CN106446278B

CN106446278B - A kind of search method to data object based on space-time database

Info

Publication number: CN106446278B
Application number: CN201610926060.5A
Authority: CN
Inventors: 林伟
Original assignee: Beijing Asiacontrol Technology Development Co Ltd
Current assignee: Beijing Asiacontrol Technology Development Co Ltd
Priority date: 2016-10-24
Filing date: 2016-10-24
Publication date: 2018-07-20
Anticipated expiration: 2036-10-24
Also published as: CN106446278A; CN108874988A; CN108984611B; CN108984611A; CN109033146A; CN108874988B

Abstract

The search method to data object based on space-time database that the present invention relates to a kind of, which is characterized in that the search method includes the following steps：Time state and spatiality for object to be managed model the management object；Administrative model according to object to be managed sets the particular community by natural language description of the object to be managed；It is retrieved with the operating status of the determination management object based on the administrative model classification for managing object and by the time attribute of natural language description and/or space attribute defined by the modelling.The space-time database of the present invention describes the data information of production monitoring management by history, real-time and three times of plan.User is not required to master computer language, it is only necessary to which the space-time data that can be understood management object by retrieving using space-time metalanguage is understood the space-time operating status of management object, not only saved computer content but also convenient for management.

Description

Retrieval method for data object based on space-time database

Technical Field

The invention relates to the field of computer retrieval, in particular to a data object retrieval method based on a time-space database.

Background

In the research results at home and abroad, most of the time-space database models are currently in the theoretical research process, and the actual realized prototype system is few. The main current time-space database models mainly include: a sequence snapshot model, a space-time cube model, a ground state correction model, a space-time composite model, an object-oriented space-time model, and the like. These models have different properties and advantages and disadvantages, but none of them meet the requirements well for the statistical industry. For example, the sequence snapshot model is composed of a series of time slices, each time slice represents a map state at different time, and it can also be understood that a group of map sets with time concept, each map has a time attribute, and the disadvantage is that it only records the state of each time slice, and for expressing the change situation of a certain time period, the presentation mode is not direct, and the data redundancy is large; the space-time cube model is composed of two space dimensions and a time dimension, and the evolution mode of the two-dimensional space can be found by analyzing along the time dimension, so that the defect that the operation of the cube becomes extremely complex along with the increase of years is overcome; in order to avoid repeated recording on each time node, the ground state modification model defines a space state at a certain time point as a ground state, namely an original state, and then records a space change part at a certain time frequency, and has the defect that the operation is complex for the change condition before the ground state is obtained; still other concepts of spatio-temporal database models, such as object-oriented concept, feature-based concept, event-based concept, etc., are all recorded by object-oriented or delta-recording, and have disadvantages that they can not satisfy more or less complicated indexes and many statistical levels of the statistical spatio-temporal database, and can not satisfy the needs of the statistical department in terms of business applicability.

The chinese patent CN103678712A discloses a disaster information spatiotemporal database, which comprises three disaster information databases, namely a disaster situation database, a disaster process database and a disaster history database, and a unified coding module hierarchically codes the received disaster information data; the attribute data management module and the spatial position data management module import disaster information attribute data and spatial position data to corresponding disaster information databases; disaster information data are transmitted among disaster information databases through a logic change module; the time-space database index module establishes an update index for the disaster information time-space database according to the time sequence; the logic change module and the time-space database index module form the basis of a disaster information time-space database, manage and maintain disaster attributes and change and maintain the time-space logic index, and prepare for warehousing, retrieval and time-space query of statistical data. The patent solves the problems of high data redundancy and low working efficiency when the spatio-temporal data storage, management and history review are realized. However, the patent has problems that: (1) the relational database is SQL or oracle, can only carry out classified retrieval through codes, and cannot carry out retrieval query through directly inputting time or space information (the quick retrieval of the historical database can only be carried out through the time information and cannot be carried out through the space information), so that the query and the calling of the database are inconvenient; (2) the search method is single, and the search can be performed only in a certain time range, but not in a certain space range, for example, all disaster information occurring in the pacific region cannot be searched.

At present, many spatial databases identify spatial information by using storage coordinate information to support the spatial information, and using a self-increment ID (Identity) to provide an index, and the search needs to be implemented by means of a spatial structure algorithm. In order to solve the above problems, many patents now use a simpler spatial index technique to simplify the field structure of the spatial database and save the occupied storage space.

Chinese patent CN102622349B discloses a method for processing a spatial location information database, which is characterized by comprising: acquiring coordinate data of a spatial position; generating a spatial position information code corresponding to the spatial position according to the coordinate data, including: space position information codes implemented in the Chinese range are defined to be divided into five sections of codes, namely country codes-area codes: and a sub-code 1: the subcode 2-additional code, the coding method principle of each section of code is that the codes are numbered from top to bottom and from left to right; and storing the spatial position information code in a spatial position information database, wherein the spatial position information code is used as the index and the position information of the spatial position in the database. In the scheme, the spatial position information code is used for replacing various data information such as database indexes, position information, permanent IDs and the like which are widely used at present, so that not only can tens of millions of spatial position information databases be conveniently managed and searched, but also the space of the databases is saved. However, the patent has problems that: (1) the accuracy and precision of the spatial description are not high enough, for example, the upper part and the bottom part of a certain device in a factory cannot be accurately and separately described; (2) the query and the call of the space analysis and the coding to the data are inconvenient, for example, the working condition of a factory boiler is checked, the coding corresponding to the boiler needs to be found, and the query cannot be directly carried out; (3) the database is not dynamically updated when the physical location update changes.

A relational database is a shareable, organized collection of relational data stored on a computer. Relational data is data represented by a relational digital model in which data is described in the form of a two-dimensional table. In an application system using a relational database to store information, there are a large number of multidimensional queries, which provide query conditions of various dimensions for a user to input, and at the same time, the user needs to retrieve required information simply, quickly, and intelligently. The optimization method for database query mainly comprises the following steps: (1) reasonably utilizing the index: for a data table in a relational database, an independent and ordered storage structure is created according to a queried field, and a directory is created for books similarly, so that space is exchanged for time, and query performance is improved; (2) redundancy relationship data: the data structure design in the relational database is required to follow certain specifications to ensure the integrity and consistency of data, reverse normalization is properly adopted, and information in other related tables is stored in a two-dimensional table in a redundant mode to reduce incidence relation during query and improve query performance; (3) and (4) separating and storing mass data: for mass data, the mass data is classified and independently stored according to certain data, for example, telephone number information is respectively stored according to the affiliated areas, so that the service logic complexity is increased, the design difficulty of an application program and the data maintenance difficulty are improved, the query range is narrowed, and the query performance can be improved.

Chinese patent CN100483411C discloses an information retrieval method in a relational database, comprising the steps of: constructing a dimensionality reduction strategy tree and placing the dimensionality reduction strategy tree in a database system, wherein the dimensionality reduction strategy tree comprises at least one child node and at least one root node, and each child node at least comprises the node number identification information, inquiry condition combination information and a lower-level child node number; when the database is searched according to the user query condition and the required data is not obtained, sequentially constructing new query conditions according to the dimension reduction strategy tree; and searching the database according to the new query condition until required data is obtained or querying a root node of the dimension reduction strategy tree to return information without the required data. The patent also discloses an information retrieval device in a relational database, comprising: the device comprises a query condition acquisition unit, a query result output unit, a strategy tree storage unit and a query operation unit. By using the method, the database retrieval efficiency can be improved. However, the patent has problems that: specific number identification information is needed when data is inquired and called, and in some fields, such as the field of industrial process control, the types and the quantity of equipment and process parameters are large, so that the data is inconvenient to inquire by using the number. This is a problem that always exists in conventional relational databases — each object must retrieve the corresponding data through a unique corresponding identification code. If a method is available to search directly from the location of the device or the time point of the process parameter, the search efficiency will be much improved.

Therefore, the technology of the application software or database commonly used at present, such as Oracle in Oracle, Microsoft SQL, Access, etc., is based on the laggard hardware environment of the 80 th of the 20 th century, and no high-resolution display, scanner, etc. is yet popularized. The establishment of the database structure must be limited by the character terminal, and the database structure is described by words. Despite the decades of improvement, the current situation of database structure building has not changed. All database structures must be built through a series of intricate definitions or assignments. The input or output of data must be programmed by a programmer in a programming language. The maintenance of the database requires that the original programmer provide the source code to maintain or upgrade the software. A specific identification must be entered when querying or calling database data. The traditional database establishment mode is used today, but the huge expenditure thereof causes the market to need a more convenient, efficient and low-cost database system. There is no database that can improve the functions of these databases and that can be configured.

Disclosure of Invention

Aiming at the defects of the prior art, the invention provides a retrieval method for data objects based on a space-time database, which is characterized by comprising the following steps:

modeling the management object according to the time state and the space state of the object to be managed;

setting specific attributes of the object to be managed, which are described by means of natural language, according to a management model of the object to be managed;

retrieving to determine an operational state of a management object based on the management model category of the management object and temporal and/or spatial attributes defined by the modeling by means of natural language.

According to a preferred embodiment, the step of retrieving for determining the operating state of the management object based on the management model category of the management object and the temporal and/or spatial attributes defined by the modeling by means of natural language comprises:

retrieving real-time data, historical data and/or planning data of at least one management object at the spatial location and/or the time based on the management model category, spatial location and/or time of the management object, thereby determining a real-time operational state, a historical operational state and/or a planned operational state of at least one of the management objects; or

Retrieving real-time data, historical data, and/or planning data of at least one management object within a spatial range and/or a temporal range based on the management model category, the spatial range, and/or the temporal range of the management object to determine a real-time operational state, a historical operational state, and/or a planned operational state of at least one of the management objects.

According to a preferred embodiment, the step of retrieving for determining the operating state of the management object based on the management model category of the management object and the temporal and/or spatial attributes defined by the modeling by means of natural language further comprises:

determining a real-time database, a historical database and/or a plan database for storing the operation data of the management object according to the type of the management model of the management object,

selecting at least one version of the management object operational data in the real-time database, the historical database and/or the planning database related to the management object according to the time attribute and/or the space attribute, and

and determining at least one management object associated with the management model, and viewing real-time data, historical data and/or plan data of the management object so as to know the real-time operation state, the historical operation state and/or the plan operation state of the management object.

According to a preferred embodiment, the management objects comprise real-time data objects stored in the real-time database, historical data objects stored in the historical database and plan data objects stored in the plan database, which are formed by classification according to time attributes,

the real-time data object comprises at least one management object formed by real-time data of the management object through the real-time data model instantiation process,

the historical data object comprises at least one management object formed by the historical data instantiation process of the historical data model and composed of historical data of the management object,

the planning data object comprises at least one management object formed by the planning data model instantiation process and composed of the planning data of the management object.

According to a preferred embodiment, the step of modelling the objects to be managed with respect to their temporal and spatial state comprises;

modeling the space of the object to be managed according to the space state of the object to be managed to establish a space model,

modeling the time of the object to be managed according to the time state of the object to be managed to establish a time model, an

Establishing the management model formed by the association of the space model and the time model of the object to be managed.

According to a preferred embodiment, the step of setting specific attributes of the object to be managed, described by means of a natural language, according to a management model of the object to be managed comprises:

the management model forms at least one management object formed by the association of the time object and the space object of the management object through an instantiation process,

the time object of the management object is formed by the time model in the instantiation process, and the space object of the management object is formed by the space model in the instantiation process.

According to a preferred embodiment, the step of setting the specific attribute described by the natural language of the object to be managed according to the management model of the object to be managed further comprises:

setting specific attributes of the management object, the specific attributes including temporal attributes, spatial attributes and/or non-spatiotemporal attributes described by means of a natural language,

the non-spatiotemporal attributes include one or more of graphical attributes, audio attributes, video attributes, and name attributes of the object to be managed.

According to a preferred embodiment, said time attributes comprise at least a time position, a start time and an end time,

when the management object is a static object, the spatial attributes at least comprise a geographic spatial range, a geographic spatial position and a spatial shape, a spatial range and a spatial position of a subspace,

and when the management object is a dynamic object, the spatial attributes at least comprise a geographic spatial position, a candidate geographic spatial position, and a spatial shape, a spatial range and a spatial position of a previous-level spatial model of the object to be managed.

According to a preferred embodiment, the temporal model and the spatial model have a multi-level structure identified by a natural language definition name,

the multi-level structure of the temporal model comprises at least one parent temporal level described by means of natural language and at least one child temporal level corresponding to the parent temporal level,

the multi-level structure of the spatial model comprises at least one parent-level spatial level described by means of natural language and at least one child-level spatial level corresponding to the parent-level spatial level.

A retrieval device for data objects based on a space-time database is characterized in that the retrieval device at least comprises a data acquisition module, at least one data calculation engine, a space-time database server, an engineering library server, a model library server and a client,

the data acquisition module is used for acquiring the operation data of the management object and sending the operation data to the engineering library server through the data calculation engine,

the model library server, on which a model library is disposed, the model library being configured to model the management object with respect to a temporal state and a spatial state of the object to be managed,

the engineering library server, on which an engineering library is set, the engineering library being configured to set specific attributes of the object to be managed, which are described by means of a natural language, according to a management model of the object to be managed,

the data calculation engine is used for performing temporal and spatial transformation on the modeled management objects according to the temporal hierarchy and the spatial hierarchy described by means of the natural language,

the space-time database server on which a space-time database is disposed, the space-time database configured to store real-time data, historical data, and/or planning data of the management object,

the client is used for searching based on the management model category of the management object and the time attribute and/or the space attribute defined by the modeling and described by the natural language to determine the running state of the management object.

The invention has the beneficial technical effects that:

the invention describes the data information of production monitoring management through three time dimensions of history, real time and plan. The invention can check and manage the running states of the object in the past, real time and planning time periods only by natural spatio-temporal meta language without the need of a user to master the computer language. Both the acquisition and processing of data are performed in a manner that naturally calls spatio-temporal functions. Any one of the actual physical object and the production event may be expressed in terms of time-space data. The position of the produced product, the time and the place of the production downtime event are required to be described by the space-time attribute information. Therefore, the retrieval method is simple and convenient and is beneficial to production management.

Drawings

FIG. 1 is a block diagram of an apparatus for retrieving data objects based on a space-time database according to the present invention.

List of reference numerals

100: the object to be managed 200: the data acquisition module 300: data calculation engine

400: the temporal database server 410: real-time database 420: history database

430: plan database 500: engineering library server 510: object definition module

520: the verification module 600: model library server 700: graphic computing engine

710: the graphics computation module 720: the scene processing module 800: client terminal

810: the caching module 820: the login module 830: refresh processing module

210: third party database

Detailed Description

The following detailed description is made with reference to the accompanying drawings.

The time-space database of the present invention refers to a database built based on data having temporal attributes and spatial attributes. The spatio-temporal database includes a historical database, a real-time database, and a planning database, which respectively store modeled historical data, real-time data, and planning data of the management object. Each data in the spatiotemporal database has temporal and spatial attributes. The same model instantiates objects, and if their temporal position, spatial position, temporal length, and spatial length are all the same, they must be the same object. Spatio-temporal is a very natural language and can be understood by a user without learning complex grammatical rules. The data checked by each production manager must be retrieved by time and place, the time-space database stores the data according to the time-space language, and users do not need to store the data and optimize the data.

Example 1

The invention provides a retrieval method of data objects based on a space-time database, which is characterized by comprising the following steps:

s1: modeling the management object according to the time state and the space state of the object to be managed;

s2: setting specific attributes of the object to be managed, which are described by means of natural language, according to a management model of the object to be managed;

s3: the search is performed to determine the operating state of the management object based on the management model category of the management object and the temporal and/or spatial attributes described by the modeling in natural language.

The following describes in detail the steps of a method for retrieving data objects based on a space-time database according to the present invention.

S1: the management object is modeled for the temporal state and the spatial state of the object to be managed.

Preferably, the temporal model, the spatial model and/or the management model are respectively established according to the temporal state and the spatial state of each object to be managed.

And modeling the time of the object to be managed according to the time state of the object to be managed so as to establish a time model.

Time model: the time model is a data model for describing the time state of the object to be managed. The time granularity of the time model is a unit for measuring the time state and the time hierarchy, and comprises a time unit and a time parameter customized based on the production condition. Preferably, the temporal model is a model describing the temporal granularity of the objects to be managed. For example, the time granularity is year, month, day, hour, minute, or second, etc. The time granularity can be customized, such as class or batch, and the like.

And (3) space model: the spatial model is a data model for describing the spatial state of an object to be managed. The spatial granularity of the spatial model is a spatial parameter for measuring spatial hierarchy and spatial position, and comprises a spatial unit and a self-defined spatial parameter based on production conditions. For example, the spatial granularity is a head office, branch office, plant, production line, process, station, or equipment, etc.

Preferably, the temporal model and the spatial model have a multi-level structure identified by a natural language definition name.

The multi-level structure of the temporal model comprises at least one parent temporal level described by means of natural language and at least one child temporal level corresponding to the parent temporal level.

For example, the time hierarchy includes years, months, days, hours, minutes, or seconds. A year is a parent level time hierarchy of a month, and a month is a child level hierarchy of a year. A month is a parent time hierarchy of days and a day is a child time hierarchy of months. A year corresponds to a plurality of months, while a month corresponds to only one year. Alternatively, the time hierarchy includes orders, work orders, and products described by natural language. The order is a parent level time hierarchy of the work order, which is a child level hierarchy of the order. One order corresponds to multiple work orders, and one work order corresponds to only one order. The work order is a parent level time hierarchy of the product, and the product is a child level hierarchy of the work order. One work order corresponds to a plurality of products, and one product corresponds to only one work order.

The time hierarchy of the present invention is not limited thereto, and includes other time hierarchies described by means of natural language. If the heating season is the parent-level time hierarchy, and the heating month is the child-level time hierarchy. One heating season includes a plurality of heating months, that is, the 2016 heating season includes 5 heating months of 2016 11 months to 2017 3 months, and the 2017 heating season includes 5 heating months of 2017 11 months to 2018 3 months. After the time hierarchy of the heating season is set for the first time, the heating season of each year automatically comprises 5 cross-year heating months, and the record condition of time confusion can not occur.

And establishing a management model formed by associating the space model and the time model of the object to be managed. The management model is a data model for describing the spatial state and the temporal state of the object to be managed. The data is associated with a temporal model and a spatial model to form a spatio-temporal model. The spatio-temporal model is a data model which effectively organizes and manages the temporal spatial data and has more complete attributes, spatial and temporal semantics. The time state and the space state of the object to be managed are monitored through the spatio-temporal model, and the running state of the object to be managed can be fully known. The invention refers to the spatio-temporal model of the management object as the management model for short.

Modeling data: each management object described is instantiated by a management model. The management model forms at least one management object formed by the association of the time object and the space object of the management object through an instantiation process. The time object of the management object is formed by the time model in the instantiation process, and the space object of the management object is formed by the space model in the instantiation process.

For example, the equipment is one object, and the equipment is one thing to process a product. The product is the time, i.e. the time the equipment processes the product. This time has a time position, a start production time and an end production time. The time location is a unique product number. If the data is real-time data, the product number, the production starting time and the elapsed time are changed into historical data after the end. Elapsed time refers to the length of time the product is produced. The scanning device then continues to produce the content for the next time (product). The present invention data models things based on three time periods of past, present and future.

The data model is divided into a real-time data model, a historical data model and a planning data model according to the time state. And is divided into data, alarm and event according to the function. There is a greater variety of data for the combination of tenses and functions. The data model has a version. In a system, multiple versions of the same model may exist. An instance of a model is one that corresponds to a certain version of the model. When the model generates a new version, the generated instance is not influenced. When a version of the model is modified, the instance generated by the version model is affected.

The data model is a model for describing data, and consists of two parts: attribute, member.

Attributes are data-specific parts, such as name, description, time, space, attributes are system-defined and user-definable. Depending on the classification, the attributes of the data model may differ, such as real-time data, with the attribute of freshness. The event data has attributes of a start time, an end time, a duration, and the like.

The members are components forming data, and the user can define different members to describe the data in the service according to needs. The member's data may use various types as defined above. From a performance point of view, the number of members is limited to a maximum of 256 members.

Preferably, for attribute names, member names are not allowed to exceed 64 characters.

The management model of the present invention specifies an associated temporal model and spatial model. The present invention refers to a data model comprising a temporal model and a spatial model, simply referred to as a spatio-temporal model. For example, in a performance model of workstation processing, the spatial attribute is the workstation and the temporal attribute is the shift. Other attributes of the data may be work order number, number of processes, number of alarms, number of reworks, etc. The data of the data models are completed through calculation models, production data changes are calculated and detected, and performance data are counted and output to the performance models at regular time.

Modeling a factory: plant modeling is plant instantiation from previous models. The real-time data model, historical data model, and planning data model distributions are instantiated as real-time data objects, historical data objects, and planning data objects. For example, the actual plant name, production line name, equipment name, process name, which is a space instantiation. The affiliation of each object is also determined during instantiation. There are several shifts of a day, how long each shift, which is a time instantiation. The actual production line has several process performance data, which is an instantiation of a performance data model. There is of course also instantiation of the calculations as these performance data are to be calculated.

The system operates: after the system runs, the background of the system automatically detects production information, records the production condition of each process and counts the production performance of each station in real time. Thus, the production conditions of a factory are recorded in real time in the time-space database. And inquiring real-time and historical production data information in the scene model when the user needs to check.

Planning: the planning generally makes a total plan such as a whole plant annual plan, and then is resolved into a whole plant month, a whole plant day, a whole plant class, a workshop month, a workshop day and a workshop class. The user need only break down the time and space into very fine levels and then monitor each time a spatio-temporal level is performed for execution or not as planned.

According to a preferred embodiment, the temporal and spatial attributes are independent. The time object is described with a hierarchy and precision, for example, the hierarchy of the year is the year, and the precision can be seconds or milliseconds. The time object has a time position on the data record, a start time and an end time of the time position. For example, 2016 is the time location. Month 3 of 2016 is also the time position, and the typical start time for the time position of month 3 of 2016 is 0 at month 3 of 2016: 00: 00.000, end time 2016, 3, 31, 23: 59: 59.999. preferably, the time location can also be customized. For example, 2016 is from a custom year (parent time hierarchy) -month (child time hierarchy), and the start time and end time are user-defined, such as 2016 from 2 days 3 to 15 days 3, month when the user indicates this time period with 2016.

S2: specific attributes of the object to be managed, which are described by means of the natural language, are set according to a management model of the object to be managed.

S21: and classifying the modeled objects to be managed into a specific model according to the time attributes. The management model comprises real-time data objects which are formed according to the time attribute in a classified mode and stored in a real-time database, historical data objects which are stored in a historical database and plan data objects which are stored in a plan database.

The historical data object, the real-time data object, and the planning data object include at least a temporal attribute and a spatial attribute. And configuring the historical data objects in the historical data objects with corresponding computer space attributes, and automatically storing the historical data in a historical database of the computer. And configuring the real-time object in the real-time data object with the corresponding computer space attribute, and automatically storing the real-time data in a real-time database of the computer. And allocating the planning objects in the planning data objects with corresponding computer space attributes, and automatically storing the planning data in a planning database of the computer.

The historical database, the real-time database and the planning database constitute a time-space database of the invention. And the historical database stores the historical data of the object to be managed according to the configured conditions. The real-time database is used for storing real-time values of the objects to be managed. The plan database is used for storing plan data of the objects to be managed. The historical database, the real-time database, and the planning database do not require user configuration. When the user uses the system, the user needs to designate a real-time server, a history server and a plan server through a configuration interface. When the time-space database system operates, each server client automatically sends data to the corresponding server. For example, the real-time data client transmits the real-time data generated by the system to the real-time data server, and the real-time data server automatically creates a table or a mapping list in the database and stores the related data in the database of the corresponding server. As do the history and planning databases. The storage mechanism of the present invention is not limited thereto, and other storage mechanisms are also included.

The real-time data object includes at least one management object formed from real-time data of the management object via a real-time data model instantiation process. The real-time data objects are instantiated according to a real-time data model. The spatial model associated with the real-time data model is instantiated as a spatial object. The temporal model associated with the real-time data model is instantiated as a temporal object. The real-time data objects of the management objects are associated with the spatial objects and the temporal objects, respectively. A management model is instantiated as a plurality of management objects. The real-time data model of one version instantiates the data structure of the real-time data of the plurality of management objects formed to be the same. And storing a plurality of management objects formed after the real-time data model is instantiated in a real-time database.

The historical data object comprises at least one management object formed by historical data of the management object through a historical data model instantiation process. And instantiating the historical data object according to the historical data model. The spatial model associated with the historical data model is instantiated as a spatial object. The temporal model associated with the historical data model is instantiated as a temporal object. The historical data object of the management object is associated with the space object and the time object, respectively. A management model is instantiated as a plurality of management objects. The data structure of the historical data of a plurality of management objects formed by instantiation of the historical data model of one version is the same. A plurality of management objects formed after instantiation of the historical data model are stored in a historical database.

The planning data object includes at least one management object formed from the planning data of the management object via a planning data model instantiation process. Instantiated as a planning data object according to the planning data model. The spatial model associated with the planning data model is instantiated as a spatial object. The temporal model associated with the planning data model is instantiated as a temporal object. The plan data object of the management object is associated with a spatial object and a temporal object, respectively. A management model is instantiated as a plurality of management objects. The data structures of the planning data of the plurality of management objects formed by instantiation of one version of the planning data model are the same. A plurality of management objects formed after instantiation of the planning data model are stored in a planning database. There is a great difference in the storage of historical data, real-time data and planning data.

The real-time data is real-time, and represents the data of the current time, and has the characteristic of freshness, that is, the refresh period of the data conforms to the time granularity. For example, if the freshness of an object is 5 seconds, its refresh period should also be 5 seconds, and if it is not refreshed within 5 seconds, the object is not fresh. Real-time data requires high real-time performance, millions of recorded real-time data can be refreshed every second, and the data which is not refreshed in time has a fresh period limit. Real-time data beyond the freshness period is converted into historical data.

Historical data tenses are historical, representing data of past time. Most of the historical data of the industrial process is time sequence data and can be compressed. The business data in the historical data can also be stored in an uncompressed mode.

The projected data tense is future, representing data at a future time. The planning data is calculated according to historical data and real-time data.

The real-time data objects stored in the real-time database are organized according to the spatial characteristics, so that the storage and the retrieval are convenient. The historical data objects and the plan data objects of the historical database and the plan database are organized according to the spatial characteristics and then according to the temporal characteristics.

According to a preferred embodiment, the data objects stored in the historian, real-time and planning databases must have temporal attributes. Time is an indivisible attribute of an object to be managed. Without time, the object may not exist. A temporal object is a description of the temporal location of an object to be managed. Preferably, the time attributes include at least a time position, a start time and an end time. The time position of the management object at least comprises the time hierarchy and the hierarchy precision of the object to be managed. For example, a production lot is a level. The information of the first batch in a certain workshop is very convenient for a user to use, and the user does not need to write a certain time period to acquire possible batch information.

According to a preferred embodiment, each management model has a unique temporal hierarchy and spatial hierarchy. Taking a plan data as an example, a production line (space) class (time) plan is described, a production line model and a class model are space models and time models for realizing definition, the model is determined to assume that the production line model has 3 production line objects in an engineering stage, and the class model has three classes, namely, A, B and C. The data model can only instantiate 3 production lines and 3 teams for 9 planning data objects, and cannot instantiate plans of other spatio-temporal levels. The temporal and spatial hierarchy of the management model is determined to be unable to be modified, and only the attributes or members are modified will a new version be generated.

Preferably, the space of the object to be managed is modeled according to the space state of the object to be managed to establish a space model. Formation of a multi-level spatial object through an instantiation process of a spatial model having a multi-level structure according to a preferred embodiment, a method of instantiation of a multi-level spatial object includes:

loading a map or a CAD (computer aided design) drawing of an object to be described as a base drawing through a space object editor; drawing a space object in the base map as a first-level space object according to the space coordinate of the object to be managed, and/or drawing a space object in the base map as a first-level space object according to a self-defined space range; and drawing a space object in the first-level space model as a second-level space model, and drawing a space object in the (n-1) th-level space model as an nth-level space model in the same way.

Specifically, the process of rendering the spatial object is as follows.

S201: and loading a map or a CAD drawing of the object to be managed as a base map through a spatial object editor. Preferably, a google map, a Baidu map or a CAD drawing of the object to be managed is loaded as a base map through a spatial object editor.

S202: and drawing a space object in the base map as a first-level space model. Preferably, a corresponding coordinate system is selected and set in the map configuration, a spatial range of the Root node is drawn based on the geographical position of the factory, and a spatial object is drawn by using a rectangle, an irregular polygonal line, a point and/or a broken line in the spatial range of the Root node to obtain the first-level spatial model.

S203: and drawing a space object in the first-level space model as a second-level space model. Preferably, a spatial object is rendered within the first-level spatial model with rectangles, irregular polylines, points and/or polylines to obtain the second-level spatial model.

S204: in the same way, a space object is drawn in the (n-1) th level space model as the nth level space model. Preferably, a spatial object is rendered within the level n-1 spatial model with rectangles, irregular polylines, points and/or polylines to obtain the level n spatial model.

S22: specific attributes of the object to be managed are set.

The specific attributes include temporal attributes, spatial attributes, and/or non-spatiotemporal attributes described by means of natural language. When data is inquired, the objects in the model can be retrieved according to time and space, and the method is convenient and quick. Preferably, the object to be managed includes a custom non-spatiotemporal attribute in addition to the temporal attribute and the spatial attribute.

Preferably, the non-spatiotemporal attributes of the object to be managed include one or more of a picture attribute, an audio attribute, a video attribute, and a name attribute of the object to be managed. Through the 'dotting' of the object to be managed, the object management of various unstructured data can be realized, and therefore unified storage and query calling are realized.

The video file is used for playing, the video file is used as a member of the graphic object and is transmitted to the display equipment, the display equipment can establish a playing area according to the outline of the graphic object, and the playing of the video can be controlled through a script function. The audio file is used for playing and provides a sound playing function. For picture types as members of a graphical object, the display device will draw the picture onto the screen when the graphical object is transferred to the display device. For the resource object of the file type, the resource object can be used for transmitting and storing the file, such as saving the process file and issuing the operation specification.

The graph type also belongs to the data type and can be used as a member of the object. The coordinates of the graphic type are pixel coordinates. Geometric types are shapes that describe geospatial objects that require configuration in terms of their properties.

According to a preferred embodiment, the temporal attribute and the spatial attribute are intrinsic attributes of the thing, and for data in the real-time database, the data is managed by a freshness period, and the data state exceeding the freshness period is failure. The time hierarchy of each data in the database is fixed and can be seconds, minutes or hours.

Preferably, the time attributes include at least a time position, a start time and an end time. The temporal location is a location relative to the parent spatial model. The spatial attributes of each datum in the database include geospatial range, geospatial location, shape, size and outline of subspace, candidate geospatial location, computer space, candidate computer spatial attributes, and the like.

Preferably, when the management object is a static object, the spatial attributes include at least a geospatial range, a geospatial location, and a spatial shape, spatial range, and spatial location of the subspace,

when the management object is a dynamic object, the spatial attributes at least comprise a geographic spatial position, a candidate geographic spatial position, and a spatial shape, a spatial range and a spatial position of a previous-level spatial model of the object to be managed. For example, a production line, spatial attributes include at least a geospatial location, a candidate geospatial location, and a shape, size, and outline of a parent spatial model of an object to be managed. The spatial extent of the subspace model cannot exceed that of the parent space model.

For example, the spatial extent of a serviceman is a workshop and the spatial location is equipment. The spatial location of the serviceman is described as: the maintainer is at a shop at equipment number 2. At this time, the spatial range is a parent space, and the spatial position is a subspace. The subspace does not exceed the spatial extent of the parent space.

According to a preferred embodiment, the order of the sequential steps S21 and S22 is not limited. That is, the specific attribute of the management model may be set after the modeled object to be managed is classified into the specific model according to the time attribute. Or setting the specific attribute of the management model, and classifying the modeled object to be managed into the specific model according to the time attribute.

According to a preferred embodiment, the data of the management objects stored in the temporal database includes historical data directly collected from the sites of the management objects, real-time data collected from the management objects through sensors, and plan data of the objects to be managed calculated from the historical data and the real-time data.

Preferably, the historical data and the planning data are generated by calculation. Preferably, the data in the planning database may also be graphically compiled for interface interaction or imported from a software interface of a third-party system. For example, after the planning data is imported through a third-party program interface or a third-party file and recognized by the system as the planning data, the planning data is calculated and generated through editing of the interface and calling of an algorithm.

S3: the search is performed to determine the operating state of the management object based on the management model category of the management object and the temporal and/or spatial attributes described by the modeling in natural language. Real-time data, historical data and/or planning data of the at least one management object within the spatial location and/or time are retrieved based on the management model category, the spatial location and/or the time of the management object, thereby determining a real-time operating state, a historical operating state and/or a planning operating state of the at least one management object.

Alternatively, real-time data, historical data, and/or planning data of at least one management object within a spatial range and/or a temporal range is retrieved based on a management model category, the spatial range, and/or the temporal range of the management object to determine a real-time operational state, a historical operational state, and/or a planned operational state of the at least one management object.

The step of retrieving for determining the operating state of the management object based on the management model category of the management object and the temporal and/or spatial attributes defined by the modeling by means of the natural language comprises:

determining a real-time database, a historical database and/or a plan database for storing the operation data of the management object according to the type of the management model of the management object;

selecting at least one version of management object operation data related to the management object in a real-time database, a historical database and/or a plan database according to the time attribute and/or the space attribute;

and determining at least one management object associated with the management model and viewing real-time data, historical data and/or plan data of the management object so as to know the real-time operation state, the historical operation state and/or the plan operation state of the management object.

Preferably, the user inputs a management model category, temporal attribute data, and/or spatial attribute data of the management object. The system determines a real-time database, a historical database, and/or a planning database that stores operational data for the managed object based on the class of management model input by the user. After determining the database, selecting at least one version of management object operation data related to the management object in the real-time database, the historical database and/or the plan database according to the time and/or the spatial position, the time range and/or the spatial range input by the user. And the user determines the version of the management object operation data according to the displayed management object operation data of at least one version and views the plurality of management objects instantiated by the version management model. The management object is selected, and the real-time data, the historical data and/or the plan data of the management object can be viewed by opening the operation data of the management object, so that the real-time operation state, the historical operation state and/or the plan operation state of the management object can be known.

The present embodiment is described by taking an example of viewing a real-time operation state of a specified management object.

The user selects the real-time data model. The system determines a real-time database based on the real-time data model of the managed object. After determining the database, at least one version of the real-time data object in the real-time database associated with the management object is selected in dependence on the temporal and/or spatial position input by the user. After the version of the real-time data object is determined, the real-time data of the management object can be checked by opening the real-time data object of the management object, so that the real-time running state of the management object is known. The method of viewing the historical operating state and the planned operating state of the management object is the same as the method of viewing the real-time operating state of the management object.

The user performs a search by entering the management model category, spatial range, and temporal range. After the database is determined, a data or batch of data within the database belonging to all management objects within the spatial and temporal range may be retrieved. Preferably, the spatial range for querying the real-time, historical and planning data is greater than or equal to the spatial range of the object, for example, the spatial hierarchy is factory-workshop-equipment, and the real-time data for querying the workshop cannot be queried by the equipment, and is necessarily queried by a workshop or a factory. The time granularity range of the query history and the plan should be the time granularity range of the data itself.

Example 2

This embodiment is a further improvement on the basis of embodiment 1, and only the improved part will be described.

According to a preferred embodiment, the planning data model builds and updates at least one planning data model containing different versions of the same temporal and spatial state according to the set version of the user. That is, the planning database stores multiple versions of planning data models. The multiple versions of the planning data model are instantiated as multiple versions of the planning data object. A version of the planning data model is selected that uses the multiple versions, thereby determining the version of the planning data object. The plan data object is the management object.

The multi-version schema of the planning data model is of great significance. A plan change corresponds to a series of sub-plan changes, and the related plans can be conveniently found out through the plan database with consistent versions. The currently formulated plan, the plan to be issued, and the plan being executed must be a certain specific plan data version. For example, during actual operation, for a production plan of a plant, two versions of the operation record may be created for the planned operation record to be generated by a plan data object. The actual production schedule data object will only select one version to execute. Therefore, there is a difference between the version of the planning data model and the version of the record of the planning data object.

The user sets and stores multiple versions of the planning data model in the computer. Due to frequent changes in the planning records, the planning database builds and updates multiple versions of the planning data object for the same time and space.

For example, at actual runtime, the planned run records for different planned data objects may be updated uniformly. For example, the plant schedule record changes, as do the team schedule records. The user can finish the plan change in a unified way, and the plan change is convenient to check. Preferably, the real-time data model comprises at least one different version of the real-time data model that is built and updated according to the same temporal state and spatial state. Different versions of the real-time data model are instantiated into different versions of the planning data object and stored in the real-time database.

The same real-time data model may have multiple versions that instantiate multiple real-time data objects. For example, a plant is upgraded, and new and old systems coexist, which are two versions of a data model used. And monitoring system objects under the new version and the old version, switching to the new version if a certain real-time data object is upgraded, and storing a historical record corresponding to each version when the historical records are stored.

According to a preferred embodiment, the historical data model builds and updates at least one version of the historical data model based on at least one different version of the real-time data model. Different versions of the historical data model are instantiated into different versions of the historical data object and stored in the historical database.

A history data object may have multiple versions of a value at the same time. When modifying a historical data object, there are several situations:

1. the historical data value version is unchanged, and the data is modified to cover the original historical record.

2. The historical data value is added with a version, namely a historical record.

3. The historical data is not allowed to be deleted.

The manner in which the historical data is modified is set by the user. If the user sets the modification mode of the historical data to be not allowed to modify the original record, the historical data is modified to generate a new version of historical data record.

And the historical data recorded by the historical data models of different versions can be played back, so that not only the historical data of each system can be seen, but also the historical transitions can be seen.

According to a preferred embodiment, the search is performed to determine the operating state of the management object on the basis of the management model category of the management object and the temporal and/or spatial attributes defined by the modeling, which are described by means of the natural language.

Preferably, the real-time data, the historical data and/or the planning data of the at least one management object in the spatial range and/or the temporal range are retrieved based on the management model category, the spatial range and/or the temporal range of the management object, so as to determine the real-time operation state, the historical operation state and/or the planning operation state of the at least one management object.

selecting at least one version of management operation data related to the management object in the real-time database, the historical database and/or the plan database according to the time attribute and/or the space attribute;

Preferably, the user inputs a management model category, a temporal range, and/or a spatial range of the management object. The system determines a real-time database, a historical database, and/or a planning database that stores operational data for the managed object based on the class of management model input by the user. After determining the database, selecting at least one version of management object operation data related to the management object in the real-time database, the historical database and/or the plan database according to the time range and/or the space range input by the user. The user determines the version of the management model according to the displayed management model of at least one version and views a plurality of management objects instantiated by the version management model. The management object is selected, and the real-time data, the historical data and/or the plan data of the management object can be viewed by opening the data of the management object, so that the real-time operation state, the historical operation state and/or the plan operation state of the management object can be known.

For example, the plant collection model version V1 supports the collection of only two parameters, temperature and humidity. On the basis of the model, a management object, namely a workshop 1 object is established, and the values of the temperature and the humidity are collected and stored in a historical database. The version of the corresponding model is also recorded in the historian. After a period of operation, the field system is upgraded to collect pressure in addition to temperature and humidity, and the collection model is upgraded to the V2 version, increasing the pressure parameters. After the object of the workshop 1 is upgraded, the workshop 1 starts to collect three values of temperature, humidity and pressure, and stores the corresponding values into a database. In this way, the history data generated in the case of different versions of the plant 1 is recorded in the history library.

For an acquisition management model, the first version only has two parameters of temperature and humidity. The second version includes three parameters, temperature, humidity and pressure.

The user retrieves the real-time data objects of both versions of the object after entering the management model category, the temporal data and/or the spatial data. The user can select one version of the real-time data object to know the real-time running state of the object. The user can also open the real-time data objects of the two versions at the same time and check the data of the two real-time data at the same time.

After the user inputs the management model category, the spatial data and the time data, the historical data base of the object and the plan data base store two versions of historical data objects and two versions of plan data objects.

The user can select one version of the historical data object of the management object, view the historical operating data of the object and know the historical operating state of the object. The user can also open the historical data objects of the two versions at the same time, view the historical operating data of the objects and review the historical change tracks of the objects.

The user can select one version of the planning data object of the management object, view the planning operation data of the object and know the planning operation state of the object. The user can also open the two versions of the planning data object at the same time, check the planning operation data of the object, check the data of the two planning databases at the same time, and review the planning change track of the object.

Example 3

This example is a further modification of examples 1 and 2, and only the modified portions will be described.

The embodiment provides a retrieval device for data objects based on a space-time database, which at least comprises a data acquisition module, at least one data calculation engine, a space-time database server, an engineering library server, a model library server and a client.

And the data acquisition module is used for acquiring the operation data of the management object and sending the operation data to the engineering library server through the data calculation engine.

The model library server is provided with a model library, and the model library is configured to model the management object according to the time state and the space state of the object to be managed.

The engineering library server is provided with an engineering library, and the engineering library is configured to set specific attributes of the object to be managed, which are described by means of natural language, according to a management model of the object to be managed.

The data computation engine is configured to perform temporal and spatial transformations on the modeled management objects according to temporal and spatial hierarchies described by means of natural language.

The space-time database server is provided with a space-time database, and the space-time database is configured to store real-time data, historical data and/or planning data of the management object.

The present embodiment describes a device for retrieving data objects based on a space-time database in detail.

As shown in fig. 1, a device for retrieving data objects based on a spatio-temporal database includes a data collection module 200, at least one data calculation engine 300, a spatio-temporal database server 400, an engineering library server 500, a model library server 600, at least one graphic calculation engine 700, and a client 800.

The data collecting module 200 is used for collecting data information of the object 100 to be managed.

The data calculation engine 300 is used for data conversion of the collected data.

A time-space database is provided on the time-space database server 400. The temporal databases include a real-time database 410, a historical database 420, and a planning database 430. The real-time database 410, the historical database 420, and the planning database 430 are used to store at least one version of a real-time data object, a historical data object, and a planning data object instantiated based on at least one version of a real-time data model, a historical data model, and a planning data model, respectively.

The model library server 600 is provided with a model library. The model library is used for establishing a time model, a space model, a management model and a calculation model, and is used by the engineering library 500.

The engineering library server 500 is provided with an engineering library. The engineering library is used for importing the model related to the solution in the model library 600 and instantiating the engineering object. Instantiated objects include spatial objects, temporal objects, management objects, and computational objects. The engineering library server 500 includes an object definition module 510 and a login verification module 520. The object definition module 510 is used to instantiate a modeled object and define the instantiation via natural language description. The object definition module 510 instantiates the real-time data model, historical data model, and planning data model as a real-time data object, historical data object, and planning data object, respectively. The login authentication module 520 is used for authenticating the user who retrieves the information.

The graphic computation engine 700 is used for graphic data computation, retrieval and graphic interactive presentation of the client 800. The graphic calculation engine 700 includes a graphic calculation module 710 and a scene processing module 720. The graphics computation module 710 is used to perform computations on the graphics data. The scene processing module 720 is configured to perform data processing on scene display. The scenario processing module 720 builds a scenario model for all devices of the production plant. The scene model is used for displaying specific attributes and operating states of the production equipment.

Client 800 is a device for interacting through a data retrieval computing model. The client 800 includes a caching module 810, a logging module 820, and a refresh processing module 830. The cache module 810 is used for caching the retrieved data and the displayed data. The login module 820 is used for inputting login information and retrieving information. The refresh processing module 830 is configured to perform refresh processing on the content displayed by the client.

The present embodiment describes a retrieval method of a retrieval apparatus for data objects based on a space-time database as follows.

The management object is modeled for the temporal state and the spatial state of the object to be managed. I.e. the temporal, spatial, management and calculation models are built within the model library on the model library server 600 for the temporal and spatial states of the objects to be managed.

Taking a production workshop as an example, a multi-level spatial model is established. The multi-level spatial model comprises a workshop model, a production line model and an equipment model. The multi-level spatial model is instantiated as a multi-level object. Preferably, a google map or a Baidu map is loaded through a spatial object editor, a corresponding coordinate system is selected and set in map configuration, a spatial range of a Root node is drawn based on the geographical position of a factory, and a spatial object is drawn in the spatial range of the Root node by using a rectangle/irregular polyline/point/polyline to obtain a first-level spatial model. Preferably, a spatial object is rendered with a rectangle/irregular polyline/point/polyline in the first level spatial model to obtain the second level spatial model … … a spatial object is rendered with a rectangle/irregular polyline/point/polyline in the nth-1 level spatial model to obtain the nth level spatial model. In this embodiment, the first-level spatial model is a workshop model, the second-level spatial model is a production line model, and the third-level spatial model is an equipment model. The space objects are respectively a workshop, a production line and equipment.

And establishing a time model. The hierarchy of the time model is fixed, including seconds, minutes, hours. The data recording time attribute is composed of a time position (position relative to the parent), a start time, and an end time. Preferably, the time parameter of the work site is freely defined according to the shift, the batch and the like, and the time parameter is stored in units of "shift, batch". For example, "2016 1 month 31 morning shift or first shift". Preferably, the time hierarchy of the job site is freely defined according to orders, work orders and products. The basic attributes of time at this time include:

time name: product numbering;

start time: the product online time;

end time: product offline time.

And establishing a real-time data model. As shown in fig. 1, an object to be managed, i.e., a collection object, is set as a device 100. The data acquisition module 200 acquires data of the object to be managed 100. The data collection module 200 includes a data collection server IOServer. The data collection module 200 sends the collected data to the data calculation engine 300. The data calculation engine 300 sends data model request information to the engineering library 500 server. The engineering library server 500 imports a real-time data model-Tag (time second) model required for real-time data to the model library server 600 in response to a request of the data calculation engine 300, and instantiates the Tag (time second) model as a Tag (time second) object according to the real-time collection data transmitted by the data calculation engine 300. The Tag (time second) object is the first real-time data object. The data calculation engine 300 sends the first real-time data object to the real-time database 410 for storage. The time hierarchy of the first real-time data object is a natural attribute hierarchy including year, month, day, hour, minute, second. That is, the real-time data in the first real-time data object includes time data for each device, such as an online status, an offline status, a current product, current parameters, and the like.

Data calculation engine 300 may also computationally convert a first real-time data object into a second real-time data object at a different time hierarchy according to a user's custom time hierarchy. The calculation engine 300 performs dynamic data processing on the real-time data of the first real-time data object according to a preset version to obtain a second real-time data object. For example, the time hierarchy in the second real-time data object is order, work order, product, so the real-time data of the second real-time data object includes production events of the product on the equipment.

The data calculation engine 300 converts the first real-time data object into a historical data object exceeding the freshness period and sends the historical data object to the historical database 420 for storage. Specifically, when the data collection module 200 is online according to the real-time data detection device, a first real-time data object is established. When the data acquisition module 200 detects that the device is off-line according to the real-time data, the event state of the device is stored as a historical data object. The data calculation engine 300 processes the real-time data object calculation into a historical data object, stores the historical data object in a historical database, and sets the state of the real-time event to zero.

The data collection module 200 also includes a third party database 210 or data import interface device. The data calculation engine 300 calculates a projected data object for the production facility based on the real-time data object and the historical data object. Or engineering library server 500 instantiates the planning data model as a first planning data object based on third-party imported planning data. The data calculation engine 300 performs dynamic data processing on the first plan data object according to a preset version to obtain a second plan data object describing a production event. The data computation engine 300 sends the second plan data object to the plan database 430 for storage. The real-time data objects are stored in a real-time database, the historical data objects are stored in a historical database, and the plan data objects are stored in a plan database.

According to a preferred embodiment, production facilities with equipment in the spatial position of the spatial model are configured to different management models. The equipment space of the production workshop is configured on different computing processes and a time-space database, all the computations of the system based on the space model of the production workshop are operated under the corresponding processes by default, and all the data are respectively operated under the corresponding library nodes. First, a parent space model is established for each production device. Secondly, a version is selected and its own real-time data model is built for each device. For example, the production device 1 is selected to have the product event as a first version of the second real-time data model, spatially located at the first device first production line. A second version of the second real-time data model is selected for the production device 2, the spatial location of which is the second device first production line. I.e. different versions of the same real-time data model are applicable to real-time data objects of both production facilities. Since the production equipment 1 and the production equipment 2 are not put into production in actual projects for the same year, although the production events of the products are recorded, the information of the production events recorded in the second version is more than that of the first version. If the production equipment 1 is upgraded, only the corresponding data model version is modified at the product event object of the production equipment 1.

The user inputs login information through the login module 810 of the client 800. The client 800 sends the login information to the verification module 520 of the engineering library server 500. After verifying the login information, the verification module 520 sends access information to the time-space database 400, and the time-space database 400 allows the client 800 to access the time-space database 400 based on the access information.

The description will be given taking the retrieval of history data as an example. The user initiates a request for historical data retrieval through the login module 810 and enters the management model category, spatial location, and time of the production facility. The graph computation module 710 determines the space to which the historical object in the geographic space belongs according to the spatial location computation. The graph computation module 710 determines a historical database based on the information that the management model is a historical data model. The graph computation module 710 establishes a connection with the target historical database of the space to which it belongs, and sends a spatiotemporal search request. And the target historical database receives the retrieval request, positions the historical data area according to the geographic space, and retrieves the historical data object according to the time. The target historian sends the retrieved historian object to the graph computation module 710. The graph computation module 710 receives the historical data object and places it into a graph data channel for presentation to the user via the scene model computed by the scene processing module 720. The method of retrieving planning data of a production facility is the same as the method of retrieving a historical database, and the retrieval is positioned first by space and then by time.

The example of retrieving real-time data is described. The user initiates a real-time data retrieval request through the login module and inputs the management model category, time and/or spatial location of the production equipment. The graph computation module 710 determines the space to which the real-time object in the geographic space belongs according to the spatial location computation. The graph computation module 710 determines the real-time database based on the information that the management model is a real-time data model. The graph computation module 710 establishes a connection with the target real-time database of the space to which it belongs, and sends a spatiotemporal search request. And the target real-time database receives the retrieval request and positions the real-time data area according to the geographic space. The target real-time database sends the retrieved real-time data objects to the graph computation module 710. The graphics computation module 710 receives real-time data objects and places them into a graphics data channel for presentation to the user via a scene model. Preferably, the login module 820 and the scene processing module 720 are provided on the client 800 to facilitate the user to retrieve and view data.

Example 4

This example is a further modification of example 1, example 2 and example 3, and only the modified portions will be described.

The user needs to realize the information monitoring of the first boiler of the first factory. The first boiler is the object to be managed. Firstly, a relevant data model of a first boiler is established, and a matched time and space model is established to locate or monitor relevant data information of the boiler. And associating the related data information with the time model and the space model to establish a management model. The time model and the space model are respectively instantiated by establishing a space-time system and a management model, and the management model is instantiated into a management object. The management object needs to select a certain spatial object of the spatial model associated with the management model and also select a certain temporal object of the temporal model associated with the management model for association. For example, the space associated with the first boiler real-time data model is a boiler space model and the associated time is a time model of the shift. The first boiler space model instantiates three space objects, boiler 1, boiler 2, and boiler 3. Boiler 1, boiler 2, boiler 3 are the self-defined space object names by means of natural language. The shift time model instantiates three time objects of early shift, middle shift and late shift. The morning shift, the middle shift and the evening shift are time object names defined by natural language. The boiler real-time data model instantiates a boiler real-time data object 1. The associated space object of the real-time data object 1 of the boiler is the boiler 1, the associated time objects are the early shift, the middle shift and the late shift, and the time model is corresponding to the shift model by default. Because the boiler 1 is a real-time data object, only real-time database space can be selected when the computer stores the configuration. And modeling and associating processes of the historical database and the plan database are realized in the same way.

The time model is established in the modeling process, for example, the class model is established immediately to complete the definition of the time period of each class, for example, the morning class time is set to six am to three pm. The establishment of the spatial model, such as the establishment of the plant model and the boiler model, is the description of the spatial shape, the spatial range and the spatial position of the plant model and the boiler model. While defining the plant model as the parent model of the boiler model. And (3) establishing a management model, wherein when the management model is established, the type of the model needs to be set firstly. The management model includes a real-time data model, a historical data model, and a planning data model. After running, the storage locations of the real-time data objects, historical data objects, and planning data objects are determined by model type: a real-time database, a historical database, a planning database. Similarly, the query is also based on the type of the management model to determine which database in the time-space database to search for data. After the system is operated, the real-time data object stores the acquired value into the real-time database, so that information including time information of 2016, 10, 21 and 21 early shifts, for example, is recorded when data are actually generated. The early shift information includes spatial shape, spatial range and spatial position information of a boiler model corresponding to the boiler 1 as spatial information, model category information of the boiler corresponding to the management model, and other attribute information of the boiler. Such as status information of the boiler 1, e.g. colour, temperature, pressure in the chamber, etc.

Meanwhile, the user can realize the state retrieval of a certain boiler at a certain moment based on the time information, the space information and the model category information.

The user selects the real-time data model. The system determines a real-time database based on the real-time data model of the managed object. After determining the database, at least one version of the boiler real-time data object in the real-time database that is associated with the management object is selected according to the time and/or the boiler name input by the user. After determining the versions of the boiler real-time data objects, one or more of the boiler model instantiated space objects boiler 1, boiler 2, boiler 3 are selected. Opening the data of the boiler 1 allows to view the real-time data of the boiler 1, thereby knowing the real-time operation state of the boiler 1.

The method of checking the historical operating state and the planned operating state of the boiler 1 is the same as the method of checking the real-time operating state of the boiler 1. The method of checking the state of the other boilers is the same as the method of checking the operation state of the boiler 1.

It should be noted that the above-mentioned embodiments are exemplary, and that those skilled in the art, having benefit of the present disclosure, may devise various arrangements that are within the scope of the present disclosure and that fall within the scope of the invention. It is to be understood by persons skilled in the art that the present description is illustrative only and not limiting upon the claims. The scope of the invention is defined by the claims and their equivalents.

Claims

1. A method for retrieving data objects based on a space-time database, the method comprising the steps of:

modeling the management object according to the time state and the space state of the object to be managed; performing data modeling on an object to be managed based on three time periods of the past, the present and the future, and establishing a management model formed by associating a space model and a time model of the object to be managed;

setting specific attributes of the object to be managed, which are described by means of natural language, according to a management model of the object to be managed; before the specific attribute is set, classifying the modeled object to be managed into a specific model according to the time attribute; retrieving to determine an operational state of a management object based on the management model category of the management object and temporal and/or spatial attributes defined by the modeling by means of natural language; wherein,

the step of retrieving based on the management model category of the management object and the temporal and/or spatial attributes described by means of natural language defined by the modeling to determine the operating state of the management object comprises:

determining at least one management object associated with the management model and viewing real-time data, historical data and/or plan data of the management object, so as to know the real-time running state, the historical running state and/or the plan running state of the management object, wherein the real-time data objects stored in a real-time database are organized according to spatial characteristics, and the historical data objects and the plan data objects of the historical database and the plan database are organized according to the spatial characteristics and then according to the temporal characteristics.

2. The method of claim 1, wherein the step of retrieving based on the management model category of the management object and the time attribute and/or the space attribute defined by the modeling through the natural language to determine the operation status of the management object comprises:

3. The method of claim 1, wherein the management objects comprise real-time data objects stored in the real-time database, historical data objects stored in the historical database, and planning data objects stored in the planning database, the real-time data objects being formed by time attribute classification,

4. A method for retrieving data objects based on a spatio-temporal database according to one of the preceding claims, characterized in that the step of modeling the managed objects with respect to their temporal and spatial state comprises;

5. The method as claimed in claim 4, wherein the step of setting the specific attributes of the object to be managed described by the natural language according to the management model of the object to be managed comprises:

6. The method for retrieving data objects based on the spatio-temporal database as claimed in claim 5, wherein the step of setting specific attributes of the object to be managed described by means of natural language according to the management model of the object to be managed further comprises:

7. The method of claim 6, wherein the temporal attributes include at least a temporal location, a start time, and an end time,

8. Method for retrieving data objects based on a spatio-temporal database according to one of claims 5 to 7,

the temporal model and the spatial model have a multi-level structure identified by a natural language definition name,

9. A retrieval device for data objects based on a space-time database is characterized in that the retrieval device at least comprises a data acquisition module, at least one data calculation engine, a space-time database server, an engineering library server, a model library server and a client,

the model base server is provided with a model base, and the model base is configured to model the management object according to the time state and the space state of the object to be managed, wherein the object to be managed is subjected to data modeling based on the past time period, the present time period and the future time period, and a management model formed by the association of the space model and the time model of the object to be managed is established;

the engineering library server is provided with an engineering library, the engineering library is configured to set specific attributes of the object to be managed, which are described by virtue of natural language, according to a management model of the object to be managed, wherein the modeled object to be managed is classified into a specific model according to time attributes before the specific attributes are set;

the client is used for searching based on the management model category of the management object and the time attribute and/or the space attribute which is defined by the modeling and described by the natural language to determine the operation state of the management object, wherein the client determines a real-time database, a historical database and/or a plan database for storing the operation data of the management object according to the management model category of the management object, and selects at least one version of the operation data of the management object in the real-time database, the historical database and/or the plan database related to the management object according to the time attribute and/or the space attribute,

the client is also capable of determining at least one management object associated with the management model and looking at real-time data, historical data and/or plan data of the management object to know the real-time running state, historical running state and/or plan running state of the management object, wherein the real-time data objects stored in the real-time database are organized according to the spatial characteristics, and the historical data objects and plan data objects of the historical database and the plan database are organized according to the spatial characteristics and then according to the temporal characteristics.