TWI673723B - Intelligent pre-diagnosis and health management system and method - Google Patents

Abstract

The invention relates to an intelligent pre-diagnosis and health management system and method. The system includes an analysis engine service management module, an intelligent prediction and health management object analysis tree module, a machine learning library module, and a file system module. After the analysis engine service management module defines the analysis tree according to the components of the machine to be monitored, the intelligent prediction and health management object analysis tree module is controlled by the analysis engine service management module to obtain the monitoring data of the machine to be monitored, and The model with the highest similarity is selected among the preset reference hypothesis model sets for modeling, and the model selection and deployment are quickly completed accordingly.

Description

Intelligent pre-diagnosis and health management system and method

The present invention relates to a pre-diagnosis system and method, and more particularly to an intelligent pre-diagnosis and health management system and method that establishes an object analysis tree for machine management and can select a prediction model in an adaptive manner according to the characteristics of the new machine.

In order to ensure the stability of the production process and improve the production rate, the manufacturing industry must carry out strict quality control on the operation status of the production machine.

In order to meet the quality requirements in the conventional technology, critical process parameters are closely monitored and observed. The so-called "key process parameters" refer to the factors most relevant to equipment failure. In practice, these factors will be monitored as important indicators for equipment maintenance pre-diagnosis. In order to improve the accuracy of the pre-diagnosis, many public technologies have been proposed for various improvements, including the applicant's proposal of a leading auxiliary parameter selection method in U.S. Patent Application No. US16 / 001,520 and equipment maintenance pre-combination with key parameters and leading auxiliary parameters. The method of diagnosis is to filter and collect the data collected by the sensor into a set of critical parameters (CP) and other characteristic parameters, and then identify from the characteristic parameter set the earliest time that affects the key parameter as the lead Auxiliary parameters (leading associated parameters (LAP)), and further use the key parameter (CP) set and the leading auxiliary parameters (LAP) to establish an equipment maintenance pre-diagnosis model that effectively enhances early warning capabilities.

In addition, the conventional technology needs to construct a separate feature database for each machine to build a predictive model. In this way, when plural types and heterogeneous machines are introduced into the machine pre-diagnosis and health management system, in addition to increasing the system's In addition to complexity, it will consume a lot of resources and costs.

Therefore, it is necessary to develop an intelligent pre-diagnosis and health management system and method to solve the maintenance and management problems of the pre-diagnosis and health management system that are faced when introducing a large number of production machines of the same type or different types.

The main purpose of the present invention is to solve the shortcomings of the conventional technology that it is difficult to maintain and manage the pre-diagnosis and health management system when introducing a large number of production machines of the same type or different types.

In order to achieve the above object, the present invention provides an intelligent pre-diagnosis and health management system, including: an analytical engine service manager (AESM) module; an intelligent prediction and health management object analysis tree (SMART prognostics and health management object analytics tree (SPHM-OAT) module, a machine learning library module, and a file system module, wherein the intelligent prediction and health management object analysis tree module (SPHM-OAT) is connected to the analysis engine service management Module (AESM) and the intelligent prediction and health management object analysis tree module (SPHM-OAT) includes a plurality of analysis trees (OAT), and each analysis tree includes a plurality of analysis tree nodes (SPHM-object) to obtain a Monitoring data of the machine to be monitored; the machine learning library module is connected to the intelligent prediction and health management object analysis tree module to provide at least one algorithm to the intelligent prediction and health management object analysis tree module (SPHM-OAT); And the file system module is connected to the intelligent prediction and health management object analysis tree module (SPHM-OAT) to provide the intelligent prediction and health management object analysis tree module Hypothesis model and a corresponding reference characteristic of the sample data.

The invention also provides a smart pre-diagnosis and health management method. The method includes a new tree establishment and similarity analysis step, and a modeling step: the new tree establishment and similarity analysis step is defined according to a component of a machine to be monitored. Produce at least one analysis tree (OAT), the analysis tree (OAT) includes a plurality of analysis tree nodes (SPHM-objoct) and a built-in reference hypothesis model of each analysis tree node and one of the corresponding feature sample data storage indicators, according to the Store the index to obtain monitoring data of a machine to be monitored from a file system, and perform a similarity analysis on the monitoring data and the characteristic sample data of the preset reference hypothesis models; and the modeling step is the following steps Either S1 or step S2, where: step S1: when the similarity analysis exceeds a threshold value, a reference hypothesis model with the highest similarity is selected from the preset reference hypothesis models to the monitoring data Perform modeling; step S2: when the similarity analysis does not exceed the threshold value, import an external hypothesis model through an expansion module to model the monitoring data.

Therefore, the effect that the present invention can achieve compared with the conventional technology lies in:

(1) The tree structure of the pre-diagnosis and health management system of the machine to be monitored is reflected through the analysis tree (OAT). Starting from the analysis tree node, the information of the monitoring points of the end component equipment is transmitted upward, and by quantifying each analysis tree The monitoring status of the nodes is analyzed in a recursive manner from bottom to top (bottom up) to gradually analyze the health status of each analytic tree node, and finally collected at the top to form an analysis tree describing the complete single specific equipment health status, and The analysis tree module is composed of a plurality of analysis trees for intelligent prediction and health management objects. The system architecture of the present invention can be universally used in any system machine equipment, which can not only simplify the introduction process of the pre-diagnosis and health management system, but also effectively utilize various computing resources, quickly complete the reference hypothesis model, and complete deployment.

(2) When a new machine is introduced into the intelligent pre-diagnosis and health management system of the present invention, the analysis engine service management module will perform similarity analysis based on the characteristic sample data of the new machine, according to intelligent prediction and health Multiple preset reference hypothesis models in the management object analysis tree module Set indicators and select appropriate reference hypothesis models from the file system module for prediction models to save system management and reference hypothesis model deployment time.

(3) If the imported monitoring data of the new machine and the feature set of the reference hypothesis model preset in the system of the present invention have a similarity lower than the specified threshold value, an external hypothesis model can be imported through the expansion module to Establish a reference hypothesis model in the intelligent prediction and health management object analysis tree module to maintain flexibility and scalability in the modeling process.

10‧‧‧ Smart Pre-diagnosis and Health Management System

20‧‧‧Analysis Engine Service Management Module

30‧‧‧Smart Prediction and Health Management Object Analysis Tree Module

31‧‧‧analysis tree

32, 33, 34‧‧‧‧analysis tree nodes

35‧‧‧ mapping table

36a‧‧‧Data pre-processing layer

36b‧‧‧Data Hypothesis

36c‧‧‧Data overall learning layer

40‧‧‧ Machine Learning Library Module

50‧‧‧File System Module

60a‧‧‧The first exchangeable application interface

60b‧‧‧Second exchangeable application program interface

60c‧‧‧Interchangeable driver interface

70‧‧‧External Machine Learning Module

80‧‧‧ External Reference Model Module

90‧‧‧ external data collection driving device

91‧‧‧external database

S100, S110, S120, S130, S131, S140, S150, S160, S170, S180, S190, S200, S210, S230, S240, S250, S260, S270, S280

CK1 ~ CK5‧‧‧Monitoring points

[Figure 1A] Schematic diagram of the intelligent pre-diagnosis and health management system according to an embodiment of the present invention.

[FIG. 1B] is a schematic diagram of a workflow architecture of an intelligent prediction and health management object analysis tree module according to an embodiment of the present invention.

[Figure 2] Schematic diagram of the operation flow of the intelligent pre-diagnosis and health management system according to an embodiment of the present invention.

[Figure 3] is a schematic diagram of an ecological architecture according to an embodiment of the present invention.

The detailed description and technical contents of the present invention are described below with reference to the drawings. The present invention provides a system architecture design pattern and method for establishing or updating a smart prediction and health management object analysis tree module (SPHM-OAT). ) For equipment health management. The system and method of the present invention can be commonly used for various types of machine equipment such as wind turbines, coal crushers, organic metal chemical vapor deposition systems (MOCVD), plasma assisted chemical vapor deposition systems (PECVD), and the like.

[Figure 1A] is a schematic diagram of the intelligent pre-diagnosis and health management system 10 according to an embodiment of the present invention, which mainly includes an analytic engine service manager (AESM) module 20, an intelligent prediction and analysis of health management objects Tree (SPHM-OAT) module 30, a machine learning library module 40, and a file system module 50. And in order to make the application of the system of the present invention more extensible, the intelligent pre-diagnosis and health management system 10 of the present invention may further include an expansion module, which is connected to the intelligent prediction and health management object analysis tree module 30, and the expansion module may include a first exchangeable application program interface 60a, a second exchangeable application program interface 60b, and an exchangeable driver program interface 60c, wherein the first exchangeable application program interface 60a is used for An external machine learning module 70 is connected, the second exchangeable application program interface 60b is used to connect an external reference model module 80, and the exchangeable driver program interface 60c is used to connect an external data collection driving device (EDCD) 90 to obtain original data set in a database 91 of a machine to be monitored.

The analysis engine service management module 20 is the core of the intelligent pre-diagnosis and health management system 10 of the present invention, and can control and manage the states of various components in the intelligent prediction and health management object analysis tree (SPHM-OAT) module 30.

Please refer to "Figure 1B". The intelligent prediction and health management object analysis tree module 30 is connected to the analysis engine service management module 20 and includes a plurality of analysis trees 31. Each analysis tree 31 includes a plurality of analysis tree nodes 33. , 34, each parse tree node 33, 34 corresponds to a key parameter (CP) and a plurality of related parameters (AP), respectively. The data source of these key parameters (CP) and related parameters (AP) may be information obtained by the sensor, or may be aggregated from the key parameters (CP) of the child nodes and other related parameters (AP). Each analysis tree node 33, 34 is linked to it by an object control table (OCB). The object control tables are used to store the calculation results of the corresponding analysis tree node 33, 34 during the analysis process, and have regular backups. And the effect of reduction. In this way, if a disaster event occurs during the analysis, the object control table can be used to quickly restore the operation. The state of the analytic tree nodes 33 and 34 is obtained from the last checkpoint, and the hierarchical integration operation is continuously performed from the sibling node to the parent node in a recursive manner from bottom to top. The analysis is completed until the analysis tree node (ie, root) at the highest level is completed. Regarding the above sibling nodes and parent nodes, for example, for a certain parse tree node 34, other parse tree nodes 34 are its sibling nodes, and these parse tree nodes 33 are its parent nodes.

According to this, on the premise that the source of the monitoring data is correct and the selection of the key parameters (CP) and related parameters (AP) is correct, the intelligent pre-diagnosis and health management system 10 of the present invention can use these analysis tree nodes 33, 34 Reflect the health status of the analysis tree nodes 33 and 34 in a timely manner, and do a good job of early warning and health management.

The Intelligent Prediction and Health Management Object Analysis Tree (SPHM-OAT) module 30 not only manages the above analysis trees 31 representing the corresponding plural types of machines, but also is responsible for the workflow on the analysis tree nodes 33 and 34 ( workflow) management. The so-called "workflow" is managed by a mapping table 35, and may include a stacked data preprocessing layer 36a, a data hypothesis layer 36b, and a data overall learning layer (data ensemble layer) 36c, the level, sequence and actual work content of the workflow can be adjusted according to requirements, and is not limited to the above.

The mapping table 35 operates through a table driven mechanism, and is selected from the machine learning library module 40 connected to the intelligent prediction and health management object analysis tree module 30 according to a preset working method from the table. At least one appropriate algorithm is used for the above-mentioned workflows such as the data pre-processing layer 36a, the data hypothesis layer 36b, or the data overall learning layer 36c. For example, algorithms applicable to the data pre-processing layer 36a may include algorithms with feature selection capabilities such as feature selection algorithms or feature extraction algorithms; algorithms applicable to the data hypothesis layer 36b Methods can include regression algorithms, autoregressive moving average models Type (autoregressive integrated moving average model (ARIMA) algorithm, relative strength index (RSI) algorithm, or other algorithms with predictive power; the working method of the overall learning layer 36c of this data is to build a group A plurality of reference hypothesis models specified by the mapping table 35 are used for voting, or according to the hierarchical integration operation currently specified by the analysis tree. In addition, the analysis engine service management module 20 also controls the workflow of each analysis tree according to the mechanism of the mapping table 35.

The file system module 50 in this embodiment may be used as a place where the system writes back files and / or stores files. The “files” mentioned above may include, for example, the intelligent prediction and health management object analysis tree module 30. Quantitative analysis information of the life cycle of the analysis tree 31, or the pre-modeling feature sample data set of the reference hypothesis model set, or the backup data when the system fails during the calculation process, or the reference hypothesis to which each analysis tree node belongs To provide the information required by the intelligent prediction and health management object analysis tree module 30 when necessary.

If necessary, the system of the present invention can be expanded by connecting external devices through the expansion module. For example, when the existing data of the machine learning library module 40 is insufficient, the first exchangeable module can be used The application program interface 60a is connected to the external machine learning module 70 to expand the existing machine learning library function; or, the second exchangeable application program interface 60b of the extension module can be connected to the external reference model module 80 to Extend the reference hypothesis model of the mapping table 35 of the intelligent prediction and health management object analysis tree module 30 and participate in the selection and deployment of an external hypothesis model in the manual mode; or by the interchangeable driver of the expansion module The program interface 60c is connected to an external data collection driving device 90, and the external data collection driving device 90 is connected to the external database 91, so the external data stored in the equipment to be monitored can be obtained through the external data collection driving device 90 Source of Library 91.

Please continue to refer to “FIG. 2”, which is a schematic diagram of the operation flow of the intelligent pre-diagnosis and health management system 10 according to an embodiment of the present invention, which mainly includes a new tree establishment and similarity analysis step and a modeling step.

With regard to the new tree establishment and similarity analysis steps, first, a new tree can be manually established, and the related tree construction information is transmitted to the intelligent prediction and health management object analysis tree module 30 through the analysis engine service management module 20 A new parse tree. Secondly, the external data collector 90 is used to further collect the data required for the analysis tree. The data includes the first n pieces of original data of the monitoring points of each end component of the machine to be monitored (5110). The `` manual method '' here refers to the classification of the first-level equipment, second-level equipment, and third-level equipment by the engineering personnel according to the up, down, first, and subsequent subordinate relationships among the components in the machine to be monitored. Based on this, the number of layers is determined to define an analysis tree specific to the ecological architecture of the machine to be monitored.

Then, the analysis engine service management module 20 starts to perform similarity analysis on the collected raw data (S120). The intelligent prediction and health management object analysis tree module 30 first uses the intelligent prediction and health management object analysis tree according to the intelligent prediction and health management object analysis tree. The location specified by the reference hypothesis model index of each analysis tree node in the memory in the module 30 and the storage index of the corresponding feature sample data requests the file system module 50 (S130) and obtains the data of the feature sample of the reference hypothesis model. Matrix (S131), and then check the similarity between the obtained sample features of the machine to be monitored and the sample feature before the reference model hypothesis provided by the file system module 50 (S140). When the similarity exceeds the threshold value and is the highest, the reference hypothesis model is used as a baseline model hypothesis model, and the workflow selected by the baseline model hypothesis model is used as a preset basic workflow (S160).

After the analysis engine service management module 20 receives the workflow-related information transmitted by the intelligent prediction and health management object analysis tree module 30 (S170), it passes the intelligent prediction and health management object analysis tree module 30. The mapping table 35, select the required algorithm from the machine learning library module 40 and complete the automatic modeling setting (S180), and then the intelligent prediction and health management objects The analysis tree module 30 adds hypothetical model indexes and workflows suitable for the machine to be monitored to the mapping table 35 (S190), and simultaneously stores the new reference hypothesis model and feature sample data into the file system module 50. The model migration is completed (S200). Finally, the file system module 50 notifies the intelligent prediction and health management object analysis tree module 30 to update the hypothesis analysis module settings of the new to-be-monitored machine in the mapping table 35, and notifies The analysis engine service management module 20 is transplanted (S210).

If the intelligent prediction and health management object analysis tree module 30 cannot find feature samples with high similarity in the file system module 50, for example, when the first n feature sample data of the machine to be monitored and the existing reference model When the similarity index values between the feature sample data before the hypothesis set modeling are all lower than the specified threshold value, the intelligent prediction and health management object analysis tree module 30 first notifies the analysis engine service management module 20 (S230) , First remind the engineering staff to carry out external expansion instructions. Then, the analysis engine service management module 20 instructs the intelligent prediction and health management object analysis tree module 30 to manually insert appropriate reference hypothesis model indicators, characteristic data set indicators, And the corresponding workflow is set to the intelligent prediction and health management object analysis tree module 30 (S240). Next, the intelligent prediction and health management object analysis tree module 30 calls the algorithms required by the external plug-in workflow from the machine learning library module 40 (S250), completes a manual modeling setting (S260), and then The external plug-in information and modeling information are written back. The reference hypothesis model indicators and feature data set indicators above specify the storage location of the file system module 50 (S270), and notify the analysis engine service management module 20 to complete the hypothesis model expansion ( S280).

The above-mentioned "similarity" is to understand whether there is a hypothesis model suitable for analyzing the machine to be monitored in the reference hypothesis model set preset in the system of the present invention. A specific comparison manner, for example, may compare the distance similarity between the feature set of the hypothetical models preset in the system before modeling and the first n pieces of original data of the machine to be monitored into the same feature space. The smaller the relative distance between the two features, the higher the similarity; otherwise, the lower the similarity. Common Similarity calculation methods can be used such as euclidean distance, mahalanobis distance, manhattan distance, minkowski distance, cosine similarity (cosine similarity) and so on. Through the similarity quantification calculation described above, a suitable hypothesis model can be selected from the preset reference hypothesis model set as the baseline prediction model of the machine to be monitored.

In the following, the system of the present invention is applied as an example to monitor an organic metal chemical vapor deposition (MOCVD) machine. Please refer to "Fig. 3" and match "Fig. 1" and "Fig. 2".

In this example, the relationship between all MOCVD equipment components and an analytic tree node is first defined according to a hierarchical structure. In Figure 3, each analytic tree node corresponds to a key parameter (CP) and a plurality of related parameters (AP). , And one of the designated health indicators (SPHM Health Indicator (SPHM-HI)) timely reflects the health status of each node of the analysis tree, and do a good job of early warning and health management.

The health index (SPHM-HI) is extensible. Examples of basic items may include the Next N-Run Fail (NRF) index, the Remaining Useful Life (RUL) index of key equipment components, General Health Indicators (HI) and other similar health indicators are not known here because their functions, types, actual quantification and analysis methods are well known to those skilled in the art.

Next, the analysis engine service management module 20 branches downward from an analysis tree node 32, and defines an exclusive MOCVD machine according to the up, down, first, and subsequent subordinate relationship between the zero groups in the MOCVD device. An ecological prediction and health management object analysis tree module 30 of Ecological Hierarchy. The root of the tree represents the MOCVD machine (that is, the analysis tree node 32), which is connected to one or several child nodes (that is, the secondary device, the analysis tree node 33), and then continued by the child nodes. Connected to one or several new child nodes (ie, a third-level device, the Analysis tree node 34). Repeatedly connecting in this way, it slowly grows downward like a tree root, thereby forming a complete intelligent prediction and health management object analysis tree module 30 (S100). It is added that only the three-level settings of the first-level device, the second-level device, and the third-level device are described here. However, in other embodiments, the number of layers may be increased or decreased according to actual conditions and requirements. The invention is not limited to this.

Continuing with "Figure 2" and "Figure 1", after the intelligent prediction and health management object analysis tree module 30 is set up, it starts from the leaf node (terminal node (ie, the analysis tree nodes 33, 34) and starts data Collect (S110) and aggregate the data in the external database 91. The leaf nodes (ie, the analysis tree nodes 33 and 34) represent the monitoring status of the end equipment components of the MOCVD machine, and the data sources are the monitoring points CK1, CK2, CK3, CK4, and CK5.

Next, the intelligent prediction and health management object analysis tree module 30 obtains the raw data of the terminal monitoring points from the external database 91 and starts the similarity analysis (S120): first, the intelligent prediction and health management object analysis tree module 30 According to the mapping table 35, a preset reference hypothesis model set is found, and according to the data feature samples (S130 and S131) before the construction of each reference hypothesis model, it is converted into the feature type with the n original data before the MOCVD end monitoring point collection. After the state, the similarity comparison is performed (S140). When the similarity is higher than a specified threshold, the reference hypothesis set with the highest similarity is selected as the baseline prediction model hypothesis (S150).

Then specify the workflow of the baseline prediction model hypothesis, introduce related algorithms (S160, S170, and S180) from the machine learning library module 40, and map the mapping in the intelligent prediction and health management object analysis tree module 30 In Table 35, the above-mentioned most similar reference hypothesis model set index and feature data set index are added to the mapping table 35 of the intelligent prediction and health management object analysis tree module 30, so that the dedicated MOCVD machine can be completed. Modeling settings (S190). Finally, will The transplanted baseline prediction model is stored in the file system module 50 (S200), and the analysis engine service management module 20 is notified to complete the automatic model migration (S210).

However, when the similarity is lower than a specified threshold value, the intelligent prediction and health management object analysis tree module 30 notifies the analysis engine service management module 20 that it cannot find similar feature sample data and corresponding hypothetical models (S230 ). The analysis engine service management module 20 is then connected to the external reference model module 80 through the second exchangeable application program interface 60b of the expansion module, and is manually imported from the external reference model module 80 to the MOCVD machine. And add indicators to the mapping table 35 of the intelligent prediction and health management object analysis tree module 30 (S240), and call the algorithm required for modeling from the machine learning library module 40 (S250). After the modeling is completed, the file system module 50 is written, and the analysis engine service management module 20 is notified to complete the manual model expansion (S280).

Finally, as shown in FIG. 3, when the intelligent pre-diagnosis and health management system 10 of the present invention starts to perform pre-diagnosis analysis on the MOCVD machine, the intelligent prediction and health management object analysis tree module 30 is based on each The workflow specified by the node in the mapping table 35, based on the characteristics of the key parameters (CP) and related parameters (AP), through a hierarchical integrated operation, quantitatively analyzes the health status of each node from the bottom up. , Finally converging to the top (root). The same can also be applied to other machines such as PECVD for pre-diagnosis analysis.

The "tree structure" emphasized by the present invention is a data concept in computer science. According to an embodiment of the present invention, the intelligent prediction and health management object analysis tree module 30 has the following characteristics: (1) A tree with only one highest-level node is called a "root" 32, which can be regarded as a wait. Monitor the status of the top layer of the machine; (2) Each node can derive more than one child node. If all the children derived from all nodes are within two, it is called a binary tree; (3) the bottom-most node at the bottom is called a "leaf" or it can be called a "terminal node" (Such as analysis tree nodes 33, 34), can be regarded as the terminal components of the machine to be monitored, including the data source of the terminal components. Monitoring points CK1, CK2, CK3, CK4, CK5; (4) Many sub-trees that are not connected are called "forests" and can be considered as multiple machines to be monitored at the same time. As can be seen from the above description, the "tree structure" is a hierarchical structure. A new machine to be monitored starts from the "root" and is connected to one or more child nodes (secondary level equipment, such as the analysis tree). Node 33), and the child nodes continue to connect to one or more new child nodes (third level equipment, such as the analysis tree node 34). Repeatedly connecting like this, slowly grows like a tree root to form a complete analysis tree (OAT). The advantage of the tree structure is that the hierarchy is clear and organized, and it can clearly show the subordinate relationship among the components in the machine to be monitored. Therefore, it is suitable for pre-diagnosis and health management of plural types of equipment.

In this way, in addition to reducing the management complexity and labor costs of introducing multiple types and heterogeneous machines into the machine pre-diagnosis and health management system, and maintaining a certain degree of accuracy in the system, further matching the automatic model selection mechanism not only simplifies The introduction process of the pre-diagnosis and health management system can make more efficient use of computing resources and quickly complete the selection and deployment of predictive models.

The present invention has been described in detail above, but the above is only a preferred embodiment of the present invention, and the scope of implementation of the present invention cannot be limited. That is, all equivalent changes and modifications made in accordance with the scope of the application of the present invention should still fall within the scope of the patent of the present invention.

Claims (6)

An intelligent pre-diagnosis and health management system includes: an analysis engine service management module; an intelligent prediction and health management object analysis tree module; the intelligent prediction and health management object analysis tree module is connected to the analysis engine service management module The intelligent prediction and health management object analysis tree module includes a plurality of analysis trees, and each analysis tree includes a plurality of analysis tree nodes to obtain monitoring data of a machine to be monitored. The analysis engine service management module is based on The components of the machine to be monitored define the analysis tree in the intelligent prediction and health management object analysis tree module; a machine learning library module, the machine learning library module is connected to the intelligent prediction and health management object analysis tree module Module to provide at least one algorithm to the intelligent prediction and health management object analysis tree module; and a file system module, the file system module is connected to the intelligent prediction and health management object analysis tree module to provide the intelligent prediction And health management object analysis tree module-a reference hypothesis model and corresponding feature sample data; Measurement and health management object tree analysis module includes a mapping table, which is based on the intelligent analysis and forecasting analysis engine health management service object mapping table management module of the module tree in order to control the workflow of these parse tree nodes. The intelligent pre-diagnosis and health management system described in item 1 of the patent application scope further includes an expansion module, which is connected to the intelligent prediction and health management object analysis tree module, and includes a first exchangeable An application program interface, a second exchangeable application program interface, and an exchangeable driver program interface, wherein the first exchangeable application program interface is used to connect an external machine learning module, and the second exchangeable application program interface is used to An external reference model module is connected, and the interchangeable driver interface is used to connect an external data collection driving device to obtain the original data of a database set in the machine equipment to be monitored. The intelligent pre-diagnosis and health management system described in item 1 of the scope of patent application, wherein each analysis tree node corresponds to a key parameter (CP) and a plurality of related parameters (AP). An intelligent pre-diagnosis and health management method. The method includes a new tree establishment and similarity analysis step, and a modeling step: the new tree establishment and similarity analysis step defines at least one analysis according to a component of a machine to be monitored. Tree, the analysis tree includes a plurality of analysis tree nodes and a built-in reference hypothesis model of each analysis tree node and one of the corresponding feature sample data storage indicators to obtain monitoring data of a machine to be monitored from a file system, and The monitoring data is subjected to a similarity analysis with the preset feature sample data of the reference hypothesis models; and the modeling step is performed in one of the following steps S1 or S2, where: step S1: when the similarity analysis When a threshold value is exceeded, a reference hypothesis model with the highest similarity is selected from the preset reference hypothesis models to model the monitoring data; Step S2: When the similarity analysis does not exceed the threshold value, An external hypothesis model is introduced through an expansion module to model the monitoring data; wherein, in the new tree establishment and similarity analysis steps, the system An analysis engine service management module is used to define the analysis tree in an intelligent prediction and health management object analysis tree module according to the components of the machine to be monitored, and the intelligent prediction and health management object analysis tree module includes a A mapping table that selects at least one algorithm from a machine learning library module connected to the intelligent prediction and health management object analysis tree module to the analysis tree nodes for a workflow management. The intelligent pre-diagnosis and health management method described in item 4 of the scope of patent application, wherein the similarity analysis is performed by converting the first n pieces of original data of the machine to be monitored into preset reference hypothesis models. The feature set before modeling is after the same feature space and the distance similarity is compared. According to the intelligent pre-diagnosis and health management method described in item 4 of the scope of patent application, in the new tree establishment and similarity analysis steps, the monitoring data and the monitoring data are combined with the intelligent prediction and health management object analysis tree module. The similarity analysis is performed on the preset feature sample data of the reference hypothesis models.