A semi-physical production simulation system and method

Abstract

This invention relates to the field of data processing technology, and in particular to a semi-physical production simulation system and method. The method includes the following steps: S1, building a digital modeling platform in a web application, collecting building data, and obtaining a data model and corresponding packaged data for each building data; S2, a model table is provided in the database, the data models are entered into the model table, the data models and packaged data of the same building data are marked accordingly, and icon templates are stored in the database; S3, the web front-end obtains the data models of the relevant data, obtains icon templates with the same number of icons, and obtains the packaged data marked for each relevant data; S4, the web front-end obtains the parsed data, enters the parsed data into the icon templates, and obtains the page charts; S5, the web client loads the page charts into the UE engine for display; this loosens the high coupling between the UE engine and the web client, improving resource utilization.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a semi-physical production simulation system and method. Background Art

[0002] Semi-physical simulation is a simulation technology that combines physical entities and mathematical models. In this technology, part of the simulated object system is introduced into the simulation loop in a physical way, while the rest is described by a mathematical model, and the real-time mathematical simulation and physical simulation are jointly simulated through the simulation calculation model. Specifically, the controller is combined with the simulation model of the control object implemented on the computer for testing. Therefore, in the semi-physical simulation process, the data needs to be processed on the Web client, and the processed data is displayed through the UE engine to obtain the mathematical model.

[0003] The UE engine is used to create video games and other virtual reality applications. Due to its powerful virtual reality functional performance, it is currently widely used in the field of construction engineering. In the field of construction engineering, it is usually necessary to establish a BIM model to display a virtual three-dimensional model of the construction engineering. The BIM model is often combined with the UE engine and the Unity3D engine to publicly display engineering data information that is consistent with the actual situation.

[0004] In the field of construction engineering, information related to engineering projects often has very strong project attributes. Project attributes are special. In different engineering projects, the information that users pay attention to is not the same. Every change in the information shows that the developed products need to be redeveloped and iterated. For the establishment of mathematical models of new products, all the data of new product designs need to be processed by the Web client and displayed through the UE engine. In the design data of new products, there will inevitably be a lot of data that is the same as that of old products. The coupling between the UE engine and the Web client is high, resulting in high data processing repetition and low resource utilization.

[0005] Therefore, there is an urgent need to provide a semi-physical production simulation system and method, which can loosen the high coupling between the UE engine and the Web client and improve resource utilization compared to the existing technology. Summary of the invention

[0006] The present invention solves the technical problems existing in the prior art and provides a semi-physical production simulation system and method.

[0007] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0008] A semi-physical production simulation system and method, comprising the following steps:

[0009] S1. Build a digital model platform in the Web application end, collect building data through the digital model platform, and process the building data to obtain a data model of each building data; package the obtained data model to obtain packaged data of the building data, and store all data models in the digital model platform;

[0010] S2. A model table is provided in the database, and the data model obtained in step S1 is entered into the model table, and the packaged data obtained in step S1 is stored in the database, and the data model and the packaged data of the same building data are marked correspondingly, and the icon template is stored in the database;

[0011] S3. The Web front end obtains the data model of the relevant data through the interface service of the digital model platform, and the relevant data is all or part of the data in the building data; according to the number of relevant data, obtains the icon template with the same number of icon placements; according to the data type of the relevant data, obtains the packaged data corresponding to the mark of each relevant data;

[0012] S4, the Web front end parses and processes the packaged data obtained in step S3 to obtain parsed data, enters the parsed data into the icon template, and obtains a page chart;

[0013] S5. The Web client loads the page chart in step S4 into the UE engine for display through the WebUI plug-in.

[0014] Furthermore, the building data in step S1 includes multiple indicators. Natural language processing technology is used to form each indicator into multiple words, and then each word of each indicator is converted into metadata, which is represented by fields to form a data model for each indicator.

[0015] Furthermore, when the digital modeling platform obtains new indicators, it compares and analyzes the new indicators with the original indicators in the database, uses natural language processing technology to divide each new indicator into multiple words, and compares the words divided into the new indicators with the words divided into the original indicators for similarity. Based on the comparison results, the original indicators are modified to form a data model for the new indicators.

[0016] Furthermore, when there are new fields in the new indicator, that is, when the new indicator obtained by the digital modeling platform contains the same words as those in the original indicator, and all the words in the original indicator are located in the new indicator, the original indicator with the same words as the new indicator will be used as a replacement indicator, and the metadata fields corresponding to all the words in the replacement indicator will be retained. The words in the new indicator that are different from the replacement indicator will be converted into metadata and represented by fields, which will be added to the back of the last field in the replacement indicator to complete the modification of the new indicator.

[0017] Furthermore, when there is a replacement field in the new indicator, that is, when the new indicator obtained by the digital modeling platform contains the same words as those in the original indicator, and only part of the words in the original indicator are located in the new indicator, the original indicator with the same words as the new indicator will be marked as a replacement indicator, and the metadata fields corresponding to the words in the replacement indicator that are the same as those in the new indicator will be retained, and the different words in the new indicator will be replaced with the positions of the different words in the replacement indicator, and the replaced fields will be marked as disabled; the fields marked as disabled will not be obtained by the Web front end, and the fields marked as disabled will not be packaged.

[0018] Furthermore, when there are fields to be deleted in the new indicator, that is, when the new indicator obtained by the digital modeling platform contains the same words as those in the original indicator, and all the words in the new indicator are located in the original indicator, the original indicator with the same words as the new indicator will be marked as a replacement indicator, and the metadata fields corresponding to the words in the replacement indicator that are the same as those in the new indicator will be retained, and the metadata fields corresponding to the words that do not exist in the new indicator will be marked as disabled.

[0019] Furthermore, in step S3, according to the data type of the relevant data, the specific method for obtaining the packaged data corresponding to each relevant data mark is:

[0020] S31, for the first indicator in the related data, obtaining the packaged data corresponding to the mark, and inputting the first indicator into the correlation model to obtain the related indicator having a strong correlation with the first indicator, and the packaged data corresponding to the mark of the related indicator;

[0021] S32, comparing the relevant indicator with other indicators in the relevant data in turn, and when there is an indicator that is the same as the indicator in the relevant data, directly calling the packaged data of the relevant indicator, and subsequently no longer performing the analysis of step S31 on the relevant indicator, and marking the corresponding packaged data as having a correlation with the packaged data of the first indicator;

[0022] S33. Use steps S31-S32 to process the next indicator, and loop steps S31-S33 until the packaged data of all indicators are obtained.

[0023] Furthermore, in step S4, when entering the parsed data into the icon template, the parsed data after parsing the packaged data with correlation marks in step S3 is set at adjacent positions in the icon template and marked with the same color.

[0024] Furthermore, the specific construction method of the correlation model in step S31 is:

[0025] S311, obtaining multiple sets of historical building data;

[0026] S312, construct a machine learning model, input all indicators in a set of historical building data into the machine learning model, perform cluster analysis, specifically use the DBSCAN algorithm, mark the indicators grouped in a cluster as related indicators, and mark each indicator with its related indicators;

[0027] S313, using the method in S312, obtaining each indicator and an indicator having correlation with the indicator;

[0028] S314, dividing each indicator and its marked correlation indicator into a data set, the data set includes a target indicator and a marked indicator, the marked indicator is the target indicator, and the marked correlation indicator is the marked indicator. When the number of marked indicators is greater than a set value, all marked indicators in the data set are set to have a strong correlation with the target indicator;

[0029] The set value satisfies the following formula:

[0030] ;

[0031] In the above formula, L represents the set value, and N represents the total number of indicators in the set of historical building data;

[0032] S315. For the historical building data of other groups obtained in S311, use the methods in steps S312-S314 to train the machine learning model to obtain a correlation model.

[0033] A semi-physical production simulation system executes a semi-physical production simulation method, including a UE engine, a Web client and a database, the Web client includes a digital model platform and a Web front end, the digital model platform interacts with the Web front end, the Web front end interacts with the UE engine and the database respectively, and the digital model platform interacts with the database.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] A Web front end and a digital modeling platform are provided in the Web client of the present invention. The Web front end and the digital modeling platform are highly loosely coupled and have no strong association with each other. The Web front end obtains the data model of the digital modeling platform through an interface service. This process is completed in real time and does not require secondary development of the original code. Real-time data interaction can be completed, and resource utilization is also improved. Moreover, when extracting relevant data, the extraction of relevant data can be completed as soon as possible based on the relevant indicators of the indicators, thereby improving the processing speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a flow chart of the method of the present invention.

[0037] Figure 2 It is a structural diagram of the system of the present invention. DETAILED DESCRIPTION

[0038] The technical solution of the present invention will be clearly described below in conjunction with the accompanying drawings. Obviously, the described embodiments are not all embodiments of the present invention, and all other embodiments obtained by ordinary technicians in the field without making creative work are within the protection scope of the present invention.

[0039] like Figure 1 As shown, the present invention provides a semi-physical production simulation method, comprising the following steps:

[0040] S1. Build a digital modeling platform in the Web application end, collect building data through the digital modeling platform, and process the building data to obtain a data model of each building data; package the obtained data model to obtain packaged data of the building data, the packaged data is in JSON format, and all data models are stored in the digital modeling platform.

[0041] Building data includes multiple indicators, including equipment name, online time, last offline time, and last maintenance time. Natural language processing technology is used to form multiple words for each indicator, and then each word of each indicator is converted into metadata and represented in the form of fields to form a data model for each indicator.

[0042] When the digital model platform obtains new indicators, the new indicators are compared and analyzed with the original indicators in the database. Natural language processing technology is used to divide each new indicator obtained by the digital model platform into multiple words, and the words divided into the new indicators are compared with the words divided into the original indicators through natural language processing technology for similarity, and then divided into the following situations for discussion:

[0043] (1) When there are new fields in the new indicator, that is, when the new indicator obtained by the digital modeling platform contains the same words as those in the original indicator, and all the words in the original indicator are located in the new indicator, the original indicator with the same words as the new indicator is marked as the replacement indicator, and the metadata fields corresponding to all the words in the replacement indicator are retained. The words in the new indicator that are different from the replacement indicator are converted into metadata and represented by fields, which are added to the end of the last field in the replacement indicator to complete the modification of the new indicator.

[0044] (2) When there is a replacement field in the new indicator, that is, when the new indicator obtained by the digital modeling platform contains the same words as those in the original indicator, and the words in the original indicator are only partially located in the new indicator, the original indicator with the same words as the new indicator will be marked as the replacement indicator, and the metadata fields corresponding to the words in the replacement indicator that are the same as the new indicator will be retained. The different words in the new indicator will be replaced with the positions of the different words in the replacement indicator, and the replaced fields will be marked as disabled; the fields marked as disabled will not be obtained by the Web front end, and the fields marked as disabled will not be packaged.

[0045] (3) When there are fields to be deleted in the new indicator, that is, when the new indicator obtained by the digital modeling platform contains the same words as those in the original indicator, and all the words in the new indicator are located in the original indicator, the original indicator with the same words as the new indicator will be marked as a replacement indicator, and the metadata fields corresponding to the words in the replacement indicator that are the same as those in the new indicator will be retained, and the metadata fields corresponding to the words that do not exist in the new indicator will be marked as disabled; the fields marked as disabled will not be obtained by the Web front end, and the fields marked as disabled will not be packaged.

[0046] S2. A model table is provided in the database. The data model obtained in step S1 is entered into the model table. The packaged data obtained in step S1 is stored in the database. The data model and packaged data of the same building data are marked correspondingly. Icon templates are stored in the database. There are multiple types of icon templates. The icon templates are classified according to the number of icons placed.

[0047] S3. The Web front end obtains the data model of the relevant data through the interface service of the digital model platform. The relevant data is all or part of the data in the building data; according to the number of relevant data, an icon template with the same number of icon placements is obtained; according to the data type of the relevant data, the packaged data corresponding to each relevant data is obtained.

[0048] According to the data type of the relevant data, the specific method for obtaining the packaged data corresponding to each relevant data mark is:

[0049] S31. For the first indicator in the relevant data, obtain the corresponding marked packaged data, and input the first indicator into the correlation model to obtain the relevant indicator with strong correlation with the first indicator, and the corresponding marked packaged data of the relevant indicator. The specific construction method of the correlation model is:

[0050] S311. Obtain multiple sets of historical building data.

[0051] S312. Build a machine learning model, input all indicators in a set of historical building data into the machine learning model, and perform cluster analysis. Specifically, use the DBSCAN algorithm to mark the indicators grouped in a cluster as correlated indicators, and mark each indicator with the indicators that are correlated with it.

[0052] S313. Using the method in S312, obtain each indicator and an indicator that is correlated with the indicator.

[0053] S314. Divide each indicator and its marked correlation indicator into a data set, the data set includes a target indicator and a marked indicator, the marked indicator is the target indicator, the marked correlation indicator is the marked indicator, and when the number of marked indicators is greater than the set value, all marked indicators in the data set are set to have a strong correlation with the target indicator.

[0054] The set value satisfies the following formula:

[0055] ;

[0056] In the above formula, L represents the set value, and N represents the total number of indicators in this group of historical building data.

[0057] S315. For the historical building data of other groups obtained in S311, use the methods in steps S312-S314 to train the machine learning model to obtain a correlation model.

[0058] S32. Compare the relevant indicators obtained in step S31 with other indicators in the relevant data in turn. When there is an indicator that is the same as the indicator in the relevant data, directly call the packaged data of the relevant indicator, and the relevant indicator will no longer be analyzed in step S31 later, and the corresponding packaged data will be marked as having a correlation with the packaged data of the first indicator.

[0059] S33. Use steps S31-S32 to process the next indicator, and loop steps S31-S33 until the packaged data of all indicators are obtained.

[0060] S4, the Web front end parses the packaged data obtained in step S3 to obtain parsed data, and enters the parsed data into the icon template to obtain a page chart. The process of parsing data is to extract data items according to the JSON data structure of the packaged data. Since JSON supports nested structures, parsing data requires recursively searching for nested sub-data, extracting the required specific data items from the parsed data, and obtaining the parsed data.

[0061] When entering the parsed data into the icon template, the parsed data after parsing the packaged data with correlation marks in step S3 is set at adjacent positions in the icon template and marked with the same color.

[0062] S5. The Web client loads the page chart in step S4 into the UE engine for display through the WebUI plug-in.

[0063] like Figure 2 As shown, the present invention also provides a semi-physical production simulation system, including a UE engine, a Web client and a database, the Web client includes a digital model platform and a Web front end, the digital model platform and the Web front end interact with each other, the Web front end interacts with the UE engine and the database respectively, and the digital model platform interacts with the database.

[0064] A Web front end and a digital modeling platform are provided in the Web client of the present invention. The Web front end and the digital modeling platform are highly loosely coupled and have no strong association with each other. The Web front end obtains the data model of the digital modeling platform through an interface service. This process is completed in real time and does not require secondary development of the original code. Real-time data interaction can be completed, and resource utilization is also improved. Moreover, when extracting relevant data, the extraction of relevant data can be completed as soon as possible based on the relevant indicators of the indicators, thereby improving the processing speed.

[0065] Finally, it should be noted that the above content is only used to illustrate the technical solution of the present invention, rather than to limit the scope of protection of the present invention. Simple modifications or equivalent substitutions of the technical solution of the present invention by ordinary technicians in this field do not deviate from the essence and scope of the technical solution of the present invention.

Claims

1. A semi-physical production simulation method, characterized in that: The following steps are involved: S1. Build a digital model platform in the Web application end, collect building data through the digital model platform, the building data includes multiple indicators, and process the building data to obtain a data model of each building data; package the obtained data model to obtain packaged data of the building data, and store all data models in the digital model platform; S2. A model table is provided in the database, and the data model obtained in step S1 is entered into the model table, and the packaged data obtained in step S1 is stored in the database, and the data model and the packaged data of the same building data are marked correspondingly, and the icon template is stored in the database; S3. The Web front end obtains the data model of the relevant data through the interface service of the digital model platform. The relevant data is all or part of the data in the building data; according to the number of relevant data, an icon template with the same number of icon placements is obtained; according to the data type of the relevant data, the packaged data corresponding to each relevant data is obtained; according to the data type of the relevant data, the specific method for obtaining the packaged data corresponding to each relevant data is as follows: S31. For the first indicator in the relevant data, obtain the packaged data corresponding to the mark, and input the first indicator into the correlation model to obtain the relevant indicator with a strong correlation with the first indicator, and the packaged data corresponding to the mark of the relevant indicator; the specific construction method of the correlation model is: S311, obtaining multiple sets of historical building data; S312, construct a machine learning model, input all indicators in a set of historical building data into the machine learning model, perform cluster analysis, specifically use the DBSCAN algorithm, mark the indicators grouped in a cluster as related indicators, and mark each indicator with its related indicators; S313, using the method in S312, obtaining each indicator and an indicator having correlation with the indicator; S314, dividing each indicator and its marked correlation indicator into a data set, the data set includes a target indicator and a marked indicator, the marked indicator is the target indicator, and the marked correlation indicator is the marked indicator. When the number of marked indicators is greater than a set value, all marked indicators in the data set are set to have a strong correlation with the target indicator; The set value satisfies the following formula: ； In the above formula, L represents the set value, and N represents the total number of indicators in the set of historical building data; S315, for the historical building data of other groups obtained in S311, use the method in steps S312-S314 to train the machine learning model, so as to obtain a correlation model; S32, comparing the relevant indicator with other indicators in the relevant data in turn, and when there is an indicator that is the same as the indicator in the relevant data, directly calling the packaged data of the relevant indicator, and subsequently no longer performing the analysis of step S31 on the relevant indicator, and marking the corresponding packaged data as having a correlation with the packaged data of the first indicator; S33, using steps S31-S32 to process the next indicator, and looping steps S31-S33 until the packaged data of all indicators are obtained; S4, the Web front end parses and processes the packaged data obtained in step S3 to obtain parsed data, enters the parsed data into the icon template, and obtains a page chart; S5. The Web client loads the page chart in step S4 into the UE engine for display through the WebUI plug-in.

2. A semi-physical production simulation method according to claim 1, characterized in that: In step S1, natural language processing technology is used to form multiple words for each indicator, and then each word of each indicator is converted into metadata, represented by fields, to form a data model for each indicator.

3. A semi-physical production simulation method according to claim 2, characterized in that: When the digital model platform obtains new indicators, it compares and analyzes the new indicators with the original indicators in the database. It uses natural language processing technology to divide each new indicator into multiple words, and compares the words that the new indicators are divided into with the words that the original indicators are divided into for similarity. Based on the comparison results, the original indicators are modified to form a data model for the new indicators.

4. A semi-physical production simulation method according to claim 3, characterized in that: When there are new fields in the new indicator, that is, when the new indicator obtained by the digital modeling platform contains the same words as those in the original indicator, and all the words in the original indicator are located in the new indicator, the original indicator with the same words as the new indicator will be used as a replacement indicator, and the metadata fields corresponding to all the words in the replacement indicator will be retained. The words in the new indicator that are different from the replacement indicator will be converted into metadata and represented by fields, which will be added to the back of the last field in the replacement indicator to complete the modification of the new indicator.

5. A semi-physical production simulation method according to claim 3, characterized in that: When there is a replacement field in the new indicator, that is, when the new indicator obtained by the digital modeling platform contains the same words as those in the original indicator, and the words in the original indicator are only partially located in the new indicator, the original indicator with the same words as the new indicator is marked as the replacement indicator, and the metadata fields corresponding to the words in the replacement indicator that are the same as the new indicator are retained, and the different words in the new indicator are replaced with the positions of the words in the replacement indicator that are different from the new indicator, and the replaced fields are marked as disabled; Fields marked as disabled will not be obtained by the Web front end, and no packaging operation will be performed on the fields marked as disabled.

6. A semi-physical production simulation method according to claim 4, characterized in that: When there are fields to be deleted in the new indicator, that is, when the new indicator obtained by the digital modeling platform contains the same words as those in the original indicator, and all the words in the new indicator are located in the original indicator, the original indicator with the same words as the new indicator will be marked as a replacement indicator, and the metadata fields corresponding to the words in the replacement indicator that are the same as those in the new indicator will be retained, and the metadata fields corresponding to the words that do not exist in the new indicator will be marked as disabled.

7. A semi-physical production simulation method according to claim 1, characterized in that: In step S4, when entering the parsed data into the icon template, the parsed data after parsing the packaged data with correlation marks in step S3 is set at adjacent positions in the icon template and marked with the same color.

8. A semi-physical production simulation system, characterized in that: A semi-physical production simulation method according to any one of claims 1 to 7 is implemented, comprising a UE engine, a Web client and a database, wherein the Web client comprises a digital model platform and a Web front end, the digital model platform interacts with the Web front end, the Web front end interacts with the UE engine and the database respectively, and the digital model platform interacts with the database.

Images