Intelligent detection method for abnormal data of meter

Abstract

The invention belongs to the technical field of abnormal data detection, and discloses an intelligent detection method for abnormal data of a meter. A hardware data set and an environment data set are collected and preprocessed, the hardware data set comprises reading change data, reading stable data and system stable data, and the environment data set comprises temperature data, humidity data and vibration data; processing the hardware data set and the environment data set to obtain operation reference values, and classifying the operation reference values to obtain a state classification result; analyzing the state classification result to obtain an equipment data decision result; according to the method, a visual interface is provided, data with abnormal equipment is checked, analyzed and processed through the visual interface, and generally, the method has the remarkable advantages of being good in auxiliary detection effect, high in auxiliary decision-making capacity and high in use convenience.

Description

Technical Field

[0001] The present invention relates to the technical field of abnormal data detection, and more specifically, to an intelligent detection method for abnormal data of a meter. Background Art

[0002] Meters generally refer to instruments used to measure the total amount (accumulated amount) of fluid flowing through a pipeline within a certain time interval, such as water meters, gas meters, electricity meters and heat meters. The total flow rate displayed by the instrument at two relevant times can be subtracted. Meters are widely used in industry, scientific research and other fields, such as machinery manufacturing industry, electronics industry, chemical industry, medical industry, etc. In these fields, meters can be used to measure machine tool accuracy, metal material size, electronic device size, chemical equipment size, density and temperature, etc. In general, meters are an indispensable measuring tool. With the rapid development of my country's economy in recent years, meters have been widely used as an important tool in daily life and work production. However, with the long-term use of meters, after a certain period of time, abnormal measurement data is often prone to occur, which not only affects daily life, but also easily causes deviations in production data in industrial production, thereby leading to economic losses.

[0003] The patent with application publication number CN116304935A discloses an abnormal data monitoring method, device and electronic device. The method obtains the number of jumps of the internal counter of the electric energy meter to be detected within a preset time interval to obtain the electric quantity data to be detected, and determines the first probability corresponding to the electric quantity data to be detected based on the Poisson distribution of the pre-constructed electric quantity data. The Poisson distribution of the electric quantity data is constructed based on the historical electric quantity data of the electric energy meter to be detected. The first probability is used to represent the probability of the electric quantity data to be detected occurring in the next preset time interval. Whether the electric quantity data to be detected is abnormal is determined by comparing the first probability with a preset probability threshold.

[0004] However, although the above-mentioned abnormal data monitoring method, device and electronic equipment obtain the first probability by collecting the power data to be detected and matching the Poisson distribution, thereby reflecting the state of the monitored power data, it is difficult to detect whether the meter data is abnormal based on the meter reading data during the use of the meter. The abnormality of the meter data will not only cause troubles in daily life, but also easily generate data deviations in the industrial production process if the meter data is abnormal, which will seriously affect industrial production and cause economic losses to the enterprise.

[0005] In view of this, the present invention proposes an intelligent detection method for abnormal data of a meter to solve the above problem. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solutions: comprising:

[0007] The data acquisition module is used to collect hardware data sets and environmental data sets and perform preprocessing. The hardware data sets include reading change data, reading stability data and system stability data. The environmental data sets include temperature data, humidity data and vibration data.

[0008] Furthermore, the methods of collecting the hardware data set and the environment data set include:

[0009] By connecting to the database, the reading values ​​of the meter within the preset time unit are collected, and the reading difference between two adjacent time points is divided by the reading value of the previous time point to obtain the reading change data;

[0010] By substituting into the calculation formula: Get the stable reading data, where is the number of reading values, is the data point to be calculated, It is the average value of the readings;

[0011] By connecting to the meter management platform, the number of failures within the preset time unit of the meter is collected and divided by the number of days in the preset time unit to obtain system stability data;

[0012] By installing temperature and humidity sensors, the temperature and humidity values ​​in the area where the meter is located are collected respectively to obtain temperature data and humidity data;

[0013] By installing a vibration monitor, the vibration values ​​in the area where the meter is located are collected to obtain vibration data;

[0014] The preprocessing methods include data cleaning, data denoising and data normalization;

[0015] The data processing module is used to process the hardware data set and the environmental data set to obtain the operation reference value, and classify them to obtain the state classification result;

[0016] Furthermore, the steps of processing the hardware data set and the environment data set include:

[0017] T1: By substituting into the calculation formula: Get the device hardware reference value, where is the reading change data, To stabilize the data of the system, , and are the corresponding weight factors respectively;

[0018] T2: By substituting into the calculation formula: Get the device environment reference value, where is the temperature data, is the humidity data, is the vibration data, , and The corresponding weight factors respectively;

[0019] T3: Collect M groups of historical equipment hardware reference values ​​and historical equipment environment reference values ​​as sample sets, and divide the sample sets into 70%M training sets, 15%M test sets, and 15%M validation sets;

[0020] Extract feature vectors based on the sample set, use keras to define the input layer and output layer, use feature vectors as input layer data, and use running reference values ​​as output layer data. Feature vectors are the weighted product of historical device hardware reference values ​​and historical device environment reference values.

[0021] Building an equipment abnormality warning model based on the sample set, training the equipment abnormality warning model to obtain a first equipment abnormality warning model, testing and verifying the first equipment abnormality warning model using a test set and a validation set, and outputting a first equipment abnormality warning model that meets a preset accuracy, wherein the equipment abnormality warning model is one of a logistic regression model, a naive Bayes model, or a support vector machine model;

[0022] T4: input the equipment hardware reference value and the equipment environment reference value into the first equipment early warning model to obtain the operation reference value;

[0023] Further, the classification methods include:

[0024] The equipment abnormality threshold interval (W1, W2) is preset. When the operating reference value is less than W1, an equipment stability signal is generated. When the operating reference value is greater than W1 and less than W2, an equipment warning signal is generated. When the operating reference value is greater than W2, an equipment abnormality signal is generated.

[0025] The device stable signal includes a group of fields representing that the device data state is stable, the device warning signal includes a group of fields representing that the device data state may be abnormal, and the device abnormal signal includes a group of fields representing that the device data state is abnormal;

[0026] Pack equipment stability signals, equipment warning signals and equipment abnormality signals to obtain status classification results;

[0027] Intelligent alarm module, used to analyze the status classification results and obtain equipment data decision results;

[0028] Furthermore, the method of analyzing the status classification results includes:

[0029] When the status classification result is a stable signal of the equipment, a stable SMS is sent to the staff through the communication unit; when the status classification result is a warning signal of the equipment, an alarm SMS is sent to the staff through the communication unit; when the status classification result is an abnormal signal of the equipment, an abnormal SMS is sent to the staff through the communication unit;

[0030] Stable SMS includes information indicating that the equipment data status is normal, reminding staff to perform regular maintenance on the meter, check the environment of the meter installation area, and take corresponding improvement measures based on the environmental conditions;

[0031] The alarm message includes an explanation of possible abnormalities in the device data status, requiring staff to immediately go to the device installation area and analyze the cause of the alarm. Based on the cause, the equipment installation environment should be optimized, the equipment should be restarted, or the equipment software version should be updated.

[0032] Abnormal SMS includes information indicating abnormalities in the device data status, requiring the device to be stopped immediately, and the damaged parts of the device to be replaced, software vulnerabilities to be fixed, or the device installation area to be adjusted according to the abnormality cause, and the device to be verified after the repair is completed;

[0033] Pack stable SMS, alarm SMS and abnormal SMS to get the equipment data decision result;

[0034] Visual feedback module, used to provide a visual interface, through which data with equipment anomalies can be viewed, analyzed and processed;

[0035] Further, S1: collecting hardware data sets and environmental data sets and performing preprocessing, the hardware data sets include reading change data, reading stability data and system stability data, and the environmental data sets include temperature data, humidity data and vibration data;

[0036] S2: Process the hardware data set and the environment data set to obtain the operation reference value, and classify them to obtain the state classification result;

[0037] S3: Analyze the status classification results to obtain the equipment data decision results;

[0038] S4: Provides a visual interface to view, analyze and process data with device anomalies.

[0039] The technical effects and advantages of the intelligent detection method for abnormal data of a meter according to the present invention are as follows:

[0040] The present invention processes the hardware data set and the environmental data set to obtain the operating reference value, which can assist the staff to intuitively view the operating status of the meter. By classifying the operating reference value, the status classification result obtained can complete the quantification of the operating status of the meter, thereby obtaining the classification of different interval states. By analyzing the status classification result, the equipment data decision result obtained can effectively assist the staff to complete the decision-making actions on the meters in different classification situations, greatly reducing the decision-making processing time required when the meter has data abnormalities. Through the established visual feedback module, the staff can view the meter data with abnormalities conveniently, quickly and in a targeted manner. Generally speaking, the present invention has the significant advantages of good auxiliary detection effect, strong auxiliary decision-making ability and high convenience of use. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of an intelligent detection system for abnormal data of a meter according to the present invention;

[0042] Figure 2 It is a schematic diagram of the intelligent detection method for abnormal data of a meter according to the present invention. DETAILED DESCRIPTION

[0043] 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 only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] Example 1

[0045] See also Figure 1 As shown, the intelligent detection system for abnormal data of a meter described in this embodiment includes:

[0046] The data acquisition module is used to collect hardware data sets and environmental data sets and perform preprocessing. The hardware data sets include reading change data, reading stability data and system stability data. The environmental data sets include temperature data, humidity data and vibration data.

[0047] Furthermore, the methods for collecting hardware data sets and environment data sets include:

[0048] By connecting to the database, the reading values ​​of the meter within the preset time unit are collected, and the reading difference between two adjacent time points is divided by the reading value of the previous time point to obtain the reading change data;

[0049] It needs to be explained that the preset time unit is month, quarter or year;

[0050] By substituting into the calculation formula: Get the stable reading data, where is the number of reading values, is the data point to be calculated, It is the average value of the readings;

[0051] It should be explained that the average value of historical readings is calculated by substituting into the formula: get;

[0052] By connecting to the meter management platform, the number of failures within the preset time unit of the meter is collected and divided by the number of days in the preset time unit to obtain system stability data;

[0053] By installing temperature and humidity sensors, the temperature and humidity values ​​in the area where the meter is located are collected respectively to obtain temperature data and humidity data;

[0054] By installing a vibration monitor, the vibration values ​​in the area where the meter is located are collected to obtain vibration data;

[0055] The preprocessing methods include data cleaning, data denoising and data normalization;

[0056] The data processing module is used to process the hardware data set and the environmental data set to obtain the operation reference value, and classify them to obtain the state classification result;

[0057] Furthermore, the steps of processing the hardware data set and the environment data set include:

[0058] T1: By substituting into the calculation formula: Get the device hardware reference value, where is the reading change data, To stabilize the data of the system, , and are the corresponding weight factors respectively;

[0059] T2: By substituting into the calculation formula: Get the device environment reference value, where is the temperature data, is the humidity data, is the vibration data, , and The corresponding weight factors respectively;

[0060] T3: Collect M groups of historical equipment hardware reference values ​​and historical equipment environment reference values ​​as sample sets, and divide the sample sets into 70%M training sets, 15%M test sets, and 15%M validation sets;

[0061] Extract feature vectors based on the sample set, use keras to define the input layer and output layer, use feature vectors as input layer data, and use running reference values ​​as output layer data. Feature vectors are the weighted product of historical device hardware reference values ​​and historical device environment reference values.

[0062] Building an equipment abnormality warning model based on the sample set, training the equipment abnormality warning model to obtain a first equipment abnormality warning model, testing and verifying the first equipment abnormality warning model using a test set and a validation set, and outputting a first equipment abnormality warning model that meets a preset accuracy, wherein the equipment abnormality warning model is one of a logistic regression model, a naive Bayes model, or a support vector machine model;

[0063] T4: input the equipment hardware reference value and the equipment environment reference value into the first equipment early warning model to obtain the operation reference value;

[0064] Further classification methods include:

[0065] The equipment abnormality threshold interval (W1, W2) is preset. When the operating reference value is less than W1, an equipment stability signal is generated. When the operating reference value is greater than W1 and less than W2, an equipment warning signal is generated. When the operating reference value is greater than W2, an equipment abnormality signal is generated.

[0066] The device stable signal includes a group of fields representing that the device data state is stable, the device warning signal includes a group of fields representing that the device data state may be abnormal, and the device abnormal signal includes a group of fields representing that the device data state is abnormal;

[0067] Pack equipment stability signals, equipment warning signals and equipment abnormality signals to obtain status classification results;

[0068] Intelligent alarm module, used to analyze the status classification results and obtain equipment data decision results;

[0069] Further, the method of analyzing the status classification results includes:

[0070] When the status classification result is a stable signal of the equipment, a stable SMS is sent to the staff through the communication unit; when the status classification result is a warning signal of the equipment, an alarm SMS is sent to the staff through the communication unit; when the status classification result is an abnormal signal of the equipment, an abnormal SMS is sent to the staff through the communication unit;

[0071] Stable SMS includes information indicating that the equipment data status is normal, reminding staff to perform regular maintenance on the meter, check the environment of the meter installation area, and take corresponding improvement measures based on the environmental conditions;

[0072] The alarm message includes an explanation of possible abnormalities in the device data status, requiring staff to immediately go to the device installation area and analyze the cause of the alarm. Based on the cause, the equipment installation environment should be optimized, the equipment should be restarted, or the equipment software version should be updated.

[0073] Abnormal SMS includes information indicating abnormalities in the device data status, requiring the device to be stopped immediately, and the damaged parts of the device to be replaced, software vulnerabilities to be fixed, or the device installation area to be adjusted according to the abnormality cause, and the device to be verified after the repair is completed;

[0074] Pack stable SMS, alarm SMS and abnormal SMS to get the equipment data decision result;

[0075] Visual feedback module, used to provide a visual interface, through which data with equipment anomalies can be viewed, analyzed and processed;

[0076] The beneficial effects of this embodiment are as follows: by processing the hardware data set and the environmental data set, the operation reference value obtained can assist the staff to intuitively view the operation status of the meter; by classifying the operation reference value, the status classification result obtained can complete the quantification of the operation status of the meter, thereby obtaining the classification of different interval states; by analyzing the status classification result, the equipment data decision result obtained can effectively assist the staff to complete the decision-making actions on the meters in different classification situations, greatly reducing the decision-making processing time required when the meter has data abnormalities; through the established visual feedback module, the staff can view the meter data with abnormalities conveniently, quickly and in a targeted manner; in general, the present invention has the significant advantages of good auxiliary detection effect, strong auxiliary decision-making ability and high convenience of use.

[0077] Example 2

[0078] See also Figure 2 As shown, the part not described in detail in this embodiment is described in Example 1, and an intelligent detection method for abnormal data of a meter is provided, including: S1: collecting a hardware data set and an environmental data set, and performing preprocessing, the hardware data set includes reading change data, reading stability data and system stability data, and the environmental data set includes temperature data, humidity data and vibration data;

[0079] S2: Process the hardware data set and the environment data set to obtain the operation reference value, and classify them to obtain the state classification result;

[0080] S3: Analyze the status classification results to obtain the equipment data decision results;

[0081] S4: Provides a visual interface to view, analyze and process data with device anomalies.

[0082] Example 3

[0083] This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the operation mode of the intelligent detection method for abnormal data of a meter provided above is implemented.

[0084] Since the electronic device introduced in this embodiment is an electronic device used to implement the intelligent detection method for abnormal data of a meter in the embodiment of the present application, based on the intelligent detection method for abnormal data of a meter introduced in the embodiment of the present application, a person skilled in the art can understand the specific implementation of the electronic device of the present embodiment and its various variations, so how the electronic device implements the method in the embodiment of the present application is not described in detail here. As long as a person skilled in the art implements the electronic device used in the intelligent detection method for abnormal data of a meter in the embodiment of the present application, it belongs to the scope of protection of the present application.

[0085] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0086] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. An intelligent detection system for abnormal data of a meter, characterized in that: include: The data acquisition module is used to collect hardware data sets and environmental data sets and perform preprocessing. The hardware data sets include reading change data, reading stability data and system stability data. The environmental data sets include temperature data, humidity data and vibration data. The data processing module is used to process the hardware data set and the environmental data set to obtain the operation reference value, and classify them to obtain the state classification result; Intelligent alarm module, used to analyze the status classification results and obtain equipment data decision results; The visual feedback module is used to provide a visual interface for viewing, analyzing and processing data with device anomalies.

2. The intelligent detection system for abnormal data of a meter according to claim 1, characterized in that: Methods for collecting hardware data sets and environment data sets include: By connecting to the database, the reading values ​​of the meter within the preset time unit are collected, and the reading difference between two adjacent time points is divided by the reading value of the previous time point to obtain the reading change data; By substituting into the calculation formula: Get the stable reading data, where is the number of reading values, is the data point to be calculated, It is the average value of the readings; By connecting to the meter management platform, the number of failures within the preset time unit of the meter is collected and divided by the number of days in the preset time unit to obtain system stability data; By installing temperature and humidity sensors, the temperature and humidity values ​​in the area where the meter is located are collected respectively to obtain temperature data and humidity data; By installing a vibration monitor, the vibration values ​​in the area where the meter is located are collected to obtain vibration data; The preprocessing methods include data cleaning, data denoising and data normalization.

3. The intelligent detection system for abnormal data of a meter according to claim 2, characterized in that: The steps for processing the hardware dataset and the environment dataset include: T1: By substituting into the calculation formula: Get the device hardware reference value, where is the reading change data, To stabilize the data of the system, , and are the corresponding weight factors respectively; T2: By substituting into the calculation formula: Get the device environment reference value, where is the temperature data, is the humidity data, is the vibration data, , and The corresponding weight factors respectively; T3: Collect M groups of historical equipment hardware reference values ​​and historical equipment environment reference values ​​as sample sets, and divide the sample sets into 70%M training sets, 15%M test sets, and 15%M validation sets; Extract feature vectors based on the sample set, use keras to define the input layer and output layer, use feature vectors as input layer data, and use running reference values ​​as output layer data. Feature vectors are the weighted product of historical device hardware reference values ​​and historical device environment reference values. Building an equipment abnormality warning model based on the sample set, training the equipment abnormality warning model to obtain a first equipment abnormality warning model, testing and verifying the first equipment abnormality warning model using a test set and a validation set, and outputting a first equipment abnormality warning model that meets a preset accuracy, wherein the equipment abnormality warning model is one of a logistic regression model, a naive Bayes model, or a support vector machine model; T4: Input the equipment hardware reference value and the equipment environment reference value into the first equipment early warning model to obtain the operation reference value.

4. The intelligent detection system for abnormal data of a meter according to claim 3, characterized in that: The classification methods include: The equipment abnormality threshold interval (W1, W2) is preset. When the operating reference value is less than W1, an equipment stability signal is generated. When the operating reference value is greater than W1 and less than W2, an equipment warning signal is generated. When the operating reference value is greater than W2, an equipment abnormality signal is generated. The device stability signal includes a group of fields representing that the device data state is stable, the device warning signal includes a group of fields representing that the device data state may be abnormal, and the device abnormal signal includes a group of fields representing that the device data state is abnormal; Pack equipment stability signals, equipment warning signals and equipment abnormality signals to obtain status classification results.

5. The intelligent detection system for abnormal data of a meter according to claim 4, characterized in that: The methods for analyzing the status classification results include: When the status classification result is a stable signal of the equipment, a stable SMS is sent to the staff through the communication unit; when the status classification result is a warning signal of the equipment, an alarm SMS is sent to the staff through the communication unit; when the status classification result is an abnormal signal of the equipment, an abnormal SMS is sent to the staff through the communication unit; Stable SMS includes information indicating that the equipment data status is normal, reminding staff to perform regular maintenance on the meter, check the environment of the meter installation area, and take corresponding improvement measures based on the environmental conditions; The alarm message includes an explanation of possible abnormalities in the device data status, requiring staff to immediately go to the device installation area and analyze the cause of the alarm. Based on the cause, the equipment installation environment should be optimized, the equipment should be restarted, or the equipment software version should be updated. Abnormal SMS includes information indicating abnormalities in the device data status, requiring the device to be stopped immediately, and the damaged parts of the device to be replaced, software vulnerabilities to be repaired, or the installation area of ​​the device to be adjusted according to the cause of the abnormality, and the device to be verified after the repair is completed; Pack stable SMS, alarm SMS and abnormal SMS to get the equipment data decision result.

6. A method for intelligently detecting abnormal data of a meter, implemented according to any one of the intelligent detection systems for abnormal data of a meter according to claims 1-5, characterized in that: S1: Collect hardware data sets and environmental data sets and perform preprocessing. The hardware data sets include reading change data, reading stability data and system stability data. The environmental data sets include temperature data, humidity data and vibration data. S2: Process the hardware data set and the environment data set to obtain the operation reference value, and classify them to obtain the state classification result; S3: Analyze the status classification results to obtain the equipment data decision results; S4: Provides a visual interface to view, analyze and process data with device anomalies.

Images