Equipment state monitoring method and system based on automatic working condition calibration and multi-dimensional signal analysis

Abstract

The invention relates to the technical field of industrial equipment state monitoring, in particular to an equipment state monitoring method and system based on automatic working condition calibration and multi-dimensional signal analysis. The full-link automatic closed-loop monitoring of the equipment operation state from calibration to diagnosis to abnormal response is realized, the real-time performance, accuracy and intelligent level of the equipment state monitoring process are remarkably improved, the problems of inaccurate working condition identification, single feature analysis dimension and insensitive abnormal state identification in the prior art are solved, and the method is suitable for popularization and application. The operation reliability of the equipment is practically improved, and the fault risk and the maintenance cost are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial equipment status monitoring, and in particular to an equipment status monitoring method and system based on automatic working condition calibration and multi-dimensional signal analysis. Background Art

[0002] With the continuous improvement of the intelligence level of industrial equipment, the traditional equipment status monitoring model that relies on manual inspections and experience-based judgment has gradually shown its shortcomings and limitations because it is difficult to meet the current industrial field's comprehensive needs for real-time, accurate and intelligent equipment status monitoring. Especially in the operation scenarios of rotating mechanical equipment such as pumps, compressors, and fans, due to the changeable and complex equipment operating status and the diversity of the characteristic dimensions that need to be monitored, higher requirements are placed on the analysis and diagnosis of the equipment health status.

[0003] However, in practical applications, current equipment condition monitoring technology still widely relies on a single type of signal (such as vibration signal) for status judgment. It lacks a dynamic identification and real-time calibration mechanism for equipment operating conditions, making it difficult to comprehensively reflect the actual operating status of the equipment from multiple dimensions, and easily causing lags and misjudgments in equipment status identification. Therefore, it is urgent to develop an intelligent monitoring technology that can integrate automatic condition calibration, standardized signal acquisition and multi-dimensional data analysis algorithms to achieve a more accurate and intelligent comprehensive evaluation of the operating status of industrial equipment.

[0004] In the existing technology, CN112613646A discloses a device status prediction method based on multi-dimensional data fusion, which performs noise reduction, wavelet packet analysis and feature extraction on the status signals within the life cycle of the equipment, and completes model training and device status prediction in a cloud-edge collaborative manner; while CN119249345A discloses an online monitoring system, which establishes an online monitoring platform through the collection and analysis of temperature, vibration and pressure signals to realize real-time diagnosis of equipment status.

[0005] However, the above-mentioned existing technical solutions have the following shortcomings: First, there is a lack of automatic real-time calibration of the equipment operating conditions, resulting in a lack of a unified reference standard for operating condition identification, which reduces the ability to accurately identify different operating states; second, data analysis is mostly macro-holistic analysis, and no refined modeling and fusion of multi-dimensional signal features are performed, which limits the sensitivity and accuracy of abnormal state identification; third, the abnormality judgment process mostly relies on preset fixed thresholds for evaluation, ignoring the dynamic feature drift caused by real-time operating condition changes, reducing the accuracy and stability of state monitoring. Summary of the Invention

[0006] The purpose of this section is to summarize some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract of the specification and the title of the invention of this application to avoid blurring the purpose of this section, the abstract of the specification and the title of the invention, and such simplifications or omissions cannot be used to limit the scope of the invention.

[0007] In view of the above existing problems, the present invention is proposed.

[0008] Therefore, the technical problem solved by the present invention is that the existing technology lacks an automatic calibration and dynamic identification mechanism for equipment working conditions, and the accuracy of multi-dimensional feature analysis is not high.

[0009] To solve the above technical problems, the present invention provides the following technical solutions: collecting operating parameters during the equipment installation phase, and establishing an operating condition reference vector based on outlier elimination and mean calculation; During the equipment operation phase, real-time operating parameters are obtained and compared with the operating condition reference vector to generate a comprehensive deviation metric, and the current operating condition label is determined through a continuity determination strategy; Based on the working condition label, the corresponding signal analysis model is called to extract the spectrum characteristics, the axis trajectory characteristics and the numerical characteristics of the operating parameters; Compare the acquired multi-dimensional features with the preset threshold to determine whether the device status is abnormal and generate alarm information; The operating condition labels, characteristic curves, abnormal categories and alarm records are output in a visual manner, and question-answering interaction based on a large language model is supported.

[0010] As a preferred solution of the equipment condition monitoring system based on automatic working condition calibration and multi-dimensional signal analysis described in the present invention, the monitoring system includes a multi-source sensor module, a data acquisition unit, an analysis and storage unit, and a display and interaction unit, wherein: The multi-source sensor module is used to collect vibration, sound, temperature and electrical parameter signals; The data acquisition unit is used to condition and digitize the collected signals; The analysis and storage unit is used to perform working condition calibration, operating state identification and multi-dimensional feature analysis, and perform abnormality determination, save operating data and support trend analysis, playback and edge computing; The display interaction unit is used to visually display the monitoring results and realize question-answering interaction and remote data communication in combination with a large language model.

[0011] As a preferred solution of the equipment condition monitoring system based on automatic working condition calibration and multi-dimensional signal analysis of the present invention, it also includes one or more processors; A memory stores operable instructions, which, when executed by the one or more processors, cause the one or more processors to perform operations, including the process of the aforementioned equipment status monitoring method based on automatic working condition calibration and multi-dimensional signal analysis.

[0012] As a preferred embodiment of a computer-readable medium for storing software described in the present invention, the software includes instructions that can be executed by one or more computers, and the instructions enable the one or more computers to perform operations through such execution, and the operations include the process of the equipment status monitoring method based on automatic working condition calibration and multi-dimensional signal analysis as described above.

[0013] Beneficial effects of the present invention: The present invention comprehensively improves the real-time, accuracy and intelligence level of equipment monitoring through the combination of automatic working condition calibration, real-time working condition dynamic identification, multi-dimensional state feature precise analysis and intelligent health assessment, and effectively makes up for the problems of insufficient single-dimensional analysis, lack of unified working condition calibration mechanism and insufficient dynamic identification ability of abnormal state in the existing technology. It can be widely used in the field of rotating mechanical equipment such as pumps, compressors, fans, etc., ensuring stable and reliable operation of equipment and reducing operation and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them: Figure 1 Schematic diagram of the process of the equipment status monitoring method based on automatic working condition calibration and multi-dimensional signal analysis shown in the present invention; Figure 2 Schematic diagram of the module unit structure of the equipment status monitoring system based on automatic working condition calibration and multi-dimensional signal analysis shown in the present invention; Figure 3 This is a working diagram of the signal processing flow module of the equipment status monitoring system based on automatic working condition calibration and multi-dimensional signal analysis shown in the present invention. DETAILED DESCRIPTION

[0015] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0016] Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without making any creative work should fall within the scope of protection of the present invention.

[0017] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0018] According to an embodiment of the present invention, Figure 1 The flowchart shown is a device condition monitoring method based on automatic working condition calibration and multi-dimensional signal analysis, which specifically includes the following steps: S1. Collect operating parameters during the equipment installation phase and establish a working condition benchmark vector based on outlier elimination and mean calculation; S2. During the equipment operation phase, real-time operating parameters are obtained and compared with the operating condition reference vector to generate a comprehensive deviation metric, and the current operating condition label is determined through a continuity determination strategy; S3. Based on the working condition label, the corresponding signal analysis model is called to extract the spectrum characteristics, axis trajectory characteristics and operating parameter numerical characteristics; S4. Compare the acquired multi-dimensional features with the preset threshold value to determine whether the device status is abnormal and generate an alarm message; S5. Output working condition labels, characteristic curves, abnormal categories and alarm records in a visual manner, and support question-answering interaction based on a large language model.

[0019] Through step S1, a stable parameter sample sequence is obtained, and then the mean of each operating parameter is accurately calculated, and finally a reliable operating condition reference vector is formed. This step ensures the accuracy and stability of the equipment's personalized reference model, provides a unified and accurate reference benchmark for subsequent operating condition identification and status analysis, and fundamentally improves the accuracy of the entire status monitoring process.

[0020] Through step S2, high-precision real-time monitoring of the equipment operating status is achieved, and the working condition switching is identified quickly and accurately, which greatly improves the timeliness and accuracy of equipment status monitoring and avoids the risks brought about by false alarms or late alarms of working condition switching under traditional methods.

[0021] Step S3 realizes the multi-dimensional feature extraction of the equipment status, ensuring the refinement and pertinence of signal processing under different working conditions, effectively avoiding the risk of missed diagnosis or misjudgment in single feature dimension analysis, and significantly enhancing the comprehensiveness and accuracy of equipment status diagnosis.

[0022] Step S4 enables rapid location and accurate diagnosis of abnormal equipment conditions, significantly improving the early warning capability of equipment failures, providing maintenance personnel with timely and clear maintenance guidance, and avoiding the risk of false alarms or missed alarms caused by the single threshold setting of traditional condition monitoring solutions.

[0023] Step S5 enables efficient interaction between equipment status information and operation and maintenance personnel, intuitively and clearly displaying equipment status and change trends, greatly improving the on-site response efficiency and decision-making ability of operation and maintenance personnel, and ultimately significantly reducing the complexity of equipment maintenance and the difficulty of personnel operation.

[0024] In summary, the present invention realizes full-link automated closed-loop monitoring of the equipment operating status from calibration to diagnosis to abnormal response through the combination of automatic operating condition calibration, multi-dimensional signal precise analysis, dynamic abnormality diagnosis and intelligent interactive feedback technology, significantly improving the real-time, accuracy and intelligence level of the equipment status monitoring process, solving the problems of inaccurate operating condition identification, single feature analysis dimension and insensitive abnormal state identification in the prior art, effectively improving the equipment operation reliability, and reducing the risk of failure and maintenance costs.

[0025] The following describes in more detail the implementation process and / or effects of certain embodiments of the present invention in conjunction with some preferred or optional examples of the present invention.

[0026] Working condition calibration: To accurately identify the operating status of industrial equipment, the operating condition benchmark value must be calibrated during the equipment installation process. Depending on the installation environment, operating condition calibration can be divided into two types: Operation process calibration (Work Mode): collect data during the online operation of the device; Idle Mode calibration: collect data when the equipment is stopped or in no-load state; The calibration results are used to build a personalized reference state model for the equipment and serve as a benchmark for subsequent operating condition identification and anomaly analysis. The specific process is as follows: ① Selection of working condition parameters Select representative equipment operating parameters to form the operating condition vector, which is expressed as:

[0027] in, is the current at time t, is the speed at time t, is the power at time t; This vector is used to describe the load characteristics of the equipment's operating status; ②Calibration mode distinction Set calibration mode parameters according to the installation scenario , choose one of the two installation modes. The values ​​include: Idle: Stop installation calibration. In this mode, the calibration equipment is in an unloaded state and the parameter reference value is low. Work: Run installation calibration. In this mode, the equipment operating status is calibrated. It is a load-carrying operating condition with a high parameter reference value level. ③ Working condition parameter preprocessing In the set calibration time window This period is the calibration process, collecting key operating condition parameter sequences, including current , speed ,power , construct the working condition vector:

[0028] In order to improve the robustness of the working condition benchmark value, an outlier elimination mechanism based on the interquartile range (IQR) is introduced. The processing steps are as follows: For any working parameter , calculate the upper and lower quartiles of a set of samples: First quartile (25%): ; Third quartile (75%): ; Define IQR as: ; Set the upper and lower limits to:

[0029] in, It is an adjustable parameter with a value range of 1.5 to 3. The default setting is ; All dissatisfaction The samples with are considered as outliers and are removed; This method is applied to 、 、 In three dimensions, each operating condition parameter is independently filtered for outliers, and the removed operating condition data is then used for subsequent benchmark value calculations to ensure that the reference model has higher stability and noise resistance; ④ Calculation of working condition benchmark value For the pre-processed operating parameters, calculate their average value within the specified time interval as the reference benchmark for this installation mode:

[0030] Thus forming a complete working condition reference vector:

[0031] in, It is The current value of the sampling point, It is The speed value of the sampling point, It is The power value of the sampling point, is the total number of sampling points, 、 、 are the mean values ​​of current, speed and power, respectively, which are used to form the working condition reference vector ; For different installation calibration processes, save the corresponding working condition reference value vector under idle condition or working condition , used for subsequent working condition identification.

[0032] Working condition identification: To achieve intelligent perception and trend analysis of equipment operating status, the embodiment of the present invention breaks down the operating condition identification process into two steps: Deviation identification from the benchmark operating condition: Determine whether the current state deviates from the historical calibration operating condition; Continuous identification of working condition changes: avoid misjudgment of instantaneous fluctuations and capture trend changes; It should be noted that the identification process is based on a unique reference model and adopts a fixed working mode (idle or work only) to ensure the consistency and rigor of the identification; ① Identification of deviations from baseline conditions Load the benchmark model in the specified mode at startup, such as idle or work:

[0033] The calibration mode remains unchanged during the operation of the device to prevent misuse and misjudgment. Collect the actual operating parameters of the equipment at the current moment:

[0034] Calculate the relative deviation for each dimension:

[0035] Define a weighted composite deviation metric:

[0036] in, 、 、 Respectively represent the relative deviation of current current, speed, power and their reference values. 、 、 is the corresponding weight (satisfying ), the weights can be set based on the experience of parameter stability, and the comprehensive deviation metric Used to evaluate the overall deviation between the current working condition and the reference working condition; like (e.g. 0.3), it means that the current state has obviously deviated from the reference working condition and is marked as another working condition (e.g. mode=idle, the current point is identified as the running working condition, and if mode=work, the current point is identified as the no-load working condition); ② Identification and confirmation of operating condition continuity Taking into account the transient fluctuations in industrial systems (such as load jumps and speed disturbances), a sliding window continuous judgment method is further adopted to improve the stability and accuracy of working condition identification; Define the sliding window size (The default setting is 60 sampling points), and the number of time points where the deviation exceeds the limit is counted in each window :

[0037] in, is the indicator function; Set the continuity threshold (The default setting is 15), the judgment logic is: like , it is confirmed that the working condition has changed; like , it is considered to be a normal fluctuation and no response is required; The final working condition identification result output is:

[0038] in, Indicates the working condition type opposite to the current mode:

[0039]

[0040] Final output label It is the working condition identification result at the current time point.

[0041] Status analysis: It should be further explained that under the identified operating conditions, a state analysis process is performed. This process uses high-frequency vibration signals and key operating parameters as inputs to calculate multiple types of state features. A benchmark model is then constructed based on the feature distribution of the historical normal operating phase for subsequent anomaly identification and health assessment. The benchmark model includes at least three analysis sub-models: vibration spectrum analysis, axis trajectory calculation, and parameter trend analysis. The output results of each model are divided into numerical features and shape features, and benchmark thresholds or reference maps are established respectively. ①Vibration spectrum analysis (shape characteristics) During the period when the operating condition is identified as the operating period, the vibration acceleration signal Perform frequency domain transformation and calculate the spectrum using Fast Fourier Transform (FFT):

[0042] in, is the sampling frequency, the extracted spectrum Characterize the vibration excitation characteristics of the equipment in operation; Select the spectrum graphs of multiple operating periods to calculate the average value and generate a reference spectrum graph , as the standard shape feature in the normal state:

[0043] ② Axis trajectory analysis (shape features) Use two axial displacement sensors (such as X and Y directions) installed on the rotating equipment to synchronously collect the axis motion trajectory , obtain the dynamic behavior characteristics of the rotor in the plane; When the working condition is identified as the running period, the trajectory data of multiple cycles is extracted , directly normalize and align them according to the time series coordinates, and calculate the average trajectory as the standard reference trajectory:

[0044] Among them, each is the axis trajectory curve within one cycle; ③ Parameter trend analysis (numerical characteristics) For time series parameters (such as vibration amplitude, temperature, etc.), it is recorded as , extracting its statistical trend characteristics over time, including: Parameter mean: ; Parameter volatility: ; The trend slope is estimated using the least squares method. Fitting a first-order straight line , slope is the trend slope, ; in, is the sample size, i.e. the total number of observations in the time series, It is The time point corresponding to the observation value, It's time Parameter observations on ; For the data under normal operating conditions, the statistical threshold interval of each numerical feature is constructed, and for any statistical trend feature, the reference mean is set to , the reference standard deviation is , define the decision interval:

[0045] in, is the confidence factor (default is 3), which is used to set the alarm threshold.

[0046] Health Assessment: To achieve real-time determination of equipment status and abnormal alarms, after the status analysis is completed, the health assessment phase begins. This phase executes corresponding abnormality recognition logic based on different types of status characteristics and outputs specific alarm information when deviations are detected. The health assessment process is divided into numerical feature and shape feature similarity recognition according to feature type. Each feature type is evaluated separately. If it is judged to be abnormal, the alarm result of that type is output and the current time, deviation degree, feature value and reference value information are recorded. ①Numerical feature anomaly identification For numerical features output by state analysis (such as parameter trend slope, mean value, and maximum value), set historical reference threshold ranges under operating conditions:

[0047] The abnormal judgment logic is:

[0048] ② Shape feature anomaly recognition For shape features such as spectrograms and axis trajectories, similarity evaluation is performed based on the normalized multi-frame vector sequence; The real-time feature shape is a vector sequence of , the standard reference shape is: , where each , the dimension is A vector (such as ( ) represents a plane trajectory point, and both the spectrum diagram and the axis trajectory diagram are plane trajectory points); Calculate the RMS value of the vector distance between frames:

[0049] Setting shape anomaly threshold (The default value is 0.3), and the abnormal judgment logic is defined as:

[0050] For various features, if the abnormal recognition logic result of a certain feature value is 1, it is judged that the indicator of this type is abnormal and an alarm of this type is issued.

[0051] Question and answer interaction: The method provided in this embodiment supports natural language question-answering interaction. By combining a built-in large language model with a data interface and prompt word templates, it automatically answers user questions, achieving intelligent interpretation of device status and decision support. ①Workflow User input: Input questions through the web interface or touch terminal; Intent recognition: Identify the question intention and match the corresponding template; Status query class → call template Q1; Alarm explanation class → call template Q2; Trend analysis class → call template Q3; Health Assessment Class → Call Template Q4; History comparison class → call template Q5; Other free classes → call template Q6; Data acquisition: call the data interface to obtain real-time or historical feature values; Prompt word construction: fill the data into the preset prompt word template; Model call: input the constructed prompt into LLM to generate the answer; Result display: returns the result to the user in natural language, and supports display modules such as linkage alarm and trend chart; ②Prompt word template Q1: Query the current device status Prompt word structure: Based on the following data, please determine the current operating status of the device and answer the user in concise and natural language: Working mode: {working mode} Current operating parameters: Current: {current} A Speed: {speed} rpm Power: {power} kW Status analysis features: Spectral similarity: {spectral similarity} Axis trajectory similarity: {Axis trajectory similarity} Parameter trend deviation: {trend deviation} Based on the above information, please briefly describe whether the equipment is in normal condition and indicate its key features; Input variables: {operating mode}, {current}, {speed}, {power}, {spectral similarity}, {axis trajectory similarity}, {trend deviation}; Applicable scenarios: Users ask questions such as "Is the device running normally now?", "What is the current status of the device?", and "Are there any problems with the operating status?" Q2: Explanation of abnormal alarm Prompt word structure: If the user asks for alarm details, please explain based on the following alarm information: Alarm type: {alarm type} Trigger time: {alarm time} Feature Type: {feature type} Current value: {current value} Normal reference value: {reference value} Deviation degree: {deviation degree} Please explain the cause of the alarm, affected parts, and recommended actions in a professional but accessible manner; Input variables: {alarm type}, {alarm time}, {feature type}, {current value}, {reference value}, {deviation degree}; Applicable scenarios: Users ask questions such as "Why is the alarm going off?", "What type of anomaly is this?", and "Is the alarm serious?" Q3: Trend change analysis Prompt word structure: Please analyze the changing trends of the following key equipment parameters over the past {time interval} and briefly explain whether their changing characteristics are abnormal: Parameter trend data: Vibration fluctuation: {vibration fluctuation} Temperature mean: {temperature mean} Trend Slope: {Trend Slope} Please indicate whether there is a clear upward or fluctuating trend and explain the possible reasons; Input variables: {time interval}, {vibration fluctuation}, {temperature mean}, {trend slope}; Applicable scenarios: Users ask questions such as "Has the status changed recently?", "Is the device operating trend normal?", and "Is there a deteriorating trend?" Q4: Health score inquiry Prompt word structure: Please summarize the health status of your device based on the following health score data: Comprehensive health score: {comprehensive score} / 100 Key Feature Scores: Vibration characteristics: {vibration score} Spectral shape: {spectral score} Parameter trend: {trend score} Please evaluate whether the current condition is good and recommend whether preventive maintenance is needed; Input variables: {comprehensive score}, {vibration score}, {spectral score}, {trend score} Applicable scenarios: Users ask questions such as "Is the device healthy?", "What is the health score?", and "Does it need maintenance now?" Q5: Historical comparative analysis Prompt word structure: Compare the current device status to a historical reference status: Current status ({current time}): Vibration amplitude: {Current vibration} Spectrum similarity: {current spectrum similarity} Working mode: {Current working condition} Historical reference status ({historical time period}): Vibration amplitude: {historical vibration} Spectrum similarity: {historical spectrum similarity} Working condition mode: {historical working condition} Please explain whether the current status is deteriorating compared to the past and indicate the possible sources of change; Input variables: {current time}, {current vibration}, {current spectrum similarity}, {current working condition}, {historical time period}, {historical vibration}, {historical spectrum similarity}, {historical working condition}; Applicable scenarios: Users ask questions like "Is it better now than last month?", "Has the status deteriorated recently?", and "Is the device showing signs of aging?" Q6: Diagnosis and Suggestion Q&A (Enhanced Free Q&A) Prompt word structure: User question: {User question} Current data supports: Working condition identification result: {working condition label} Abnormal feature list: {abnormal list} Alarm record: {alarm information} Parameter trend summary: {trend summary} Please combine the data and fault knowledge to generate a natural language answer, including the judgment result, basis, and recommended actions; Input variables: {user question}, {working condition label}, {abnormality list}, {alarm information}, {trend summary}; Applicable scenarios: Users ask questions such as "What do you think of this status?", "Will there be problems later?", and "Should we shut down the system immediately?" The scoring calculation method for the aforementioned health assessment can be performed using methods and means in the existing technology and will not be described in detail in this example.

[0052] Reference Figure 2 and Figure 3 The present invention provides an equipment status monitoring system based on automatic working condition calibration and multi-dimensional signal analysis. The monitoring system includes a multi-source sensor module, a data acquisition unit, an analysis and storage unit, a display and interaction unit, a communication interface module, and a power supply module, wherein: The multi-source sensor module is installed on the monitored equipment to synchronously collect vibration, temperature, acoustics, current, speed, and power signals, and output analog signals through BNC, RS-485, and 4-20mA standard physical interfaces to obtain comprehensive operating information covering both high and low frequencies. The data acquisition unit includes a sensor interface, power supply isolation, analog conditioning, and analog / digital conversion circuits. It is used to filter, amplify, and sample analog signals from multi-source sensor modules to form a digital signal stream. The digital signal stream is then sent to the analysis and storage unit in real time via the SPI, CAN, and Ethernet buses. The analysis and storage unit integrates an embedded processor and a local thermal database to perform hierarchical processing on digital signals, including: The working condition automatic calibration module generates working condition reference vectors according to the shutdown or running mode during the installation phase; The working condition identification module outputs working condition labels in real time during the operation phase based on the comprehensive deviation measurement and sliding window algorithm; The state analysis model module calls spectrum analysis, axis trajectory analysis, and parameter trend analysis models according to working conditions to extract shape and value multidimensional features; The health assessment module performs threshold determination on numerical features and similarity determination on shape features, and outputs health scores and alarm records; The data interface service module provides the evaluation results to the display unit and the communication interface module through a unified data interface, enabling local and remote access; The display interaction unit is a touch screen or an external HDMI display, which communicates with the analysis and storage unit through the serial port and HDMI interface. It is used to display the working condition label, characteristic curve, alarm information and health score in the form of curves, spectrum diagrams, axis trajectory diagrams, etc. in real time, and supports local query and confirmation by users; The communication interface module, consisting of the USB, RS-485, Ethernet, HDMI, and audio interfaces built into the analysis and storage unit, is used to send alarm codes and control instructions to the host computer and PLC (RS-485), synchronize monitoring data, configuration parameters, and logs to the remote monitoring platform (Ethernet), provide firmware upgrades, data export or debugging (USB), and output audible and visual alarm prompts (audio); The power module provides DC12V / 24V regulated power supply and is equipped with overvoltage and undervoltage protection circuits to provide safe and reliable energy for the data acquisition unit, analysis and storage unit, and display and interaction unit; The multi-source sensor module is connected to the data acquisition unit through a standard cable, the data acquisition unit is connected to the analysis and storage unit through a bus interface, and the analysis and storage unit is connected to the display unit via a serial port and HDMI. All modules are installed in an integrated chassis and internal connections are achieved through a PCB backplane and cables.

[0053] It should be noted that the present invention collects equipment operation data in real time through multi-source sensors such as vibration, temperature, and sound, and sends the data to the analysis and storage unit after numerical conversion and preprocessing by the data acquisition unit. The system automatically identifies the current working conditions, calls corresponding professional models such as spectrum analysis, axis trajectory, and health factors to extract state features, and completes the equipment status assessment; the analysis results are transmitted to the display interaction unit via the data interface for graphical display, and at the same time support intelligent body workflow linkage to achieve full-process closed-loop monitoring from signal acquisition, state recognition to result display and intelligent feedback.

[0054] In an optional embodiment, the analysis and storage unit is used to perform working condition calibration and identification, status analysis, and health assessment on the collected multi-dimensional raw data, including the following functions: (1) Working condition automatic calibration module Working condition calibration: By analyzing working condition parameters (current, speed, power), working condition benchmark calibration is performed during the installation process. If the installation is offline, the parameter benchmark values ​​of the no-load working condition are calibrated. If the installation is online, the parameter benchmark values ​​of the operating condition are calibrated. Working condition identification: During operation, if the threshold of the baseline value of the working condition calibration parameter changes by more than 20%, it is considered that the working condition has changed, and the operating status of the equipment (such as no-load, running) is automatically identified; (2) State analysis model module Select an appropriate model for signal processing based on the equipment operating conditions, including: Spectrum calculation: Fast Fourier transform (FFT) analyzes spectrum characteristics and obtains spectrum graph; Axis trajectory analysis: obtain axis position, axis trajectory shape, inner equivalent circle, and outer equivalent circle; Parameter analysis and calculation: parameter trend, parameter over-limit, parameter volatility; After calculating the features, the state analysis model module takes the operating condition data based on the working condition identification results and calibrates the normal baseline values ​​for the output features of the analysis model. It also sets thresholds for numerical features (such as parameter trends, parameter limits, parameter volatility, axis position, axis inner equivalent circle, and axis outer equivalent circle radius), and sets standard shape references for shape features (such as frequency spectrum and axis trajectory shape). (3) Health Assessment Module The output features of the state analysis model are input into the evaluation logic to make qualitative (normal / abnormal) and quantitative (scoring) judgments on the current equipment state, generate evaluation results and alarm records, and save them in the analysis database. The evaluation logic is divided into numerical and shape types according to the feature type: numerical features calculate the deviation between the real-time features and the threshold set by the benchmark value, and it is considered abnormal if it exceeds the set degree; shape features compare the similarity between the real-time shape and the standard shape reference set by the benchmark value, and it is considered abnormal if the similarity is lower than the set degree; For each type of state analysis model feature, if an anomaly is identified, an alarm of that type will be output.

[0055] In a preferred embodiment, the display interaction unit receives the analysis results in real time and presents them visually through a Web interface or a touch display terminal. The displayed content includes sensor data curves, spectrum diagrams, axis trajectory diagrams, equipment operating status, alarm information, working condition identification results, etc., and provides a unified data access interface to the outside world, supporting operations such as data reading, alarm status acquisition, and equipment status query, which is convenient for intelligent workflow access.

[0056] Furthermore, the display interaction unit can trigger regularized processes (such as fault Q&A, knowledge retrieval, and alarm linkage) based on changes in the equipment's operating status or active user requests. After the locally deployed large language model calls the data interface to query data, it performs prompt word enhancement (user question + query result) to generate a natural language answer and output it to the user interaction window.

[0057] In the application of the above embodiments, an embodiment of the present invention discloses an equipment status monitoring system based on automatic working condition calibration and multi-dimensional signal analysis, which further includes one or more processors and memories.

[0058] The memory is used to store operable instructions, which, when executed by the one or more processors, cause the one or more processors to perform operations, including the process of the device condition monitoring method based on automatic working condition calibration and multi-dimensional signal analysis of the aforementioned embodiment, especially Figure 1 The process of the method shown.

[0059] Other aspects disclosed in the embodiments of the present invention further provide a computer-readable medium storing software, wherein the software includes instructions that can be executed by one or more computers, and the execution of these instructions causes the one or more computers to perform operations, including the process of the equipment condition monitoring method based on automatic working condition calibration and multi-dimensional signal analysis of the aforementioned embodiment, especially Figure 1 The process of the method shown.

[0060] It should be appreciated that embodiments of the present invention can be implemented or carried out by computer hardware, a combination of hardware and software, or by computer instructions stored in a non-transitory computer-readable memory.

[0061] The method may be implemented in a computer program using standard programming techniques, including a non-transitory computer-readable storage medium configured with a computer program, wherein the storage medium so configured causes a computer to operate in a specific and predefined manner.

[0062] Each program can be implemented in a high-level procedural or object-oriented programming language to communicate with a computer system, however, the program can be implemented in assembly or machine language if desired.

[0063] In any case, the language may be a compiled or interpreted language.

[0064] Furthermore, the program can be run on an application specific integrated circuit programmed for this purpose.

[0065] The processes described herein (or variations and / or combinations thereof) may be performed under the control of one or more computer systems configured with executable instructions and may be implemented as code (e.g., executable instructions, one or more computer programs, or one or more applications) that collectively executes on one or more processors, by hardware, or a combination thereof. The computer program includes a plurality of instructions that can be executed by one or more processors.

[0066] Further, the method may be implemented in any type of computing platform operably connected to a suitable computer, including but not limited to a personal computer, a minicomputer, a mainframe, a workstation, a network or distributed computing environment, a separate or integrated computer platform, or in communication with a charged particle tool or other imaging device.

[0067] Aspects of the present invention may be implemented in machine-readable code stored on a non-transitory storage medium or device, whether removable or integrated into a computing platform, such as a hard disk, optical read and / or write storage media, RAM, ROM, etc., such that it can be read by a programmable computer and, when the storage medium or device is read by the computer, can be used to configure and operate the computer to perform the processes described herein.

[0068] Additionally, the machine-readable code, or portions thereof, can be transmitted over a wired or wireless network.

[0069] The invention described herein includes these and other various types of non-transitory computer-readable storage media when such media include instructions or programs that implement the steps described above in conjunction with a microprocessor or other data processor.

[0070] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A device status monitoring method based on automatic working condition calibration and multi-dimensional signal analysis, characterized in that: include: During the equipment installation phase, operating parameters are collected and a working condition benchmark vector is established based on outlier elimination and mean calculation; During the equipment operation phase, real-time operating parameters are obtained and compared with the operating condition reference vector to generate a comprehensive deviation metric, and the current operating condition label is determined through a continuity determination strategy; Based on the working condition label, the corresponding signal analysis model is called to extract the spectrum characteristics, the axis trajectory characteristics and the numerical characteristics of the operating parameters; Compare the acquired multi-dimensional features with the preset threshold to determine whether the device status is abnormal and generate alarm information; The operating condition labels, characteristic curves, abnormal categories and alarm records are output in a visual manner, and question-answering interaction based on a large language model is supported.

2. The equipment status monitoring method based on automatic working condition calibration and multi-dimensional signal analysis according to claim 1 is characterized in that: The method for constructing the working condition reference vector includes: The original sample sequences of current, speed and power are collected respectively within the calibration time window; The interquartile range outlier removal mechanism is applied to each type of sample, and an adjustable coefficient of 1.5 to 3 is used to filter outliers; The mean of the samples after elimination is calculated to obtain the parameter reference values ​​of the shutdown condition and the operating condition, which are the condition reference vectors.

3. The equipment status monitoring method based on automatic working condition calibration and multi-dimensional signal analysis according to claim 1 or 2, characterized in that: The method for generating the comprehensive deviation metric comprises: At startup, the variance of each operating parameter during the historical stable period is read, and weights are assigned based on the principle that the smaller the variance, the higher the weight; The relative deviation between the real-time operating condition parameters and the operating condition reference vector is calculated item by item, and the deviation is multiplied by the corresponding weight and normalized to form a comprehensive deviation metric.

4. The equipment status monitoring method based on automatic working condition calibration and multi-dimensional signal analysis according to claim 3 is characterized in that: The continuity determination strategy includes: Establish a first-in-first-out buffer with a length of 60 sampling points and update the deviation judgment results in real time; Counting the number of samples with deviation exceeding the limit in the buffer zone; When the number of samples exceeding the limit reaches 15, confirm that the working condition has been switched and update the working condition label immediately; When the number of samples exceeding the limit is less than 15 points, it is considered normal fluctuation and the original working condition label remains unchanged.

5. The equipment status monitoring method based on automatic working condition calibration and multi-dimensional signal analysis according to claim 4 is characterized in that: The method for acquiring the spectrum characteristics includes: The original vibration signal is first weighted by a window function and band-pass filtered to suppress the endpoint effect and exclude irrelevant frequency bands; The vibration amplitude spectrum is calculated using fast Fourier transform, and the amplitude spectra of multiple time periods are averaged under operating conditions to generate a reference spectrum diagram; The similarity between the real-time amplitude spectrum and the reference spectrum is used as the shape feature.

6. The equipment status monitoring method based on automatic working condition calibration and multi-dimensional signal analysis according to claim 4 is characterized in that: The method for obtaining the axis trajectory feature includes: Two orthogonal displacement sensors are used to synchronously collect the X and Y displacement data of the rotor to form a plane trajectory; Perform time alignment and coordinate normalization on the trajectory sequences of multiple complete rotor cycles to eliminate the influence of speed fluctuations; The average curve of each normalized trajectory is calculated as the standard reference trajectory, which is the axis trajectory shape feature.

7. The equipment status monitoring method based on automatic working condition calibration and multi-dimensional signal analysis according to claim 4 is characterized in that: The extraction of numerical features of mean, volatility and trend slope of real-time operating parameters includes: A sliding observation window with configurable length is established for each real-time operating parameter to collect continuous sampling values ​​within the window; averaging the sampled values ​​within the observation window to obtain the instantaneous mean value of the parameter in the current window to characterize the steady-state level; Calculate the standard deviation of the sampled values ​​within the same window and use it as a parameter volatility indicator to measure short-term dispersion; The least square method is used to perform a linear fit on the curve of the sampling value in the window changing with time. The slope of the fitted line is the trend slope, which is used to reflect the rising and falling trend of the parameter. Combine the mean, volatility, and trend slope obtained above to form the numerical feature vector at the current moment.

8. A monitoring system implemented by the equipment status monitoring method based on the automatic working condition calibration and multi-dimensional signal analysis according to claim 1, characterized in that: The monitoring system includes a multi-source sensor module, a data acquisition unit, an analysis and storage unit, and a display and interaction unit, wherein: The multi-source sensor module is used to collect vibration, sound, temperature and electrical parameter signals; The data acquisition unit is used to condition and digitize the collected signals; The analysis and storage unit is used to perform working condition calibration, operating state identification and multi-dimensional feature analysis, and perform abnormality determination, save operating data and support trend analysis, playback and edge computing; The display interaction unit is used to visually display the monitoring results and realize question-answering interaction and remote data communication in combination with a large language model.

9. The equipment status monitoring system based on automatic working condition calibration and multi-dimensional signal analysis according to claim 8 is characterized in that: Also includes one or more processors; A memory storing operable instructions, wherein when the instructions are executed by the one or more processors, the one or more processors perform operations, wherein the operations include the process of the equipment status monitoring method based on automatic working condition calibration and multi-dimensional signal analysis as described in any one of claims 1 to 7.

10. A computer-readable medium storing software, characterized in that: The software includes instructions that can be executed by one or more computers, and the instructions, through such execution, enable the one or more computers to perform operations, and the operations include the process of the equipment status monitoring method based on automatic working condition calibration and multi-dimensional signal analysis as described in any one of claims 1 to 7.

Images