Overhaul processing equipment data supervision system and method based on smart factory

Abstract

The invention discloses an overhaul processing equipment data supervision system and method based on a smart factory, and the method comprises the steps: extracting a detection report of the smart factory for a produced impeller in history, and carrying out the analysis to obtain a feature object; performing an expected test in actual work based on the feature object, calculating a first expected value and a second expected value of impeller operation, and constructing an expected parameter list; saliency indexes under the current supervision strategy are quantified, and a sampling index list is formed; and performing correlation analysis based on the expected parameter list and the sampling index list, and when a correlation coefficient is smaller than a preset correlation expected threshold value, manually adjusting the supervision strategy until the correlation coefficient corresponding to the adjusted supervision strategy is not smaller than the preset correlation expected threshold value. According to the invention, a datamation decision basis can be provided for a sampling inspection strategy.

Description

Technical Field

[0001] This invention belongs to the field of impeller monitoring data analysis technology, and in particular relates to a data monitoring system and method for maintenance and processing equipment based on a smart factory. Background Technology

[0002] In recent years, underwater inspection robots have been widely used in river basins with multiple hydropower stations to replace manual diving for underwater inspections of facilities such as trash racks, dam foundations, and tailrace channels. These robots operate in environments with high sediment content and high flow velocities for extended periods, resulting in severe wear and tear on their propeller impellers and sealing components, which affects inspection efficiency and maneuverability.

[0003] Because these robots are distributed across multiple cascade hydropower stations, the maintenance and processing processes are fragmented and lack unified supervision, resulting in significant differences in impeller lifespan, difficulty in achieving expected impeller operating efficiency, and unrepresentative impeller quality monitoring. Therefore, it is necessary to design a data monitoring system and method for maintenance and processing equipment based on a smart factory to address these issues. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a data monitoring system and method for maintenance and processing equipment based on a smart factory, which aims to solve the problem of effective monitoring of impellers working in different cascade hydropower stations after production is completed, and to adaptively update the monitoring strategy.

[0005] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A data monitoring method for maintenance and processing equipment based on a smart factory includes the following steps: S1. Integrate the production of underwater robot impellers in different environments to form a smart factory with unified manufacturing and quality supervision. Extract the inspection reports of the impellers produced by the smart factory in history and analyze them to obtain characteristic objects. S2, Based on the feature object, perform expected tests in actual work, analyze the impeller vibration amplitude and impeller thrust value with one working cycle as the benchmark, calculate the first expected value and the second expected value of impeller operation, and construct a list of expected parameters; S3 quantifies the significant indicators under the current regulatory strategy to form a list of sampling indicators; S4. Based on the expected parameter list and the sampling index list, perform correlation analysis. When the correlation coefficient is less than the preset expected correlation threshold, manually adjust the regulatory strategy until the correlation coefficient of the adjusted regulatory strategy is not less than the preset expected correlation threshold.

[0006] Furthermore, by integrating the production of underwater robot impellers in different environments, a smart factory with unified manufacturing and quality supervision is formed. Inspection reports of the impellers produced in the smart factory throughout history are extracted and analyzed to obtain characteristic objects, specifically: By integrating the production of underwater robot impellers in different environments, a smart factory with unified manufacturing and quality supervision is formed. The inspection reports of the impellers produced in the smart factory in history are extracted. The inspection reports are data analysis reports formed by the pre-shipment inspection of the impeller's form and position tolerances, dynamic balance and mechanical properties. The test report is analyzed, and the impeller is judged based on the design specifications of form and position tolerances, dynamic balance and mechanical properties. Test results within the design specification range are considered to be in normal operation. Impellers operating at different water depths are classified into categories including [a1, a2, ..., a...]. n ]; where a1, a2, ..., a n These represent the maximum water depth at which the first, second, ..., nth impellers operate. Based on the water depth at which different impellers operate, different sealing tests are performed. In this test, compressed air is introduced into the impeller shaft seal, and the impeller is placed at the maximum water depth at which it operates. The number of bubbles generated within the test time T is observed, and impellers with a value less than a preset threshold are considered to have qualified sealing. All impellers that have normal operation and qualified sealing in a number of historical test reports are taken as feature objects, and the test reports corresponding to the feature objects are taken as feature reports. Furthermore, based on the aforementioned characteristic object, expected tests are conducted in actual operation. Using one working cycle as a benchmark, the impeller vibration amplitude and impeller thrust are analyzed to calculate the first and second expected values ​​of impeller operation, and a list of expected parameters is constructed, specifically as follows: Based on the aforementioned feature objects, expected tests are conducted in actual operation. Any feature object x is denoted as gx, and the underwater robot actually operating within a work cycle is obtained from the underwater robot's work log, denoted as [(s1,f1),(s2,f2),…,(st,ft)]; where t is a time marker; s1, s2,…,st represent the impeller vibration amplitude at time points 1, 2,…,t respectively; and f1, f2,…,ft represent the impeller thrust values ​​at time points 1, 2,…,t respectively. Observe t time points within one cycle of the underwater robot's actual operation, and determine whether the impeller vibration amplitude falls within the reasonable operating range at those t time points. The time points within a period are marked, and based on the t' time points marked out of the t time points within that period, the variance of the impeller vibration amplitude is calculated. The variance of the impeller vibration amplitude As the first expected value for impeller operation; where t'≤t; The theoretical variation curve of the impeller thrust value of the underwater robot gx over time is plotted with different time points as independent variables and the theoretical impeller thrust value corresponding to that time point as dependent variable, denoted as s1; the actual variation curve is plotted with the impeller thrust values ​​at the t time points as base data according to the above method, denoted as s2; the similarity dtw between the theoretical variation curve and the actual variation curve is obtained by the curve similarity calculation method, and the similarity dtw is used as the second expected value of impeller operation; Based on the actual water depth at which different underwater robots operate, the impellers corresponding to the underwater robots are clustered, and the average value of the first expected value of the impeller operation in the cluster is recorded as the first water depth parameter. The average value of the second expected value of the impeller operation in the cluster is denoted as the second water depth parameter dtw”; the first water depth parameter and the second water depth parameter corresponding to different clusters are standardized respectively to form a list of expected parameters identified by water depth; Furthermore, the significant indicators under the current regulatory strategy are quantified to form a list of sampling indicators, specifically: Monitoring the impellers that have actually completed production includes: selecting impellers whose inspection results show normal operation and satisfactory sealing; selecting samples according to the current monitoring strategy; wherein, for any selected impeller q, its actual working underwater robot is denoted as gq; and calculating the variance of the vibration amplitude of impeller q according to the method described in step S2. The similarity dtw between the theoretical and actual variation curves of the impeller q is calculated according to the method described in step S3. q ; Based on the actual working water depth of different samples, the number of impellers at the same water depth was collected, and the total number of impellers was denoted as d. The average variance of the vibration amplitude of each of the d impellers was calculated. The average value dtw' of the similarity between the theoretical and actual variation curves; the average value of the variance of the vibration amplitude based on the number of impellers d and the vibration amplitude of the impellers. The average similarity dtw' between the theoretical and actual variation curves is used to calculate the first vibration significance index K1 and the second thrust significance index K2 for impeller samples operating at different water depths; where, , ; The first vibration significance index and the second thrust significance index corresponding to different water depths were standardized and arranged in descending order to form a list of sampling indices identified by water depth.

[0007] Furthermore, based on the list of expected parameters and the list of sampling indicators, a correlation analysis is performed. When the correlation coefficient is less than a preset expected correlation threshold, the regulatory strategy is manually adjusted until the correlation coefficient of the adjusted regulatory strategy is not less than the preset expected correlation threshold. Specifically: Based on the expected parameter list and the sampling index list, list elements with the same water depth identifier in the expected parameter list and the sampling index list are extracted according to the water depth identifier, and arranged in descending order to form a strategy selection list. The correlation coefficient between the sampling index list and the strategy selection list is calculated using the Pearson correlation coefficient method. When the correlation coefficient is less than a preset expected correlation threshold, the regulatory strategy is manually adjusted until the correlation coefficient of the adjusted regulatory strategy is not less than the preset expected correlation threshold.

[0008] A data monitoring system for maintenance and processing equipment based on a smart factory includes an impeller inspection report module, a desired parameter list construction module, a sampling index list construction module, and a strategy update module. The impeller inspection report module is used to integrate the production of underwater robot impellers in different environments, forming a smart factory with unified manufacturing and quality supervision. It extracts the inspection reports of the impellers produced in the history of the smart factory and analyzes them to obtain feature objects. The expected parameter list construction module is used to perform expected tests in actual work based on the feature object, analyze the impeller vibration amplitude and impeller thrust value based on one working cycle, calculate the first expected value and the second expected value of impeller operation, and construct the expected parameter list. The sampling indicator list construction module is used to quantify the significant indicators under the current regulatory strategy and form a sampling indicator list; The strategy update module is used to perform correlation analysis based on the expected parameter list and the sampling indicator list. When the correlation coefficient is less than the preset expected correlation threshold, the regulatory strategy is adjusted manually until the correlation coefficient of the adjusted regulatory strategy is not less than the preset expected correlation threshold.

[0009] Furthermore, the impeller detection report module includes a data acquisition unit and a feature selection unit; The data acquisition unit is used to collect the detection data of the impeller and provides storage function; The feature selection unit is used to select impellers that have normal operation and qualified sealing performance from several historical test reports as feature objects.

[0010] Furthermore, the expected parameter list construction module includes a first expected calculation unit, a second expected calculation unit, and a list construction unit; The first expectation calculation unit is used to calculate the variance of the impeller vibration amplitude and use the variance of the impeller vibration amplitude as the first expectation value of the impeller operation; The second expectation calculation unit is used to obtain the similarity dtw between the theoretical change curve and the actual change curve through the curve similarity calculation method, and to use the similarity dtw as the second expectation value of the impeller operation; The list construction unit is used to construct a list of expected parameters based on the first expected value and the second expected value.

[0011] Furthermore, the strategy update module includes a judgment unit and an update unit; The judgment unit is used to calculate the correlation coefficient between the sampling index list and the strategy selection list using the Pearson correlation coefficient method, and to judge the correlation with the preset expected threshold. The update unit is used to manually adjust the regulatory strategy when the correlation coefficient is less than the preset correlation expectation threshold, until the correlation coefficient corresponding to the adjusted regulatory strategy is not less than the preset correlation expectation threshold.

[0012] The beneficial effects of this invention are: This solution integrates the production of underwater robot impellers under different environments into a smart factory, unifying manufacturing and quality inspection standards. Based on historical inspection reports, it selects impellers that are operating normally and have satisfactory sealing as feature objects, establishing an expected parameter model with impeller vibration amplitude variance and thrust variation curve similarity as dual indicators to comprehensively reflect the impeller's operating status. Combining the sample coverage at different water depths and the overall working level, it quantifies the first vibration significance index and the second thrust significance index, providing data-driven decision-making basis for sampling strategies. Through correlation analysis between expected parameters and sampling indicators, it triggers manual optimization to form a closed-loop regulatory mechanism of detection-evaluation-adjustment, effectively improving detection accuracy and the rationality of sampling strategies, reducing regulatory bias, and possessing adaptability and scalability for detecting key components under multiple operating conditions. Attached Figure Description

[0013] Figure 1 This is a flowchart illustrating the present invention. Detailed Implementation

[0014] Example 1: like Figure 1 As shown, a data monitoring method for maintenance and processing equipment based on a smart factory includes the following steps: S1. Integrate the production of underwater robot impellers in different environments to form a smart factory with unified manufacturing and quality supervision. Extract the inspection reports of the impellers produced by the smart factory in history and analyze them to obtain characteristic objects. Based on the aforementioned characteristic object, expected tests are conducted in actual work. Taking one working cycle as a benchmark, the impeller vibration amplitude and impeller thrust value are analyzed, the first and second expected values ​​of impeller operation are calculated, and a list of expected parameters is constructed. S3 quantifies the significant indicators under the current regulatory strategy to form a list of sampling indicators; S4. Based on the expected parameter list and the sampling index list, perform correlation analysis. When the correlation coefficient is less than the preset expected correlation threshold, manually adjust the regulatory strategy until the correlation coefficient of the adjusted regulatory strategy is not less than the preset expected correlation threshold.

[0015] Furthermore, by integrating the production of underwater robot impellers in different environments, a smart factory with unified manufacturing and quality supervision is formed. Inspection reports of the impellers produced in the smart factory throughout history are extracted and analyzed to obtain characteristic objects, specifically: By integrating the production of underwater robot impellers in different environments, a smart factory with unified manufacturing and quality supervision is formed. The inspection reports of the impellers produced in the smart factory in history are extracted. The inspection reports are data analysis reports formed by the pre-shipment inspection of the impeller's form and position tolerances, dynamic balance and mechanical properties. The test report is analyzed, and the impeller is judged based on the design specifications of form and position tolerances, dynamic balance and mechanical properties. Test results within the design specification range are considered to be in normal operation. Impellers operating at different water depths are classified into categories including [a1, a2, ..., a...]. n ]; where a1, a2, ..., a n These represent the maximum water depth at which the first, second, ..., nth impellers operate. Based on the water depth at which different impellers operate, different sealing tests are performed. In this test, compressed air is introduced into the impeller shaft seal, and the impeller is placed at the maximum water depth at which it operates. The number of bubbles generated within the test time T is observed, and impellers with a value less than a preset threshold are considered to have qualified sealing. All impellers that have normal operation and qualified sealing in a number of historical test reports are taken as feature objects, and the test reports corresponding to the feature objects are taken as feature reports. Furthermore, based on the aforementioned characteristic object, expected tests are conducted in actual operation. Using one working cycle as a benchmark, the impeller vibration amplitude and impeller thrust are analyzed to calculate the first and second expected values ​​of impeller operation, and a list of expected parameters is constructed, specifically as follows: Based on the aforementioned feature objects, expected tests are conducted in actual operation. Any feature object x is denoted as gx, and the underwater robot actually operating within a work cycle is obtained from the underwater robot's work log, denoted as [(s1,f1),(s2,f2),…,(st,ft)]; where t is a time marker; s1, s2,…,st represent the impeller vibration amplitude at time points 1, 2,…,t respectively; and f1, f2,…,ft represent the impeller thrust values ​​at time points 1, 2,…,t respectively. Observe t time points within one cycle of the underwater robot's actual operation, and determine whether the impeller vibration amplitude falls within the reasonable operating range at those t time points. The time points within a period are marked, and based on the t' time points marked out of the t time points within that period, the variance of the impeller vibration amplitude is calculated. The variance of the impeller vibration amplitude As the first expected value for impeller operation; where t'≤t; The theoretical variation curve of the impeller thrust value of the underwater robot gx over time is plotted with different time points as independent variables and the theoretical impeller thrust value corresponding to that time point as dependent variable, denoted as s1; the actual variation curve is plotted with the impeller thrust values ​​at the t time points as base data according to the above method, denoted as s2; the similarity dtw between the theoretical variation curve and the actual variation curve is obtained by the curve similarity calculation method, and the similarity dtw is used as the second expected value of impeller operation; Based on the actual water depth at which different underwater robots operate, the impellers corresponding to the underwater robots are clustered, and the average value of the first expected value of the impeller operation in the cluster is recorded as the first water depth parameter. The average value of the second expected value of the impeller operation in the cluster is denoted as the second water depth parameter dtw”; the first water depth parameter and the second water depth parameter corresponding to different clusters are standardized respectively to form a list of expected parameters identified by water depth; Example 2: The significant indicators under the current regulatory strategy are quantified to form a list of sampling indicators, as follows: Monitoring the impellers that have actually completed production includes: selecting impellers whose inspection results show normal operation and satisfactory sealing; selecting samples according to the current monitoring strategy; wherein, for any selected impeller q, its actual working underwater robot is denoted as gq; and calculating the variance of the vibration amplitude of impeller q according to the method described in step S2. The similarity dtw between the theoretical and actual variation curves of the impeller q is calculated according to the method described in step S3. q ; Based on the actual working water depth of different samples, the number of impellers at the same water depth was collected, and the total number of impellers was denoted as d. The average variance of the vibration amplitude of each of the d impellers was calculated. The average value dtw' of the similarity between the theoretical and actual variation curves; the average value of the variance of the vibration amplitude based on the number of impellers d and the vibration amplitude of the impellers. The average similarity dtw' between the theoretical and actual variation curves is used to calculate the first vibration significance index K1 and the second thrust significance index K2 for impeller samples operating at different water depths; where, , ; The first vibration significance index and the second thrust significance index corresponding to different water depths were standardized and arranged in descending order to form a list of sampling indices identified by water depth.

[0016] Furthermore, based on the list of expected parameters and the list of sampling indicators, a correlation analysis is performed. When the correlation coefficient is less than a preset expected correlation threshold, the regulatory strategy is manually adjusted until the correlation coefficient of the adjusted regulatory strategy is not less than the preset expected correlation threshold. Specifically: Based on the expected parameter list and the sampling index list, list elements with the same water depth identifier in the expected parameter list and the sampling index list are extracted according to the water depth identifier, and arranged in descending order to form a strategy selection list. The correlation coefficient between the sampling index list and the strategy selection list is calculated using the Pearson correlation coefficient method. When the correlation coefficient is less than a preset expected correlation threshold, the regulatory strategy is manually adjusted until the correlation coefficient of the adjusted regulatory strategy is not less than the preset expected correlation threshold.

[0017] In this context, the expected parameter list represents the expected values ​​of impellers operating at different water depths, and the sampling index list represents a monitoring process for the currently completed impeller production. In practice, a fixed number of impellers operating at different water depths need to be monitored. Some companies choose random sampling, while others select sampling based on rules, but the quantity remains constant. The quality of the sampling may be insufficient to measure all impellers produced in this batch, thus requiring adjustments to the monitoring strategy. This application calculates the expected values ​​of the impellers and uses the number of impellers operating at a certain water depth to represent the coverage of that water depth in the sampling, while using the average variance of the impeller vibration amplitude... The average value of the similarity between the theoretical and actual change curves, dtw', reflects the overall working level of the impeller at that water depth. The significance of the regulatory strategy is quantified by the ratio of the coverage scale to the overall working level, which has a certain degree of rationality. The larger the coverage scale, the lower the overall working level, indicating that the impellers working at that water depth should have a higher sampling strategy. In practice, the reduction of the number of impellers sampled in this part will largely lead to deviations in the overall regulation of the impellers.

[0018] Furthermore, the impeller is tested by using the variance of the impeller vibration amplitude and the similarity between the theoretical and actual variation curves as representatives. Since the impeller is working underwater, abnormal vibration often causes local wear, which is an important indicator that can lead to impeller and even underwater robot failures. The thrust is studied by calculating the similarity over a period of time using the dynamic time warping method to ensure that the impeller thrust is not much different from the theoretical thrust. This is a conservative prediction method for impeller blades. In practice, the magnitude of the impeller thrust can also be used to predict the battery.

[0019] Furthermore, the list of expected parameters and the list of sampling indicators are a two-column list, with the list symbol being the water depth and the list elements being the average of the variances of the impeller's vibration amplitude. The average similarity dtw' of the standardized numerical values ​​and the theoretical and actual variation curves is not the same as a general single-column list. Therefore, when calculating the Pearson correlation coefficient, it is necessary to include the average variance of the impeller vibration amplitude. The correlation between the column containing the standardized values ​​and the column containing the standardized elements of the first vibration significance index K1 is calculated. The other column is also calculated accordingly. The two are added together to obtain the correlation coefficient for judgment.

[0020] Example 3: A data monitoring system for maintenance and processing equipment based on a smart factory includes an impeller inspection report module, a desired parameter list construction module, a sampling index list construction module, and a strategy update module. The impeller inspection report module is used to integrate the production of underwater robot impellers in different environments, forming a smart factory with unified manufacturing and quality supervision. It extracts the inspection reports of the impellers produced in the history of the smart factory and analyzes them to obtain feature objects. The expected parameter list construction module is used to perform expected tests in actual work based on the feature object, analyze the impeller vibration amplitude and impeller thrust value based on one working cycle, calculate the first expected value and the second expected value of impeller operation, and construct the expected parameter list. The sampling indicator list construction module is used to quantify the significant indicators under the current regulatory strategy and form a sampling indicator list; The strategy update module is used to perform correlation analysis based on the expected parameter list and the sampling indicator list. When the correlation coefficient is less than the preset expected correlation threshold, the regulatory strategy is adjusted manually until the correlation coefficient of the adjusted regulatory strategy is not less than the preset expected correlation threshold.

[0021] Furthermore, the impeller detection report module includes a data acquisition unit and a feature selection unit; The data acquisition unit is used to collect the detection data of the impeller and provides storage function; The feature selection unit is used to select impellers that have normal operation and qualified sealing performance from several historical test reports as feature objects.

[0022] Furthermore, the expected parameter list construction module includes a first expected calculation unit, a second expected calculation unit, and a list construction unit; The first expectation calculation unit is used to calculate the variance of the impeller vibration amplitude and use the variance of the impeller vibration amplitude as the first expectation value of the impeller operation; The second expectation calculation unit is used to obtain the similarity dtw between the theoretical change curve and the actual change curve through the curve similarity calculation method, and to use the similarity dtw as the second expectation value of the impeller operation; The list construction unit is used to construct a list of expected parameters based on the first expected value and the second expected value.

[0023] Furthermore, the strategy update module includes a judgment unit and an update unit; The judgment unit is used to calculate the correlation coefficient between the sampling index list and the strategy selection list using the Pearson correlation coefficient method, and to judge the correlation with the preset expected threshold. The update unit is used to manually adjust the regulatory strategy when the correlation coefficient is less than the preset correlation expectation threshold, until the correlation coefficient corresponding to the adjusted regulatory strategy is not less than the preset correlation expectation threshold.

Claims

1. A method for monitoring maintenance and processing equipment data in a smart factory, characterized in that, Includes the following steps: S1 integrates the production of underwater robot impellers in different environments to form a smart factory with unified manufacturing and quality supervision. It extracts the inspection reports of the impellers produced in the history of the smart factory and analyzes them to obtain characteristic objects. S2, based on the feature object, perform expected tests in actual work, analyze the impeller vibration amplitude and impeller thrust value with one working cycle as the benchmark, calculate the first and second expected values ​​of impeller operation, and construct a list of expected parameters; S3 quantifies the significant indicators under the current regulatory strategy to form a list of sampling indicators; S4. Based on the list of expected parameters and the list of sampled indicators, perform correlation analysis. When the correlation coefficient is less than the preset expected threshold, the regulatory strategy is adjusted manually until the correlation coefficient of the adjusted regulatory strategy is not less than the preset expected threshold.

2. The method for monitoring maintenance and processing equipment data in a smart factory according to claim 1, characterized in that, Step S1 is as follows: By integrating the production of underwater robot impellers in different environments, a smart factory with unified manufacturing and quality supervision is formed. The inspection reports of the impellers produced in the smart factory in history are extracted. The inspection reports are data analysis reports formed by the pre-shipment inspection of the impeller's form and position tolerances, dynamic balance and mechanical properties. The test report is analyzed, and the impeller is judged based on the design specifications of form and position tolerances, dynamic balance and mechanical properties. Test results within the design specification range are considered to be in normal operation. Impellers operating at different water depths are classified into categories including [a1, a2, ..., a...]. n ]; where a1, a2, ..., a n These represent the maximum water depth at which the first, second, ..., nth impellers operate. Based on the water depth at which different impellers operate, different sealing tests are performed. In this test, compressed air is introduced into the impeller shaft seal, and the impeller is placed at the maximum water depth at which it operates. The number of bubbles generated within the test time T is observed, and impellers with a value less than a preset threshold are considered to have qualified sealing. All impellers that have normal operation and qualified sealing in a number of historical test reports are taken as feature objects, and the test reports corresponding to the feature objects are taken as feature reports.

3. The method for monitoring maintenance and processing equipment data in a smart factory according to claim 2, characterized in that, Step S2 is as follows: Based on the aforementioned feature objects, expected tests are conducted in actual operation. Any feature object x is denoted as gx, and the underwater robot actually operating within a work cycle is obtained from the underwater robot's work log, denoted as: [(s1,f1),(s2,f2),…,(st,ft)]; where t is a time marker; s1, s2,…,st represent the impeller vibration amplitude at time points 1, 2,…,t respectively; and f1, f2,…,ft represent the impeller thrust values ​​at time points 1, 2,…,t respectively. Observe t time points within one cycle of the underwater robot's actual operation, and determine whether the impeller vibration amplitude falls within the reasonable operating range at those t time points. The time points within a period are marked, and based on the t' time points marked out of the t time points within that period, the variance of the impeller vibration amplitude is calculated. The variance of the impeller vibration amplitude As the first expected value for impeller operation; where t'≤t; The theoretical variation curve of the impeller thrust value of the underwater robot gx over time is plotted with different time points as independent variables and the theoretical impeller thrust value corresponding to that time point as dependent variable, denoted as s1; the actual variation curve is plotted with the impeller thrust values ​​at the t time points as base data according to the above method, denoted as s2; the similarity dtw between the theoretical variation curve and the actual variation curve is obtained by the curve similarity calculation method, and the similarity dtw is used as the second expected value of impeller operation; Based on the actual water depth at which different underwater robots operate, the impellers corresponding to the underwater robots are clustered, and the average value of the first expected value of the impeller operation in the cluster is recorded as the first water depth parameter. The average value of the second expected value of the impeller operation in the cluster is denoted as the second water depth parameter dtw”; the first water depth parameter and the second water depth parameter corresponding to different clusters are standardized respectively to form a list of expected parameters identified by water depth.

4. The method for monitoring maintenance and processing equipment data in a smart factory according to claim 3, characterized in that, Step S3 includes monitoring the impellers that have actually been produced, specifically: Impellers with normal operation and satisfactory sealing results are selected. Samples are selected according to the current regulatory strategy. For any selected impeller q, the underwater robot actually operating within it is denoted as gq. The variance of the vibration amplitude of impeller q is calculated using the method described in step S2. The similarity dtw between the theoretical and actual variation curves of the impeller q is calculated according to the method described in step S3. q .

5. A method for monitoring maintenance and processing equipment data in a smart factory according to claim 4, characterized in that, Step S3 also includes: Based on the actual working water depth of different samples, the number of impellers at the same water depth was collected, and the total number of impellers was denoted as d. The average variance of the vibration amplitude of each of the d impellers was calculated. The average similarity dtw' between the theoretical and actual variation curves is used; based on the average variance σ' of the number of impellers d and the vibration amplitude of the impellers, and the average similarity dtw' between the theoretical and actual variation curves, the first vibration significance index K1 and the second thrust significance index K2 for impeller samples working at different water depths are calculated; where, , ; The first vibration significance index and the second thrust significance index corresponding to different water depths were standardized and arranged in descending order to form a list of sampling indicators identified by water depth.

6. A method for monitoring maintenance and processing equipment data in a smart factory according to claim 5, characterized in that, Step S4 is as follows: Based on the expected parameter list and the sampling index list, list elements with the same water depth identifier are extracted from both lists and arranged in descending order to form a strategy selection list. The correlation coefficient between the sampling index list and the strategy selection list is calculated using the Pearson correlation coefficient method. When the correlation coefficient is less than the preset expected threshold, the regulatory strategy is manually adjusted until the correlation coefficient of the adjusted regulatory strategy is not less than the preset expected threshold.

7. A data monitoring system for maintenance and processing equipment based on a smart factory, employing the data monitoring method for maintenance and processing equipment based on a smart factory as described in any one of claims 1-6, characterized in that: This includes an impeller inspection report module, a desired parameter list construction module, a sampling index list construction module, and a strategy update module; The impeller inspection report module is used to integrate the production of underwater robot impellers in different environments, forming a smart factory with unified manufacturing and quality supervision. It extracts inspection reports of impellers produced in the smart factory in history and analyzes them to obtain characteristic objects. The expected parameter list construction module is used to perform expected tests in actual work based on feature objects. It analyzes the impeller vibration amplitude and impeller thrust value based on one working cycle, calculates the first and second expected values ​​of impeller operation, and constructs the expected parameter list. The sampling indicator list construction module is used to quantify the significant indicators under the current regulatory strategy and form a sampling indicator list; The strategy update module is used to perform correlation analysis based on the expected parameter list and the sampled indicator list. When the correlation coefficient is less than the preset expected correlation threshold, the regulatory strategy is adjusted manually until the correlation coefficient of the adjusted regulatory strategy is not less than the preset expected correlation threshold.

8. A data monitoring system for maintenance and processing equipment based on a smart factory, as described in claim 7, is characterized in that... The determination of key parameters for each hydroelectric device in step S2 includes the following steps: The impeller inspection report module includes a data acquisition unit and a feature selection unit; The data acquisition unit is used to collect the detection data of the impeller and provides storage function; The feature selection unit is used to select impellers that have normal operation and qualified sealing performance from several historical test reports as feature objects.

9. A data monitoring system for maintenance and processing equipment based on a smart factory, as described in claim 7, is characterized in that, The expected parameter list construction module includes a first expected calculation unit, a second expected calculation unit, and a list construction unit; The first expected calculation unit is used to calculate the variance of the impeller vibration amplitude, and the variance of the impeller vibration amplitude is used as the first expected value of the impeller operation. The second expectation calculation unit is used to obtain the similarity dtw between the theoretical change curve and the actual change curve through the curve similarity calculation method, and the similarity dtw is used as the second expectation value of the impeller operation. The list building unit is used to construct a list of expected parameters based on the first and second expected values.

10. A data monitoring system for maintenance and processing equipment based on a smart factory, as described in claim 7, is characterized in that... The strategy update module includes a judgment unit and an update unit; The judgment unit is used to calculate the correlation coefficient between the sampled index list and the strategy selection list using the Pearson correlation coefficient method, and to judge the correlation with the preset expected threshold. The update unit is used to manually adjust the regulatory strategy when the correlation coefficient is less than the preset expected correlation threshold, until the correlation coefficient of the adjusted regulatory strategy is not less than the preset expected correlation threshold.

Images