Insulation anomaly detection method and system based on multi-dimensional time series data
Through the insulation abnormality detection method based on multi-dimensional time sequence data, the overheating or supercooling problems caused by uneven temperature distribution in the heat treatment process are solved, and the precise control of the insulation time and the improvement of data detection efficiency are achieved.
Patent Information
- Application Number
- CN202411836007.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-12
- Publication Date
- 2025-04-04
AI Technical Summary
In the heat treatment process, due to the complex shape of the workpiece and different heat capacity, the temperature distribution of the material during the insulation process is uneven, overheating or supercooling occurs, which affects the quality of the workpiece and increases the difficulty of temperature consistency analysis and detection.
The insulation abnormality detection method based on multi-dimensional timing data is adopted, through multi-point temperature monitoring and intelligent data analysis, multi-dimensional timing data is obtained, sliding windows are created, relevant parameters of temperature data are calculated, and the abnormal data points are compared with the preset threshold value, and data visualization and alarm prompts are realized.
The precise calculation of the insulation time is realized, ensuring that the insulation data in different parts of the workpiece meets the process requirements, improving data detection efficiency and reliability, and reducing quality problems caused by temperature inhomogeneity.
Smart Images
Figure CN119760600A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of data anomaly detection, and in particular to a method and system for detecting thermal insulation anomalies based on multi-dimensional time series data. Background Art
[0002] Heat treatment is a process that changes the internal structure of a material and its physical and mechanical properties through processes such as heating, insulation and cooling. Each process uses specific temperatures, insulation times and cooling methods according to the characteristics and performance requirements of different materials, thereby changing the hardness, strength, toughness and corrosion resistance of the material. Therefore, in the heat treatment process, precise control of temperature and insulation time plays an important role in ensuring product quality and reducing production costs. However, in actual applications, the complex shapes, sizes and heat capacities of different workpieces result in uneven temperature distribution of the material during the insulation process, and local overheating or overcooling of the workpiece, which affects the quality of the workpiece and increases the difficulty of temperature consistency analysis and detection. Summary of the invention
[0003] In view of this, the present disclosure proposes a method and system for detecting insulation anomalies based on multi-dimensional time series data. The data anomalies of each dimension are comprehensively calculated, and the insulation time is accurately calculated through multi-point temperature monitoring and intelligent data analysis, and it can ensure that the insulation data of different parts of the heated workpiece meet the requirements of the process setting. Multi-point data is automatically collected through the data acquisition gateway, and the data is stored in real time and directly used for data analysis, which effectively improves the efficiency and reliability of data detection.
[0004] According to one aspect of the present application, a method for detecting thermal insulation anomalies based on multidimensional time series data is provided, comprising: acquiring multidimensional time series data;
[0005] Performing data analysis of each point respectively according to the multi-dimensional time series data;
[0006] The data analysis results are recorded to complete data visualization.
[0007] In a possible implementation, when acquiring the multidimensional time series data, the multidimensional time series data is acquired from a time series database based on job task configuration information;
[0008] Among them, the operation task configuration information includes: operation start time, operation end time and data collection points.
[0009] In a possible implementation, when performing data analysis on each point according to the multi-dimensional time series data:
[0010] Creating a sliding window based on the multidimensional time series data table;
[0011] Respectively calculating relevant parameters of the temperature data in each of the sliding windows;
[0012] The temperature data and related parameters in the sliding window are compared with the temperature threshold and the threshold of the related parameters to obtain abnormal data points.
[0013] In a possible implementation, recording the data analysis results to complete data visualization includes: recording the data analysis results;
[0014] Visually display the time series data and mark abnormal data points;
[0015] And an alarm prompt is given for the data abnormal points on the visual interface.
[0016] In a possible implementation, in the database, each acquisition point corresponds to a separate time series data table for storing the time series data of the point.
[0017] In a possible implementation, the temperature threshold and thresholds of related parameters are obtained according to preset values in the process parameter configuration, including:
[0018] Any one or a combination of preset temperature upper and lower limit thresholds, preset temperature variance thresholds, and preset temperature Z-Score thresholds.
[0019] In a possible implementation, when no data anomaly is detected in each of the sliding windows, the time series data is recorded and data visualization is performed.
[0020] According to another aspect of the present disclosure, a temperature-insulating abnormality detection system based on multi-dimensional time series data is provided, comprising: a data acquisition module, a data analysis module, and a data display module;
[0021] The data acquisition module is configured to acquire multi-dimensional time series data;
[0022] The data analysis module is configured to perform data analysis of each point respectively according to the multi-dimensional time series data;
[0023] The data display module is configured to record the data analysis results and achieve data visualization.
[0024] According to another aspect of the present disclosure, a thermal insulation anomaly detection device based on multi-dimensional time series data is provided, comprising: a processor; a memory for storing processor executable instructions; wherein the processor is configured to execute the above method.
[0025] According to another aspect of the present disclosure, a non-volatile computer-readable storage medium is provided, on which computer program instructions are stored, wherein the computer program instructions implement the above method when executed by a processor.
[0026] Further features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments, features, and aspects of the disclosure and, together with the description, serve to explain the principles of the disclosure.
[0028] Figure 1 A flow chart showing a method for detecting thermal insulation anomalies based on multi-dimensional time series data according to an embodiment of the present application is shown;
[0029] Figure 2 A structural block diagram of thermal insulation anomaly detection based on multi-dimensional time series data according to an embodiment of the present application is shown;
[0030] Figure 3 A flow chart of a data acquisition module according to an embodiment of the present application is shown;
[0031] Figure 4 A flow chart of a data analysis module according to an embodiment of the present application is shown;
[0032] Figure 5 A schematic diagram of a multi-dimensional time series data analysis process according to an embodiment of the present application is shown;
[0033] Figure 6 A data analysis visualization page according to an embodiment of the present application is shown. DETAILED DESCRIPTION
[0034] Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.
[0035] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.
[0036] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present disclosure.
[0037] The present invention discloses heat treatment insulation abnormality detection based on multi-sensor and multi-dimensional time series data, and performs time series analysis on multi-point temperature data in real time through multi-point temperature monitoring and intelligent data analysis; collects time series data of multiple measuring points through data acquisition gateway, stores it in the time series database and directly uses it for data analysis, effectively improving data detection efficiency. And according to different process requirements and based on different analysis parameters, it can ensure that the insulation data of different parts of the heated workpiece meet the requirements of process settings.
[0038] Example 1
[0039] Figure 1 A flow chart of a method for detecting thermal insulation anomalies based on multi-dimensional time series data according to an embodiment of the present disclosure is shown. Figure 1 As shown, the method includes:
[0040] Step S1: Obtain a multi-dimensional time series data table. Time series data refers to data collected at different times, which is used to describe the state of temperature data that changes over time.
[0041] Step S2: Analyze the data of each point separately according to the multidimensional time series data table. Specifically, create a sliding window according to the obtained multidimensional time series data table. For example, if the whole process of heat treatment of the workpiece is 24 hours, then the time series data table within the whole 24 hours can be obtained, that is, the temperature points stored in chronological order in 24 hours. When multiple sensors are used to monitor the temperature at different positions of the workpiece at the same time, the time series data of each sensor in the same time period is obtained. Each sensor corresponds to a collection point, and each collection point has a separate time series data table to save the time series data of the point, and the time series data of the same time period are calculated and analyzed in the same sliding window.
[0042] The sliding window is used to segment the multidimensional time series data table, intercept the sequence fragments and reshape the original time series data into samples of specified length.
[0043] Step S3: Record the abnormal situation of the data analysis and complete data visualization. In this embodiment, anomaly detection is performed on the time series data in each sliding window. Anomaly detection includes anomalies of the temperature-related parameters of the data in this window compared with the temperature threshold preset by the process parameter configuration, and also includes anomalies of a certain temperature point in this time window.
[0044] The job task configuration information refers to the configuration information for processing the workpiece, including the job start time, job end time, and the data points required for analysis. The process parameter configuration refers to the appropriate process parameters set according to product requirements and process characteristics, among which the preset temperature thresholds include: temperature upper and lower limit thresholds, temperature variance threshold, temperature Z-Score threshold Z threshold .
[0045] like Figure 2 The figure shows a block diagram of the insulation anomaly detection structure based on multi-dimensional time series data of an embodiment of the present application, including: a high-precision temperature sensor, a multi-channel temperature controller, a data acquisition gateway, a data storage module, time series data storage, a process parameter configuration module, an operation task configuration module, a time series data analysis module and a time series data visualization display module.
[0046] The heated workpiece is placed in a heating furnace, and a plurality of high-precision temperature sensors are provided to collect the temperature of the heated workpiece in real time. It should be noted that in a specific embodiment of the present disclosure, the number of temperature sensors is set to 4; in fact, the number of temperature sensors can be adjusted accordingly according to the actual application environment, and in this embodiment, the number and model of sensors are not limited to specific parameters.
[0047] When the heating furnace starts to heat up and the heat treatment operation begins, the temperature sensor transmits the collected data to the multi-channel temperature controller, and the data acquisition gateway reads the temperature data of all the temperature sensors at the collection points through the multi-channel temperature controller. Among them, the temperature data includes the information of each sensor collection point, the data collection time and the collected temperature value.
[0048] The data acquisition gateway sends the read real-time temperature data to the data storage module. After receiving the real-time data, the data storage module stores the data in the time series database. In the time series database, there is a separate time series data table for each collection point to save the time series data of the collection point.
[0049] When the workpiece is heated to the specified temperature, it enters the insulation stage. The temperature of each part of the workpiece to be processed needs to be consistent and needs to be maintained within the specified upper and lower ranges, that is, the preset upper and lower temperature thresholds. The temperature data of each point is maintained within the upper and lower temperature thresholds, and the temperature data difference between each point is also maintained within the preset range.
[0050] Continuous data collection is performed at a set first time interval (1 min / time) until the current work processing is completed; the time interval can be modified accordingly according to actual conditions, and no specific parameters are set for the data collection interval in this disclosure. Repeat the above collection process until the current work is completed.
[0051] Read the current job task configuration information, obtain the job start time, job end time and required collection points, read the multi-dimensional time series data from the time series database according to the job start time and job end time; and read the current process parameter configuration, read the different upper and lower temperature limits and insulation time required for each collection point in the process configuration, and set the following thresholds according to the process parameters: temperature upper and lower limit thresholds, temperature variance threshold, and temperature Z-Score threshold.
[0052] Create a sliding window. Preferably, in this embodiment, the sliding window size is set to 10. Create a window starting from the first data point, and slide down 1 data point in turn to obtain multiple windows. Calculate temperature-related parameters in each window, including: mean, standard deviation and Z-Score, compare the temperature-related parameters and temperature points with the threshold, and judge the abnormality of the data in the window; if the temperature is abnormal or the insulation time requirement is not met, an alarm prompt will pop up on the display page. Then display the time series data in the form of a temperature curve.
[0053] It should be noted that the window size and sliding size when creating a sliding window are both preset values in the process parameters. In actual applications, the process parameters corresponding to different processes and the required sliding window sizes may also be different. The window size and sliding size can be adjusted according to the specific situation.
[0054] Specifically, starting from the first data point, a sliding time window of size N is created, and the mean μ of N sensors is calculated within the time window. i :
[0055]
[0056] Among them, T i,j is the jth temperature data point of the ith sensor in the window.
[0057] Compared with conventional temperature upper and lower limit judgments, variance can quantify temperature fluctuations at different measurement points or in different time periods. When the variance of a certain window suddenly increases, it indicates equipment failure or uneven heating of the workpiece.
[0058] According to the mean μ i Calculate the standard deviation σ of the sensor within each window i :
[0059]
[0060] Z-Score can accurately identify extreme temperature points, respond to temperature changes in real time, and detect small changes in temperature trends in advance.
[0061] The calculation formula of the Z-Score of the collection point is as follows:
[0062]
[0063] Among them, Z i,j is the Z-Score of the j-th data point of the ith sensor in the window.
[0064] Compare the result obtained by the above calculation with the threshold value and check the current temperature point T i,jWhether the temperature exceeds the upper and lower temperature thresholds;
[0065] Check whether the variance of the current temperature point exceeds the temperature variance threshold;
[0066] The variance is obtained based on the square of the standard deviation.
[0067] Check whether the absolute value of Z at the current temperature point exceeds the temperature threshold Z threshold ; Any temperature point with a value exceeding the corresponding threshold value among the upper and lower limits of temperature, temperature variance and temperature Z-Score is considered as an abnormal point and marked in data visualization.
[0068] Anomaly detection through variance and Z-Score can effectively monitor temperature fluctuations and identify abnormal points to ensure the stability and consistency of the heat treatment process. Variance can quantify temperature fluctuations at different temperature points or in different time periods. If the variance of a certain window suddenly increases, it may mean equipment failure or uneven heating of the workpiece, which will lead to different heat treatment effects in different parts of the workpiece, resulting in problems such as uneven hardness and excessive internal stress. Z-Score can accurately identify extreme temperature points. In the heat treatment process, temperatures that are too high or too low will have an adverse effect on the performance of the workpiece. Z-Score can respond to temperature changes in real time. Z-Score is more sensitive and can detect small changes in temperature trends in advance. If this local temperature anomaly is not discovered in time, it may cause the workpiece to overheat or overcool in that part, affecting the quality of the workpiece.
[0069] After the comparison in the current window is completed, the time window moves to the next period and the above comparison process is repeated until all data processing is completed.
[0070] Finally, all the temperature time series data are displayed on the data analysis page ( Figure 6 ) is displayed as a continuous curve, such as Figure 6 As shown in the figure, the minimum and maximum insulation temperatures correspond to the upper and lower temperature thresholds. The page also includes visualization of conventional temperature measuring component data such as furnace temperature, material temperature thermocouple, quenching medium temperature, and temperature control thermocouple.
[0071] A separate temperature curve is drawn for each point of time series data. Multiple temperature curves can be drawn into a temperature analysis chart, and temperature abnormal points can be marked separately to facilitate comparative analysis, so that operators can identify temperature abnormal areas in a timely manner.
[0072] According to another aspect of the present disclosure, there is also provided a thermal insulation anomaly detection system based on multi-dimensional time series data, comprising: a data acquisition module, a data analysis module, and a data display module;
[0073] like Figure 3The data collection flow chart is shown. Data is collected once a minute. The data collection gateway reads the current temperature values of all temperature sensors through a multi-channel temperature controller. The data of each temperature sensor is saved in different data tables of the time series database to determine whether the current heating result is completed. Data collection continues until the task is completed.
[0074] Data analysis process Figure 4 As shown, the time series data after the above acquisition process is stored in the time series database, the job task configuration is read, and the time series data is read from the time series database according to the job start time and job end time in the configuration information for subsequent analysis. At the same time, the process configuration used by the job is read to obtain the upper and lower temperature threshold information and insulation time required by the acquisition point. Through multi-dimensional time series data analysis, it is determined whether there is an abnormal temperature or the insulation time is insufficient, and then the time series data is visualized in the form of a continuous curve, the abnormal point is marked in the curve, and an alarm prompt of the abnormal situation pops up on the display page.
[0075] The specific process of further multidimensional time series data analysis is as follows Figure 5 As shown in the figure, the sliding time window is divided according to the data collection interval, and the mean, standard deviation and Z-Score in each time window are calculated and compared with the threshold, and then all data processing is completed when sliding to the next time window. Abnormal data is recorded for subsequent display module data visualization.
[0076] Furthermore, according to another aspect of the present disclosure, a thermal insulation anomaly detection device based on multi-dimensional time series data is also provided, comprising a processor and a memory for storing processor executable instructions, wherein the processor is configured to implement any of the above-mentioned methods when executing the executable instructions.
[0077] Here, it should be noted that the number of processors can be one or more
[0078] The memory, as a computer-readable storage medium, can be used to store software programs, computer executable programs and various modules, such as the module corresponding to the insulation anomaly detection method based on multi-dimensional time series data in the embodiment of the present disclosure. The processor executes various functional applications and data processing of the device by running the software programs or modules stored in the memory.
[0079] According to another aspect of the present disclosure, a non-volatile computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, any of the above-mentioned methods is implemented.
[0080] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.
Claims
1. A method for detecting thermal insulation anomalies based on multidimensional time series data, characterized in that: include: Obtain multi-dimensional time series data; Performing data analysis of each point respectively according to the multi-dimensional time series data; The data analysis results are recorded to complete data visualization.
2. The method for detecting thermal insulation anomalies based on multidimensional time series data according to claim 1, characterized in that: When acquiring the multi-dimensional time series data, the multi-dimensional time series data is acquired from a database based on the job task configuration information; Among them, the operation task configuration information includes: operation start time, operation end time and data collection points.
3. The method for detecting thermal insulation anomalies based on multidimensional time series data according to claim 1, characterized in that: When analyzing the data of each point respectively according to the multi-dimensional time series data: Creating a sliding window based on the multidimensional time series data; Respectively calculating relevant parameters of the temperature data in each of the sliding windows; The temperature data and related parameters in the sliding window are compared with the temperature threshold and the threshold of the related parameters to obtain abnormal data points.
4. The method for detecting thermal insulation anomalies based on multidimensional time series data according to claim 1, characterized in that: The step of recording the results of the data analysis and completing data visualization includes: Recording the data analysis results; Visually display the time series data and mark abnormal data points; And an alarm prompt is given for the data abnormal points on the visual interface.
5. The method for detecting thermal insulation anomalies based on multidimensional time series data according to claim 2, characterized in that: In the database, each acquisition point corresponds to a separate time series data table for storing the time series data of the point.
6. The method for detecting thermal insulation anomalies based on multidimensional time series data according to claim 3, characterized in that: The temperature threshold and the thresholds of related parameters are obtained according to the preset values in the process parameter configuration, including: Any one or a combination of preset temperature upper and lower limit thresholds, preset temperature variance thresholds, and preset temperature Z-Score thresholds.
7. The method for detecting thermal insulation anomalies based on multidimensional time series data according to claim 4, characterized in that: Also includes: When no data anomaly is detected in each of the sliding windows, the time series data is recorded and data visualization is performed.
8. A thermal insulation anomaly detection system based on multi-dimensional time series data, characterized in that: include: Data acquisition module, data analysis module, data display module; The data acquisition module is configured to acquire multi-dimensional time series data; The data analysis module is configured to perform data analysis of each point respectively according to the multi-dimensional time series data; The data display module is configured to record the data analysis results and achieve data visualization.
9. A thermal insulation anomaly detection device based on multi-dimensional time series data, characterized in that: include: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to implement the method described in any one of claims 1 to 6 when executing the executable instructions.
10. A non-volatile computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.