Data acquisition method based on electric energy meter

Abstract

The invention relates to the field of data processing, in particular to a data acquisition method based on an electric energy meter, which comprises the following steps: deploying an intelligent electric energy meter in a workshop to acquire power consumption data of equipment, transmitting the data to a central database, and predicting future electric quantity by adopting an improved ARIMA model based on preprocessed data. According to the model, relative fluctuation sensitivity, equipment state correction and an abnormal significance index are introduced on the basis of traditional ARIMA, and dynamic suppression of prediction residual errors is achieved. And after the final predicted value is compared with the actual power consumption data, when the deviation exceeds a preset threshold value, the system automatically triggers early warning, and timely notifies maintenance personnel to check and repair equipment abnormity. According to the invention, the data is acquired through the intelligent electric energy meter, and the real-time prediction and early warning of the abnormal power consumption of the industrial equipment are realized by combining the improved ARIMA model and the dynamic residual suppression.

Description

Technical Field

[0001] The invention relates to the field of data processing, and in particular to a data acquisition method based on an electric energy meter. Background Art

[0002] With the continuous improvement of the level of industrial automation, real-time monitoring and status evaluation of production equipment play a vital role in ensuring workshop safety and improving production efficiency. In industrial workshops, various production equipment are equipped with electricity meters to record electricity consumption data in real time. Under normal circumstances, the power consumption of equipment during operation is stable and periodic; but once the equipment fails or is subject to external interference, its power consumption will suddenly change and surge abnormally, which may cause safety accidents or production interruptions. Therefore, how to accurately judge the operating status of equipment through real-time collection and predictive analysis of power data has become a technical problem that needs to be solved urgently.

[0003] In the existing technology, the ARIMA (Autoregressive Integrated Moving Average) model is often used in electricity data prediction because it can better capture the trend and periodicity of time series data. The model first determines the stationarity of the data through the ADF test, and then uses methods such as autoregression, difference and moving average to establish a mathematical model to predict future electricity consumption. However, the ARIMA model has certain limitations: it assumes that the data series is stable and linear. When encountering nonlinear mutations such as equipment failures, the prediction effect is often not ideal, and the model fitting accuracy is easily affected by abnormal data. In addition, linear modeling that relies solely on historical data is difficult to reflect the instantaneous fluctuations caused by changes in equipment status, which in turn affects the timeliness of fault warnings. Summary of the invention

[0004] In view of the problem that the above-mentioned ARIMA model is easily affected by abnormal data in model fitting accuracy, the present invention proposes a data acquisition method based on an electric energy meter, including: obtaining historical power data of an industrial workshop, and training an ARIMA prediction model based on the historical power data; obtaining a real-time power time series of a set length and inputting it into the trained ARIMA prediction model to obtain a predicted value; calculating the difference between the predicted value and the real-time power, and in response to the difference being greater than a set threshold, the system notifies maintenance personnel to perform a fault inspection; the ARIMA prediction model also includes processing the original residual to obtain an abnormal dynamic suppression residual , ,in Indicates time The original residual of represents the natural exponential function; Indicates time The abnormal significance index is positively correlated with the modified fluctuation sensitivity and the absolute value of the original residual, and negatively correlated with the root mean square of the prediction error of the set number of electricity data; the modified fluctuation sensitivity is , ,in Indicates relative volatility sensitivity; Indicates time The load level collected; Represents the median of the load level corresponding to all acquisition moments; is the difference between adjacent loads; the relative fluctuation sensitivity is positively correlated with the difference between the power at the corresponding collection moment and the power at the previous collection moment, and is negatively correlated with the standard deviation of the power at multiple consecutive adjacent collection moments.

[0005] The present invention realizes accurate prediction and anomaly detection of industrial workshop electricity data by constructing an improved ARIMA model based on the historical data and real-time data of the electricity meter. Compared with the existing solution that only uses the traditional ARIMA model or simple threshold comparison, the present invention can effectively suppress the abnormal residual caused by sudden load changes, thereby greatly improving the accuracy and real-time performance of fault warning, ensuring that equipment maintenance is more timely and preventive measures are more effective, and significantly reducing the risks of safety accidents and economic losses.

[0006] Furthermore, the relative fluctuation sensitivity is obtained by: ; in Indicates relative volatility sensitivity; Indicates time The collected power data; Indicates time The collected power data; Indicates the set sliding window length; Indicates time The collected power data; Before The sliding average of the time points.

[0007] The present invention captures tiny fluctuations in electricity data in real time by adopting a sliding window and comparing previous means, and can discover potential anomalies earlier than traditional methods, thereby improving the sensitivity of data processing and the robustness of the prediction model, providing more accurate basic data for the subsequent calculation of the anomaly significance index, and significantly improving the delay and misjudgment problems in the prior art in capturing anomalies.

[0008] Furthermore, the method for obtaining the sliding average value is specifically as follows: ; in Before The sliding average of time points; Indicates the set sliding window length; Indicates time Collected power data.

[0009] Furthermore, the calculation method of the abnormal significance index is: ; in Indicates time The abnormal significance index of Indicates time Modified volatility sensitivity; Indicates time The original residual of Indicates the length of the set error window; Indicates time The original residual of Before The root mean square of the prediction error at each time point.

[0010] The present invention calculates the abnormal significance index, comprehensively considers the corrected fluctuation sensitivity, original residual and prediction error root mean square, and realizes quantitative evaluation of the abnormal degree of electricity data. Compared with the traditional ARIMA model method that only relies on residual judgment, the present invention can effectively distinguish normal fluctuations from real abnormal phenomena, thereby realizing precise adjustment of abnormal dynamic suppression residuals, and improving the accuracy and reliability of industrial workshop equipment fault warning.

[0011] Furthermore, obtaining historical electricity consumption data of industrial workshops also includes: deploying smart meters in the industrial workshops to collect electricity consumption of production equipment and setting the collection frequency range to 1 to 3 times / 10 minutes; transmitting the electricity consumption data to the workshop database through the RS485 communication interface; and collecting a set number of continuous electricity consumption from the workshop database to obtain historical electricity consumption data.

[0012] The present invention ensures the frequency and continuity of data collection by deploying smart meters in industrial workshops, transmitting data using the RS485 communication interface, and extracting continuous power records from the workshop database. Compared with the traditional method where data collection is not real-time or comprehensive enough, the present invention provides a high-precision and high-reliability power data collection solution, laying a solid data foundation for subsequent prediction model training and anomaly detection.

[0013] Furthermore, it also includes using the power load tester PQ5200 to obtain the load level.

[0014] Furthermore, the ARIMA forecasting model also includes using ADF to test the stationarity of the historical electricity data, using autocorrelation function and partial autocorrelation function graph to determine the autoregressive order and the moving average order, using maximum likelihood estimation method to fit the model, and using Ljung-Box to test noise.

[0015] The present invention ensures the scientific determination of model parameters and the high accuracy of model fitting by adopting a series of strict statistical methods such as ADF test, ACF and PACF graph analysis, maximum likelihood estimation and Ljung-Box test.

[0016] Furthermore, the set threshold value ranges from 15 to 25.

[0017] The technical effects of the present invention are: The present invention constructs an intelligent anomaly detection method based on electricity meter data collection, and innovatively improves the traditional ARIMA model. Specifically, by collecting real-time and historical electricity data from industrial workshops, using the ARIMA model to predict electricity, key indicators such as relative fluctuation sensitivity, equipment status correction, and abnormal significance index are introduced to achieve dynamic suppression of prediction residuals. This method comprehensively considers the periodic characteristics of the equipment during normal operation and the sudden load changes under abnormal conditions. Through multi-dimensional data processing, it significantly improves the sensitivity and accuracy of anomaly detection, effectively distinguishes normal fluctuations from fault warnings, and thus provides more timely and accurate warning information for equipment maintenance, reduces potential safety risks, and optimizes production scheduling management. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1 The figure schematically shows a flow chart of a data collection method based on an electric energy meter according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0020] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0021] Concentrator operation data management method embodiment: like Figure 1 As shown, the concentrator operation data management method of the present invention includes: S1. Obtain electricity consumption data of industrial workshop production equipment based on electricity meters.

[0022] As the core equipment for recording power consumption, the data of the electric energy meter can reflect the operating status and energy consumption of the equipment. In an industrial workshop, when the production equipment is operating normally, the power consumption is usually kept within a stable range. However, when the equipment fails or is subject to external interference, the power consumption of some equipment may suddenly increase, causing the overall power consumption data to deviate from the normal range. This abnormality may indicate equipment failure or potential safety hazards. Therefore, in this embodiment, when the data of the production equipment is collected based on the electric energy meter, the predicted power of the production equipment is obtained by data prediction means and then compared with the actual power consumption to obtain abnormal data, so as to avoid greater losses or safety accidents caused by long-term abnormal operation of the equipment. It not only ensures real-time data monitoring of industrial workshops and rapid response to abnormal events, but also provides administrators with a decision-making basis for production plans, effectively improving the level of intelligence of industrial workshop equipment management.

[0023] First, smart meters can be deployed in industrial workshops to collect power consumption. In this embodiment, the collection frequency of power consumption can be set to 1 time / 10 minutes, and then the accumulated power consumption data can be regularly transmitted to the workshop database through the communication interface of the power meter (such as RS485 or Modbus protocol). In addition to being deployed in industrial workshops, the above-mentioned smart meters can also be deployed on large-scale production equipment or key production equipment to monitor the relevant data of a specific equipment. Pull the common data from the workshop database The historical electricity data is recorded as ,have: ; in, Indicates at time The collected power data; Indicates The power data at the time of collection, It can be set to an empirical value of 1000. It should be noted that: since the collection frequency is set to 1 time / 10 minutes in this embodiment, the corresponding data is collected 144 times a day. Here, a total of 1000 historical data are collected, including data for a total of 7 consecutive days, and as much as possible for a week, so as to provide sufficient data sources for subsequent data prediction. The implementer can also set the data according to the shift schedule of the industrial workshop and the working status of the production equipment. Set it to 2000, 3000, or 4000 experience points, etc.

[0024] S2. Obtain relative fluctuation sensitivity based on the electricity data near any historical collection time; correct the relative fluctuation sensitivity according to the equipment status to obtain the corrected fluctuation sensitivity; obtain the abnormal significance index to quantify the significance of the abnormality, and calculate the final abnormal dynamic suppression residual based on the abnormal significance index to obtain the improved ARIMA model.

[0025] In the industrial workshop scenario of this embodiment, with the orderly advancement of the production plan, the fixed operation time of each device and the fixed working hours of the workers, the power consumption of the industrial workshop on each working day will not be too different, that is, in the long run, the power data of the industrial workshop has a certain periodic regularity. Therefore, considering the time series characteristics of the data collected by the power meter of the industrial workshop, the ARIMA (autoregressive integrated moving average) model can be used in this embodiment to predict the power data.

[0026] The ARIMA model can capture the trend and periodicity of the data, which includes the autoregressive order , difference order And the sliding average order The algorithm first uses the ADF (Augmented Dickey-Fuller) test to determine the stationarity of the sequence. The difference order when the sequence is stationary is ; Then determine the autocorrelation function (ACF) and partial autocorrelation function (PACF) graphs and ; Finally, the maximum likelihood estimation method is used to fit the model. The general form of the ARIMA model is: ; in, is the lag operator; is an autoregressive polynomial; ; is white noise. For example, if there is a time series whose fitting result is ARIMA(1,1,0), then the model can be expressed as: .

[0027] S2.1. Obtain relative fluctuation sensitivity based on the electricity data near any historical collection time.

[0028] Although the ARIMA model performs well in predicting electricity consumption in industrial workshops, it assumes that the time series is stable and linear, and cannot effectively capture nonlinear mutations caused by equipment failure or external interference, such as a surge in electricity consumption. However, these mutations are closely related to anomalies. Therefore, in this embodiment, an abnormal dynamic suppression residual is constructed to enhance the sensitivity and robustness of the model to anomalies.

[0029] For historical power data of industrial workshops Generally speaking, it has the following characteristics: when each device is operating normally, the power consumption presents a certain periodic pattern on weekdays and weekends; but when the equipment fails, the power consumption will show a short-lasting and isolated surge, and the anomaly is usually closely related to the operating status of the equipment.

[0030] In order to capture the characteristics of mutations, an indicator can be designed to quantify the dynamic trend of electricity consumption and highlight the magnitude and speed of abnormalities, as follows: ; in Indicates relative volatility sensitivity; Indicates time The collected power data; Indicates time The collected power data; Indicates the set sliding window length, which can be set to an empirical value of 10 in this embodiment; Indicates time The collected power data; , indicating the front The sliding average of time points; Before The local standard deviation at a time point reflects the Normal fluctuation level nearby.

[0031] Indicates time The sensitivity of the changes in the collected power data relative to the historical fluctuation level. is small, the local standard deviation is stable, then Also smaller; abnormal, If the It also increased significantly, highlighting the mutation characteristics.

[0032] S2.2. Correct the relative fluctuation sensitivity according to the equipment status to obtain the corrected fluctuation sensitivity.

[0033] The relative volatility sensitivity It can capture the power data that may be sudden changes in the power time series, but it does not consider the impact of the equipment status on the abnormality. That is, when a high load or load mutation suddenly occurs in some equipment in the industrial workshop, it causes fluctuations in the power time series. At this time, based on the fluctuation sensitivity The acquired mutation power data may be caused by device abnormality rather than normal fluctuation. Therefore, the mutation sensitivity can be further corrected by introducing device status information, as shown below: ; in Indicates the modified volatility sensitivity; Indicates relative volatility sensitivity; , which represents the load change; Represents a logarithmic function, which is used to amplify the nonlinearity of load changes; Indicates time The equipment load level collected; It represents the reference load when the equipment is operating normally. In this embodiment, the median of all historical load levels of the equipment can be used as the reference load.

[0034] It should be noted that: when the energy meter only collects power data of a single device, the load level here refers to the single device; when the energy meter collects power data for the entire industrial workshop, the load level here refers to the ,in Indicates the first The load level of each device, accordingly, It represents the sum of the reference loads of all equipment in the industrial workshop. The above load level can be obtained through power metering equipment such as power load tester (PQ5200), and the specific acquisition process will not be repeated here.

[0035] When the device operates normally according to the reference load, and Smaller, at this time ; and when the load surges ( ) or load mutation ( Larger), The overall increase, is amplified, indicating that the mutation is more likely to be abnormal.

[0036] S2.3. Obtain anomaly significance index to quantify the significance of anomalies, and calculate the final anomaly dynamic suppression residual based on the anomaly significance index to obtain the improved ARIMA model.

[0037] For the moment Collected power data , the forecast value based on the ARIMA model is , the corresponding residual In order to quantify the significance of the anomaly and ultimately achieve dynamic suppression of the model residual, an anomaly significance index is designed in this embodiment , the specific calculation method is: ; in Indicates time The abnormal significance index of Indicates time Modified volatility sensitivity; Indicates time The residual of represents the length of the error window, which can be set to an empirical value of 5 in this embodiment; Indicates time The residual. It means before The root mean square error (RMSE) of the prediction at each time point can reflect the stability of the prediction. When the industrial workshop equipment operates normally, the prediction error is small and the RMSE is stable. Small; when an abnormality occurs, Sudden increase, and Larger, at this time Significantly magnified, highlighting abnormal features.

[0038] So far, the time has been obtained The abnormal significance index of the ARIMA model is finally used to calculate the abnormal dynamic suppression residual of the ARIMA model based on the abnormal significance index. , specifically: ; in Indicates time The abnormal dynamic suppression residual of Indicates the time in the initial ARIMA model The original residual of represents the natural exponential function; Indicates time The abnormal significance index.

[0039] When the power sequence is normal, Smaller, at this time Close to 0, , retaining the normal residual information in the time series; and when there may be anomalies in the power time series, the anomaly significance index Increase, at this time Close to 0, the original residual is suppressed, that is, , reducing the interference of abnormal phenomena on model training, making model training more focused on normal modes, thereby improving the ability to predict future mutations.

[0040] Exemplary explanation: During the model training process, assume that the total power of the equipment in the industrial workshop is 100kW at noon every day, that is, the hourly power data at noon increases by 100kW·h. Then, under normal circumstances, the prediction result of the ARIMA model will increase at a rate of 100 per hour, that is, the residual value of the predicted value remains around 0 and can pass the Ljung-Box test; but there may be equipment failure on a certain day or for several consecutive days, causing the total power to increase to 110kW, resulting in a gradual increase in the residual (the residual of the hourly power data at noon may be around 10). At this time, due to the small power change, the model is still It can identify periodic patterns, that is, the subsequent model may consider that the change in electricity consumption is normal, and thus adjust the parameters to adapt to the abnormal change, causing the subsequent model's prediction ability for normal electricity consumption to deteriorate; and the abnormal dynamic suppression residual of this embodiment can suppress the impact of abnormal phenomena. For example, in the above example, the residual 10 can be adjusted to be reduced or even close to 0, thereby retaining normal model training information, so that the model training can focus more on normal training data. The training model obtained by the abnormal dynamic suppression residual can reflect the electricity data under normal conditions, so that the status of the equipment can be evaluated based on the gap between the predicted value and the actual value of the electricity meter.

[0041] S3. When collecting power data based on the electric energy meter, the power consumption of the industrial workshop is predicted according to the improved ARIMA model, and the equipment status is evaluated based on the predicted power data and the actual power data.

[0042] The historical electricity data obtained in step S1 is input into the ARIMA model after improving the abnormal dynamic suppression residual in step S2 for training. The model dynamically suppresses the model residual by introducing relative fluctuation sensitivity, equipment status correction and abnormal significance index, thereby enhancing the ability to capture sudden abnormalities.

[0043] Collect the real-time electricity data time series of the electric energy meter. In this embodiment, the length of the real-time electricity data time series can be set to 150. The implementer can also set the length of the real-time electricity data time series according to the production planning of the industrial workshop and the frequency of electricity meter data collection to ensure that there are enough original data to obtain more accurate prediction values. Then input the collected real-time electricity data time series into the trained ARIMA model to output the current time prediction value; then compare the real-time electricity data with the prediction value. When the error exceeds the preset threshold, it is judged that the equipment may be abnormal (such as failure, load mutation, etc.), and then trigger the early warning mechanism. In this embodiment, the preset threshold can be set to the empirical value of 20; the final early warning signal can be notified to the on-site maintenance personnel in a timely manner through SMS or mobile APP, etc., to guide them to conduct on-site inspections and troubleshooting.

Claims

1. A data collection method based on an electric energy meter, characterized in that: The method comprises: Obtain historical power data of the industrial workshop, and train the ARIMA prediction model based on the historical power data; obtain the real-time power time series of a set length and input it into the trained ARIMA prediction model to obtain a predicted value; calculate the difference between the predicted value and the real-time power, and in response to the difference being greater than a set threshold, the system notifies the maintenance personnel to conduct a fault inspection; The ARIMA forecasting model also includes processing the original residual to obtain abnormal dynamic suppression residual , ,in Indicates time The original residual of represents the natural exponential function; Indicates time The abnormal significance index of The abnormal significance index is positively correlated with the modified fluctuation sensitivity and the absolute value of the original residual, and negatively correlated with the root mean square of the prediction error of the set amount of electricity data; the modified fluctuation sensitivity is , ,in Indicates relative volatility sensitivity; Indicates time The load level collected; Represents the median of the load level corresponding to all acquisition moments; is the difference between adjacent loads; The relative fluctuation sensitivity is positively correlated with the difference between the electric quantity at the corresponding collection moment and the electric quantity at the previous collection moment, and is negatively correlated with the standard deviation of the electric quantity at multiple consecutive adjacent collection moments.

2. The data collection method based on electric energy meter according to claim 1 is characterized in that: The method for obtaining the relative fluctuation sensitivity is specifically as follows: ； in Indicates relative volatility sensitivity; Indicates time Collected power data; Indicates time Collected power data; Indicates the set sliding window length; Indicates time Collected power data; Before The sliding average of the time points.

3. The data collection method based on electric energy meter according to claim 2 is characterized in that: The method for obtaining the sliding average value is specifically as follows: ； in Before The sliding average of time points; Indicates the set sliding window length; Indicates time Collected power data.

4. The data collection method based on electric energy meter according to claim 1 is characterized in that: The calculation method of the abnormal significance index is: ； in Indicates time The abnormal significance index of Indicates time Modified volatility sensitivity; Indicates time The original residual of Indicates the length of the set error window; Indicates time The original residual of Before The root mean square of the prediction error at each time point.

5. The data collection method based on electric energy meter according to claim 1 is characterized in that: Obtain historical power data of industrial workshops, including: Deploy smart meters in industrial workshops to collect electricity consumption of production equipment and set the collection frequency range to 1 to 3 times / 10 minutes; Transmit power consumption data to the workshop database via RS485 communication interface; A set number of continuous power quantities are collected from the workshop database to obtain historical power quantity data.

6. The data collection method based on electric energy meter according to claim 1 is characterized in that: It also includes obtaining the load level using the power load tester PQ5200.

7. The data collection method based on electric energy meter according to claim 1 is characterized in that: The ARIMA forecasting model also includes using ADF to test the stationarity of the historical electricity data, using autocorrelation function and partial autocorrelation function graph to determine the autoregressive order and the moving average order, using maximum likelihood estimation method to fit the model, and using Ljung-Box to test noise.

8. The data collection method based on electric energy meter according to claim 1 is characterized in that: The value range of the set threshold is 15~25.

Images