Predictive Maintenance System and Method for Industrial Equipment Based on In-Storage Computing Technology

Abstract

This invention discloses a predictive maintenance system and method for industrial equipment based on in-memory computing technology. The early warning method divides the prediction and maintenance risk control stage in the implementation phase of industrial equipment management prediction and maintenance into several prediction and maintenance parameter assessment risk points. Then, when each prediction and maintenance risk point occurs, its operational progress parameters are compared with the corresponding risk point early warning standard level value. When the operational progress parameters of a prediction and maintenance risk point exceed the corresponding risk point early warning standard level value, a corresponding abnormal early warning prompt is issued. This invention also discloses a corresponding early warning system. The predictive maintenance system and method for industrial equipment based on in-memory computing technology of this invention can provide abnormal early warning prompts for each prediction and maintenance risk control stage before the implementation of prediction and maintenance, thereby improving the prediction and maintenance risk control effect of industrial equipment management prediction and maintenance implementation.

Description

Technical Field

[0001] This invention relates to the field of industrial equipment management, and in particular to a predictive maintenance system and method for industrial equipment based on in-store computing technology. Background Technology

[0002] In factory production, a large number of industrial equipment, such as motors and water pumps, are used. After prolonged operation, these devices are prone to various malfunctions, thus requiring maintenance. Traditional equipment maintenance is usually based on experience, scheduling maintenance on a fixed schedule, making it difficult to predict malfunctions in advance.

[0003] The concept of Computing in Memory (CIM) technology can be traced back to the 1990s. With the development of cloud computing and artificial intelligence (AI) applications in recent years, slow data transfer and high energy consumption have become key bottlenecks in computing centers, especially given the massive data deluge. Retrieving data from external storage units often takes hundreds or thousands of times longer than the computation time, with 60%-90% of the energy consumed in the process being wasted, resulting in very low energy efficiency. The "storage wall" has become a major obstacle to data computing applications. The biggest challenge in accelerating deep learning is the frequent movement of data between computing and storage units. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of existing technologies, this invention provides a predictive maintenance system and method for industrial equipment based on in-memory computing technology.

[0005] The technical solution adopted in this invention is a predictive maintenance method for industrial equipment based on in-memory computing technology. It divides the prediction and maintenance risk control stage in the implementation phase of industrial equipment management prediction and maintenance into several prediction and maintenance parameter assessment risk points. Then, when each prediction and maintenance risk point occurs, its operation progress parameters are compared with the corresponding risk point warning standard level value. When the operation progress parameters of the prediction and maintenance risk point exceed the corresponding risk point warning standard level value, a corresponding abnormal warning prompt is issued.

[0006] Preferably, the specific steps include:

[0007] S1: Collect vibration data of industrial equipment through intelligent vibration sensors, use part of the data as a training set and part as a test set, preprocess the data, including data preprocessing and feature extraction, filter and organize the data, identify working condition information in the data, and remove non-essential variables.

[0008] S2: The processed data is imported into a deep neural network model for training. The trained deep neural network model is stored in an in-memory computing chip. The prediction and maintenance risk control stage in the industrial equipment management prediction and maintenance implementation stage is divided into several prediction and maintenance parameter assessment risk points. Then, based on the progress of industrial equipment management prediction and maintenance, the risk point early warning standard level value of each prediction and maintenance risk point, as well as the prediction and maintenance implementation early warning standard level value of the prediction and maintenance implementation stage are generated.

[0009] S3: Collects real-time vibration data and transmits it to the intelligent data acquisition and analysis box. The integrated storage and computing chip analyzes the data and compares it with the historical database. When a predicted and maintenance risk point occurs, it obtains the operation progress parameters of that predicted and maintenance risk point. Then, it compares the operation progress parameters of that predicted and maintenance risk point with the corresponding risk point warning standard level value. When the operation progress parameters exceed the corresponding risk point warning standard level value, it issues the corresponding abnormal warning prompt and sends the final result to the predictive maintenance management platform.

[0010] S4: During the prediction and maintenance implementation phase, obtain the corresponding operational progress parameters, then compare the operational progress parameters of the prediction and maintenance implementation phase with the corresponding prediction and maintenance implementation early warning standard level value, and issue the corresponding abnormal early warning prompt when the operational progress parameters exceed the corresponding prediction and maintenance implementation early warning standard level value. Based on the content of the predictive maintenance management platform, the staff will analyze the operation status of the industrial equipment and formulate a predictive maintenance plan.

[0011] Preferably, in step S2, a risk assessment level value corresponding to each predicted and maintenance risk point is generated based on the prediction and maintenance progress of industrial equipment management; in step S3, when a predicted and maintenance risk point occurs, the operation progress parameter of the predicted and maintenance risk point is compared with the corresponding risk assessment level value, and a corresponding risk warning is issued when the operation progress parameter is less than the corresponding risk assessment level value.

[0012] Preferably, in step S2, when the operational progress parameter of the previous predicted and maintained risk point is less than its corresponding risk point warning standard level value, the risk point warning standard level value of the next predicted and maintained risk point is adjusted according to the operational progress parameter and risk point warning standard level value of the previous predicted and maintained risk point.

[0013] Preferably, step S4 specifically includes the following steps:

[0014] S41: After the submitted information for the prediction and maintenance implementation phase is approved, obtain the preliminary risk level and progress risk level for the prediction and maintenance implementation phase.

[0015] S42: Compare the preliminary risk level with the corresponding prediction and maintenance implementation early warning standard level. If the preliminary risk level is greater than the prediction and maintenance implementation early warning standard level, issue the corresponding abnormal early warning prompt; otherwise, proceed to the next step.

[0016] S43: Compare the progress risk level with the corresponding prediction and maintenance implementation early warning standard level. If the progress risk level is greater than the prediction and maintenance implementation early warning standard level, issue the corresponding abnormal early warning prompt; otherwise, proceed to the next step.

[0017] S44: Entering the risk control stage of industrial equipment management.

[0018] Preferably, the early warning standard level for prediction and maintenance implementation is 95% of the progress of prediction and maintenance in industrial equipment management.

[0019] Preferably, in step S2, the prediction and maintenance risk control stage in the industrial equipment management prediction and maintenance implementation stage is divided into prediction and maintenance start risk point, prediction and maintenance mid-term risk point, prediction and maintenance modification risk point, and prediction and maintenance completion risk point.

[0020] Preferably, the risk warning standard level for the predicted and maintenance start risk point is 40% of the predicted and maintenance progress level, and the risk assessment level for the predicted and maintenance start risk point is 20% of the predicted and maintenance progress level; the risk warning standard level for the predicted and maintenance mid-term risk point and the predicted and maintenance modification risk point is 70% of the predicted and maintenance progress level, and the risk assessment level for the predicted and maintenance mid-term risk point and the predicted and maintenance modification risk point is 50% of the predicted and maintenance progress level; the risk warning standard level for the predicted and maintenance completion risk point is 80% of the predicted and maintenance progress level, and the risk assessment level for the predicted and maintenance completion risk point is 60% of the predicted and maintenance progress level.

[0021] This invention also discloses a predictive maintenance system for industrial equipment based on in-memory computing technology. It implements the aforementioned method for predictive and maintenance implementation control and abnormal early warning alerts, specifically including:

[0022] The Industrial Equipment Management Prediction and Maintenance Division Module is used to divide the prediction and maintenance risk control stage in the implementation phase of industrial equipment management prediction and maintenance into several prediction and maintenance parameter assessment risk points.

[0023] The industrial equipment management prediction and maintenance risk point monitoring module is used to obtain the operational progress parameters of the predicted and maintenance risk point and the degree of prediction and maintenance progress of the industrial equipment management when the predicted and maintenance risk point occurs.

[0024] The industrial equipment management prediction and maintenance control module is used to input the operational progress parameters of predicted and maintenance risk points into a pre-established prediction and maintenance risk early warning model and obtain the prediction and maintenance risk early warning results output by the prediction and maintenance risk early warning model. The prediction and maintenance risk early warning model generates a risk point early warning standard level value corresponding to the predicted and maintenance risk point based on the degree of prediction and maintenance progress. Then, it compares the operational progress parameters of the predicted and maintenance risk point with the corresponding risk point early warning standard level value. When the operational progress parameters are greater than the corresponding risk point early warning standard level value, it outputs the corresponding prediction and maintenance risk score and issues an abnormal early warning prompt as the prediction and maintenance risk early warning result.

[0025] Preferably, the prediction and maintenance risk warning model is established based on prediction and maintenance parameters of various dimensions and historical prediction and maintenance information; the prediction and maintenance parameters of various dimensions include prediction and maintenance risk points, prediction and maintenance risk point start and end times, prediction and maintenance risk factors, prediction and maintenance human work content, and prediction and maintenance progress; the prediction and maintenance risk warning model calculates the corresponding risk point warning standard level value based on an artificial neural network model.

[0026] Compared with existing technologies, the prediction and maintenance implementation control abnormal early warning method and system of this invention have the following advantages:

[0027] In this invention, the prediction and maintenance risk control stage within the industrial equipment management prediction and maintenance implementation phase is divided into several prediction and maintenance parameter assessment risk points. The operational progress parameters for each prediction and maintenance risk point are controlled, preventing industrial equipment management from exceeding expectations in prediction and maintenance risk control. This, to a certain extent, avoids the problem of excessive workload in prediction and maintenance risk control during the implementation phase, thereby improving the control effectiveness of industrial equipment management prediction and maintenance implementation and contributing to improved industrial equipment management quality. Secondly, based on the progress of prediction and maintenance, and historical data or experience values, this invention pre-calculates and generates early warning standard levels for risk points and prediction and maintenance implementation, resulting in better risk prediction and control effects for each prediction and maintenance risk point and implementation stage, thus improving the control effectiveness of industrial equipment management prediction and maintenance implementation. Attached Figure Description

[0028] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a first flowchart of the predictive maintenance method for industrial equipment based on in-memory computing technology in Embodiment 1 of the present invention;

[0030] Figure 2 This is a second flowchart of the predictive maintenance method for industrial equipment based on in-memory computing technology in Embodiment 1 of the present invention;

[0031] Figure 3 This is a first system structure diagram of the predictive maintenance system for industrial equipment based on in-memory computing technology in Embodiment 3 of the present invention;

[0032] Figure 4 This is a second system structure diagram of the predictive maintenance system for industrial equipment based on in-memory computing technology in Embodiment 3 of the present invention;

[0033] Figure 5 This is a third system structure diagram of the predictive maintenance system for industrial equipment based on in-memory computing technology in Embodiment 3 of the present invention;

[0034] Figure 6 This is the fourth system structure diagram of the predictive maintenance system for industrial equipment based on in-memory computing technology in Embodiment 3 of the present invention. Detailed Implementation

[0035] It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. The following describes the application in further detail with reference to the accompanying drawings and specific embodiments.

[0036] Example 1:

[0037] This embodiment discloses a predictive maintenance method for industrial equipment based on in-memory computing technology.

[0038] A predictive maintenance method for industrial equipment based on in-memory computing technology divides the prediction and maintenance risk control phase of industrial equipment management and prediction and maintenance implementation into several prediction and maintenance parameter assessment risk points. Then, when each predicted and maintenance risk point occurs, its operational progress parameters are compared with the corresponding risk point warning standard level value. When the operational progress parameters of a predicted and maintenance risk point exceed the corresponding risk point warning standard level value, a corresponding abnormal warning prompt is issued. Combined with... Figure 1 As shown, the specific steps include the following:

[0039] S1: Collect vibration data of industrial equipment through intelligent vibration sensors, use part of the data as a training set and part as a test set, preprocess the data, including data preprocessing and feature extraction, filter and organize the data, identify working condition information in the data, and remove non-essential variables.

[0040] S2: The processed data is imported into a deep neural network model for training. The trained deep neural network model is stored in an in-memory computing chip. The prediction and maintenance risk control stage in the industrial equipment management prediction and maintenance implementation stage is divided into several prediction and maintenance parameter assessment risk points. Then, based on the progress of industrial equipment management prediction and maintenance, the risk point early warning standard level value of each prediction and maintenance risk point, as well as the prediction and maintenance implementation early warning standard level value of the prediction and maintenance implementation stage are generated.

[0041] S3: Collects real-time vibration data and transmits it to the intelligent data acquisition and analysis box. The integrated storage and computing chip analyzes the data and compares it with the historical database. When a predicted and maintenance risk point occurs, it obtains the operation progress parameters of that predicted and maintenance risk point. Then, it compares the operation progress parameters of that predicted and maintenance risk point with the corresponding risk point warning standard level value. When the operation progress parameters exceed the corresponding risk point warning standard level value, it issues the corresponding abnormal warning prompt and sends the final result to the predictive maintenance management platform.

[0042] S4: During the prediction and maintenance implementation phase, obtain the corresponding operational progress parameters, then compare the operational progress parameters of the prediction and maintenance implementation phase with the corresponding prediction and maintenance implementation early warning standard level value, and issue the corresponding abnormal early warning prompt when the operational progress parameters exceed the corresponding prediction and maintenance implementation early warning standard level value. Based on the content of the predictive maintenance management platform, the staff will analyze the operation status of the industrial equipment and formulate a predictive maintenance plan.

[0043] In this invention, the prediction and maintenance risk control stage within the industrial equipment management prediction and maintenance implementation phase is divided into several prediction and maintenance parameter assessment risk points. The operational progress parameters for each prediction and maintenance risk point are controlled, preventing industrial equipment management from exceeding expectations in prediction and maintenance risk control. This, to a certain extent, avoids the problem of excessive workload in prediction and maintenance risk control during the implementation phase, thereby improving the control effectiveness of industrial equipment management prediction and maintenance implementation and contributing to improved industrial equipment management quality. Secondly, based on the progress of prediction and maintenance, and historical data or experience values, this invention pre-calculates and generates early warning standard levels for risk points and prediction and maintenance implementation, resulting in better risk prediction and control effects for each prediction and maintenance risk point and implementation stage, thus improving the control effectiveness of industrial equipment management prediction and maintenance implementation.

[0044] In the specific implementation process, in step S2, risk assessment level values ​​corresponding to each predicted and maintenance risk point are generated based on the prediction and maintenance progress of industrial equipment management; in step S2, when a predicted and maintenance risk point occurs, the operation progress parameter of the predicted and maintenance risk point is compared with the corresponding risk assessment level value, and a corresponding risk warning is issued when the operation progress parameter is less than the corresponding risk assessment level value.

[0045] In actual management, to avoid increasing the workload of predicting and maintaining risks, some industrial equipment manufacturers procure equipment and raw materials whose specifications do not conform to preset rules, making it difficult to guarantee the quality of industrial equipment management. Therefore, this invention also sets risk point assessment level values ​​and can issue corresponding risk warning prompts when the operation progress parameters are lower than the corresponding risk point assessment level value. That is, it can issue risk warnings when the operation progress parameters are significantly different from the estimated score, so as to remind relevant personnel to check, thereby helping to improve the quality of industrial equipment management.

[0046] In the specific implementation process, combined with Figure 2 As shown, step S4 specifically includes the following steps:

[0047] S41: After the submitted information for the prediction and maintenance implementation phase is approved, obtain the preliminary risk level and progress risk level for the prediction and maintenance implementation phase; the early warning standard level for prediction and maintenance implementation is 95% of the progress level of prediction and maintenance.

[0048] S42: Compare the preliminary risk level with the corresponding prediction and maintenance implementation early warning standard level. If the preliminary risk level is greater than the prediction and maintenance implementation early warning standard level, issue the corresponding abnormal early warning prompt; otherwise, proceed to the next step.

[0049] S43: Compare the progress risk level with the corresponding prediction and maintenance implementation early warning standard level. If the progress risk level is greater than the prediction and maintenance implementation early warning standard level, issue the corresponding abnormal early warning prompt; otherwise, proceed to the next step.

[0050] S44: Entering the risk control stage of industrial equipment management.

[0051] In the prediction and maintenance implementation phase, this invention first verifies the submitted information for prediction and maintenance implementation, and after the submitted information is approved, it conducts scoring and control of the prediction and maintenance implementation phase, and can issue corresponding abnormal early warning prompts, thereby improving the control effect of prediction and maintenance implementation in industrial equipment management. Secondly, this invention applies two different risk controls to the operational progress parameters in the prediction and maintenance implementation phase, which is conducive to improving the control effect of prediction and maintenance implementation in industrial equipment management.

[0052] In the specific implementation process, in step S2, the prediction and maintenance risk control stage in the industrial equipment management prediction and maintenance implementation stage is divided into prediction and maintenance start risk point, prediction and maintenance mid-term risk point, prediction and maintenance modification risk point, and prediction and maintenance completion risk point.

[0053] In this invention, each prediction and maintenance risk point is divided according to the prediction and maintenance risk control stage of industrial equipment management. This makes the divided prediction and maintenance risk points compatible with the actual prediction and maintenance risk control process of industrial equipment management, thereby improving the control effect of each prediction and maintenance risk point and the control effect of the implementation of prediction and maintenance in industrial equipment management.

[0054] In the specific implementation process, the risk warning standard level for the prediction and maintenance of initial risk points is 40% of the prediction and maintenance progress, and the risk assessment level for the prediction and maintenance of initial risk points is 20% of the prediction and maintenance progress; the risk warning standard level for the prediction and maintenance of intermediate and modified risk points is 70% of the prediction and maintenance progress, and the risk assessment level for the prediction and maintenance of intermediate and modified risk points is 50% of the prediction and maintenance progress; the risk warning standard level for the prediction and maintenance of completed risk points is 80% of the prediction and maintenance progress, and the risk assessment level for the prediction and maintenance of completed risk points is 60% of the prediction and maintenance progress.

[0055] In this invention, by setting the risk warning standard level value and risk assessment level value for each predicted and maintained risk point, the risk prediction and control effect for each predicted and maintained risk point and the prediction and maintenance implementation stage is improved. It can issue risk warnings when the operation progress parameters are significantly inconsistent with the estimated scores, so as to remind relevant personnel to check the materials and risk prediction information. This can improve the control effect of industrial equipment management prediction and maintenance implementation and help improve the quality of industrial equipment management.

[0056] Example 2:

[0057] This embodiment, based on Embodiment 1, discloses a method for updating the standard level value of risk point early warning.

[0058] In step S2 of this embodiment, when the operation progress parameter of the previous predicted and maintained risk point is less than its corresponding risk point warning standard level value, the risk point warning standard level value of the next predicted and maintained risk point is adjusted according to the operation progress parameter of the previous predicted and maintained risk point and the corresponding risk point warning standard level value.

[0059] In this invention, the risk warning standard level value of the next predicted and maintained risk point can be adjusted based on the difference between the operational progress parameters of the previous predicted and maintained risk point and the risk point warning standard level value. This makes the final prediction and maintenance implementation score of the industrial equipment management more consistent with the degree of prediction and maintenance progress, thereby enabling better risk prediction and control of each predicted and maintained risk point and improving the control effect of prediction and maintenance implementation in industrial equipment management.

[0060] For example, the pre-set risk warning standard level for the predicted and maintenance start risk point is 40% of the predicted and maintenance progress; the risk warning standard level for the predicted and maintenance mid-term risk point and the predicted and maintenance modification risk point is 70% of the predicted and maintenance progress; and the risk warning standard level for the predicted and maintenance completion risk point is 80% of the predicted and maintenance progress. Therefore, when the operational progress parameter for the predicted and maintenance start risk point is 25% of the predicted and maintenance progress, the difference between this and the risk warning standard level is 15%. Thus, in this embodiment, the risk warning standard level for the next predicted and maintenance risk point, the predicted and maintenance mid-term risk point, and the predicted and maintenance modification risk point is adjusted to 75% of the predicted and maintenance progress. Conversely, when the operational progress parameter for the predicted and maintenance start risk point is 38% of the predicted and maintenance progress, the difference between this and the risk warning standard level is 2%. Therefore, in this embodiment, the risk warning standard level for the next predicted and maintenance risk point, the predicted and maintenance mid-term risk point, and the predicted and maintenance modification risk point is adjusted to 69% of the predicted and maintenance progress.

[0061] Specifically, when the difference between the operational progress parameters of the current predicted and maintained risk point and the risk point warning standard level is greater than 10%, the risk point warning standard level for the next predicted and maintained risk point will be adjusted upwards; when the difference is less than 5%, the risk point warning standard level for the next predicted and maintained risk point will be adjusted downwards. The specific adjustment values ​​will be determined based on historical data or the corresponding industrial equipment management. This ensures that the final predicted and maintained implementation score for the industrial equipment management more closely matches the degree of predicted and maintained progress.

[0062] Example 3:

[0063] Based on Embodiment 1, this embodiment further discloses an early warning system for predicting and maintaining abnormal implementation and control.

[0064] Combination Figures 3 to 6As shown, the predictive maintenance system for industrial equipment based on in-memory computing technology specifically includes: an industrial equipment management prediction and maintenance division module, which is used to divide the prediction and maintenance risk control stage in the implementation stage of industrial equipment management prediction and maintenance into several prediction and maintenance parameter assessment risk points;

[0065] The industrial equipment management prediction and maintenance risk point monitoring module is used to obtain the operational progress parameters of the predicted and maintenance risk point and the degree of prediction and maintenance progress of the industrial equipment management when the predicted and maintenance risk point occurs.

[0066] The industrial equipment management prediction and maintenance control module is used to input the operation progress parameters of the prediction and maintenance risk points into the pre-established prediction and maintenance risk warning model and obtain the prediction and maintenance risk warning results output by the prediction and maintenance risk warning model.

[0067] The prediction and maintenance risk early warning model generates a risk point early warning standard level value for each predicted and maintained risk point based on the progress of prediction and maintenance. Then, it compares the operational progress parameter of the predicted and maintained risk point with the corresponding risk point early warning standard level value. When the operational progress parameter exceeds the corresponding risk point early warning standard level value, it outputs a corresponding prediction and maintenance risk score and issues an abnormal warning as the prediction and maintenance risk early warning result. In this embodiment, the prediction and maintenance risk early warning model is established based on prediction and maintenance parameters of various dimensions and historical prediction and maintenance information. The prediction and maintenance parameters of each dimension include the predicted and maintained risk point, the start and end time of the predicted and maintained risk point, the predicted and maintained risk factors, the human work content of the prediction and maintenance, and the progress of the prediction and maintenance. The prediction and maintenance risk early warning model calculates the corresponding risk point early warning standard level value using an artificial neural network model.

[0068] In this invention, the prediction and maintenance risk control stage within the industrial equipment management prediction and maintenance implementation phase is divided into several prediction and maintenance parameter assessment risk points through an industrial equipment management prediction and maintenance segmentation module. Combined with an industrial equipment management prediction and maintenance risk point monitoring module and an industrial equipment management prediction and maintenance control module, the operational progress parameters of each prediction and maintenance risk point are controlled. This prevents industrial equipment management from exceeding the estimated prediction and maintenance risks, thereby mitigating the workload issues associated with prediction and maintenance risk control during the implementation phase. This improves the control effectiveness of industrial equipment management prediction and maintenance implementation and contributes to enhancing the quality of industrial equipment management. Secondly, the prediction and maintenance risk early warning model in this invention pre-calculates corresponding risk point early warning standard levels and prediction and maintenance implementation early warning standard levels based on the progress of prediction and maintenance, as well as historical data or experience values. This results in better risk prediction and control for each prediction and maintenance risk point and prediction and maintenance implementation phase, further enhancing the control effectiveness of industrial equipment management prediction and maintenance implementation.

[0069] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0070] The above descriptions are merely embodiments of the present invention. Commonly known structures and characteristics of the solutions are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field to which the invention pertains as of the application date or priority date, are aware of all prior art in that field, and have the ability to apply conventional experimental methods as of that date. Those skilled in the art can, under the guidance of this application, improve and implement this solution in combination with their own capabilities. Some typical known structures or methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention. These should also be considered within the scope of protection of the present invention, and will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A method for predictive maintenance of industrial equipment based on memory computing technology, characterized in that: The prediction and maintenance risk control stage in the implementation stage of industrial equipment management prediction and maintenance is divided into a plurality of prediction and maintenance parameter evaluation risk points, and when each prediction and maintenance risk point occurs, the operation progress parameter thereof is compared with the corresponding risk point early warning standard level value, and when the operation progress parameter of the prediction and maintenance risk point exceeds the corresponding risk point early warning standard level value, a corresponding abnormal early warning prompt is given; The method specifically comprises the following steps: S1: Collect vibration data of the industrial equipment through an intelligent vibration sensor, use a part of the data as a training set and a part as a test set, pre-process the data, including data pre-processing and feature extraction, filter and arrange the data, identify working condition information in the data, and eliminate unimportant variables; S2: Import the processed data into a deep neural network model for training, store the trained deep neural network model in a storage-computing integrated chip, and divide the prediction and maintenance risk control stage in the implementation stage of industrial equipment management prediction and maintenance into a plurality of prediction and maintenance parameter evaluation risk points; then generate risk point early warning standard level values of each prediction and maintenance risk point and a prediction and maintenance implementation early warning standard level value of the implementation stage of the prediction and maintenance according to the prediction and maintenance progress degree of the industrial equipment management; S3: Collect real-time vibration data, transmit the data to an intelligent data collection and analysis box, analyze the data by the storage-computing integrated chip, and compare the data with historical data; when a prediction and maintenance risk point occurs, obtain an operation progress parameter of the prediction and maintenance risk point, compare the operation progress parameter of the prediction and maintenance risk point with the corresponding risk point early warning standard level value, and when the operation progress parameter is greater than the corresponding risk point early warning standard level value, give a corresponding abnormal early warning prompt, and send the final result to a predictive maintenance management platform; S4: Obtain an operation progress parameter of the implementation stage of the prediction and maintenance, compare the operation progress parameter of the implementation stage of the prediction and maintenance with the corresponding prediction and maintenance implementation early warning standard level value, and when the operation progress parameter is greater than the corresponding prediction and maintenance implementation early warning standard level value, give a corresponding abnormal early warning prompt; a worker makes a judgment on the operation of the industrial equipment according to the content of the predictive maintenance management platform, and formulates a predictive maintenance scheme; In step S2, risk point evaluation grade values of each prediction and maintenance risk point are generated according to the prediction and maintenance progress degree of the industrial equipment management; In step S3, when a prediction and maintenance risk point occurs, compare the operation progress parameter of the prediction and maintenance risk point with the corresponding risk point evaluation grade value, and when the operation progress parameter is less than the corresponding risk point evaluation grade value, give a corresponding risk early warning prompt; In step S2, when the operation progress parameter of a previous prediction and maintenance risk point is less than the corresponding risk point early warning standard level value, adjust the risk point early warning standard level value of a next prediction and maintenance risk point according to the operation progress parameter and the risk point early warning standard level value of the previous prediction and maintenance risk point; In step S4, the following steps are specifically included: S41: After the submission information of the prediction and maintenance implementation phase is passed, the preliminary risk level and the progress risk level of the prediction and maintenance implementation phase are obtained; S42: The preliminary risk level is compared with the corresponding prediction and maintenance implementation warning standard level value. If the preliminary risk level is greater than the prediction and maintenance implementation warning standard level value, a corresponding abnormal warning prompt is issued; otherwise, the next step is entered; S43: The progress risk level is compared with the corresponding prediction and maintenance implementation warning standard level value. If the progress risk level is greater than the prediction and maintenance implementation warning standard level value, a corresponding abnormal warning prompt is issued; otherwise, the next step is entered; S44: The industrial equipment management risk control phase is entered; The prediction and maintenance implementation warning standard level value is 95% of the prediction and maintenance progress degree of the industrial equipment management; In step S2, the prediction and maintenance risk control phase in the prediction and maintenance implementation phase of the industrial equipment management is divided into a prediction and maintenance start risk point, a prediction and maintenance mid-term risk point, a prediction and maintenance modification risk point, and a prediction and maintenance completion risk point; The risk point warning standard level value of the prediction and maintenance start risk point is 40% of the prediction and maintenance progress degree, and the risk point evaluation level value of the prediction and maintenance start risk point is 20% of the prediction and maintenance progress degree. The risk point warning standard level value of the prediction and maintenance mid-term risk point and the prediction and maintenance modification risk point is 70% of the prediction and maintenance progress degree, and the risk point evaluation level value of the prediction and maintenance mid-term risk point and the prediction and maintenance modification risk point is 50% of the prediction and maintenance progress degree. The risk point warning standard level value of the prediction and maintenance completion risk point is 80% of the prediction and maintenance progress degree, and the risk point evaluation level value of the prediction and maintenance completion risk point is 60% of the prediction and maintenance progress degree.

2. The predictive maintenance system for industrial equipment based on the memory computing integrated technology according to claim 1, wherein The system is applied to the industrial equipment predictive maintenance method based on the storage-computing integrated technology in claim 1, and specifically comprises: An industrial equipment management prediction and maintenance division module, configured to divide a prediction and maintenance risk control phase in a prediction and maintenance implementation phase of industrial equipment management into a plurality of prediction and maintenance parameter evaluation risk points; An industrial equipment management prediction and maintenance risk point monitoring module, configured to obtain an operation progress parameter of the prediction and maintenance risk point and a prediction and maintenance progress degree of the industrial equipment management when the prediction and maintenance risk point occurs; An industrial equipment management prediction and maintenance control module, configured to input the operation progress parameter of the prediction and maintenance risk point into a prediction and maintenance risk warning model established in advance and obtain a prediction and maintenance risk warning result output by the prediction and maintenance risk warning model; The prediction and maintenance risk early warning model generates a risk point early warning standard level value of a corresponding prediction and maintenance risk point according to a prediction and maintenance progress degree, then compares an operation progress parameter of the prediction and maintenance risk point with the corresponding risk point early warning standard level value, and when the operation progress parameter is greater than the corresponding risk point early warning standard level value, outputs a corresponding prediction and maintenance risk score and issues an abnormal early warning prompt as a prediction and maintenance risk early warning result. 3.The industrial equipment predictive maintenance system based on the computing and storage integrated technology of claim 2, wherein: The prediction and maintenance risk early warning model is established according to prediction and maintenance dimensional parameters and historical prediction and maintenance information; the prediction and maintenance dimensional parameters include a prediction and maintenance risk point, a prediction and maintenance risk point start and end time, a prediction and maintenance risk factor, a prediction and maintenance human work content and a prediction and maintenance progress degree; The prediction and maintenance risk early warning model calculates the corresponding risk point early warning standard level value according to an artificial neural network model.

Images