Electric power data real-time online acquisition and analysis method and system

Abstract

The invention discloses an electric power data real-time online collection and analysis method and system, and relates to the technical field of electric power data collection and analysis, and the method comprises the steps: collecting real-time industrial electric power data and real-time industrial production data, obtaining real-time equipment electric power information based on the real-time industrial electric power data and the real-time industrial production data, and carrying out the real-time online collection and analysis of the electric power data according to the real-time equipment electric power information; the method comprises the following steps: obtaining a real-time equipment operation state index, training through an ARIMA time sequence model to obtain a real-time equipment state prediction model, inputting the real-time equipment operation state index into the trained real-time equipment state prediction model to obtain a predicted predicted equipment state index, and determining the real-time equipment state based on a predicted equipment state final index threshold value. Whether the predicted equipment state final index meets the standard or not is judged, early warning is given out before the power consumption abnormity of the industrial equipment occurs, an equipment maintenance plan is made, and the situation that some parts of the equipment age along with time, and consequently the power consumption abnormity of the equipment cannot be found in time is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of power data collection and analysis, and in particular to a method and system for real-time online collection and analysis of power data. Background Art

[0002] At present, industrial electricity consumption in my country accounts for more than 60% of the total electricity consumption in the society. Factory electricity consumption has become one of the main sources of electricity consumption in my country. There are a large number of high-energy consumption processing equipment in the factory. If the processing equipment has abnormalities, it will often lead to higher energy consumption and waste of electricity resources. Therefore, it is necessary to collect the power data of the processing equipment in real time online and analyze the data so that the staff can find faulty processing equipment more quickly.

[0003] However, the number of processing equipment in existing large factories is huge and they are widely distributed, making it difficult to quickly and accurately obtain the power data of the processing equipment. Even if the factory adopts regular equipment inspection and maintenance plans, during the operation of the equipment, some parts of the equipment will still naturally wear and age over time, resulting in many processing equipment with abnormal power consumption that cannot be discovered. Therefore, it is necessary to provide a real-time online power data collection and analysis method and system to solve the above-mentioned problems. Summary of the invention

[0004] In order to solve the above technical problems, a method and system for real-time online collection and analysis of power data are provided. This technical solution solves the problem raised in the above background technology that the number of processing equipment in existing large factories is huge and they are widely distributed, making it difficult to quickly and accurately obtain the power data of the processing equipment. Even if the factory adopts regular equipment inspection and maintenance plans, during the operation of the equipment, some parts of the equipment will still naturally wear and age over time, resulting in many processing equipment with abnormal power consumption cannot be discovered.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:

[0006] A method for real-time online collection and analysis of power data, comprising:

[0007] Collect real-time industrial power data, real-time industrial production data, historical industrial power data and historical industrial production data;

[0008] The real-time industrial production data includes real-time industrial equipment information, real-time production scale information and real-time production environment information;

[0009] Based on the real-time industrial power data and the real-time industrial production data, real-time device power information is acquired, where the real-time device power information is used to describe the power information of the industrial equipment under the current power quality;

[0010] According to the real-time device power information, a real-time device operation status index is obtained;

[0011] Inputting the real-time equipment operation status index into the trained real-time equipment status prediction model to obtain a predicted equipment status index;

[0012] Based on the predicted equipment status index and real-time production environment information, obtain the predicted equipment status final index;

[0013] Based on historical industrial power data and historical industrial production data, obtain historical power consumption information and abnormal event information;

[0014] Based on historical power consumption information and abnormal event information, obtain the equipment abnormality index;

[0015] The equipment abnormality index is used as the threshold value for predicting the final index of the equipment status;

[0016] Based on the predicted equipment status final index threshold, determine whether the predicted equipment status final index meets the standard. If so, continue to collect equipment power data. If not, analyze the operating status of the industrial equipment based on the predicted equipment status final index and formulate an equipment maintenance plan.

[0017] In an optional embodiment, the step of constructing the real-time device status prediction model specifically includes:

[0018] Based on historical industrial production data and historical industrial power data, obtain historical industrial equipment information, historical production scale information, historical power quality information and historical power consumption information;

[0019] Integrate historical industrial equipment information, historical production scale information, historical power quality information and historical power consumption information to obtain equipment status prediction data;

[0020] Perform data standardization on the equipment status prediction data to obtain a standard sample set of equipment status prediction;

[0021] Divide the equipment state prediction standard sample set according to a preset ratio to obtain a training set, a validation set, and a test set of equipment state prediction standard data;

[0022] Based on the training set, validation set and test set of the equipment status prediction standard data, the ARIMA time series model is trained to obtain the real-time equipment status prediction model.

[0023] In an optional embodiment, obtaining a real-time device operation status index according to the real-time device power information specifically includes:

[0024] According to the real-time industrial equipment information and real-time power quality information, the real-time operating status influencing factors of a single industrial equipment in an operating cycle under different working modes are obtained;

[0025] According to the real-time equipment power information, the real-time useful power and real-time apparent power of a single industrial equipment within an operation cycle are obtained;

[0026] Based on the real-time useful power and real-time apparent power, as well as the real-time operating status influencing factors of a single industrial device in an operating cycle under different working modes, the real-time equipment operating status index is obtained;

[0027] The calculation formula of the real-time operation status influencing factor is:

[0028]

[0029] In the formula, γ represents the real-time operating status influencing factor, Q m Represents the real-time power quality parameter, P m Indicates the rated power of the device in different working modes, P e Indicates the rated power of the device in different working modes;

[0030] The calculation formula of real-time useful power is:

[0031] P real =V real I real cosθ;

[0032] Where P real Indicates real-time useful power, V real Indicates the effective voltage of industrial equipment in actual operation, I real It represents the effective current in the actual operation of industrial equipment, and cosθ represents the power factor;

[0033] The calculation formula for real-time apparent power is:

[0034] S fac =V fac I fac ;

[0035] In the formula, S fac Indicates real-time apparent power, V fac Indicates the total voltage of industrial equipment during operation, I fac Indicates the total current of industrial equipment during operation;

[0036] The calculation formula of the real-time equipment operation status index is:

[0037]

[0038] In the formula, Fpow Indicates the real-time device operation status index.

[0039] In an optional embodiment, based on the predicted equipment status index and the real-time production environment information, obtaining the predicted equipment status final index specifically includes:

[0040] According to the real-time production environment information, obtain the real-time temperature index, real-time humidity index and real-time vibration parameters;

[0041] Based on the real-time temperature index, the real-time humidity index, the real-time vibration parameter and the predicted equipment status index, the predicted equipment status final index is obtained;

[0042] Among them, the calculation formula for predicting the final index of equipment status is:

[0043]

[0044] Where W τ Represents the predicted equipment status final index, T i Represents the real-time temperature index of the i-th time unit, H i Represents the real-time humidity index of the i-th time unit, V i represents the real-time vibration parameter of the i-th time unit, F ast (τ) represents the predicted equipment status index of the τth industrial equipment, β 0 represents the error term, β 1 Indicates the weight coefficient of the real-time temperature index, β 2 Represents the weight coefficient of the real-time humidity index, β 3 represents the weight coefficient of the real-time vibration parameter, and j represents the equipment operation cycle.

[0045] In an optional embodiment, obtaining a device abnormality index based on historical power consumption information and abnormal event information specifically includes:

[0046] Based on the historical power consumption information, obtain the historical power consumption information of all workshops, and further obtain the historical power consumption information of a single device;

[0047] Based on the abnormal event information, abnormal device information, abnormal time node information and abnormal environment information are obtained, wherein the abnormal device information, abnormal time node information and abnormal environment information correspond to each other one by one;

[0048] Based on the abnormal device information and abnormal time node information, the power information and environmental information of the abnormal device in a period of operation before the abnormality occurs are collected and marked to obtain the marked device power information and marked environmental information;

[0049] Based on the identification device power information and identification environment information, the device abnormality index is obtained.

[0050] In an optional embodiment, obtaining a device abnormality index based on the identification device power information and the identification environment information specifically includes:

[0051] Based on statistical methods, the power information of the identification equipment and the identification environment information are converted into curve statistical graphs to obtain the identification power curve statistical graph and the identification environment curve statistical graph;

[0052] Based on historical industrial power data and historical industrial production data, obtain statistical graphs of standard power curves and standard environmental curves of industrial equipment under normal operating conditions;

[0053] Overlap the statistical graph of the identification power curve and the statistical graph of the standard power curve, calculate the distance between the Y-axis coordinate points when the identification power curve and the standard power curve are at the same X-axis coordinate, and obtain the equipment power difference value;

[0054] Overlap the identification environment curve statistical graph and the standard environment curve statistical graph, calculate the distance between the Y-axis coordinate points when the identification environment curve and the standard environment curve are at the same X-axis coordinate, and obtain the equipment environment difference value;

[0055] Based on the equipment power difference value and the equipment environment difference value, obtain the equipment abnormality index;

[0056] The calculation formula of the equipment abnormality index is:

[0057]

[0058] In the formula, δ represents the equipment abnormality index, M μ Indicates the abnormal power parameter per unit time in the μth operation cycle, N μ represents the standard power parameter per unit time in the μth operation cycle, Y μ represents the abnormal environmental parameters per unit time in the μth operating cycle, Z μ It represents the standard environmental parameters per unit time in the μth operating cycle, μ represents the unit time in the operating cycle, and π represents the operating cycle.

[0059] Furthermore, a real-time online power data collection and analysis system is proposed to implement the above analysis method, including:

[0060] A main control module, wherein the main control module is used to perform data standardization processing on the equipment state prediction data to obtain a standard sample set for equipment state prediction, divide the standard sample set for equipment state prediction according to a preset ratio, obtain a training set, a validation set and a test set for the standard data for equipment state prediction, train an ARIMA time series model based on the training set, the validation set and the test set for the standard data for equipment state prediction, obtain a real-time equipment state prediction model, collect power information and environmental information of the abnormal equipment for a period of operation cycle before the abnormality occurs based on the abnormal equipment information and the abnormal time node information, and mark them, convert the identification equipment power information and the identification environmental information into a curve statistical graph based on a statistical method, obtain an identification power curve statistical graph and an identification environment curve statistical graph, overlap the identification power curve statistical graph and the standard power curve statistical graph, calculate the distance between the Y-axis coordinate points when the identification power curve and the standard power curve are at the same X-axis coordinate, and judge whether the predicted equipment state final index meets the standard based on the predicted equipment state final index threshold;

[0061] A data acquisition module, the data acquisition module is used to acquire real-time industrial power data, real-time industrial production data, historical industrial power data, historical industrial production data, real-time equipment power information, equipment status prediction data, historical workshop power information, historical equipment power information, abnormal equipment information, abnormal time node information, abnormal environment information, identification equipment power information, identification environment information, real-time equipment operation status index, predicted equipment status index, predicted equipment status final index, equipment abnormality index, real-time operation status influencing factor, real-time useful power, real-time apparent power, real-time equipment operation status index, equipment power difference value and equipment environment difference value;

[0062] A data transmission module, which is used to transmit the data collected by the data acquisition module to the main control module;

[0063] The display module is used to display the identification power curve statistical graph, the standard power curve statistical graph, the identification environment curve statistical graph and the standard environment curve statistical graph, predict the final equipment status index, analyze the operating status of industrial equipment, and formulate equipment maintenance plans.

[0064] In an optional embodiment, the main control module specifically includes:

[0065] A data processing unit, wherein the calculation unit is used to perform data standardization processing on the device state prediction data to obtain a device state prediction standard sample set, divide the device state prediction standard sample set according to a preset ratio, obtain a training set, a verification set and a test set of the device state prediction standard data, collect power information and environmental information of the abnormal device for a period of operation cycle before the abnormality occurs based on the abnormal device information and the abnormal time node information, and mark them, and based on a statistical method, convert the identification device power information and the identification environment information into a curve statistical graph to obtain an identification power curve statistical graph and an identification environment curve statistical graph, overlap the identification power curve statistical graph and the standard power curve statistical graph, and calculate the distance between the Y-axis coordinate points of the identification power curve and the standard power curve when they are at the same X-axis coordinate;

[0066] A training unit, wherein the training unit is used to train the ARIMA time series model based on a training set, a validation set, and a test set of equipment state prediction standard data to obtain a real-time equipment state prediction model;

[0067] A judgment unit is used to judge whether the predicted device status index meets the standard based on the device abnormality threshold.

[0068] In an optional embodiment, the data acquisition module specifically includes:

[0069] a first acquisition unit, the first acquisition unit being used to acquire real-time industrial power data, real-time industrial production data, historical industrial power data, historical industrial production data, real-time equipment power information, equipment status prediction data, historical workshop power information, historical equipment power information, abnormal equipment information, abnormal time node information, abnormal environment information, identification equipment power information, and identification environment information;

[0070] The second acquisition unit is used to obtain the real-time equipment operation status index, the predicted equipment status index, the predicted equipment status index, the equipment abnormality index, the real-time operation status influencing factor, the real-time useful power, the real-time apparent power, the real-time equipment operation status index, the equipment power difference value and the equipment environment difference value.

[0071] In an optional embodiment, the display module specifically includes:

[0072] A first display unit, the first display unit is used to display a statistical graph of a marked power curve, a statistical graph of a standard power curve, a statistical graph of a marked environment curve, and a statistical graph of a standard environment curve;

[0073] The second display unit is used to display the predicted equipment status final index, analyze the operating status of industrial equipment, and formulate equipment maintenance plan.

[0074] Compared with the prior art, the present invention has the following beneficial effects:

[0075] In a real-time online power data collection and analysis method and system proposed in this scheme, an ARIMA time series model is used for training to obtain a real-time equipment status prediction model. The real-time equipment operation status index is input into the trained real-time equipment status prediction model to obtain a predicted equipment status index. Based on the predicted equipment status final index threshold, it is determined whether the predicted equipment status final index meets the standard. Before the industrial equipment has abnormal power consumption, an early warning is issued and an equipment maintenance plan is formulated to avoid the aging of some parts of the equipment over time, which may cause the abnormal power consumption of the equipment to fail to be discovered in time.

[0076] The scheme proposes a real-time online collection and analysis method and system for electric power data. The method overlaps the identification power curve statistical graph and the standard power curve statistical graph, calculates the distance between the Y-axis coordinate points when the identification power curve and the standard power curve are at the same X-axis coordinate, and obtains the equipment power difference value. The identification environment curve statistical graph and the standard environment curve statistical graph are overlapped, calculates the distance between the Y-axis coordinate points when the identification environment curve and the standard environment curve are at the same X-axis coordinate, and obtains the equipment environment difference value. This enables the staff to analyze the operating status of the industrial equipment more intuitively, and make timely adjustments to the parameters of the industrial equipment to avoid abnormalities of the industrial equipment due to load or equipment parts problems. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 A flowchart of a method for real-time online collection and analysis of power data proposed by the present invention;

[0078] Figure 2 This is a flowchart for obtaining the final index of the predicted device state in the present invention;

[0079] Figure 3 This is a flow chart for obtaining the device abnormality index in the present invention;

[0080] Figure 4 This is a block diagram of the system structure proposed by the present invention. DETAILED DESCRIPTION

[0081] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0082] Reference Figure 1-4 As shown, a real-time online power data collection and analysis method includes:

[0083] Collect real-time industrial power data, real-time industrial production data, historical industrial power data and historical industrial production data;

[0084] The real-time industrial production data includes real-time industrial equipment information, real-time production scale information and real-time production environment information;

[0085] Based on the real-time industrial power data and the real-time industrial production data, real-time device power information is acquired, where the real-time device power information is used to describe the power information of the industrial equipment under the current power quality;

[0086] According to the real-time device power information, a real-time device operation status index is obtained;

[0087] Inputting the real-time equipment operation status index into the trained real-time equipment status prediction model to obtain a predicted equipment status index;

[0088] Based on the predicted equipment status index and real-time production environment information, obtain the predicted equipment status final index;

[0089] Based on historical industrial power data and historical industrial production data, obtain historical power consumption information and abnormal event information;

[0090] Based on historical power consumption information and abnormal event information, obtain the equipment abnormality index;

[0091] The equipment abnormality index is used as the threshold value for predicting the final index of the equipment status;

[0092] Based on the predicted equipment status final index threshold, determine whether the predicted equipment status final index meets the standard. If so, continue to collect equipment power data. If not, analyze the operating status of the industrial equipment based on the predicted equipment status final index and formulate an equipment maintenance plan.

[0093] Specifically, real-time industrial power data is used to describe and monitor the power consumption in the industrial production process, including real-time information on the power consumption of industrial equipment, machines and production lines. Real-time industrial production data is used to describe and monitor various key indicators and parameters in the industrial production process, including real-time industrial equipment information, real-time production scale information and real-time production environment information. Real-time industrial equipment information is used to describe the basic information of industrial equipment. Real-time production scale information is used to describe the production plan and scheduling of the factory, as well as production efficiency and output. Real-time production environment information is used to describe the environment in which industrial equipment is located during production. When collecting real-time equipment power information, sensors can be installed on industrial equipment to obtain real-time power information of industrial equipment more quickly, including current, voltage and power consumption of industrial equipment. Real-time equipment operation status index is used to describe the power status index of industrial equipment during operation, taking into account the current, voltage and other parameters of industrial equipment. The predicted equipment status index is the operation status index of the real-time equipment operation status of industrial equipment collected through monitoring in the future one or more operation cycles. The predicted equipment status final index is the predicted equipment status index of industrial equipment after being disturbed by the real-time production environment, that is, the comprehensive operation status reflected by the operation status of industrial equipment after being disturbed by external factors. In order to improve the timeliness of the system's response to abnormal industrial equipment status, a value of 0.7 to 0.9 of the equipment abnormality index can be used as the threshold for predicting the final index of the equipment status, but this will increase the frequency of the system's warning of abnormal industrial equipment status.

[0094] Furthermore, the steps of constructing the real-time device status prediction model specifically include:

[0095] Based on historical industrial production data and historical industrial power data, obtain historical industrial equipment information, historical production scale information, historical power quality information and historical power consumption information;

[0096] Integrate historical industrial equipment information, historical production scale information, historical power quality information and historical power consumption information to obtain equipment status prediction data;

[0097] Perform data standardization on the equipment status prediction data to obtain a standard sample set of equipment status prediction;

[0098] The equipment state prediction standard sample set is divided according to a preset ratio to obtain a training set, a validation set, and a test set of the equipment state prediction standard data. The preset ratio is usually 8:1:1.

[0099] Based on the training set, validation set and test set of the equipment status prediction standard data, the ARIMA time series model is trained to obtain the real-time equipment status prediction model.

[0100] Specifically, historical industrial equipment information, historical production scale information, historical power quality information and historical electricity consumption information are integrated to obtain equipment status prediction data, which is used to collect time series data of historical industrial equipment information, historical production scale information, historical power quality information and historical electricity consumption information to ensure that the data has a consistent time interval, and when the data is integrated, the time interval is ensured to be within an operating cycle of the industrial equipment. Then, missing data and outliers are processed, and a preliminary time series analysis is performed on the time series of each data. According to the characteristics of the time series data, a suitable ARIMA model is selected, and the ARIMA model is fitted with the training set of the equipment status prediction standard data, and the maximum likelihood estimation (MLE) is used to determine the autoregressive order. AR(p), difference order I(d) and moving average order MA(q), diagnose the fitted ARIMA model, including checking whether the residual meets the characteristics of white noise, use the trained ARIMA model to verify the validation set of the equipment status prediction standard data, and adjust the model's hyperparameters during the verification process, such as the autoregressive order AR(p), difference order I(d) and moving average order MA(q) in the ARIMA model, evaluate the performance of the model under different parameter combinations, and select the best parameters. Finally, use the test set of the equipment status prediction standard data to evaluate the performance of the model under different parameter combinations, evaluate the generalization ability and prediction performance of the final model, and obtain the final qualified real-time equipment status prediction model.

[0101] Among them, the expression formula of the autoregressive order AR(p) and moving average order MA(q) of the real-time equipment status prediction model is:

[0102]

[0103] Where, X σ represents the value of σ at the current time point of the time series of the equipment status prediction data, and are the autoregressive coefficients of the autoregressive order AR(p), and are the parameters of the moving average order MA(q), c and e represent the intercept terms of the autoregressive order AR(p) and the moving average order MA(q), respectively. σ-1 , X σ-2 and X σ-ω represents the value of the time series of equipment status prediction data at σ-1, σ-2 and σ-ω, α σ-1 , α σ-2 and Represents the white noise error term of the time series of equipment status prediction data at σ-1, σ-2 and σ-ω, ∈ σ and α σ Both represent white noise error terms, which represent random fluctuations that cannot be explained by the model;

[0104] It can be understood that the real-time device operation status index is input into the trained real-time device status prediction model to obtain the predicted device status index. The calculation formula of the predicted device status index is:

[0105]

[0106] In the formula, F ast (τ) represents the predicted equipment status index, represents the predicted value at the time node (t+h), represents the intercept of the model, and h represents the prediction time step.

[0107] Further, according to the real-time device power information, a real-time device operation status index is obtained, which specifically includes:

[0108] According to the real-time industrial equipment information and real-time power quality information, the real-time operating status influencing factors of a single industrial equipment in an operating cycle under different working modes are obtained;

[0109] According to the real-time equipment power information, the real-time useful power and real-time apparent power of a single industrial equipment within an operation cycle are obtained;

[0110] Based on the real-time useful power and real-time apparent power, as well as the real-time operating status influencing factors of a single industrial device in an operating cycle under different working modes, the real-time equipment operating status index is obtained;

[0111] The calculation formula of the real-time operation status influencing factor is:

[0112]

[0113] In the formula, γ represents the real-time operating status influencing factor, Q m Represents the real-time power quality parameter, P m Indicates the rated power of the device in different working modes, P e Indicates the rated power of the device in different working modes;

[0114] The calculation formula of real-time useful power is:

[0115] P real =V real I real cosθ;

[0116] Where P real Indicates real-time useful power, V real Indicates the effective voltage of industrial equipment in actual operation, I realIt represents the effective current in the actual operation of industrial equipment, and cosθ represents the power factor;

[0117] The calculation formula for real-time apparent power is:

[0118] S fac =V fac I fac ;

[0119] In the formula, S fac Indicates real-time apparent power, V fac Indicates the total voltage of industrial equipment during operation, I fac Indicates the total current of industrial equipment during operation;

[0120] The calculation formula of the real-time equipment operation status index is:

[0121]

[0122] In the formula, F pow Indicates the real-time device operation status index.

[0123] Specifically, the power factor cosθ can be estimated or measured based on the nature of the equipment and actual measurements. The real-time operating status influencing factor is used to describe the factors that affect the performance of the system or equipment during real-time operation. The real-time equipment operating status index is an index obtained through analysis and comprehensive evaluation of the equipment operation data obtained in real time by various sensors and monitoring systems. It can reflect the current operating status of the equipment, including current, voltage and power. Real-time useful power refers to the part of the circuit that actually completes power conversion, which means the part of the circuit that can do work, that is, the actual conversion efficiency of electric energy. Real-time apparent power refers to the product of voltage and current in the circuit. It reflects the total power transmission capacity of the circuit, including the sum of actual useful power and reactive power (virtual work).

[0124] Furthermore, based on the predicted equipment status index and the real-time production environment information, the predicted equipment status final index is obtained, specifically including:

[0125] According to the real-time production environment information, obtain the real-time temperature index, real-time humidity index and real-time vibration parameters;

[0126] Based on the real-time temperature index, the real-time humidity index, the real-time vibration parameter and the predicted equipment status index, the predicted equipment status final index is obtained;

[0127] Among them, the calculation formula for predicting the final index of equipment status is:

[0128]

[0129] Where W τRepresents the predicted equipment status final index, T i Represents the real-time temperature index of the i-th time unit, H i Represents the real-time humidity index of the i-th time unit, V i represents the real-time vibration parameter of the i-th time unit, F ast (τ) represents the predicted equipment status index of the τth industrial equipment, β 0 represents the error term, β 1 Indicates the weight coefficient of the real-time temperature index, β 2 Represents the weight coefficient of the real-time humidity index, β 3 represents the weight coefficient of the real-time vibration parameter, and j represents the equipment operation cycle.

[0130] Specifically, the real-time temperature index, real-time humidity index and real-time vibration parameters are all external factors of industrial equipment. The real-time temperature index is the temperature index of the production environment in which the industrial equipment is located. The real-time humidity index is the humidity index of the production environment in which the industrial equipment is located. The real-time vibration parameter is the vibration impact of the vibration factors in the production environment in which the industrial equipment is located on the industrial equipment.

[0131] Furthermore, based on the historical power consumption information and abnormal event information, the device abnormality index is obtained, including:

[0132] Based on the historical power consumption information, obtain the historical power consumption information of all workshops, and further obtain the historical power consumption information of a single device;

[0133] Based on the abnormal event information, abnormal device information, abnormal time node information and abnormal environment information are obtained, wherein the abnormal device information, abnormal time node information and abnormal environment information correspond to each other one by one;

[0134] Based on the abnormal device information and abnormal time node information, the power information and environmental information of the abnormal device in a period of operation before the abnormality occurs are collected and marked to obtain the marked device power information and marked environmental information;

[0135] Based on the identification device power information and identification environment information, the device abnormality index is obtained.

[0136] Furthermore, based on the identification device power information and identification environment information, the device abnormality index is obtained, specifically including:

[0137] Based on statistical methods, the power information of the identification equipment and the identification environment information are converted into curve statistical graphs to obtain the identification power curve statistical graph and the identification environment curve statistical graph;

[0138] Based on historical industrial power data and historical industrial production data, obtain statistical graphs of standard power curves and standard environmental curves of industrial equipment under normal operating conditions;

[0139] Overlap the statistical graph of the identification power curve and the statistical graph of the standard power curve, calculate the distance between the Y-axis coordinate points when the identification power curve and the standard power curve are at the same X-axis coordinate, and obtain the equipment power difference value;

[0140] Overlap the identification environment curve statistical graph and the standard environment curve statistical graph, calculate the distance between the Y-axis coordinate points when the identification environment curve and the standard environment curve are at the same X-axis coordinate, and obtain the equipment environment difference value;

[0141] Based on the equipment power difference value and the equipment environment difference value, obtain the equipment abnormality index;

[0142] The calculation formula of the equipment abnormality index is:

[0143]

[0144] In the formula, δ represents the equipment abnormality index, M μ Indicates the abnormal power parameter per unit time in the μth operation cycle, N μ represents the standard power parameter per unit time in the μth operation cycle, Y μ represents the abnormal environmental parameters per unit time in the μth operating cycle, Z μ It represents the standard environmental parameters per unit time in the μth operating cycle, μ represents the unit time in the operating cycle, and π represents the operating cycle.

[0145] Furthermore, a real-time online power data collection and analysis system is proposed to implement the above analysis method, including:

[0146] A main control module, wherein the main control module is used to perform data standardization processing on the equipment state prediction data to obtain a standard sample set for equipment state prediction, divide the standard sample set for equipment state prediction according to a preset ratio, obtain a training set, a validation set and a test set for the standard data for equipment state prediction, train an ARIMA time series model based on the training set, the validation set and the test set for the standard data for equipment state prediction, obtain a real-time equipment state prediction model, collect power information and environmental information of the abnormal equipment for a period of operation cycle before the abnormality occurs based on the abnormal equipment information and the abnormal time node information, and mark them, convert the identification equipment power information and the identification environmental information into a curve statistical graph based on a statistical method, obtain an identification power curve statistical graph and an identification environment curve statistical graph, overlap the identification power curve statistical graph and the standard power curve statistical graph, calculate the distance between the Y-axis coordinate points when the identification power curve and the standard power curve are at the same X-axis coordinate, and judge whether the predicted equipment state final index meets the standard based on the predicted equipment state final index threshold;

[0147] A data acquisition module, the data acquisition module is used to acquire real-time industrial power data, real-time industrial production data, historical industrial power data, historical industrial production data, real-time equipment power information, equipment status prediction data, historical workshop power information, historical equipment power information, abnormal equipment information, abnormal time node information, abnormal environment information, identification equipment power information, identification environment information, real-time equipment operation status index, predicted equipment status index, predicted equipment status final index, equipment abnormality index, real-time operation status influencing factor, real-time useful power, real-time apparent power, real-time equipment operation status index, equipment power difference value and equipment environment difference value;

[0148] A data transmission module, which is used to transmit the data collected by the data acquisition module to the main control module;

[0149] The display module is used to display the identification power curve statistical graph, the standard power curve statistical graph, the identification environment curve statistical graph and the standard environment curve statistical graph, predict the final equipment status index, analyze the operating status of industrial equipment, and formulate equipment maintenance plans.

[0150] Furthermore, the main control module specifically includes:

[0151] A data processing unit, wherein the calculation unit is used to perform data standardization processing on the device state prediction data to obtain a device state prediction standard sample set, divide the device state prediction standard sample set according to a preset ratio, obtain a training set, a verification set and a test set of the device state prediction standard data, collect power information and environmental information of the abnormal device for a period of operation cycle before the abnormality occurs based on the abnormal device information and the abnormal time node information, and mark them, and based on a statistical method, convert the identification device power information and the identification environment information into a curve statistical graph to obtain an identification power curve statistical graph and an identification environment curve statistical graph, overlap the identification power curve statistical graph and the standard power curve statistical graph, and calculate the distance between the Y-axis coordinate points of the identification power curve and the standard power curve when they are at the same X-axis coordinate;

[0152] A training unit, wherein the training unit is used to train the ARIMA time series model based on a training set, a validation set, and a test set of equipment state prediction standard data to obtain a real-time equipment state prediction model;

[0153] A judgment unit is used to judge whether the predicted device status index meets the standard based on the device abnormality threshold.

[0154] Furthermore, the data acquisition module specifically includes:

[0155] a first acquisition unit, the first acquisition unit being used to acquire real-time industrial power data, real-time industrial production data, historical industrial power data, historical industrial production data, real-time equipment power information, equipment status prediction data, historical workshop power information, historical equipment power information, abnormal equipment information, abnormal time node information, abnormal environment information, identification equipment power information, and identification environment information;

[0156] The second acquisition unit is used to obtain the real-time equipment operation status index, the predicted equipment status index, the predicted equipment status index, the equipment abnormality index, the real-time operation status influencing factor, the real-time useful power, the real-time apparent power, the real-time equipment operation status index, the equipment power difference value and the equipment environment difference value.

[0157] Furthermore, the display module specifically includes:

[0158] A first display unit, the first display unit is used to display a statistical graph of a marked power curve, a statistical graph of a standard power curve, a statistical graph of a marked environment curve, and a statistical graph of a standard environment curve;

[0159] The second display unit is used to display the predicted equipment status final index, analyze the operating status of industrial equipment, and formulate equipment maintenance plan.

[0160] In summary, the advantages of the present invention are: training through the ARIMA time series model to obtain a real-time equipment status prediction model, by inputting the real-time equipment operation status index into the trained real-time equipment status prediction model, to obtain a predicted equipment status index, and based on the predicted equipment status final index threshold, to judge whether the predicted equipment status final index meets the standard, before the industrial equipment has abnormal power consumption, an early warning is issued, and an equipment maintenance plan is formulated to avoid the aging of some parts of the equipment over time, resulting in the failure to timely discover the abnormal power consumption of the equipment; by converting the identification equipment power information and the identification environment information into a curve statistical graph based on a statistical method, a marking power curve statistical graph and a marking environment curve statistical graph are obtained, based on historical industrial power The power data and historical industrial production data are used to obtain the standard power curve statistical graph and the standard environment curve statistical graph of industrial equipment under normal operating conditions. The identification power curve statistical graph and the standard power curve statistical graph are overlapped, and the distance between the Y-axis coordinate points of the identification power curve and the standard power curve when they are at the same X-axis coordinate is calculated to obtain the equipment power difference value. The identification environment curve statistical graph and the standard environment curve statistical graph are overlapped, and the distance between the Y-axis coordinate points of the identification environment curve and the standard environment curve when they are at the same X-axis coordinate is calculated to obtain the equipment environment difference value, so that the staff can analyze the operating status of the industrial equipment more intuitively and make timely adjustments to the parameters of the industrial equipment to avoid abnormalities of the industrial equipment due to load, equipment parts or environmental problems.

[0161] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for real-time online collection and analysis of power data, characterized in that: include: Collect real-time industrial power data, real-time industrial production data, historical industrial power data and historical industrial production data; The real-time industrial production data includes real-time industrial equipment information, real-time production scale information and real-time production environment information; Based on the real-time industrial power data and the real-time industrial production data, real-time device power information is acquired, where the real-time device power information is used to describe the power information of the industrial equipment under the current power quality; According to the real-time device power information, a real-time device operation status index is obtained; Inputting the real-time equipment operation status index into the trained real-time equipment status prediction model to obtain a predicted equipment status index; Based on the predicted equipment status index and real-time production environment information, obtain the predicted equipment status final index; Based on historical industrial power data and historical industrial production data, obtain historical power consumption information and abnormal event information; Based on historical power consumption information and abnormal event information, obtain the equipment abnormality index; The equipment abnormality index is used as the threshold value for predicting the final index of the equipment status; Based on the predicted equipment status final index threshold, determine whether the predicted equipment status final index meets the standard. If so, continue to collect equipment power data. If not, analyze the operating status of the industrial equipment based on the predicted equipment status final index and formulate an equipment maintenance plan.

2. A method for real-time online collection and analysis of power data according to claim 1, characterized in that: The steps of constructing the real-time device status prediction model specifically include: Based on historical industrial production data and historical industrial power data, obtain historical industrial equipment information, historical production scale information, historical power quality information and historical power consumption information; Integrate historical industrial equipment information, historical production scale information, historical power quality information and historical power consumption information to obtain equipment status prediction data; Perform data standardization on the equipment status prediction data to obtain a standard sample set of equipment status prediction; Divide the equipment state prediction standard sample set according to a preset ratio to obtain a training set, a validation set, and a test set of equipment state prediction standard data; Based on the training set, validation set and test set of the equipment status prediction standard data, the ARIMA time series model is trained to obtain the real-time equipment status prediction model.

3. The method for real-time online collection and analysis of power data according to claim 1, characterized in that: According to the real-time device power information, a real-time device operation status index is obtained, which specifically includes: According to the real-time industrial equipment information and real-time power quality information, the real-time operating status influencing factors of a single industrial equipment in an operating cycle under different working modes are obtained; According to the real-time equipment power information, the real-time useful power and real-time apparent power of a single industrial equipment within an operation cycle are obtained; Based on the real-time useful power and real-time apparent power, as well as the real-time operating status influencing factors of a single industrial device in an operating cycle under different working modes, the real-time equipment operating status index is obtained; The calculation formula of the real-time operation status influencing factor is: In the formula, γ represents the real-time operating status influencing factor, Q m Represents the real-time power quality parameter, P m Indicates the rated power of the device in different working modes, P e Indicates the rated power of the device in different working modes; The calculation formula of real-time useful power is: P real =V real I reat cosθ; Where P real Indicates real-time useful power, V real Indicates the effective voltage of industrial equipment in actual operation, I real It represents the effective current in the actual operation of industrial equipment, and cosθ represents the power factor; The calculation formula for real-time apparent power is: S fac =V fac I fac ; In the formula, S fac Indicates real-time apparent power, V fac Indicates the total voltage of industrial equipment during operation, I fac Indicates the total current of industrial equipment during operation; The calculation formula of the real-time equipment operation status index is: In the formula, F pow Indicates the real-time device operation status index.

4. The method for real-time online collection and analysis of power data according to claim 1, characterized in that: Based on the predicted equipment status index and real-time production environment information, the predicted equipment status final index is obtained, including: According to the real-time production environment information, obtain the real-time temperature index, real-time humidity index and real-time vibration parameters; Based on the real-time temperature index, the real-time humidity index, the real-time vibration parameter and the predicted equipment status index, the predicted equipment status final index is obtained; Among them, the calculation formula for predicting the final index of equipment status is: Where W τ Represents the predicted equipment status final index, T i Represents the real-time temperature index of the i-th time unit, H i Represents the real-time humidity index of the i-th time unit, V i represents the real-time vibration parameter of the i-th time unit, F ast (τ) represents the predicted equipment status index of the τth industrial equipment, β0 represents the error term, β1 represents the weight coefficient of the real-time temperature index, β2 represents the weight coefficient of the real-time humidity index, β3 represents the weight coefficient of the real-time vibration parameter, and j represents the equipment operation cycle.

5. The method for real-time online collection and analysis of power data according to claim 1, characterized in that: Based on historical power consumption information and abnormal event information, obtain the device abnormality index, including: Based on the historical power consumption information, obtain the historical power consumption information of all workshops, and further obtain the historical power consumption information of a single device; Based on the abnormal event information, abnormal device information, abnormal time node information and abnormal environment information are obtained, wherein the abnormal device information, abnormal time node information and abnormal environment information correspond to each other one by one; Based on the abnormal device information and abnormal time node information, the power information and environmental information of the abnormal device in a period of operation before the abnormality occurs are collected and marked to obtain the marked device power information and marked environmental information; Based on the identification device power information and identification environment information, the device abnormality index is obtained.

6. A method for real-time online collection and analysis of power data according to claim 5, characterized in that: Based on the power information and environment information of the identified equipment, the equipment abnormality index is obtained, including: Based on statistical methods, the power information of the identification equipment and the identification environment information are converted into curve statistical graphs to obtain the identification power curve statistical graph and the identification environment curve statistical graph; Based on historical industrial power data and historical industrial production data, obtain statistical graphs of standard power curves and standard environmental curves of industrial equipment under normal operating conditions; Overlap the statistical graph of the identification power curve and the statistical graph of the standard power curve, calculate the distance between the Y-axis coordinate points when the identification power curve and the standard power curve are at the same X-axis coordinate, and obtain the equipment power difference value; Overlap the identification environment curve statistical graph and the standard environment curve statistical graph, calculate the distance between the Y-axis coordinate points when the identification environment curve and the standard environment curve are at the same X-axis coordinate, and obtain the equipment environment difference value; Based on the equipment power difference value and the equipment environment difference value, obtain the equipment abnormality index; The calculation formula of the equipment abnormality index is: In the formula, δ represents the equipment abnormality index, M μ Indicates the abnormal power parameter per unit time in the μth operation cycle, N μ represents the standard power parameter per unit time in the μth operation cycle, Y μ represents the abnormal environmental parameters per unit time in the μth operating cycle, Z μ It represents the standard environmental parameters per unit time in the μth operating cycle, μ represents the unit time in the operating cycle, and π represents the operating cycle.

7. A real-time online power data acquisition and analysis system, used to implement the analysis method according to any one of claims 1 to 6, characterized in that: include: A main control module, wherein the main control module is used to perform data standardization processing on the equipment state prediction data to obtain a standard sample set for equipment state prediction, divide the standard sample set for equipment state prediction according to a preset ratio, obtain a training set, a validation set and a test set for the standard data for equipment state prediction, train an ARIMA time series model based on the training set, the validation set and the test set for the standard data for equipment state prediction, obtain a real-time equipment state prediction model, collect power information and environmental information of the abnormal equipment for a period of operation cycle before the abnormality occurs based on the abnormal equipment information and the abnormal time node information, and mark them, convert the identification equipment power information and the identification environmental information into a curve statistical graph based on a statistical method, obtain an identification power curve statistical graph and an identification environment curve statistical graph, overlap the identification power curve statistical graph and the standard power curve statistical graph, calculate the distance between the Y-axis coordinate points when the identification power curve and the standard power curve are at the same X-axis coordinate, and judge whether the predicted equipment state final index meets the standard based on the predicted equipment state final index threshold; A data acquisition module, the data acquisition module is used to acquire real-time industrial power data, real-time industrial production data, historical industrial power data, historical industrial production data, real-time equipment power information, equipment status prediction data, historical workshop power information, historical equipment power information, abnormal equipment information, abnormal time node information, abnormal environment information, identification equipment power information, identification environment information, real-time equipment operation status index, predicted equipment status index, predicted equipment status final index, equipment abnormality index, real-time operation status influencing factor, real-time useful power, real-time apparent power, real-time equipment operation status index, equipment power difference value and equipment environment difference value; A data transmission module, which is used to transmit the data collected by the data acquisition module to the main control module; The display module is used to display the identification power curve statistical graph, the standard power curve statistical graph, the identification environment curve statistical graph and the standard environment curve statistical graph, predict the final equipment status index, analyze the operating status of industrial equipment, and formulate equipment maintenance plans.

8. The real-time online power data acquisition and analysis system according to claim 7, characterized in that: The main control module specifically includes: A data processing unit, the data processing unit is used to perform data standardization processing on the device state prediction data to obtain a device state prediction standard sample set, divide the device state prediction standard sample set according to a preset ratio to obtain a training set, a verification set and a test set of the device state prediction standard data, based on the abnormal device information and the abnormal time node information, collect the power information and environmental information of the abnormal device in a period of operation before the abnormality occurs, and mark them, based on a statistical method, convert the identification device power information and the identification environment information into a curve statistical graph to obtain an identification power curve statistical graph and an identification environment curve statistical graph, overlap the identification power curve statistical graph and the standard power curve statistical graph, and calculate the distance between the Y-axis coordinate points of the identification power curve and the standard power curve when they are at the same X-axis coordinate; A training unit, wherein the training unit is used to train the ARIMA time series model based on a training set, a validation set, and a test set of equipment state prediction standard data to obtain a real-time equipment state prediction model; A judgment unit is used to judge whether the predicted device status index meets the standard based on the device abnormality threshold.

9. The real-time online power data acquisition and analysis system according to claim 7, characterized in that: The data acquisition module specifically includes: a first acquisition unit, the first acquisition unit being used to acquire real-time industrial power data, real-time industrial production data, historical industrial power data, historical industrial production data, real-time equipment power information, equipment status prediction data, historical workshop power information, historical equipment power information, abnormal equipment information, abnormal time node information, abnormal environment information, identification equipment power information, and identification environment information; The second acquisition unit is used to obtain the real-time equipment operation status index, the predicted equipment status index, the predicted equipment status index, the equipment abnormality index, the real-time operation status influencing factor, the real-time useful power, the real-time apparent power, the real-time equipment operation status index, the equipment power difference value and the equipment environment difference value.

10. The real-time online power data acquisition and analysis system according to claim 7, characterized in that: The display module specifically includes: A first display unit, the first display unit is used to display a statistical graph of a marked power curve, a statistical graph of a standard power curve, a statistical graph of a marked environment curve, and a statistical graph of a standard environment curve; The second display unit is used to display the predicted equipment status final index, analyze the operating status of industrial equipment, and formulate equipment maintenance plan.

Images