expanding day by day. The current power grid visualization platforms have problems such as incomplete information transmission, inaccurate and untimely prediction.

[0053] For example, a power outage in the main grid of the power grid affects both the distribution network associated with the main grid downward and the low-voltage side associated with the distribution network downward. However, the existing information management and control technologies do not consider the upstream and downstream association relationships, resulting in incomplete and inaccurate perception of each event, thus causing low efficiency of power grid system management and control, untimely response, and affecting the safety and stability of the entire power grid operation;

[0054] In the present invention, through the data acquisition module and the data screening module, data is integrated and processed based on cloud collaboration technology. The data acquisition module can access various types of multi-source heterogeneous data such as sensing front-ends and business systems, breaking data islands. Through the big data real-time refresh unit and the replacement unit, new data is incorporated into old data, thus expanding the data content stored in the Internet of Things database. The data screening module can quickly screen and analyze a large amount of power grid data through its efficient data processing ability, so as to realize the real-time monitoring and information analysis of the power grid operation status, and ensure the security and stability of data transmission. It effectively solves the problem that existing information control technologies do not consider the upstream and downstream correlation relationships, resulting in incomplete and inaccurate perception of each event, thus causing low control efficiency of the power grid system, untimely response, and affecting the safety and stability of the entire power grid operation.

[0055] In addition, in this embodiment, due to the lack of a prediction module in the traditional power grid visualization platform, it is difficult to specifically compare the power flow differences between the previous quarter and the next quarter, and thus formulate a suitable power distribution plan. The present invention analyzes various factors such as historical power consumption data, weather conditions, and economic development trends through the next quarter power consumption prediction analysis module, and uses technologies such as statistical method, comparison method, and correlation analysis method to build models and analyze data, and can relatively accurately provide the power consumption demand prediction for the next quarter. Based on the prediction results, power enterprises can more reasonably allocate power resources, ensure the stable operation of the power grid at different time periods, reduce operation costs at the same time, and optimize the allocation of power resources. Specifically, by constructing a power supply information acquisition module covering all voltage levels of the main grid, distribution network, and users, all power flow events at all voltage levels can be perceived, so as to better monitor the dynamics of the entire power supply network, identify potential faults or problems, and more accurately predict possible risks.

[0056] Embodiment 2

[0057] In this embodiment, as Figure 3 shown, the power grid data visualization platform includes a data receiving unit, a big data modeling unit, and a visualization display interface. In the power grid data visualization platform, the data receiving unit receives the data information reported from the data screening unit and the next quarter power consumption prediction analysis module. The big data modeling unit establishes line charts, bar charts, column charts, stacked charts, pie charts, and histograms of different types of data, and conveys the established model charts to the visualization display interface, and the model charts are displayed through the visualization display interface. In this embodiment, the data receiving unit receives the classified data information of the power grid data screening module, and the big data modeling unit establishes display graphs of different types of data, and transmits the established model charts to the visualization display interface, and the model charts are displayed through the visualization display interface, which is convenient for displaying different types of data visualization model charts and intuitively showing the power consumption situation of the city in this quarter and the prediction of the power consumption situation of the city in the next quarter.

[0058] Example 3

[0059] As Figure 4 shown, in the electricity consumption prediction and analysis module for the next quarter, the big data analysis and prediction unit predicts future trends through historical data, and uses methods such as statistical method, comparison method, and correlation analysis method to establish a model to analyze and predict data. The seasonal autoregressive integrated moving average model is adopted. For non-stationary time series containing trends or seasons, appropriate successive differences and seasonal differences must be performed to eliminate the trend influence, and then an ARIMA model is established for the new stationary series formed for analysis, and the information analyzed is used to predict the changes in power grid big data. Table 1 below shows the evaluation index parameters of the ARIMA model prediction data.

[0060] Table 1 Evaluation Index Parameters of ARIMA Model Prediction Data

[0061]

[0062] In this embodiment, the prediction and analysis module obtains historical data, establishes a seasonal mathematical model using various calculation methods, establishes a model for the historical data to predict and analyze the data, finds the correlation between the data, understands the relationship between different variables, and uses the analyzed information to predict the changes in power grid big data. The results are transmitted to the power grid data visualization platform through the data transmission unit, realizing the comparison and fusion of the newly obtained data and the old data in real time. Through the cooperation between the terminal device and the cloud, the terminal transmits the collected data to the cloud, and the cloud can transmit relevant instructions or analysis results back to the terminal after processing, further ensuring the reliability of the prediction after the visualization platform perceives, transmits, and processes.

[0063] In summary, first, through the power grid information collection module, the present invention can comprehensively and accurately collect the system power data of each voltage level of the power grid, providing a solid data foundation for subsequent analysis and prediction. Secondly, the power grid data display and visualization module can establish an intuitive and clear visualization model diagram based on the collected data. This model diagram can present complex power data in a graphical manner, helping users quickly understand and analyze the data. Furthermore, the electricity consumption prediction and analysis module for the next quarter will deeply analyze and reasonably schedule the data selected by the user, and display the analysis results in real time and accurately on the visualization platform. This platform can realize the comparison and integration of newly acquired data and old data in real time. Specifically, through the close cooperation between the terminal device and the cloud, the terminal device timely transmits the collected data to the cloud, and after a series of professional and efficient processing by the cloud, relevant instructions or analysis results can be quickly transmitted back to the terminal. This end-cloud collaborative working mode further ensures the reliability of the visualization platform in predicting after perceiving, transmitting, and processing data, greatly improving the convenience for users to obtain real-time power grid information, while enhancing the efficiency of information collection, effectively solving many problems existing in the prior art, and providing strong support for the efficient operation and scientific management of the power grid.

[0064] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor limit the invention to the specific embodiments described.

Claims

1. A big data visualization electric energy metering analysis and prediction platform based on end-cloud collaboration, characterized in that: include: Power grid data collection module, power grid data screening module, power grid data visualization platform, and next quarter electricity consumption forecast analysis module; Signal connections are established between the power grid data acquisition module, the power grid data screening module, and the power grid data visualization platform; at the same time, the power grid data screening module realizes a two-way connection with the next quarter's electricity consumption forecast analysis module, and a signal connection is established between the power grid data acquisition module and the power grid data visualization platform. Through the cooperation between the terminal device and the cloud, the terminal transmits the collected data to the cloud, and the cloud can transmit the relevant instructions or analysis results back to the terminal after processing, so as to realize efficient data interaction and collaborative work; the power grid data acquisition module includes a data acquisition unit, a big data real-time refresh unit, an Internet of Things database and a data transmission unit, the Internet of Things database includes a data receiving and fusion unit and a replacement unit, the power grid data screening module includes a data statistics unit and a data classification unit, the power grid data visualization platform includes a data receiving unit, a big data modeling unit, and a visualization display interface, and the next quarter's electricity consumption forecast analysis module includes a big data analysis and prediction unit and a data transmission unit.

2. According to the big data visualization electric energy metering analysis and prediction platform based on end-cloud collaboration as described in claim 1, it is characterized in that: The Internet of Things database is used to store data collected by the data collection unit, the real-time refresh unit is used to refresh the data information on the cloud platform in real time, and the data transmission unit is used to transfer the collected data to the data screening unit.

3. According to the big data visualization electric energy metering analysis and prediction platform based on end-cloud collaboration as described in claim 1, it is characterized in that: The data receiving and fusion unit receives the collected data, integrates the new data into the old data in the Internet of Things database, and performs simple processing and classification on the data. The replacement unit block deletes the old data and retains the content of the new data.

4. According to the big data visualization electric energy metering analysis and prediction platform based on end-cloud collaboration as described in claim 1, it is characterized in that: In the power grid data screening module, the power grid information is counted by the data statistics unit, and the big data is classified by the data classification unit. The classification of the data classification unit adopts a decision tree algorithm to classify the same type of data into one category.

5. According to the big data visualization electric energy metering analysis and prediction platform based on end-cloud collaboration as described in claim 1, it is characterized in that: The data receiving unit is used to receive information transmitted by the data screening module, and the big data modeling unit is used to establish line charts, column charts, bar charts, stacked charts, pie charts, and histograms of different types of data, and pass the results to the visualization display interface. The visualization display interface receives the model diagram established by the big data modeling unit and displays it.

6. According to the big data visualization electric energy metering analysis and prediction platform based on end-cloud collaboration as described in claim 1, it is characterized in that: The next quarter's electricity consumption forecasting and analysis module performs a comprehensive analysis based on the data transmitted by the data screening unit and in combination with the city's electricity consumption in the previous quarter, seasonal factors, upstream and downstream events, etc., to predict the next quarter's electricity consumption and identify the peak electricity consumption period, generate forecast data and transmit it to the power grid data visualization platform.

7. According to the big data visualization electric energy metering analysis and prediction platform based on end-cloud collaboration as described in claim 1, it is characterized in that: The big data analysis and prediction unit is applied to multivariate analysis modeling of data to predict the electricity consumption in the next quarter through historical data. The data transmission unit is applied to transmit the results to the power grid data visualization platform. Through the cooperation between the terminal device and the cloud, the terminal transmits the collected data to the cloud, and the cloud returns the relevant instructions or analysis results to the terminal after processing.

8. According to the big data visualization electric energy metering analysis and prediction platform based on end-cloud collaboration as described in claim 1, it is characterized in that: The big data analysis and prediction unit adopts the seasonal autoregressive integrated moving average model ARIMA. The general expression of the seasonal autoregressive integrated moving average model is, where p is the autoregressive order, d is the difference order, and q is the moving average order; Its algorithm formula is based on a combination of three models: autoregression, integration and moving average.

9. According to the big data visualization electric energy metering analysis and prediction platform based on end-cloud collaboration as described in claim 1, it is characterized in that: The model combination method comprises the following steps: S1: Autoregressive part: The formula of the autoregressive model AR(p) is: In the formula, y t is the value of the time series at time, φ1, φ2, …, φ p is the autoregressive coefficient, y t-1 ,y t-2 ,…,y t-p is the lagged value of the time series, ∈ t is a white noise error term, representing random fluctuations that cannot be explained by the model at time instant, usually assumed to be ∈ t The mean is 0 and the variance is σ 2 Normal distribution of S2: Moving average part: The formula of the moving average model MA(q) is: y t =∈ t +θ1∈ t-1 +θ2∈ t-2 +…+θ q ∈ t-q (2) In the formula, θ1, θ2, …, θ q is the moving average coefficient, ∈ t-1 ,∈ t-2 ,…,∈ t-q is the white noise error term at the past moment; S3: Integral part: The second-order difference formula used in the integral part algorithm is: Δ 2 y t =Δy t -Δy t-1 =y t -2y t-1 +y t-2 (3) S4: Combining the autoregressive, moving average and integral parts, we get the formula for the ARIMA(p,m,q) model: F(B)D d y t =Θ(B)∈ t (4) Where B is the lag operator, B k y t =y t-k , Φ(B)=1-φ1B-φ2B 2 -…-φ p B p is an autoregressive polynomial, θ(B)=1+θ1B+θ2B 2 +…+θ q B q is the moving average polynomial; The autoregressive polynomial introduces a non-stationary random signal sequence {δ n }, let the mth difference be d m δ m =δ n (n>m), if {γ n } is a stationary ARMA(p,q) sequence, that is: c n -b1γ n-1 -b2c n-2 -…-b p c n-p =W n -a1W n-1 -a2W n-2 -…-a q W n-q (5) Then the non-stationary random signal sequence {δ n The signal model obtained as shown in formula (5) is called the m-order summation ARMA (p, q) model and is represented by the symbol ARIMA (p, m, q); According to dδ n =δ n -δ n-1 =(1-D)·δ n , introduce the difference operator d and the delay operator D, the relationship is d = (1-D); and Get {δ n The m-order difference sequence of} is expressed as: d m d n =(1-D) m d n =c n (7) Then the time domain expression of the ARIMA (p, m, q) signal model can be written as: B(D)(1-D) m δ n =A(d)W n (8) S5: Determine by processing the time series for stationarity and estimating model parameters θ j By knowing the values ​​of parameters such as p, q, m, d, and D, the ARIMA model can be used to predict and analyze electric energy data.

10. Application of a big data visualization electric energy metering analysis and prediction platform based on end-cloud collaboration as described in any one of claims 1 to 9 in the field of big data visualization technology.

Images