An intelligent safety detection system and method for industrial equipment

Abstract

The present invention relates to the field of equipment detection technology, and specifically to an intelligent safety detection system and method for industrial equipment. The intelligent safety detection system for industrial equipment includes a dynamic monitoring and analysis module, an anomaly identification and processing module, a causal relationship analysis module, and a stability monitoring module. In the present invention, by capturing and analyzing sound and vibration data at a detailed level, continuous comparison and evaluation of the equipment operating status are achieved, the ability to capture subtle anomalies of the equipment is enhanced, and early prediction of potential faults is achieved. In particular, in the analysis of frequency offset and amplitude changes in abnormal data, the risk of faults can be more accurately assessed and identified. At the same time, in-depth analysis of causal relationships further enables the precise location of operating variables that lead to performance degradation, providing a basis for taking specific maintenance measures, significantly improving the effectiveness of preventive maintenance, effectively reducing unexpected failures and maintenance costs, and ensuring the efficiency and safety of the production process.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment detection, and in particular to an intelligent safety detection system and method for industrial equipment. Background Art

[0002] The field of equipment detection technology focuses on the development and implementation of systems and methods for monitoring, diagnosing, and maintaining industrial equipment. These systems typically include sensors, data acquisition hardware, analysis software, and communication technologies, which are used to collect and analyze equipment operating data in real time. The goal is to improve equipment reliability, safety, and efficiency by identifying potential failures or performance degradations at an early stage. These systems play a key role in preventive maintenance, fault diagnosis, and system optimization.

[0003] Intelligent safety monitoring systems for industrial equipment use integrated intelligent technologies to automatically monitor the safety status of industrial equipment, identifying and reporting potential safety risks in real time. These systems typically include multiple sensors, such as temperature, pressure, and vibration, along with a data processing unit that analyzes sensor data and issues warnings or automatically takes action based on pre-set safety parameters. These systems are crucial for ensuring the safety of industrial production environments, helping to prevent accidents and ensure the safety of personnel and equipment. They can also reduce downtime and repair costs caused by equipment failures.

[0004] Existing technologies lack the ability to conduct comprehensive, in-depth data analysis, relying primarily on fixed parameters and simple fault diagnosis logic. They lack the ability to dynamically adapt and learn, limiting their effectiveness in identifying complex or uncommon fault patterns. Furthermore, existing systems often fail to fully leverage collected data to predict trends in equipment performance, resulting in their unrealized potential for preventive maintenance and fault prediction. For example, in the face of rapidly changing production environments, the system is unable to instantly adjust its monitoring and response strategies, resulting in inaccurate or inadequate responses to potential issues, increasing safety risks and losses. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an intelligent safety detection system and method for industrial equipment.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: An intelligent safety detection system for industrial equipment includes:

[0007] The dynamic monitoring and analysis module collects equipment sound and vibration data, records data changes every second through sensors, compares them with the equipment's standard performance data, analyzes the equipment's current operating status, and generates a dynamic record of equipment performance;

[0008] The abnormality identification and processing module uses the dynamic records of the equipment performance to conduct in-depth analysis of the equipment data, captures data points that deviate from normal operating parameters in real time, calculates the frequency offset and amplitude change of the data points, analyzes the abnormal characteristics of the data points based on the calculation results, uses statistical techniques to evaluate the significance of the abnormal signals, identifies potential mechanical failures or operational abnormalities, and generates equipment failure prediction records;

[0009] The causal relationship analysis module constructs a relationship chart based on the equipment failure prediction record to represent the connection strength between the operating variables, analyzes the correlation between the operating variables and equipment performance degradation, identifies the key operating variables that cause equipment performance degradation based on the analysis results, quantitatively evaluates the impact of the key operating variables, and generates a comprehensive impact analysis record of the key variables;

[0010] The stability monitoring module continuously monitors the equipment operation data based on the comprehensive impact analysis records of the key variables and calculates the data information entropy. If the information entropy increases abnormally, it determines the change as an indicator of equipment stability degradation, conducts equipment status safety analysis, and generates equipment safety detection and evaluation records.

[0011] As a further solution of the present invention, the steps for obtaining the dynamic record of equipment performance are:

[0012] The sensor collects the sound data and vibration data of the device every second, and uses the formula:

[0013]

[0014] calculate Comprehensive device data value at the moment , get the weighted original data set, where and is the weighting coefficient, which is used to adjust the influence weight of sound and vibration data. and They are Observation values of equipment sound and vibration data at all times;

[0015] Based on the weighted original data set, the standard performance data of the equipment is compared and the formula is used.

[0016]

[0017] Calculate the performance deviation percentage of the device per second , used to standardize the deviation size and generate a performance deviation percentage data set, where yes Standard performance data at all times;

[0018] Analyze the performance deviation percentage data set, using the formula,

[0019]

[0020] Computing devices in Equipment stability index at the moment , get the dynamic record of equipment performance, where is the sensitivity coefficient, which is used to adjust the effect of deviation on the stability index.

[0021] As a further solution of the present invention, the steps for calculating the frequency offset and amplitude change of the data points are:

[0022] Combined with the dynamic recording of the equipment performance, recording timestamps and performance indicators, the formula is used:

[0023]

[0024] Calculates the adjusted time interval between consecutive data points , and output time series data, where It is The timestamp of each data point, is the time scaling factor, which is used to adjust the sensitivity of the time interval;

[0025] According to the time series data, Fourier transform is applied for analysis and converted into frequency domain, using the formula,

[0026]

[0027] Calculate the frequency offset value , and obtain the frequency offset data set, where is the frequency adjustment factor, which is used to enhance the sensitivity of frequency offset;

[0028] Based on the frequency offset data set, the formula is used,

[0029]

[0030] Calculate the amplitude change for each frequency offset , generate the amplitude change characteristic analysis results, where is the mean frequency offset of the frequency offset dataset, is the amplitude adjustment coefficient, which is used to adjust the sensitivity of amplitude changes.

[0031] As a further solution of the present invention, the steps for obtaining the equipment failure prediction record are:

[0032] Combined with the analysis results of the amplitude change characteristics, an abnormal threshold is set to identify abnormal amplitudes, and data points exceeding the abnormal threshold are marked as abnormal. The formula is used.

[0033]

[0034] Generate a set of anomalous data points ,in, is the amplitude change value of a single data point, It is a preset abnormal amplitude threshold used to distinguish between normal and abnormal operating states. is the threshold adjustment factor, which is used to adjust the flexibility of the anomaly definition;

[0035] Perform statistical analysis on the abnormal data point set and use the formula:

[0036]

[0037] Calculating significance indices , evaluate the significance of abnormal signals and generate abnormal signal significance evaluation results, where, is a single data point in the set of abnormal data points, and Represent the mean and standard deviation of the data point set, respectively, and are used to calculate the deviation of the data points;

[0038] According to the abnormal signal significance evaluation result, the formula is used:

[0039]

[0040] Calculate the predicted probability of failure , generate equipment failure prediction records, where, is the model sensitivity coefficient, which is used to adjust the response speed of the prediction model.

[0041] As a further solution of the present invention, the steps for analyzing the correlation between the operating variables and the equipment performance degradation are:

[0042] Based on the equipment failure prediction record, the operating variable data is extracted, the linear relationship between the operating variables is evaluated, and the formula is used.

[0043]

[0044] Calculated variables and Pearson correlation coefficient between , generates a correlation matrix between the operational variables, where and are the observed values of the two manipulated variables, is the sample size, which represents the total number of observation points;

[0045] Based on the correlation matrix between the operational variables, the formula is adopted,

[0046]

[0047] Calculate the operating variable and The connection weight , construct and output a relationship graph, where is to adjust the threshold, is the adjustment factor, which is used to amplify or reduce the impact of the correlation;

[0048] The dynamic relationship between the operating variables is analyzed using the relationship graph to identify the key connections that lead to performance degradation. The formula is used.

[0049]

[0050] Calculate the operating variable Total connection strength in the relationship diagram , revealing the operating variables Importance among all variables, set key strength judgment value ,like Greater than , then mark the operation variable For key operating variables, generate key operating variable identification results.

[0051] As a further solution of the present invention, the steps for obtaining the comprehensive impact analysis record of key variables are:

[0052] Analyze the key operating variable identification results and use a multiple linear regression model to quantify the impact of key operating variables on equipment performance. Using the formula,

[0053]

[0054] Calculate predicted values for equipment performance , and get the regression model analysis results, where It is The key operating variables, is the corresponding regression coefficient, which affects the quantitative degree of device performance;

[0055] Based on the regression model analysis results, the formula is used:

[0056]

[0057] Calculate key operating variables Contribution percentage , indicating a variable The relative influence on equipment performance, to obtain quantitative evaluation results of key variables;

[0058] Comprehensively analyze the quantitative evaluation results of the key variables, record the actual impact of each key variable on equipment performance, and generate a comprehensive impact analysis record of key variables.

[0059] As a further solution of the present invention, the steps for obtaining the equipment safety detection and evaluation record are:

[0060] Based on the comprehensive impact analysis records of the key variables, the equipment operation data is continuously monitored and the formula is used.

[0061]

[0062] Calculating information entropy , used to evaluate the randomness of data and obtain the information entropy calculation results, where Represents a data point The frequency of occurrence, Indicates the number of data points;

[0063] Based on the information entropy calculation result, it is compared with the predetermined security threshold and the formula is used.

[0064]

[0065] Computing Equipment Stability Index , used to quantify the comparison result between information entropy and predetermined security threshold, if >1 indicates that the equipment stability decreases, and the equipment stability status analysis results are obtained, where: Indicates the preset safety threshold;

[0066] According to the equipment stability status analysis results, the equipment safety assessment is carried out using the formula:

[0067]

[0068] Calculating risk level , reflects the current safety status of the equipment and generates equipment safety inspection and assessment records, among which, represents a fixed coefficient used to increase the sensitivity of risk assessment.

[0069] An intelligent safety detection method for industrial equipment includes the following steps:

[0070] S1: Sensors collect the device's sound and vibration data every second, calculate the comprehensive device data value, compare it with the device's standard performance data, analyze performance deviations, and calculate the device stability index to obtain a dynamic record of device performance;

[0071] S2: combining the dynamic performance records of the device, recording the timestamp and performance indicators, applying Fourier transform to analyze, converting to the frequency domain, calculating the frequency offset value and the amplitude change value of the frequency offset, and generating the amplitude change characteristic analysis result;

[0072] S3: Based on the amplitude change characteristic analysis results, set an abnormal threshold, identify abnormal amplitudes, perform statistical analysis, evaluate the significance of abnormal signals, calculate and predict the probability of failure, and generate equipment failure prediction records;

[0073] S4: Based on the equipment failure prediction record, extracting operating variable data, evaluating the linear relationship between the operating variables, calculating the connection weights between the operating variables, analyzing the dynamic associations between the operating variables, identifying the key connections that cause performance degradation, and generating key operating variable identification results;

[0074] S5: Analyze the key operating variable identification results, use a multivariate linear regression model to quantify the impact of the key operating variables on equipment performance, calculate the contribution percentage of the key operating variables, record the actual impact of each key variable on equipment performance, and generate a comprehensive impact analysis record of the key variables;

[0075] S6: Based on the comprehensive impact analysis record of the key variables, continuously monitor the equipment operation data, calculate the information entropy and evaluate the randomness of the data, compare it with the predetermined safety threshold, calculate the equipment stability index, perform equipment safety assessment, and generate equipment safety detection and assessment records.

[0076] Compared with the prior art, the advantages and positive effects of the present invention are:

[0077] In the present invention, by capturing and analyzing sound and vibration data at a detailed level, continuous comparison and evaluation of the equipment operating status is achieved, the ability to capture subtle equipment anomalies is enhanced, and early prediction of potential faults is achieved. In particular, in the analysis of frequency offset and amplitude changes in abnormal data, fault risks can be more accurately evaluated and identified. At the same time, in-depth analysis of cause-and-effect relationships further enables the precise location of operating variables that lead to performance degradation, providing a basis for taking specific maintenance measures, significantly improving the effectiveness of preventive maintenance, effectively reducing unexpected failures and maintenance costs, and ensuring the efficiency and safety of the production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 is a system flow chart of the present invention;

[0079] Figure 2 A flowchart for obtaining dynamic records of device performance of the present invention;

[0080] Figure 3 Flowchart for calculating frequency offset and amplitude change of data points of the present invention;

[0081] Figure 4 A flowchart for obtaining equipment fault prediction records of the present invention;

[0082] Figure 5 This is a flow chart for analyzing the correlation between operating variables and equipment performance degradation in the present invention;

[0083] Figure 6 A flowchart for obtaining comprehensive impact analysis records of key variables of the present invention;

[0084] Figure 7 This is a flowchart for obtaining equipment safety inspection and evaluation records of the present invention. DETAILED DESCRIPTION

[0085] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0086] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0087] See also Figure 1 , an intelligent safety detection system for industrial equipment includes:

[0088] The dynamic monitoring and analysis module collects equipment sound and vibration data, records data changes every second through sensors, compares them with the equipment's standard performance data, analyzes the equipment's current operating status, and generates a dynamic record of equipment performance;

[0089] The abnormality identification and processing module dynamically records equipment performance and conducts in-depth analysis of equipment data. It captures data points that deviate from normal operating parameters in real time, calculates the frequency offset and amplitude change of the data points, analyzes the abnormal characteristics of the data points based on the calculation results, and uses statistical techniques to evaluate the significance of abnormal signals, identifying potential mechanical failures or operational anomalies and generating equipment failure prediction records.

[0090] The causal relationship analysis module, based on equipment failure prediction records, constructs a relationship chart to represent the connection strength between operating variables, analyzes the correlation between operating variables and equipment performance degradation, identifies the key operating variables that cause equipment performance degradation based on the analysis results, quantitatively evaluates the impact of key operating variables, and generates a comprehensive impact analysis record of key variables;

[0091] The stability monitoring module continuously monitors equipment operation data based on the comprehensive impact analysis records of key variables and calculates the data information entropy. If the information entropy increases abnormally, it determines the change as an indicator of equipment stability degradation, conducts equipment status safety analysis, and generates equipment safety detection and assessment records.

[0092] Dynamic records of equipment performance include sound data, vibration data and equipment operation status analysis records; equipment fault prediction records include frequency offset calculation results, amplitude change value calculation results and abnormal signal statistical evaluation records; key variable comprehensive impact analysis records include correlation strength evaluation results, key variable identification records and impact quantification analysis records; equipment safety detection and evaluation records include information entropy value calculation results, stability analysis records and equipment safety analysis results.

[0093] See also Figure 2 , the steps for obtaining dynamic records of equipment performance are:

[0094] The sensor collects the sound data and vibration data of the device every second, and uses the formula:

[0095]

[0096] calculate Comprehensive device data value at the moment , get the weighted original data set, where and is the weighting coefficient, which is used to adjust the influence weight of sound and vibration data. and They are Observation values of equipment sound and vibration data at all times;

[0097] Based on the weighted original data set, compared with the equipment standard performance data, the formula is used.

[0098]

[0099] Calculate the performance deviation percentage of the device per second , used to standardize the deviation size and generate a performance deviation percentage data set, where yes Standard performance data at all times;

[0100] To analyze the performance deviation percentage data set, we use the formula,

[0101]

[0102] Computing devices in Equipment stability index at the moment , get the dynamic record of equipment performance, where is the sensitivity coefficient, which is used to adjust the effect of deviation on the stability index.

[0103] The sensor collects data at a specific moment Data:

[0104] (decibel)

[0105] (acceleration)

[0106] set up , , substituting into the formula, we get:

[0107]

[0108] Results This means that at this moment, the comprehensive sound and vibration data value of the equipment is 18 units, which can be used for subsequent comparison with standard performance data to determine whether the equipment is in normal operating condition.

[0109] Equipment standard performance data is Under the same conditions:

[0110]

[0111] Combined with the above calculation results, substitute into the formula:

[0112]

[0113] The resulting performance deviation percentage is 0.1 or 10%, representing equipment performance that is 10% below standard performance, which helps identify possible maintenance needs or performance degradation of the equipment.

[0114] Setting the sensitivity coefficient , used to amplify the impact of deviation and ensure the sensitivity of the calculation results.

[0115] Substituting into the formula, we get:

[0116]

[0117] Results This means that at this moment, the stability index of the device is 0.623. The closer this value is to 1, the more stable the device is. It is used to monitor and analyze the operating efficiency and safety of the device.

[0118] See also Figure 3 , the calculation steps for the data point frequency offset and amplitude change are:

[0119] Combined with the dynamic recording of equipment performance, record timestamps and performance indicators, use the formula:

[0120]

[0121] Calculates the adjusted time interval between consecutive data points , and output time series data, where It is The timestamp of each data point, is the time scaling factor, which is used to adjust the sensitivity of the time interval;

[0122] According to the time series data, Fourier transform is applied for analysis and converted into frequency domain, using the formula,

[0123]

[0124] Calculate the frequency offset value , and obtain the frequency offset data set, where is the frequency adjustment factor, which is used to enhance the sensitivity of frequency offset;

[0125] Based on the frequency offset data set, the formula is used,

[0126]

[0127] Calculate the amplitude change for each frequency offset , generate the amplitude change characteristic analysis results, where is the mean frequency offset of the frequency offset dataset, is the amplitude adjustment coefficient, which is used to adjust the sensitivity of amplitude changes.

[0128] set up Second, Second.

[0129] set up , used to increase the sensitivity of the time interval.

[0130] Substituting into the formula, we get:

[0131]

[0132] The obtained time interval Seconds. This means that after adjustment, the interval between data points is 2 seconds, indicating that the frequency of data collection is accelerated to capture faster performance changes.

[0133] set up , to improve the calculation sensitivity of frequency offset, then:

[0134]

[0135] Calculated frequency offset Hz, which means the frequency of performance changes between data points is 5 times per second, which can help detect small changes in device performance earlier.

[0136] : The average value of frequency offset. The collected frequency offsets are 2.5, 4, 5, and 3, and the average value is calculated. is 4Hz.

[0137] set up , combined with the above calculation results, substitute into the formula and we get:

[0138]

[0139] Calculated amplitude change This indicates that the current frequency deviation has increased by approximately 1.25 times relative to the average frequency deviation. Increased amplitude variation indicates that the device is experiencing anomalies or performance fluctuations, requiring further analysis to confirm potential mechanical failures or operational anomalies.

[0140] See also Figure 4 ,The steps for obtaining equipment failure prediction records are:

[0141] Combined with the analysis results of amplitude change characteristics, an abnormal threshold is set to identify abnormal amplitudes, and data points exceeding the abnormal threshold are marked as abnormal. The formula is used.

[0142]

[0143] Generate a set of anomalous data points ,in, is the amplitude change value of a single data point, It is a preset abnormal amplitude threshold used to distinguish between normal and abnormal operating states. is the threshold adjustment factor, which is used to adjust the flexibility of the anomaly definition;

[0144] Perform statistical analysis on the abnormal data point set and use the formula,

[0145]

[0146] Calculating significance indices , evaluate the significance of abnormal signals and generate abnormal signal significance evaluation results, where, is a single data point in the set of abnormal data points, and Represent the mean and standard deviation of the data point set, respectively, and are used to calculate the deviation of the data points;

[0147] According to the abnormal signal significance evaluation results, the formula is used.

[0148]

[0149] Calculate the predicted probability of failure , generate equipment failure prediction records, where, is the model sensitivity coefficient, which is used to adjust the response speed of the prediction model.

[0150] Combined with the above calculation results .

[0151] set up and , to improve the flexibility of anomaly detection.

[0152] Substituting into the formula:

[0153]

[0154] Since 1.25 is less than 2.0, this data point is not marked as an anomaly. The abnormality is significant but does not exceed the adjusted abnormality threshold of 2.0, indicating that the data points are still within the normal range under the current environment, which helps to avoid false abnormality alerts and enhance the practicality and reliability of the system.

[0155] Based on the abnormal data point set , calculate the mean , calculate the standard deviation is 0.5.

[0156] Taking data points 1.5, 2.0, and 2.5 as an example, then:

[0157]

[0158]

[0159] Significance indicators Indicates that the abnormal data point has a significant change from the mean, indicating a potential equipment failure or operational anomaly and requiring further investigation.

[0160] : Model sensitivity coefficient, used to adjust the response speed of the prediction model, set to 1.0.

[0161] Combining the above calculation results and substituting them into the formula, we get:

[0162]

[0163] The predicted failure probability A value very close to 1 indicates that the device is highly likely to have failed based on the anomaly significance of the current data point. This high probability prompts the operations team to conduct immediate inspection and maintenance to prevent the failure from developing or damaging the device.

[0164] See also Figure 5 , the analysis steps of the correlation between operating variables and equipment performance degradation are:

[0165] Based on the equipment failure prediction records, the operating variable data are extracted, the linear relationship between the operating variables is evaluated, and the formula is used.

[0166]

[0167] Calculated variables and Pearson correlation coefficient between , generates a correlation matrix between the operational variables, where and are the observed values of the two manipulated variables, is the sample size, which represents the total number of observation points;

[0168] Based on the correlation matrix between the operating variables, the formula is adopted.

[0169]

[0170] Calculate the operating variable and The connection weight , construct and output a relationship graph, where is to adjust the threshold, is the adjustment factor, which is used to amplify or reduce the impact of the correlation;

[0171] Use relationship charts to analyze the dynamic associations between operating variables, identify key connections that lead to performance degradation, and use formulas to

[0172]

[0173] Calculate the operating variable Total connection strength in the relationship diagram , revealing the operating variables Importance among all variables, set key strength judgment value ,like Greater than , then mark the operation variable For key operating variables, generate key operating variable identification results.

[0174] Represents the operation variable Values: [2, 3, 4].

[0175] Represents the operation variable Values: [5, 6, 7].

[0176] =3, substitute into the formula and calculate:

[0177]

[0178]

[0179]

[0180]

[0181]

[0182] Calculate the Pearson correlation coefficient:

[0183]

[0184]

[0185]

[0186] The obtained correlation coefficient Indicator operation variable and Complete linear correlation means that the manipulated variables and Together they affect device performance.

[0187] set up:

[0188]

[0189]

[0190] Substituting into the formula, we get:

[0191]

[0192]

[0193] Connection strength Indicates that there is a strong correlation between the two variables. This value is used in the chart to represent the width of the edge between the two variables.

[0194] In addition to operating variables and , and operational variables , combined with the above formula, we can calculate .

[0195] Substituting into the formula, we get:

[0196]

[0197] The total connection strength obtained indicator variables The overall connection strength with other variables is high, indicating that it has a strong centrality in the network and may be a key variable affecting device performance.

[0198] set up is 0.8, and 0.923 is greater than 0.8, so the operating variable Recorded as key operational variables.

[0199] See also Figure 6 , the steps for obtaining the comprehensive impact analysis records of key variables are:

[0200] Analyze the key operating variable identification results and use the multivariate linear regression model to quantify the impact of key operating variables on equipment performance. Using the formula,

[0201]

[0202] Calculate predicted values for equipment performance , and get the regression model analysis results, where It is The key operating variables, is the corresponding regression coefficient, which affects the quantitative degree of device performance;

[0203] Based on the regression model analysis results, the formula is used.

[0204]

[0205] Calculate key operating variables Contribution percentage , indicating a variable The relative influence on equipment performance, to obtain quantitative evaluation results of key variables;

[0206] Comprehensively analyze the quantitative evaluation results of key variables, record the actual impact of each key variable on equipment performance, and generate a comprehensive impact analysis record of key variables.

[0207] According to the results of key operating variables identification, three operating variables are determined Affects device performance.

[0208] The observation values are:

[0209]

[0210]

[0211]

[0212] Set the coefficients of the regression model to

[0213] For a specific data point, , substituting into the formula, we get:

[0214]

[0215]

[0216]

[0217] The predicted value of equipment performance is 19.4. The specific numerical impact of each variable on performance according to its regression coefficient is shown, indicating the importance of each variable in affecting equipment performance.

[0218] Combining the above results, calculate the sum of the absolute values of all regression coefficients:

[0219]

[0220] Calculate the contribution percentage of each variable:

[0221] for :

[0222]

[0223] for :

[0224]

[0225] for :

[0226]

[0227] The percentage reflects the relative importance of each operating variable in affecting the performance of the equipment. has the greatest impact, accounting for about 53.57%, followed by variables , followed by , helps to understand which operating variables have the greatest impact on performance, so that these variables can be adjusted or controlled in priority to improve equipment performance.

[0228] Finally, comprehensive processing is performed to record the actual impact of each key variable on equipment performance, and then the comprehensive impact analysis record of key variables is output.

[0229] See also Figure 7 , the steps for obtaining equipment safety inspection and assessment records are as follows:

[0230] Based on the comprehensive impact analysis records of key variables, continuous monitoring of equipment operation data, using formulas,

[0231]

[0232] Calculating information entropy , used to evaluate the randomness of data and obtain the information entropy calculation results, where Represents a data point The frequency of occurrence, Indicates the number of data points;

[0233] Based on the information entropy calculation results, it is compared with the predetermined security threshold and the formula is used.

[0234]

[0235] Computing Equipment Stability Index , used to quantify the comparison result between information entropy and predetermined security threshold, if >1 indicates that the equipment stability decreases, and the equipment stability status analysis results are obtained, where: Indicates the preset safety threshold;

[0236] According to the results of equipment stability status analysis, equipment safety assessment is carried out using the formula:

[0237]

[0238] Calculating risk level , reflects the current safety status of the equipment and generates equipment safety inspection and assessment records, among which, represents a fixed coefficient used to increase the sensitivity of risk assessment.

[0239] Taking device temperature data (unit: Celsius) as an example, the data set collected through continuous monitoring is:

[0240]

[0241] Calculate the probability of each data point occurring:

[0242]

[0243]

[0244]

[0245] Substituting into the formula, we get:

[0246]

[0247]

[0248] The information entropy value is , indicating that the uncertainty and complexity of the data set are relatively medium, and the randomness of the equipment operating status is within a relatively normal range.

[0249] Predetermined safety threshold Set it to 1.0, combine the above calculation results, and substitute into the formula to get:

[0250]

[0251] Stability Index , slightly greater than 1, indicating that the stability of the device has slightly decreased, and attention should be paid to risks or further inspections should be conducted.

[0252] set up =0.1, continue to calculate the risk level:

[0253]

[0254] A risk level of 0.6 (not exceeding 1) indicates that the equipment is in a low-risk state and preliminary maintenance checks should be considered to prevent possible failures.

[0255] An intelligent safety detection method for industrial equipment includes the following steps:

[0256] S1: Sensors collect the device's sound and vibration data every second, calculate the comprehensive device data value, compare it with the device's standard performance data, analyze performance deviations, and calculate the device stability index to obtain a dynamic record of device performance;

[0257] S2: Combine the dynamic records of device performance, record the timestamp and performance indicators, apply Fourier transform to analyze, convert to the frequency domain, calculate the frequency offset value and the amplitude change value of the frequency offset, and generate the amplitude change characteristic analysis results;

[0258] S3: Based on the results of the amplitude change characteristic analysis, set the abnormal threshold, identify the abnormal amplitude, perform statistical analysis, evaluate the significance of the abnormal signal, calculate and predict the possibility of failure, and generate equipment failure prediction records;

[0259] S4: Based on the equipment failure prediction records, extract the operating variable data, evaluate the linear relationship between the operating variables, calculate the connection weights between the operating variables, analyze the dynamic association between the operating variables, identify the key connections that cause performance degradation, and generate key operating variable identification results;

[0260] S5: Analyze the key operating variable identification results, use the multivariate linear regression model to quantify the impact of key operating variables on equipment performance, calculate the contribution percentage of key operating variables, record the actual impact of each key variable on equipment performance, and generate a comprehensive impact analysis record of key variables;

[0261] S6: Based on the comprehensive impact analysis records of key variables, continuously monitor equipment operation data, calculate information entropy and evaluate the randomness of the data, compare it with the predetermined safety threshold, calculate the equipment stability index, perform equipment safety assessment, and generate equipment safety detection and assessment records.

[0262] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. An intelligent safety detection system for industrial equipment, characterized by: The system comprises: The dynamic monitoring and analysis module collects equipment sound and vibration data, records data changes every second through sensors, compares them with the equipment's standard performance data, analyzes the equipment's current operating status, and generates a dynamic record of equipment performance; The abnormality identification and processing module uses the dynamic records of the equipment performance to conduct in-depth analysis of the equipment data, captures data points that deviate from normal operating parameters in real time, calculates the frequency offset and amplitude change of the data points, analyzes the abnormal characteristics of the data points based on the calculation results, uses statistical techniques to evaluate the significance of the abnormal signals, identifies potential mechanical failures or operational abnormalities, and generates equipment failure prediction records; The causal relationship analysis module constructs a relationship chart based on the equipment failure prediction record to represent the connection strength between the operating variables, analyzes the correlation between the operating variables and equipment performance degradation, identifies the key operating variables that cause equipment performance degradation based on the analysis results, quantitatively evaluates the impact of the key operating variables, and generates a comprehensive impact analysis record of the key variables; The stability monitoring module continuously monitors the equipment operation data based on the comprehensive impact analysis records of the key variables, calculates the data information entropy, and if the information entropy increases abnormally, determines the change as an indicator of equipment stability degradation, conducts equipment status safety analysis, and generates equipment safety detection and assessment records; The calculation steps for the frequency offset and amplitude change of the data point are: Combined with the dynamic recording of the equipment performance, recording timestamps and performance indicators, the formula is used: Calculates the adjusted time interval between consecutive data points , and output time series data, where It is The timestamp of each data point, is the time scaling factor, which is used to adjust the sensitivity of the time interval; According to the time series data, Fourier transform is applied for analysis and converted into frequency domain, using the formula, Calculate the frequency offset value , and obtain the frequency offset data set, where is the frequency adjustment factor, which is used to enhance the sensitivity of frequency offset; Based on the frequency offset data set, the formula is used, Calculate the amplitude change for each frequency offset , generate the amplitude change characteristic analysis results, where is the mean frequency offset of the frequency offset dataset, is the amplitude adjustment coefficient, which is used to adjust the sensitivity of amplitude changes; The steps for obtaining the equipment failure prediction record are: Combined with the analysis results of the amplitude change characteristics, an abnormal threshold is set to identify abnormal amplitudes, and data points exceeding the abnormal threshold are marked as abnormal. The formula is used. Generate a set of anomalous data points ,in, is the amplitude change value of a single data point, It is a preset abnormal amplitude threshold used to distinguish between normal and abnormal operating states. is the threshold adjustment factor, which is used to adjust the flexibility of the anomaly definition; Perform statistical analysis on the abnormal data point set and use the formula: Calculating significance indices , evaluate the significance of abnormal signals and generate abnormal signal significance evaluation results, where, is a single data point in the set of abnormal data points, and Represent the mean and standard deviation of the data point set, respectively, and are used to calculate the deviation of the data points; According to the abnormal signal significance evaluation result, the formula is used: Calculate the predicted probability of failure , generate equipment failure prediction records, where, is the model sensitivity coefficient, which is used to adjust the response speed of the prediction model; The analysis steps for the correlation between the operating variables and equipment performance degradation are as follows: Based on the equipment failure prediction record, the operating variable data is extracted, the linear relationship between the operating variables is evaluated, and the formula is used. Calculated variables and Pearson correlation coefficient between , generates a correlation matrix between the operational variables, where and are the observed values of the two manipulated variables, is the sample size, which represents the total number of observation points; Based on the correlation matrix between the operational variables, the formula is adopted, Calculate the operating variable and The connection weight , construct and output a relationship graph, where is to adjust the threshold, is the adjustment factor, which is used to amplify or reduce the impact of the correlation; The dynamic relationship between the operating variables is analyzed using the relationship graph to identify the key connections that lead to performance degradation. The formula is used. Calculate the operating variable Total connection strength in the relationship diagram , revealing the operating variables Importance among all variables, set key strength judgment value ,like Greater than , then mark the operation variable For key operating variables, generate key operating variable identification results; The steps for obtaining the comprehensive impact analysis record of key variables are as follows: Analyze the key operating variable identification results and use a multiple linear regression model to quantify the impact of key operating variables on equipment performance. Using the formula, Calculate predicted values for equipment performance , and get the regression model analysis results, where It is The key operating variables, is the corresponding regression coefficient, which affects the quantitative degree of device performance; Based on the regression model analysis results, the formula is used: Calculate key operating variables Contribution percentage , indicating a variable The relative influence on equipment performance, to obtain quantitative evaluation results of key variables; Comprehensively analyze the quantitative evaluation results of the key variables, record the actual impact of each key variable on equipment performance, and generate a comprehensive impact analysis record of key variables.

2. The intelligent safety detection system for industrial equipment according to claim 1, characterized in that: The steps for obtaining the dynamic record of equipment performance are as follows: The sensor collects the sound data and vibration data of the device every second, and uses the formula: calculate Comprehensive device data value at the moment , get the weighted original data set, where and is the weighting coefficient, which is used to adjust the influence weight of sound and vibration data. and They are Observation values of equipment sound and vibration data at all times; Based on the weighted original data set, the standard performance data of the equipment is compared and the formula is used. Calculate the performance deviation percentage of the device per second , used to standardize the deviation size and generate a performance deviation percentage data set, where yes Standard performance data at all times; Analyze the performance deviation percentage data set, using the formula, Computing devices in Equipment stability index at the moment , get the dynamic record of equipment performance, where is the sensitivity coefficient, which is used to adjust the effect of deviation on the stability index.

3. The intelligent safety detection system for industrial equipment according to claim 1, characterized in that: The steps for obtaining the equipment safety inspection and assessment records are as follows: Based on the comprehensive impact analysis records of the key variables, the equipment operation data is continuously monitored and the formula is used. Calculating information entropy , used to evaluate the randomness of data and obtain the information entropy calculation results, where Represents data points The frequency of occurrence, Indicates the number of data points; Based on the information entropy calculation result, it is compared with the predetermined security threshold and the formula is used. Computing Equipment Stability Index , used to quantify the comparison result between information entropy and predetermined security threshold, if >1 indicates that the equipment stability decreases, and the equipment stability status analysis results are obtained, where: Indicates the preset safety threshold; According to the equipment stability status analysis results, the equipment safety assessment is carried out using the formula: Calculating risk level , reflects the current safety status of the equipment and generates equipment safety inspection and assessment records, among which, represents a fixed coefficient used to increase the sensitivity of risk assessment.

4. An intelligent safety detection method for industrial equipment, characterized in that: The intelligent safety detection system for industrial equipment according to any one of claims 1 to 3 comprises the following steps: The sensor collects the equipment's sound and vibration data every second, calculates the comprehensive equipment data value, compares it with the equipment's standard performance data, analyzes performance deviations, and calculates the equipment stability index to obtain a dynamic record of equipment performance; Combined with the dynamic recording of the device performance, the timestamp and performance indicators are recorded, and Fourier transform is applied for analysis, converted to the frequency domain, and the frequency offset value and the amplitude change value of the frequency offset are calculated to generate the amplitude change characteristic analysis result; Based on the analysis results of the amplitude change characteristics, an abnormal threshold is set, abnormal amplitudes are identified, statistical analysis is performed, the significance of abnormal signals is evaluated, the probability of failure is calculated and predicted, and a device failure prediction record is generated; extracting operating variable data based on the equipment failure prediction record, evaluating the linear relationship between the operating variables, calculating the connection weights between the operating variables, analyzing the dynamic associations between the operating variables, identifying the key connections that cause performance degradation, and generating key operating variable identification results; Analyze the key operating variable identification results, use a multivariate linear regression model to quantify the impact of the key operating variables on equipment performance, calculate the contribution percentage of the key operating variables, record the actual impact of each key variable on equipment performance, and generate a comprehensive impact analysis record of the key variables; Based on the comprehensive impact analysis records of the key variables, the equipment operation data is continuously monitored, the information entropy is calculated and the randomness of the data is evaluated, compared with the predetermined safety threshold, the equipment stability index is calculated, the equipment safety assessment is performed, and the equipment safety detection and assessment record is generated.

Images