Method and system for validating injection molding machine sensor data and determining data normality

Abstract

The application discloses a method and system for verifying injection molding machine sensor data and judging data normality, relates to the technical field of data processing, and comprises the following steps: acquiring sensor data of an injection molding machine in an idle state through multiple sensors; clustering the sensor data and comparing the clustering result with a preset standard clustering mode to perform deviation analysis and anomaly detection on the sensor data and obtain a first verification result; filtering the sensor data and comparing the sensor data with corresponding historical data to detect potential abnormal patterns and obtain a second verification result; determining the time offset between the sensor data, performing time synchronization on the sensor data according to the time offset, and then performing error analysis to identify sensors with abnormal readings and obtain a third verification result; and finally generating a comprehensive evaluation report and pushing an early warning message. The application can verify whether the sensor data is abnormal from multiple angles, and improves the data accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to a method and system for verifying sensor data of an injection molding machine and judging data normality. BACKGROUND

[0002] With the continuous improvement of the intelligence and automation level of injection molding machines, more and more sensors are integrated into the injection molding machine system. The above-mentioned sensors are responsible for monitoring the running state, process parameters and safety performance of the injection molding machine, providing the injection molding machine with various intelligent functions such as real-time monitoring, fault early warning, production efficiency analysis, etc. However, the performance and data reliability of the sensors directly affect the stable operation of the injection molding machine system. If the sensors themselves have faults, abnormal data or decreased precision, it will produce false data output, leading to incorrect judgments and control decisions by the management platform, which may cause product quality problems or equipment damage. SUMMARY

[0003] The main purpose of the embodiments of the present application is to provide a method and system for verifying sensor data of an injection molding machine and judging data normality, so as to verify whether the sensor data of the injection molding machine is abnormal.

[0004] To achieve the above-mentioned purpose, one aspect of the embodiments of the present application provides a method for verifying sensor data of an injection molding machine and judging data normality, which comprises the following steps:

[0005] Data of the injection molding machine in an idle state obtained by a plurality of sensors are respectively taken as different sensor data;

[0006] Each of the sensor data is clustered, and the clustering result is compared with a preset standard clustering mode, so as to perform deviation analysis and anomaly detection on each of the sensor data, and obtain a first verification result;

[0007] Each of the sensor data is filtered, and each of the sensor data is compared with corresponding historical data, so as to detect potential abnormal patterns and obtain a second verification result;

[0008] The time offset between each of the sensor data is determined, each of the sensor data is time-synchronized according to the time offset, and error analysis is performed on each of the sensor data to identify the sensor with abnormal readings, and a third verification result is obtained;

[0009] A comprehensive evaluation report and a push warning information are generated according to the state data of each of the sensors, the first verification result, the second verification result and the third verification result.

[0010] In some embodiments, the data obtained by the plurality of sensors when the injection molding machine is in an idle state as different sensor data respectively comprises the following steps:

[0011] The data obtained by the plurality of sensors when the injection molding machine is in an idle state as initial data;

[0012] The initial data is subjected to field integrity check, and all time markers in the initial data are uniformly converted into Unix time format, and the type of the initial data is converted into a target data type, to obtain standardized data;

[0013] The data missing rate of each variable in the standardized data is calculated, and the variables with a data missing rate lower than a first missing threshold are subjected to data completion by a high-order polynomial curve fitting method; the variables with a data missing rate higher than a second missing threshold are marked and removed, and relevant warning information is recorded, to obtain the sensor data.

[0014] In some embodiments, the clustering of each of the sensor data and the comparison of the clustering result with a preset standard clustering mode are performed to analyze the deviation and detect the anomaly of each of the sensor data, to obtain a first verification result, comprising the following steps:

[0015] Each of the sensor data is clustered by using a probability model based on expectation maximization;

[0016] The similarity measure between the clustering result and the preset standard clustering mode is calculated to analyze the deviation and detect the anomaly of each of the sensor data, to obtain the first verification result.

[0017] In some embodiments, the method further comprises the following steps:

[0018] The sampling interval, mean and variance of each of the sensor data are calculated to obtain sampling characteristics;

[0019] The sampling characteristics are subjected to standardization processing, and each of the sensor data is mapped into a normal distribution space with zero mean and unit variance, and then the sensors with statistical feature anomaly are identified according to the difference of each of the sensor data from the central data range;

[0020] The sensors identified as the statistical feature anomaly are added to the first verification result.

[0021] In some embodiments, the filtering of each of the sensor data and the comparison of each of the sensor data with corresponding historical data are performed to detect potential abnormal patterns, to obtain a second verification result, comprising the following steps:

[0022] a recursive filtering algorithm based on Bayesian estimation is used to perform dynamic state estimation on each of the sensor data to filter out high-frequency random noise;

[0023] the filtered sensor data is compared with corresponding historical data to detect potential abnormal patterns, and the second verification result is obtained.

[0024] In some embodiments, the method further comprises the following steps:

[0025] The coupling and the degree of cooperative change of the sensor data between different sensors on the same injection molding machine are evaluated to obtain a data correlation degree;

[0026] For the sensors with a data correlation degree lower than a correlation threshold, an abnormal early warning information is issued to prompt the existence of functional failure or data anomaly.

[0027] In some embodiments, the generation of a comprehensive evaluation report and the pushing of early warning information according to the state data of each sensor, the first verification result, the second verification result and the third verification result comprises the following steps:

[0028] According to the state data of each sensor, the first verification result, the second verification result and the third verification result, the statistical significance and the confidence interval of each sensor data are evaluated, and then the abnormal detection result is determined and the corresponding processing suggestion is generated;

[0029] According to the state data of each sensor, the first verification result, the second verification result and the third verification result, a data trend chart, a complex network correlation analysis atlas and a visualization marking map based on abnormal point detection are generated;

[0030] According to the abnormal detection result, the processing suggestion, the data trend chart, the complex network correlation analysis atlas and the visualization marking map based on abnormal point detection, the comprehensive evaluation report is generated;

[0031] When it is detected that each of the sensor data triggers an abnormal pattern or an index breaks through a set confidence interval, an early warning mechanism is triggered and the early warning information is pushed to a maintenance personnel.

[0032] To achieve the above-mentioned purpose, another aspect of the embodiments of the present application proposes a system for verifying injection molding machine sensor data and judging data normality, the system comprising:

[0033] a data acquisition unit configured to acquire data of an injection molding machine in an idle state through a plurality of sensors as different sensor data respectively;

[0034] The first verification unit is configured to cluster each of the sensor data, compare the clustering result with a preset standard clustering mode, perform deviation analysis and anomaly detection on each of the sensor data, and obtain a first verification result;

[0035] The second verification unit is configured to filter each of the sensor data, compare each of the sensor data with corresponding historical data, detect a potential abnormal pattern, and obtain a second verification result;

[0036] The third verification unit is configured to determine a time offset between each of the sensor data, perform time synchronization on each of the sensor data according to the time offset, perform error analysis on each of the sensor data to identify the sensor with a reading abnormality, and obtain a third verification result;

[0037] The evaluation and early warning unit is configured to generate a comprehensive evaluation report and push early warning information according to the state data of each of the sensors, the first verification result, the second verification result, and the third verification result.

[0038] To achieve the above object, another aspect of the embodiment of the present application provides an electronic device, which comprises a memory and a processor, the memory stores a computer program, and the processor implements the above method when executing the computer program.

[0039] To achieve the above object, another aspect of the embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the above method.

[0040] The embodiment of the present application at least has the following beneficial effects:

[0041] The application can obtain data of the injection molding machine in an idle state through multiple sensors as different sensor data respectively, cluster each sensor data, and compare the clustering result with a preset standard clustering mode to perform deviation analysis and anomaly detection on each sensor data, to obtain a first verification result; filter each sensor data, and compare each sensor data with corresponding historical data to detect a potential abnormal pattern, to obtain a second verification result; determine a time offset between each sensor data, perform time synchronization on each sensor data according to the time offset, and perform error analysis on each sensor data to identify a sensor with a reading abnormality, to obtain a third verification result; and generate a comprehensive evaluation report and push an early warning information according to the state data of each sensor, the first verification result, the second verification result and the third verification result. The application can verify whether the sensor data is abnormal from multiple angles through clustering analysis, filtering comparison, time synchronization and error analysis of the sensor data, to improve the accuracy of the sensor data; and the generation of the comprehensive evaluation report and the push of the early warning information can make the maintenance personnel more clearly understand the state of each sensor, to improve the efficiency of maintenance. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0043] Figure 1 The flowchart of the method for verifying sensor data of an injection molding machine and judging data normality provided by the embodiments of the present application;

[0044] Figure 2 The structural schematic diagram of the system for verifying sensor data of an injection molding machine and judging data normality provided by the embodiments of the present application;

[0045] Figure 3 The hardware structural schematic diagram of an electronic device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0046] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not intended to limit the present application. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with embodiments of the present application. They are only examples of systems and methods consistent with some aspects of the embodiments of the present application as detailed in the appended claims.

[0047] It can be understood that the terms "first", "second", and the like used in the present application can be used herein to describe various concepts, but unless specifically stated, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of the present application, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information. Depending on the context, the word "if" as used herein can be interpreted as "when" or "when" or "in response to determining".

[0048] The terms "at least one", "multiple", "each", "any" and the like used in the present application include one, two or more than two, multiple includes two or more than two, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0049] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0050] Before the embodiments of the present application are described in detail, first, some nouns and related technologies involved in the embodiments of the present application are described, and the nouns and related technologies involved in the embodiments of the present application are applicable to the following explanations:

[0051] Injection molding machine: a machine device for manufacturing plastic products by injecting hot and molten thermoplastic material into a mold.

[0052] End-side system: a processing system deployed on a terminal device for data acquisition and preliminary processing.

[0053] Unix time format: a time representation of the number of seconds elapsed from January 1, 1970 UTC.

[0054] High-order polynomial curve fitting: a mathematical method for fitting data points with a polynomial function.

[0055] Expectation-Maximization (EM): An iterative algorithm used to find the maximum likelihood estimate in a probabilistic model.

[0056] Bayesian Estimation: A statistical inference method based on Bayes' theorem.

[0057] Recursive Filtering Algorithm: An iterative algorithm for noise filtering of time series data.

[0058] Dynamic Time Warping (DTW): An algorithm for measuring the similarity of two time series.

[0059] Physical Quantity Terms:

[0060] RPM: Revolutions Per Minute, the unit of speed.

[0061] Mpa: Megapascal, a unit of pressure.

[0062] MAC Address: Media Access Control Address, the hardware address of network equipment.

[0063] Data Analysis Terms:

[0064] Pearson Correlation Coefficient: A statistical indicator used to measure the degree of linear correlation between two variables.

[0065] 3σ Rule: A statistical rule of standard deviation in normal distribution.

[0066] Isolation Forest Algorithm: An anomaly detection algorithm.

[0067] SVM (Support Vector Machine): A machine learning algorithm.

[0068] Data Format Standards:

[0069] ISO 8601: International standard for date and time representation.

[0070] Jetson: An embedded AI computing platform.

[0071] Related Technologies of the Present Application:

[0072] Sensor Monitoring System with Preset Parameters: Traditional monitoring systems require manual input of each sensor's model, parameters and configuration before deployment. These systems rely on predefined rules and thresholds to determine the status of the sensor.

[0073] Semi-automated data collection and reporting: Some systems can automatically collect data, but data analysis and report generation still require human intervention, or only provide basic automation functions, and cannot achieve full-process automated management.

[0074] Fixed model anomaly detection: Many existing systems use fixed statistical models or simple algorithms for anomaly detection, lacking the ability to learn and adapt to new data.

[0075] Defects of related technologies:

[0076] 1. Lack of adaptability, difficult to support diversified sensor types: Due to the reliance on preset sensor parameters, the system cannot automatically diagnose when faced with new or unknown types of sensors, requiring manual configuration and increasing maintenance costs.

[0077] 2. Low degree of automation, cumbersome monitoring and reporting process: Data analysis and report generation need to be performed manually, which is time-consuming and labor-intensive. Delayed anomaly monitoring is prone to occur. Real-time monitoring and feedback of sensor status cannot be achieved, which may lead to the use of incorrect sensor data in subsequent applications.

[0078] 3. Lack of continuous learning and pattern updating ability: Using fixed models or rules, it cannot be updated according to historical data and new trends, resulting in a decrease in diagnostic accuracy; not fully exploiting long-term accumulated sensor data, unable to improve the intelligent level of the system.

[0079] Therefore, the embodiments of the present application provide a method and system for verifying sensor data of an injection molding machine and judging data normality. The technical scheme of the present application comprises: acquiring data of the injection molding machine in an idle state by a plurality of sensors as different sensor data respectively; clustering each sensor data and comparing the clustering result with a preset standard clustering pattern to perform deviation analysis and anomaly detection on each sensor data, obtaining a first verification result; filtering each sensor data and comparing each sensor data with corresponding historical data to detect potential abnormal patterns, obtaining a second verification result; determining the time offset between each sensor data, performing time synchronization on each sensor data according to the time offset, and performing error analysis on each sensor data to identify sensors with reading errors, obtaining a third verification result; generating a comprehensive evaluation report and pushing an early warning information according to the state data of each sensor, the first verification result, the second verification result and the third verification result. By clustering analysis, filtering comparison, time synchronization and error analysis of sensor data, the present application can verify whether the sensor data is abnormal from multiple angles, thereby improving the accuracy of the sensor data; the generation of a comprehensive evaluation report and the pushing of early warning information can make the maintenance personnel more clearly understand the state of each sensor, improving the efficiency of maintenance.

[0080] The embodiment of the present application provides a method for verifying sensor data of an injection molding machine and judging data normality, and relates to the technical field of data processing. The method for verifying sensor data of an injection molding machine and judging data normality provided by the embodiment of the present application can be applied to a terminal, can be applied to a server, and can also be software running in the terminal or the server. In some embodiments, the terminal can be a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, a vehicle-mounted terminal and the like, but is not limited thereto; the server end can be configured as an independent physical server, can be configured as a server cluster or a distributed system formed by multiple physical servers, can also be configured as a cloud server providing basic cloud computing services such as cloud service, cloud database, cloud computing, cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN and big data and artificial intelligence platform, and the server can also be a node server in a blockchain network; the software can be an application for implementing the method for verifying sensor data of an injection molding machine and judging data normality, and the like, but is not limited to the above forms.

[0081] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present application can also be practiced in a distributed computing environment, in which tasks are performed by remote processing devices connected by a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0082] Reference Figure 1 The embodiment of the present application provides a method for verifying sensor data of an injection molding machine and judging data normality, which can include but is not limited to S100 to S140, and specifically as follows:

[0083] S100: acquiring data when the injection molding machine is in an idle state as different sensor data through multiple sensors.

[0084] Exemplarily, multiple sensors can be arranged on an injection molding machine, the types of the sensors can be the same or different, and when it is detected that the injection molding machine is in an idle state, the data collected by the sensors at this time can be acquired as sensor data.

[0085] Further, S100 can include S101-S103:

[0086] S101: Obtain data when the injection molding machine is in an idle state as initial data through a plurality of sensors;

[0087] S102: Perform field integrity checking on the initial data, uniformly convert all time markers in the initial data into Unix time format, and convert the type of the initial data into a target data type to obtain standardized data;

[0088] S103: Calculate the data missing rate of each variable in the standardized data, and use a high-order polynomial curve fitting method to complete the data for the variable with a data missing rate lower than a first missing threshold; mark and remove the variable with a data missing rate higher than a second missing threshold, and record relevant warning information to obtain the sensor data.

[0089] In some optional embodiments, the data acquisition and preprocessing of the present embodiment can include the following steps:

[0090] Real-time monitoring of the state of the injection molding machine: The end-side system perceives the running state of the injection molding machine in real time, judges whether the injection molding machine is in an idle state, and starts data acquisition only when the injection molding machine is idle to avoid interference in the running state. The collected data is transmitted to the cloud processing system in real time through a wireless communication network.

[0091] Data format verification and standardization: The cloud system performs integrity checking on necessary fields of the sensor data, including time markers, device unique identifiers, and variable data, and uniformly converts all time markers into a universal Unix time format to establish a unified time reference system. At the same time, the data is type-verified and converted to ensure that all variable data is in a processable numerical type.

[0092] Missing value judgment: Calculate the data missing rate of each variable. For variables with a low missing rate (≤3%), use a high-order polynomial curve fitting method to complete the data; for variables with a high missing rate, mark and remove them from subsequent analysis, while recording relevant warning information.

[0093] S110: Cluster each sensor data and compare the clustering result with a preset standard clustering mode to perform deviation analysis and anomaly detection on each sensor data to obtain a first verification result.

[0094] Specifically, the clustering result can judge whether the sensor data is similar to the preset standard clustering mode, and if it is not similar to any standard clustering mode, it can be considered that the sensor data is abnormal.

[0095] Further, S110 can include S111-S112:

[0096] S111: clustering each of the sensor data using a probability model based on expectation maximization;

[0097] S112: calculating a similarity measure between the clustering result and the preset standard clustering pattern to perform deviation analysis and anomaly detection on each of the sensor data, to obtain the first verification result.

[0098] In some optional embodiments, the sensor state analysis and verification of the present embodiment can include the following steps:

[0099] Clustering analysis of sensor data and comparison with standard pattern:

[0100] Clustering and dividing the current sensor data using a probability model based on expectation maximization, and comparing and analyzing with the preset standard clustering pattern.

[0101] The present embodiment can maintain a standard clustering pattern library classified by sensor type, and each type of sensor has its corresponding statistical characteristic distribution model.

[0102] By calculating the similarity measure between the clustering result of the current sensor and the standard pattern, deviation analysis and anomaly detection are performed:

[0103] If the clustering result deviates significantly from the standard pattern, it is marked as an abnormal sensor.

[0104] For sensors that have been running stably for a long time and have consistent clustering results, their data can be used as supplementary samples for updating the standard clustering pattern.

[0105] Standard pattern library management: store the respective standard clustering statistical characteristic models according to the sensor type classification.

[0106] Further, the present embodiment can also include S113-S115:

[0107] S113: calculating the sampling interval, mean and variance of each of the sensor data to obtain the sampling characteristics;

[0108] S114: standardizing the sampling characteristics and then mapping each of the sensor data to a normal distribution space with zero mean and unit variance, and then identifying the sensor with statistical characteristic anomaly according to the difference of each of the sensor data from the central data range;

[0109] S115: adding the sensor identified as the statistical characteristic anomaly to the first verification result.

[0110] Specifically, the sampling characteristic analysis can include calculating the sampling interval, mean value and variance of each sensor. By standardizing the sampling characteristics and mapping them to a normal distribution space with zero mean and unit variance, the degree of deviation from the central tendency is used to identify sensors with significantly abnormal statistical characteristics.

[0111] S120: filtering each of the sensor data and comparing each of the sensor data with corresponding historical data to detect potential abnormal patterns, to obtain a second verification result.

[0112] It can be understood that the embodiment can filter out interference data such as abnormal values or noise values in the sensor data, and then compare with the data collected by the same sensor historically to determine whether there is an abnormality.

[0113] Further, S120 can include S121-S122:

[0114] S121: using a recursive filtering algorithm based on Bayesian estimation to dynamically estimate the state of each of the sensor data to filter out high-frequency random noise;

[0115] S122: comparing each of the filtered sensor data with corresponding historical data to detect potential abnormal patterns to obtain the second verification result.

[0116] In some optional embodiments, the multi-dimensional data analysis of the embodiment can include the following steps:

[0117] Data smoothing: using a recursive filtering algorithm based on Bayesian estimation to dynamically estimate the state of the sensor data to filter out high-frequency random noise and improve the authenticity and stability of the signal.

[0118] Self-comparison analysis: comparing the current sensor data with its historical data, establishing a time series model containing a self-feedback mechanism, and deeply mining the time dependence and trend of the data to detect potential abnormal patterns.

[0119] Further, the embodiment can further include the following steps S123-S124:

[0120] S123: evaluating the coupling and collaborative change degree of the sensor data between different sensors on the same injection molding machine to obtain a data correlation degree;

[0121] S124: issuing an abnormal early warning information for the sensor with a data correlation degree lower than a correlation threshold, prompting a functional failure or data abnormality.

[0122] Specifically, the inter-sensor cross analysis can include: evaluating the data coupling and the degree of coordinated change between different sensors on the same injection molding machine. For sensors with a correlation degree below a threshold, an abnormality warning is issued, suggesting that there may be a functional failure or data anomaly.

[0123] S130: determining a time offset between each of the sensor data, time synchronizing each of the sensor data according to the time offset, and performing error analysis on each of the sensor data to identify the sensor with a reading anomaly, to obtain a third verification result.

[0124] In some optional embodiments, the time synchronization and fault diagnosis of the present embodiment can include the following steps:

[0125] Time consistency analysis: based on data characteristics, an optimal reference sensor is automatically selected, a nonlinear sequence comparison algorithm for calculating the minimum matching cost between time sequences is used, and the time sequences of each sensor are analyzed in depth to accurately calculate the time offset and delay between sensors.

[0126] Data alignment and error analysis: according to the calculated time offset, all sensor data are accurately aligned on the time axis and compared under a unified time reference system. Then, the differences between the aligned sensor data are calculated, and the sensors with possible reading anomalies are identified using an outlier detection method in statistics.

[0127] S140: generating a comprehensive evaluation report and pushing an early warning information according to the state data of each of the sensors, the first verification result, the second verification result, and the third verification result.

[0128] Specifically, the present embodiment can generate comprehensive evaluation reports in different forms and visually display them, and when an anomaly is detected, it can push early warning information to maintenance personnel in various forms.

[0129] Further, S140 can include S141-S144:

[0130] S141: evaluating the statistical significance and confidence interval of each of the sensor data according to the state data of each of the sensors, the first verification result, the second verification result, and the third verification result, and further determining an anomaly detection result and generating a corresponding processing suggestion;

[0131] S142: generating a data trend chart, a complex network correlation analysis chart, and a visual marker chart based on outlier detection according to the state data of each of the sensors, the first verification result, the second verification result, and the third verification result;

[0132] S143: generating the comprehensive evaluation report according to the abnormality detection result, the processing suggestion, the data trend chart, the complex network correlation analysis graph and the visualization marking graph based on the abnormal point detection;

[0133] S144: when detecting that each of the sensor data triggers an abnormal mode or an index breaks a set confidence interval, triggering a pre-warning mechanism and pushing the pre-warning information to a maintenance personnel.

[0134] In some optional embodiments, the report generation and pre-warning of the present embodiment can include the following steps:

[0135] Automatic report generation:

[0136] A comprehensive list of sensor states is constructed, covering multi-dimensional information such as sampling frequency, parameter configuration and running situation.

[0137] The data analysis results are comprehensively summarized, including multivariate statistical comparison analysis, time series synchronicity evaluation and collaborative correlation research across sensor data.

[0138] A comprehensive evaluation report of data quality is provided, which details the abnormality detection results and corresponding processing suggestions, and evaluates the data reliability based on statistical significance and confidence interval.

[0139] Data trend charts, complex network correlation analysis graphs and visualization marking graphs based on abnormal point detection are automatically generated, supporting multi-scale data mining and visual analysis.

[0140] Intelligent pre-warning mechanism:

[0141] A hierarchical and progressive pre-warning threshold system is set, and the differential and higher order derivative changes of key indicators are used to monitor potential nonlinear dynamic behavior in real time.

[0142] When an abnormal mode or an index breaks a set confidence interval, an advanced pre-warning mechanism is automatically triggered, supporting adaptive alarm level adjustment.

[0143] Through multiple channels such as short message, email, real-time push, etc., the maintenance personnel can be reliably and instantly notified, and detailed abnormality description and suggested coping strategies are provided.

[0144] The historical data of pre-warning events are recorded and persistently stored, supporting time series-based fault tracing, pattern recognition and root cause analysis, and continuously optimizing the pre-warning model by using machine learning algorithms.

[0145] In summary, the embodiment provides a method for verifying sensor data of an injection molding machine and judging data normality, which can comprehensively verify, analyze and judge the sensor data, ensure the reliability of the data, timely find sensor abnormalities, reduce fault false reports, and improve the safety and reliability of the operation of the injection molding machine.

[0146] Next, the scheme of the embodiment of the application will be described and explained in detail in combination with specific application examples.

[0147] Specifically, the embodiment can include the following schemes:

[0148] Phase one: intelligent data acquisition and preprocessing.

[0149] 1. Real-time monitoring of the state of the injection molding machine:

[0150] An embedded end-side system is installed on the injection molding machine, which is configured with a high-performance processor (Jetson) and a WiFi communication module. The running parameters of the injection molding machine are monitored in real time, including motor speed, hydraulic pressure, and mold temperature. Whether the injection molding machine is in an idle state is judged by a pre-set threshold value, for example, when the motor speed is lower than 100 RPM, the hydraulic pressure is lower than 5 MPa, and the mold temperature is stable, the injection molding machine is considered to be in an idle state. Only in the idle state of the injection molding machine, the end-side system starts the collection of sensor data, avoiding the influence of mechanical vibration and electromagnetic interference on the data in the running state. The collected data includes various physical quantities, including temperature, pressure, humidity, displacement, current, and voltage, which are obtained in real time by sensors installed at various parts of the injection molding machine. The end-side system transmits the collected data to the cloud processing system in real time through Wi-Fi, such as Wi-Fi.

[0151] 2. Data format verification and standardization:

[0152] After the cloud processing system receives the sensor data, it first checks the integrity of the necessary fields of the data:

[0153] Time stamp: verify whether the time stamp exists and the format conforms to the ISO 8601 standard, such as "2024-10-15T08:00:00Z".

[0154] Device unique identifier: check whether the device ID or MAC address is valid and compare it with the registered device list.

[0155] Variable data: confirm whether the data of all key sensors exists and the data format is correct.

[0156] Convert all time markers to Unix timestamps (in milliseconds) to establish a unified time reference system for subsequent data processing and analysis.

[0157] Type checking for variable data to ensure it is a processable numerical type. If an abnormal data type is found, try to convert it; if it cannot be converted, record an exception and exclude it from the dataset.

[0158] 3. Missing value judgment:

[0159] Calculate the proportion of missing data for each variable, the formula is:

[0160] Missing rate = total data points / missing data points × 100%;

[0161] For variables with low missing rate (≤3%), use least squares method to fit a quintic polynomial, and use the fitted curve to predict the missing values. For variables with high missing rate (>3%), mark and exclude from subsequent analysis, and generate warning information, suggesting that there may be sensor failure or communication problems.

[0162] Phase two: Sensor state analysis and verification.

[0163] 4. Cluster analysis and standard mode comparison of sensor data:

[0164] Use Gaussian Mixture Model (GMM) based on Expectation Maximization (EM) algorithm to cluster and divide the current sensor data:

[0165] Data preparation: Standardize the sensor data to eliminate the influence of dimension.

[0166] Model training: Estimate the parameters of GMM using EM algorithm, including the mean, covariance matrix and mixing coefficient of each Gaussian distribution.

[0167] Cluster division: According to the trained model, map the data points to the corresponding cluster.

[0168] Compare the clustering results with the models in the standard clustering mode library. The standard mode library is classified by sensor type and stores the respective statistical feature distribution model.

[0169] Deviation analysis and anomaly detection:

[0170] Calculate the Kullback-Leibler divergence (used as a similarity measure) between the current clustering result and the standard mode.

[0171] If the similarity measure exceeds the set threshold (such as 0.1), it is marked as an abnormal sensor.

[0172] For sensors that have been running stably for a long time and have consistent clustering results, their data can be used as supplementary samples for the standard mode, and online learning algorithms can be used to update the standard mode library, improving the accuracy of the model.

[0173] 5. Standard pattern library management:

[0174] Data storage: The standard pattern library uses a MySQL database for management, storing information including sensor types, models, clustering model parameters, etc.

[0175] Model update: The model is updated regularly every week based on new data, using a Bayesian update mechanism to balance the influence of new and old data.

[0176] 6. Sampling characteristic analysis:

[0177] Calculate the time difference between the two adjacent data, get the sampling interval sequence, analyze its stability. Statistics of sensor data mean and variance, get the distribution characteristics of data. Finally, the sampling characteristics are standardized, mapped to the normal distribution space of zero mean and unit variance: identify the statistical characteristics of the sensor that is significantly abnormal, if the standardized value exceeds ±3, mark it as abnormal, prompt further inspection.

[0178] Phase three: Multidimensional data analysis.

[0179] 7. Data smoothing processing:

[0180] Use Kalman filter algorithm based on Bayesian estimation to perform dynamic state estimation on sensor data. Define the state transition equation and observation equation of the system, assume that the process noise and observation noise are Gaussian distributed. For each time point, perform prediction and update steps to get the optimal state estimation. Finally, get smooth sensor data, effectively filter out high-frequency random noise.

[0181] 8. Self-comparison analysis:

[0182] Based on time series model, compare the current data of sensor with historical data. Use ARIMA model to train the time series model of sensor data. And predict the future data trend, compare with the actual data, calculate the prediction error. When the prediction error exceeds the threshold, identify the potential abnormal pattern, prompt possible failure.

[0183] 9. Cross analysis between sensors:

[0184] Evaluate the data coupling and collaborative change between different sensors on the same injection molding machine. Calculate the Pearson correlation coefficient or mutual information between sensors to judge the correlation between variables. Set the correlation threshold (such as 0.6), sensors below the threshold are marked as abnormal. For sensors with low correlation, issue an abnormal warning, indicating possible performance degradation or failure.

[0185] Phase four: Time synchronization and fault diagnosis.

[0186] 10. Time consistency analysis:

[0187] Using Dynamic Time Warping (DTW) algorithm, analyze the time series of each sensor, calculate the time offset and delay. Automatically select the sensor with stable data and low noise as the reference (such as the mold temperature sensor in this example). Calculate the minimum matching path between the time series of the reference sensor and other sensors to get the time offset. If the time offset exceeds the set threshold (such as 1 second), consider that there is a time synchronization problem.

[0188] 11. Data alignment and error analysis:

[0189] According to the calculated time offset, align the data of all sensors in time to ensure comparison under a unified time reference system. Calculate the difference between the aligned sensor data, use or 3σ principle to identify abnormal readings. Finally, use the Isolation Forest algorithm learning method to detect sensors that may have abnormal readings.

[0190] Phase five: intelligent report generation and early warning push.

[0191] 12. Automatic report generation, including:

[0192] 12.1 Basic information:

[0193] Sampling frequency: average sampling rate, standard deviation of sampling interval, sampling stability evaluation.

[0194] Parameter configuration: sensor model, range, accuracy, calibration date, etc.

[0195] Running situation: online status, data integrity, abnormal record.

[0196] 12.2 Comprehensive summary of data analysis results:

[0197] Multivariate statistical comparison analysis: provide a comparison table of statistical characteristics of each sensor data.

[0198] Time series synchronization evaluation: show the distribution graph of time offset.

[0199] Cross-sensor data correlation study: generate correlation matrix heat map.

[0200] 12.3 Provide a comprehensive evaluation report on data quality:

[0201] Abnormal detection results: list abnormal sensors, abnormal types, and occurrence times in detail.

[0202] Handling suggestions: provide possible reasons and recommended solutions for each abnormality.

[0203] Data reliability evaluation: evaluate the reliability of data based on confidence interval.

[0204] 12.4 Automatically generate other visualization charts:

[0205] Data trend chart: Time series chart of sensor data.

[0206] Correlation analysis map: Correlation network diagram between sensors.

[0207] Abnormal point marking chart: Highlight the location and severity of abnormal data points.

[0208] Support user-defined report content and format, provide PDF, HTML and other formats for export.

[0209] 13. Intelligent early warning mechanism:

[0210] Set up a hierarchical and progressive early warning threshold system to monitor key indicators in real time:

[0211] First level warning (prompt): The indicator is close to the threshold, suggesting attention.

[0212] Second level warning (warning): The indicator exceeds the threshold, measures need to be taken.

[0213] Third level warning (serious): The indicator deviates seriously, which may cause equipment failure and needs to be handled immediately.

[0214] Use the differential and higher order derivative changes of key indicators to monitor the mutation and trend of data. Calculate the first order derivative (change rate) and second order derivative (acceleration) of the key indicators. Identify the nonlinear changes of the indicators through differentiation and higher order derivatives to predict potential risks.

[0215] When abnormal patterns or indicators break through the set confidence interval are detected, the advanced early warning mechanism is automatically triggered. According to the degree of abnormality, the warning level is dynamically adjusted. Through SMS, email, real-time push and other ways, immediately notify the relevant personnel. And provide detailed information including abnormal description, impact range, suggested coping strategies, etc.

[0216] Store the warning events in the database for easy query and analysis. Based on time series, trace back the historical warning data to analyze the failure patterns. Use support vector machine (SVM) to train the warning data to optimize the warning model. Finally, combine with association rule mining to find out the root cause of the abnormality.

[0217] The technical scheme of the embodiment at least includes the following beneficial effects:

[0218] The embodiment is directed to the monitoring and diagnosis of injection molding machine sensors, and proposes a self-adaptive, end-to-end automated solution with continuous learning capability. The embodiment significantly improves the efficiency and accuracy of injection molding machine sensor monitoring by supporting any type of sensor, automatically completing monitoring and reporting, continuously updating the model library, and using advanced anomaly detection algorithms.

[0219] The embodiment has excellent adaptability and can support a variety of sensor types: The embodiment can automatically identify and diagnose different models and types of injection molding machine sensors without pre-inputting sensor parameters. By maintaining a standard cluster pattern library classified by sensor type, the embodiment can adaptively analyze data from new sensors and automatically diagnose unknown or new sensors. The adaptability of the embodiment greatly improves flexibility and is suitable for monitoring various injection molding machine sensors, reducing the cumbersome process of manual configuration.

[0220] The embodiment realizes end-to-end automated monitoring and reporting, simplifying the management process of sensors: When the injection molding machine is in an idle state, data collection is automatically started to ensure data quality. The collected data is uploaded to the cloud in real time through a wireless network, and data format verification, preprocessing, analysis, and report generation are automatically completed. When an anomaly is detected, the embodiment can automatically generate a report containing data analysis results and processing suggestions, and notify maintenance personnel immediately through SMS, email, etc. This full-process automation not only improves work efficiency, but also ensures that abnormal problems can be discovered and handled in a timely manner, ensuring the normal operation of the injection molding machine.

[0221] The embodiment has the ability of continuous learning and pattern updating, improving the accuracy and intelligence level of diagnosis: For sensors that have been running normally and stably for a long time, the embodiment can include their data in the standard cluster pattern library, constantly enrich and update the model, and then the embodiment can optimize itself according to historical data and new trends, adapt to changes in the injection molding machine operating environment and device state, improve the accuracy of diagnosis, and reduce false positives and false negatives.

[0222] Reference Figure 2 The embodiment of the application also provides a system for verifying injection molding machine sensor data and judging data normality, which can implement the method for verifying injection molding machine sensor data and judging data normality described above. The system comprises:

[0223] A data acquisition unit is configured to acquire data of an injection molding machine in an idle state through a plurality of sensors as different sensor data.

[0224] The first verification unit is configured to cluster each of the sensor data, compare the clustering result with a preset standard clustering mode, perform deviation analysis and anomaly detection on each of the sensor data, and obtain a first verification result.

[0225] The second verification unit is configured to filter each of the sensor data, compare each of the sensor data with corresponding historical data, detect a potential abnormal pattern, and obtain a second verification result.

[0226] The third verification unit is configured to determine a time offset between each of the sensor data, perform time synchronization on each of the sensor data according to the time offset, perform error analysis on each of the sensor data to identify the sensor with a reading abnormality, and obtain a third verification result.

[0227] The evaluation and early warning unit is configured to generate a comprehensive evaluation report and push early warning information according to the state data of each of the sensors, the first verification result, the second verification result, and the third verification result.

[0228] It can be understood that the content in the above method embodiments is applicable to the present system embodiments, the present system embodiments specifically implement the functions of the above method embodiments, and achieve the same beneficial effects as the above method embodiments.

[0229] The present application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor implements the above method for verifying sensor data of an injection molding machine and judging data normality when executing the computer program. The electronic device can be any intelligent terminal, such as a tablet computer or a vehicle-mounted computer.

[0230] It can be understood that the content in the above method embodiments is applicable to the present device embodiments, the present device embodiments specifically implement the functions of the above method embodiments, and achieve the same beneficial effects as the above method embodiments.

[0231] Please refer to Figure 3 , Figure 3 The hardware structure of the electronic device of another embodiment is illustrated, which includes:

[0232] The processor 301 can be implemented in a general-purpose CPU (Central Processing Unit), a microprocessor, an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits, and is configured to execute a related program to implement the technical solutions provided by the present application.

[0233] The memory 302 can be implemented in the form of a Read Only Memory (ROM), a static storage device, a dynamic storage device, or a Random Access Memory (RAM), etc. The memory 302 can store an operating system and other application programs, and when the technical solutions provided by the embodiments of the present specification are implemented by software or firmware, the related program codes are stored in the memory 302 and are called and executed by the processor 301 to perform the method for verifying sensor data of an injection molding machine and judging normality of data according to the embodiments of the present application;

[0234] The input / output interface 303 is configured to realize information input and output.

[0235] The communication interface 304 is configured to realize communication interaction between the device and other devices, and the communication can be realized by a wired manner (for example, a USB, a network cable, etc.) or a wireless manner (for example, a mobile network, WIFI, Bluetooth, etc.).

[0236] The bus 305 is configured to transmit information between various components (for example, the processor 301, the memory 302, the input / output interface 303, and the communication interface 304) of the device.

[0237] The processor 301, the memory 302, the input / output interface 303, and the communication interface 304 are connected to each other through the bus 305 to realize communication connection between the device.

[0238] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the above method for verifying sensor data of an injection molding machine and judging normality of data.

[0239] It can be understood that the contents in the above method embodiments are applicable to the present storage medium embodiments. The present storage medium embodiments specifically implement the same functions as the above method embodiments, and achieve the same beneficial effects as the above method embodiments.

[0240] The memory is a non-transitory computer readable storage medium, which can be used to store non-transitory software programs and non-transitory computer executable programs. In addition, the memory can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state memory device. In some embodiments, the memory can optionally include a memory remotely arranged relative to the processor, and these remote memories can be connected to the processor through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0241] The embodiments described in the present application are for more clearly illustrating the technical solutions of the present application, and do not constitute a limitation on the technical solutions provided by the present application. Those skilled in the art can know that, with the evolution of technology and the appearance of new application scenarios, the technical solutions provided by the present application are also applicable to similar technical problems.

[0242] Those skilled in the art can understand that the technical solutions shown in the figures do not constitute a limitation on the present application, and can include more or fewer steps than shown in the figures, or combine certain steps, or different steps.

[0243] The system embodiments described above are merely illustrative, and the units described as separate components can or can not be physically separated, that is, can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the present embodiment.

[0244] Those skilled in the art can understand that all or some of the steps in the above disclosed method, the functional modules / units in the system, and the device can be implemented as software, firmware, hardware, and appropriate combinations thereof.

[0245] The terms "first", "second", "third", "fourth" and the like (if any) in the specification of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0246] It should be understood that, in the application, "at least one" refers to one or more, and "multiple" refers to two or more. "And / or" is used to describe the association relationship of the associated objects, which means that there can be three relationships, for example, "A and / or B" can represent three cases of only A, only B, and A and B existing at the same time, wherein A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects before and after it. "At least one of the following" or similar expressions means any combination of these items, including any combination of single or multiple items. For example, at least one of a, b or c can represent a, b, c, "a and b", "a and c", "b and c", or "a and b and c", wherein a, b, and c can be single or multiple.

[0247] In several embodiments provided in the application, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiments described above are only illustrative, for example, the division of the above units is only a logical function division, and actual implementation can have another division manner, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed units can be indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0248] The units described above as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple network units. According to actual needs, part or all of the units can be selected to achieve the purpose of the embodiment scheme.

[0249] In addition, the functional units in each embodiment of the application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0250] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in other words, the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes multiple instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program storage media.

[0251] The preferred embodiments of the embodiments of the present application are described above with reference to the accompanying drawings, and are not limited to the scope of the embodiments of the present application. Any modifications, equivalent replacements and improvements made by those skilled in the art without departing from the scope and essence of the embodiments of the present application shall be within the scope of the embodiments of the present application.

Claims

1. A method of verifying sensor data of an injection molding machine and judging data normality, characterized in that, The method comprises the following steps: obtaining data of an injection molding machine in an idle state by a plurality of sensors as different sensor data respectively; performing clustering on each of the sensor data and comparing the clustering results with a preset standard clustering mode to perform deviation analysis and anomaly detection on each of the sensor data, and obtaining a first verification result; filtering each of the sensor data and comparing each of the sensor data with corresponding historical data to detect potential abnormal patterns, and obtaining a second verification result; determining the time offset between each of the sensor data, performing time synchronization on each of the sensor data according to the time offset, and performing error analysis on each of the sensor data to identify the sensor with a reading abnormality, and obtaining a third verification result; generating a comprehensive evaluation report and pushing an early warning information according to the state data of each of the sensors, the first verification result, the second verification result and the third verification result; The generation of the comprehensive evaluation report and the pushing of the early warning information according to the state data of each of the sensors, the first verification result, the second verification result and the third verification result comprises the following steps: evaluating the statistical significance and confidence interval of each of the sensor data according to the state data of each of the sensors, the first verification result, the second verification result and the third verification result, and further determining an anomaly detection result and generating a corresponding processing suggestion; wherein the state data includes sampling frequency, parameter configuration and running situation; generating a data trend chart, a complex network correlation analysis graph and a visualization marker graph based on anomaly point detection according to the state data of each of the sensors, the first verification result, the second verification result and the third verification result; generating the comprehensive evaluation report according to the anomaly detection result, the processing suggestion, the data trend chart, the complex network correlation analysis graph and the visualization marker graph based on anomaly point detection; When it is detected that each of the sensor data triggers an abnormal pattern or an index breaks through a set confidence interval, triggering an early warning mechanism and pushing the early warning information to a maintenance personnel.

2. The method of validating injection molding machine sensor data and determining data normality of claim 1, wherein, The obtaining of data of an injection molding machine in an idle state by a plurality of sensors as different sensor data respectively comprises the following steps: obtaining data of the injection molding machine in an idle state by a plurality of the sensors as initial data; performing field integrity check on the initial data, uniformly converting all time markers in the initial data into Unix time format, and converting the type of the initial data into a target data type to obtain standardized data; calculating the data missing rate of each variable in the standardized data, using a high-order polynomial curve fitting method to complete data for the variable with a data missing rate lower than a first missing threshold, marking and removing the variable with a data missing rate higher than a second missing threshold, and recording relevant warning information to obtain the sensor data.

3. The method of validating injection molding machine sensor data and determining data normality of claim 1, wherein, The sensor data is clustered, and the clustering result is compared with a preset standard clustering mode to perform deviation analysis and anomaly detection on the sensor data, and a first verification result is obtained. The sensor data is clustered by using a probability model based on expectation maximization. The similarity between the clustering result and the preset standard clustering mode is calculated to perform deviation analysis and anomaly detection on the sensor data, and the first verification result is obtained.

4. The method of validating injection molding machine sensor data and determining data normality of claim 3, wherein, The method further includes the following steps: The sampling interval, mean value and variance of each sensor data are calculated to obtain sampling characteristics. The sampling characteristics are standardized, and each sensor data is mapped to a normal distribution space with zero mean and unit variance. Then, the sensors with statistical feature anomalies are identified according to the difference of each sensor data from the central data range. The sensors identified as having statistical feature anomalies are added to the first verification result.

5. The method of validating injection molding machine sensor data and determining data normality of claim 1, wherein, The sensor data is filtered, and each sensor data is compared with corresponding historical data to detect potential abnormal patterns, and a second verification result is obtained. A recursive filtering algorithm based on Bayesian estimation is used to dynamically estimate the state of each sensor data to filter out high-frequency random noise. The filtered sensor data is compared with the corresponding historical data to detect potential abnormal patterns, and the second verification result is obtained.

6. The method of validating injection molding machine sensor data and determining data normality of claim 5, wherein, The method further includes the following steps: The coupling and collaborative change degree of the sensor data between different sensors on the same injection molding machine are evaluated to obtain a data correlation degree. If the data correlation degree of a sensor is lower than a correlation threshold, an abnormal early warning information is sent out to prompt a functional failure or data anomaly.

7. A system for verifying sensor data of an injection molding machine and judging normality of the data, characterized in that The system includes: A data acquisition unit configured to acquire data of an injection molding machine in an idle state by multiple sensors as different sensor data; A first verification unit configured to cluster each sensor data and compare the clustering result with a preset standard clustering mode to perform deviation analysis and anomaly detection on each sensor data, and obtain a first verification result; A second verification unit configured to filter each sensor data and compare each sensor data with corresponding historical data to detect potential abnormal patterns, and obtain a second verification result; A third verification unit configured to determine the time offset between each sensor data, perform time synchronization on each sensor data according to the time offset, and perform error analysis on each sensor data to identify sensors with reading anomalies, and obtain a third verification result; An evaluation and early warning unit configured to generate a comprehensive evaluation report and push early warning information according to the state data of each sensor, the first verification result, the second verification result and the third verification result. The generating of the comprehensive evaluation report and the pushing of the early warning information according to the state data of each sensor, the first verification result, the second verification result and the third verification result comprises the following steps: According to the state data of each sensor, the first verification result, the second verification result and the third verification result, the statistical significance and the confidence interval of each sensor data are evaluated, and then the abnormal detection result is determined and the corresponding processing suggestion is generated; wherein the state data comprises sampling frequency, parameter configuration and running situation; According to the state data of each sensor, the first verification result, the second verification result and the third verification result, a data trend chart, a complex network correlation analysis atlas and a visual marking chart based on abnormal point detection are generated; According to the abnormal detection result, the processing suggestion, the data trend chart, the complex network correlation analysis atlas and the visual marking chart based on abnormal point detection, the comprehensive evaluation report is generated; When it is detected that each sensor data triggers an abnormal mode or an index breaks through a set confidence interval, an early warning mechanism is triggered and the early warning information is pushed to a maintenance personnel.

8. An electronic device, comprising: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method of any one of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 8. The computer program is executed by the processor to implement the method of any one of claims 1 to 6.

Images