A full-dimensional digital operation decision-making method, system and storage medium

Abstract

The present invention relates to the field of digital information processing, and specifically to a full-dimensional digital operation decision-making method and system. It collects the full-dimensional data sets of all user terminals and divides them into several data subsets corresponding to different attributes, and stores all the data subsets in the container units subordinate to the data storage array; docks the application program set of the cloud platform with the data storage array to implement the matching task processing of the data subsets, and returns the task processing results to the container units; loads the container units into the modeling engine to implement the modeling operation of the modeling engine; returns the modeling results of the modeling engine to the operation decision-making task thread of the user terminal, so as to send control instructions to the device terminals subordinate to the user terminal, providing reliable decision-making guidance for determining the working state control strategy of the device terminals. The present invention can improve the operation efficiency of the device terminals and the operation coordination between different device terminals, as well as improve the overall operation efficiency of the park and reduce the accident probability of the park.

Description

Technical Field

[0001] The present invention relates to the field of digital information processing, and particularly to a full-dimensional digital operation decision-making method, system and storage medium. Background Art

[0002] Smart industrial parks connect a large number of user terminals and device terminals to the cloud platform through the Internet of Things. The user terminals can be terminals such as smart phones or computers held by the internal staff of the park, and the device terminals can be sensing devices distributed at different locations within the park and production devices performing different production processes. As the control party sinking to the specific production processes within the park, the user terminal can directly control the corresponding production devices to achieve the efficient and accurate operation of the production devices.

[0003] The main function of a smart industrial park is to manage and control all production devices within the park in a coordinated manner. Currently, the relationship between user terminals and device terminals within the park is only limited to one-to-one management and control, that is, the user terminal can only control the operation of the device terminal based on a single data source. The above method cannot ensure that the device terminal can adapt to the complex and changeable environment within the park. For example, when the power supply within the park is unstable and / or there are multiple fire hazard factors, the user terminal will not be able to accurately adjust the working state of the device terminal according to multi-dimensional data such as power supply and fire hazards within the park, and cannot provide an accurate and reliable operation decision-making plan for the efficient operation of the device terminals within the park and the coordinated operation between different device terminals. It can be seen that how to use the full-dimensional data within the park to achieve reliable management of the device terminal by the user terminal, the efficient operation of the device terminal itself, and the coordinated operation between different device terminals is of great significance for maintaining the overall operation efficiency of the park and reducing the probability of production accidents in the park. Summary of the Invention

[0004] In order to avoid affecting the operation efficiency of the device terminal and the operation coordination between different device terminals by using single data for operation decision-making management of the device terminal, and to improve the overall operation efficiency of the park and reduce the probability of accidents in the park, the present invention provides a full-dimensional digital operation decision-making method, and the method includes the following steps:

[0005] Monitor all user terminals within the park, collect the full-dimensional data sets of all user terminals respectively, and divide the full-dimensional data sets into several data subsets based on the data attributes of the full-dimensional data sets; based on the storage state characteristics of the data storage array, store all the data subsets into the corresponding container units under the data storage array respectively;

[0006] Dock the application program set of the cloud platform with the data storage array to perform matching task processing on the data subsets; based on the trust status of the matching task processing results, return and store the matching task processing results to the corresponding container units;

[0007] Select matching container units from the data storage array based on the modeling requirement characteristics of the modeling engine and load them into the modeling engine; perform change processing on the matching container units based on the modeling state characteristics of the modeling engine;

[0008] Return the modeling result of the modeling engine to the operation decision task thread of the client based on the running state characteristics of the client; send a control instruction to the device end subordinate to the client based on the execution state characteristics of the operation decision task thread.

[0009] Preferably, monitor all clients in the park, collect the full-dimensional data sets of all clients respectively, divide the full-dimensional data sets into several data subsets based on the data attributes of the full-dimensional data sets; based on the storage state characteristics of the data storage array, store all data subsets into the corresponding container units subordinate to the data storage array respectively, specifically:

[0010] Determine the software tools in the normal state in each client based on the work logs of all clients in the park; determine the monitoring operation parameters for the software tools in the normal state based on the interface bandwidth of the software tools in the normal state in the client, so as to collect the full-dimensional data set from all software tools in the normal state in each client;

[0011] Mark all data based on the type attributes and timeliness attributes of all data subordinate to the full-dimensional data set, so as to divide the full-dimensional data set into several data subsets;

[0012] Determine the data read / write busy / idle attributes of all container units based on the data read / write states of all container units subordinate to the data storage array; identify all container units subordinate to the data storage array in the available state based on the data read / write busy / idle attributes; store all data subsets into the corresponding available container units respectively based on the data type compatibility characteristics and storage capacity of the available container units.

[0013] Preferably, dock the application program set of the cloud platform with the data storage array to perform matching task processing on the data subsets; return and store the matching task processing result to the corresponding container unit based on the trustworthiness state of the matching task processing result, specifically:

[0014] Verify all the applications installed in the cloud platform, determine the data processing permission information of all the applications, and integrate the applications that meet the preset data processing permission requirements into an application set; based on the gateway status of the network where the cloud platform is located, dock the application set with the data storage array; and perform matching task processing on the data subset based on the running threads of all the applications under the application set;

[0015] Obtain the thread running error occurrence parameters of the matching task processing, and based on the thread running error occurrence parameters, determine whether the matching task processing result is credible;

[0016] After cleaning the container unit where the data subset is located, return and store all the credible matching task processing results to the corresponding container unit.

[0017] Preferably, based on the modeling requirement characteristics of the modeling engine, select a matching container unit from the data storage array and load it into the modeling engine; based on the modeling status characteristics of the modeling engine, perform change processing on the matching container unit, specifically:

[0018] Parse the knowledge graph of the modeling engine to determine the metadata type characteristics required by the modeling engine in the current modeling process, and use this as the modeling requirement characteristics; compare the metadata type characteristics with the container units in the data storage array that store the matching processing results to determine the container units matching the current modeling process, and load the image of the matching container unit into the modeling engine;

[0019] Monitor the modeling process of the modeling engine to obtain the modeling progress status characteristics of the modeling engine; based on the modeling progress status characteristics, clear the stored content of the used matching container units.

[0020] Preferably, based on the running status characteristics of the user terminal, return the modeling result of the modeling engine to the operation decision task thread of the user terminal; based on the execution status characteristics of the operation decision task thread, send a control instruction to the device terminal under the user terminal, specifically:

[0021] Parse the operation decision running instruction received by the user terminal to obtain the operation decision running environment status characteristics of the user terminal, and use this to determine whether the user terminal meets the preset operation decision running environment conditions; if it meets the preset operation decision running environment conditions, return the modeling result of the modeling engine to the operation decision task thread of the user terminal;

[0022] Extract the complete operation decision result output by the operation decision task thread based on the execution progress status characteristics of the operation decision task thread; based on the complete operation decision result, send a work behavior control instruction to the device end subordinate to the user end.

[0023] On the other hand, the present invention provides a full-dimensional digital operation decision system, and the system includes the following modules:

[0024] A data collection and division module, which is used to monitor all user ends in the park, collect the full-dimensional data sets of all user ends respectively, and divide the full-dimensional data sets into several data subsets based on the data attributes of the full-dimensional data sets;

[0025] A data storage control module, which is used to store all data subsets into the corresponding container units subordinate to the data storage array respectively based on the storage status characteristics of the data storage array;

[0026] A data processing control module, which is used to dock the application program set of the cloud platform with the data storage array to perform matching task processing on the data subsets; based on the trust status of the matching task processing result, return the matching task processing result to the corresponding container unit for storage;

[0027] A modeling engine control module, which is used to select a matching container unit from the data storage array and load it into the modeling engine based on the modeling requirement characteristics of the modeling engine;

[0028] A container change processing module, which is used to perform change processing on the matching container unit based on the modeling status characteristics of the modeling engine;

[0029] An operation decision execution module, which is used to return the modeling result of the modeling engine to the operation decision task thread of the user end based on the running status characteristics of the user end;

[0030] An instruction issuing module, which is used to send a control instruction to the device end subordinate to the user end based on the execution status characteristics of the operation decision task thread.

[0031] Preferably, the data collection and division module is used to monitor all user ends in the park, collect the full-dimensional data sets of all user ends respectively, and divide the full-dimensional data sets into several data subsets based on the data attributes of the full-dimensional data sets, specifically:

[0032] Based on the work logs of all user terminals in the park, determine the software tools in normal state in each user terminal; based on the interface bandwidth of the software tools in normal state at the user terminal, determine the monitoring operation parameters for the software tools in normal state, so as to collect a full-dimensional data set from all software tools in normal state in each user terminal;

[0033] Based on the type attributes and timeliness attributes of all data under the full-dimensional data set, all data are marked, so as to divide the full-dimensional data set into a number of data subsets;

[0034] The data storage control module is used to store all data subsets in corresponding container units under the data storage array based on the storage state characteristics of the data storage array, specifically:

[0035] Based on the data read and write status of all container units under the data storage array, determine the data read and write busy and idle attributes of all container units; based on the data read and write busy and idle attributes, identify all container units in the data storage array that are in an available state; based on the data type compatibility characteristics and storage capacity of all container units in an available state, store all data subsets in corresponding container units that are in an available state.

[0036] Preferably, the data processing control module is used to connect the application set of the cloud platform with the data storage array to perform matching task processing on the data subset; based on the credibility status of the matching task processing result, the matching task processing result is returned and stored in the corresponding container unit, specifically:

[0037] Verify all applications installed in the cloud platform, determine the data processing permission information of all applications, and integrate the applications that meet the preset data processing permission requirements into an application set; based on the gateway status of the network where the cloud platform is located, connect the application set with the data storage array; and based on the running threads of all applications under the application set, perform matching task processing on the data subset;

[0038] Obtaining thread running error occurrence parameters of the matching task processing, and judging whether the matching task processing result is credible based on the thread running error occurrence parameters;

[0039] After cleaning the container unit where the data subset is located, all credible matching task processing results are returned and stored in the corresponding container unit.

[0040] Preferably, the modeling engine control module is used to select matching container units from the data storage array and load them into the modeling engine based on the modeling requirement characteristics of the modeling engine, specifically:

[0041] Parse and process the knowledge graph of the modeling engine to determine the metadata type features required by the modeling engine in the current modeling process, and use this as the modeling requirement features; compare the metadata type features with the container units stored in the data storage array with matching processing results to determine the container units matching the current modeling process, and load the images of the matching container units into the modeling engine;

[0042] The container change processing module is used to perform change processing on the matching container units based on the modeling state features of the modeling engine. Specifically:

[0043] Monitor the modeling process of the modeling engine to obtain the modeling progress state features of the modeling engine; based on the modeling progress state features, clear the stored content of the used matching container units;

[0044] The operation decision execution module is used to return the modeling results of the modeling engine to the operation decision task thread of the user terminal based on the operation state features of the user terminal. Specifically:

[0045] Parse and process the operation decision running instructions received by the user terminal to obtain the operation decision running environment state features of the user terminal, and use this to determine whether the user terminal meets the preset operation decision running environment conditions; if the preset operation decision running environment conditions are met, return the modeling results of the modeling engine to the operation decision task thread of the user terminal;

[0046] The instruction issuing module is used to send control instructions to the device terminals subordinate to the user terminal based on the execution state features of the operation decision task thread. Specifically:

[0047] Based on the execution progress state features of the operation decision task thread, extract the complete operation decision results output by the operation decision task thread; based on the complete operation decision results, send work behavior control instructions to the device terminals subordinate to the user terminal.

[0048] In addition, the present invention also provides a computer-readable storage medium, on which a computer program is stored, and the computer program realizes the method described above when executed by a processor.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] Monitor all the client devices in the industrial park, collect the full-dimensional data sets of each client device, and divide the full-dimensional data sets into several data subsets based on the data attributes of the full-dimensional data sets; based on the storage state characteristics of the data storage array, store all the data subsets into the corresponding container units under the data storage array. Each production operation area in the industrial park is equipped with a client device, which can be but is not limited to a terminal such as a smart phone or a computer held by the staff responsible for the operation area, and it can actively or passively collect the full-dimensional data set corresponding to the production operation area; among them, the full-dimensional data can be but is not limited to the water, electricity and gas supply data of the production operation area, the operation status data of the production equipment, the logistics distribution data of the production raw materials, the on-the-job status data of the employees, etc. These full-dimensional data can represent the production process status in different aspects such as the production environment, production materials, and human resources in the industrial park. From the above introduction of the full-dimensional data set, it can be seen that the full-dimensional data set contains data of different types of attributes in the industrial park. In order to avoid data crosstalk during the process of subsequent application programs processing the full-dimensional data, it is necessary to classify and categorize the full-dimensional data set in advance. Based on the attributes of each piece of data under the full-dimensional data set, divide it into several data subsets to avoid other data with different category attributes being mixed into each data subset and affecting the credibility of the task processing results of the application program. In addition, the communication bandwidth of the client device itself is limited. If the cloud platform directly retrieves the data subset from the client device for the application program to process, it will cause a large communication load on the client device, affect the normal operation of the client device and even cause the client device to crash. Therefore, use the data storage array as a data storage transfer node between the client device and the cloud platform, and store different data subsets separately through all the container units in the data storage array to avoid crosstalk between different data subsets during the storage process and improve the storage independence and security of the data subsets.

[0051] Dock the application program set of the cloud platform with the data storage array to perform matching task processing on the data subset; based on the trust status of the matching task processing result, return and store the matching task processing result to the corresponding container unit. The cloud platform has high computing power and memory resource support. Compared with the user side, it can stably run multiple application programs simultaneously and achieve parallel operation of multiple application programs. Since the data sources required during the task processing of different application programs are different, in order to accurately provide data sources for the task processing of all application programs, the application program set of the cloud platform is docked with the data storage array, and each application program under the application program set can obtain a suitable data subset for processing matching tasks to ensure the correctness of the task processing of the application program. In addition, thread running errors will inevitably occur during the task processing of the application program, resulting in incorrect running results of the corresponding thread, and continuously accumulating during the thread running process, thus affecting the credibility of the final task processing result. The modeling operation of the modeling engine is based on the task processing results of all application programs in the cloud platform. In order to ensure the correct modeling of the modeling engine, it is necessary to determine whether the matching task processing result of the application program is credible, and return and store the credible matching task processing result to the container unit where its corresponding data subset is located, so that the modeling engine can accurately retrieve the required matching task processing result for modeling operations at any time.

[0052] Based on the modeling requirement characteristics of the modeling engine, select a matching container unit from the data storage array and load it into the modeling engine; based on the modeling status characteristics of the modeling engine, perform change processing on the matching container unit. The modeling engine can be, but is not limited to, an engine that uses the computing power and memory resources of the cloud platform for modeling operations, so that there is no need for the modeling engine to have its own resource library, realizing the lightweight of the modeling engine. The modeling operation of the modeling engine is implemented according to the corresponding knowledge graph, and the knowledge graph represents the process of the modeling operation. In order to provide a reliable modeling data source for the entire modeling operation of the modeling engine, it is necessary to first determine the modeling requirement characteristics of the modeling engine, compare them with the matching processing results stored in all container units in the data storage array, determine the data sources required for each modeling process of the modeling engine, and thus accurately load the container unit into the modeling engine, facilitating the modeling engine to directly and accurately retrieve data from the container unit for modeling. Also, when the modeling engine has completed a certain modeling process and no longer needs to retrieve data from the container unit, the container unit can be changed at this time to avoid long-term occupation of the container unit and improve the recycling efficiency of the container unit.

[0053] Based on the operating status characteristics of the user terminal, return the modeling results of the modeling engine to the operation decision task thread of the user terminal; based on the execution status characteristics of the operation decision task thread, send control instructions to the device terminal subordinate to the user terminal. The user terminal is directly connected to the device terminal used to execute the production process in a control manner, so that the user terminal can independently adjust the operating status of the device terminal and coordinate the operating relationships between different device terminals. In order to improve the control and adjustment performance of the user terminal over the device terminal, use the modeling results generated by the modeling engine to guide the control strategy of the user terminal for the device terminal. For this purpose, based on the operation decision status during the process of the user terminal selecting and executing different control strategies for the device terminal, select the corresponding modeling results from the modeling engine and return them to the operation decision task thread of the user terminal. After the user terminal completes the operation decision task thread, based on the complete operation decision results, instruct the user terminal to send work behavior control instructions to the subordinate device terminals, so that the device terminals can adjust their own working status, so as to match the environmental conditions of the park and / or coordinate with other device terminals in the work process. Brief Description of the Drawings

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following-described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. Among them:

[0055] Figure 1 It is a flowchart of a full-dimensional digital operation decision method provided by the present invention.

[0056] Figure 2 It is a structural diagram of a full-dimensional digital operation decision system provided by the present invention. Detailed Embodiments

[0057] In order to make the above-mentioned objects, features, and advantages of the present invention more obvious and understandable, the following will make a detailed description of the specific embodiments of the present invention in conjunction with the drawings. It can be understood that the specific embodiments described here are only used to explain the present invention, rather than limiting the present invention. In addition, it should be noted that for the sake of description, only parts related to the present invention are shown in the drawings, rather than all structures. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0058] The terms "comprising" and "having" and any variations thereof in the present invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally further include steps or units not listed, or may optionally further include other steps or units inherent to these processes, methods, products or devices.

[0059] Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The phrase appears in various places in the specification and does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0060] Please refer to Figure 1 As shown, the present invention provides a full-dimensional digital operation decision-making method, which includes the following steps:

[0061] S100, monitor all user terminals in the park, collect the full-dimensional data sets of all user terminals respectively, divide the full-dimensional data sets into several data subsets based on the data attributes of the full-dimensional data sets; based on the storage state characteristics of the data storage array, store all data subsets into the corresponding container units under the data storage array respectively.

[0062] Further, monitoring all user terminals in the park, collecting the full-dimensional data sets of all user terminals respectively, dividing the full-dimensional data sets into several data subsets based on the data attributes of the full-dimensional data sets; based on the storage state characteristics of the data storage array, storing all data subsets into the corresponding container units under the data storage array respectively, specifically:

[0063] Based on the working logs of all user terminals in the park, determine the software tools in a normal state in each user terminal; based on the interface bandwidth of the software tools in a normal state at the user terminal, determine the monitoring operation parameters for the software tools in a normal state, so as to collect the full-dimensional data sets from all software tools in a normal state in each user terminal;

[0064] Based on the type attributes and timeliness attributes of all data under the full-dimensional data set, mark all data, thereby dividing the full-dimensional data set into several data subsets;

[0065] Determine the data read / write busy / idle attributes of all container units under the data storage array based on the data read / write status of all container units under the data storage array; identify all container units in an available state under the data storage array based on the data read / write busy / idle attributes; and store all data subsets into the corresponding available container units respectively based on the data type compatibility characteristics and storage capacities of all available container units.

[0066] In each production operation area within the industrial park, a user terminal will be arranged. The user terminal can be, but is not limited to, a terminal such as a smart phone or computer held by the staff responsible for the operation area, or a server terminal responsible for all device terminals within the operation area. It can actively or passively collect the full-dimensional data set corresponding to the production operation area, that is, the user terminal can actively collect the corresponding operation data from the water, electricity, and gas lines and device terminals corresponding to the operation area or regularly receive the operation data fed back by the water, electricity, and gas lines and device terminals corresponding to the operation area, etc., so as to integrate all operation data into a full-dimensional data set. Among them, the full-dimensional data can be, but is not limited to, the water, electricity, and gas supply data of the production operation area, the operation status data of production equipment, the production raw material logistics distribution data, the employee on-duty status data, etc. These full-dimensional data can represent the production process status in different aspects such as the production environment, production materials, and human resources in the industrial park. The user terminal realizes active or passive collection of the full-dimensional data set through the software tool installed on itself, and the working state of the software tool directly affects the reliability of the collected full-dimensional data. When the software tool works abnormally, there will be a large amount of interference noise in the collected full-dimensional data. First, based on the respective work logs of all user terminals, determine the software tools in a normal working state within each user terminal, and then only listen to and extract the full-dimensional data set for the software tools in a normal working state. Also, based on the interface bandwidth of the software tools in a normal working state on the user terminal, determine parameters such as the listening operation frequency of the software tools in a normal working state. Generally speaking, the larger the interface bandwidth of the software tool, the greater the listening operation frequency of the software tool, making full use of the interface bandwidth of the interface tool to improve the extraction speed of the full-dimensional data set of the software tool. From the above introduction of the full-dimensional data set, it can be seen that the full-dimensional data set contains data of different types of attributes in the industrial park. In order to avoid data crosstalk during the subsequent processing of the full-dimensional data by the application program, it is necessary to pre-classify and categorize the full-dimensional data set. Based on the attributes of each data item under the full-dimensional data set, divide it into several data subsets to prevent other data with different category attributes from being mixed into each data subset and affecting the credibility of the task processing results of the application program.

[0067] In addition, the communication bandwidth of the user side itself is limited. If the cloud platform directly retrieves data subsets from the user side for application processing, it will cause a relatively large communication load on the user side, affecting the normal operation of the user side and even causing the user side to crash. Therefore, a data storage array is used as a data storage transfer node between the user side and the cloud platform, and all container units (i.e., independent storage units) in the data storage array store different data subsets separately, avoiding crosstalk between different data subsets during storage and improving the storage independence and security of the data subsets. Specifically, based on the data read / write operation processes of all container units under the data storage array, it is determined whether a container unit is performing a data read / write operation. If so, it is determined that the container unit is in a busy state of data read / write; if not, it is determined that the container unit is in an idle state of data read / write, and the container units in the idle state of data read / write are determined as available container units to limit the container units for subsequent storage of data subsets in the data storage array. Then, based on the data type compatibility characteristics and storage capacity of all available container units, each data subset is stored in a container unit with sufficient storage capacity and compatible with the data subset type.

[0068] S200, dock the application program set of the cloud platform with the data storage array to perform matching task processing on the data subsets; based on the trust status of the matching task processing results, return and store the matching task processing results to the corresponding container units.

[0069] Furthermore, dock the application program set of the cloud platform with the data storage array to perform matching task processing on the data subsets; based on the trust status of the matching task processing results, return and store the matching task processing results to the corresponding container units, specifically:

[0070] Verify all the application programs installed in the cloud platform to determine the data processing permission information of all the application programs, and thus integrate the application programs that meet the preset data processing permission requirements into an application program set; based on the gateway status of the network where the cloud platform is located, dock the application program set with the data storage array; and perform matching task processing on the data subsets based on the running threads of all the application programs under the application program set.

[0071] Obtain the thread running error occurrence parameters of the matching task processing, and based on the thread running error occurrence parameters, determine whether the matching task processing results are trustworthy.

[0072] After cleaning the container units where the data subsets are located, return and store all the trustworthy matching task processing results to the corresponding container units.

[0073] The cloud platform has high computing power and memory resource support. Compared with the user side, it can stably run multiple application programs simultaneously and achieve parallel operation of multiple application programs. In order to uniformly manage all application programs in the cloud platform, first verify all the application programs installed in the cloud platform to determine the data processing permission information of all application programs. The data processing permission information can be, but is not limited to, the quantity permissions of computing power resources and memory resources that the application program is allowed to call during data processing. Application programs whose allowed called computing power resources and memory resources do not exceed the preset quantity threshold (i.e., application programs that meet the preset data processing permission requirements) are integrated into an application program set to achieve unified resource retrieval management for all subordinate application programs in the set. Then, based on the available bandwidth of each gateway under the network where the cloud platform is located (i.e., the gateway status), the application programs in the application program set are docked with the container units in the data storage array through the corresponding gateways, and based on the running threads of all application programs subordinate to the application program set, matching task processing is performed on the data subsets in the container units to ensure the accurate processing of the data subsets by the application programs.

[0074] In addition, thread operation errors inevitably occur during the task processing of the application program, resulting in incorrect running results of the corresponding thread, and continuously accumulating during the thread operation process, thus affecting the credibility of the final task processing result. The modeling operation of the modeling engine is based on the task processing results of all application programs in the cloud platform as basic data. In order to ensure the correct modeling of the modeling engine, it is necessary to determine whether the matching task processing results of the application programs are credible, and return the credible matching task processing results to be stored in the container unit where their corresponding data subsets are located, facilitating the modeling engine to accurately retrieve the required matching task processing results for modeling operations at any time. At the same time, cleaning processing is also performed on the container units where the data subsets are located, that is, clearing the original stored data subsets in the container units to avoid crosstalk between the data subsets and the matching task processing results in the container units.

[0075] S300, based on the modeling requirement characteristics of the modeling engine, select a matching container unit from the data storage array and load it into the modeling engine; based on the modeling status characteristics of the modeling engine, perform change processing on the matching container unit.

[0076] Furthermore, based on the modeling requirement characteristics of the modeling engine, select a matching container unit from the data storage array and load it into the modeling engine; based on the modeling status characteristics of the modeling engine, perform change processing on the matching container unit, specifically as follows:

[0077] Parse the knowledge graph of the modeling engine to determine the metadata type characteristics required by the modeling engine during the current modeling process, and use this as the modeling requirement characteristics; compare the metadata type characteristics with the container units in the data storage array that store the matching processing results to determine the container units that match the current modeling process, and load the images of the matching container units into the modeling engine;

[0078] Monitor the modeling process of the modeling engine to obtain the modeling progress status characteristics of the modeling engine; based on the modeling progress status characteristics, clear the stored content of the used matching container units.

[0079] The modeling engine can be, but is not limited to, an engine that uses the computing power and memory resources of the cloud platform for modeling operations, so that there is no need for the modeling engine to have its own resource library, realizing the lightweight of the modeling engine. The modeling operation of the modeling engine is implemented according to the corresponding knowledge graph, and the knowledge graph represents the process of the modeling operation. In order to provide a reliable modeling data source for the entire modeling operation of the modeling engine, it is necessary to first determine the modeling requirement characteristics of the modeling engine (that is, the data sources required for each modeling process included in the entire modeling operation of the modeling engine), and compare them with the matching processing results stored in all container units in the data storage array to determine the data sources required for each modeling process of the modeling engine, so as to accurately load the images of the container units into the modeling engine, facilitating the modeling engine to directly and accurately retrieve data from the container units for modeling. Also, when the modeling engine has completed a certain modeling process and no longer needs to retrieve data from the container unit, the stored content of the container unit can be cleared at this time to avoid the long-term occupation of the container unit and improve the recycling efficiency of the container unit.

[0080] S400, based on the running state characteristics of the user side, return the modeling results of the modeling engine to the operation decision task thread of the user side; based on the execution state characteristics of the operation decision task thread, send a control instruction to the device side subordinate to the user side.

[0081] Furthermore, based on the running state characteristics of the user side, return the modeling results of the modeling engine to the operation decision task thread of the user side; based on the execution state characteristics of the operation decision task thread, send a control instruction to the device side subordinate to the user side, specifically:

[0082] Parse the operation decision running instruction received by the user side to obtain the operation decision running environment state characteristics of the user side, and use this to judge whether the user side meets the preset operation decision running environment conditions; if it meets the preset operation decision running environment conditions, return the modeling results of the modeling engine to the operation decision task thread of the user side;

[0083] Extract the complete operation decision result output by the operation decision task thread based on the execution progress status feature of the operation decision task thread; send a work behavior control instruction to the device side subordinate to the user side based on the complete operation decision result.

[0084] The user side is directly controlled and connected to the device side used to execute the production process, so that the user side can independently adjust the operating state of the device side and coordinate the operating relationship between different device sides. In order to improve the control and adjustment performance of the user side over the device side, the modeling result generated by the modeling engine is used to guide the control strategy of the user side for the device side. For this purpose, the operation decision operation instruction received by the user side is parsed and processed to obtain the operation decision operation environment status feature of the user side (such as the performance parameters of the operation decision operation hardware environment and software environment of the user side), so as to judge whether the user side meets the preset operation decision operation environment conditions, that is, whether it meets the performance requirements of the preset operation decision operation hardware environment and software environment. When the performance requirements of the preset operation decision operation hardware environment and software environment are met, the modeling result of the modeling engine is returned to the operation decision task thread of the user side, so that the user side can use the modeling result as the basis for operation decision. Also, based on the execution progress status feature of the operation decision task thread, extract the complete operation decision result output by the operation decision task thread, so as to instruct the user side to send a work behavior control instruction to the subordinate device side, so that the device side can adjust its own working state, so as to match the environmental conditions of the park and / or coordinate with other device sides in the work process.

[0085] Please refer to Figure 2 As shown, the present invention provides a full-dimensional digital operation decision system, which includes the following modules:

[0086] The data collection and division module is used to monitor all user sides in the park, collect the full-dimensional data sets of all user sides respectively, and divide the full-dimensional data sets into several data subsets based on the data attributes of the full-dimensional data sets;

[0087] The data storage control module is used to store all data subsets into the corresponding container units subordinate to the data storage array respectively based on the storage status feature of the data storage array;

[0088] The data processing control module is used to dock the application program set of the cloud platform with the data storage array, so as to perform matching task processing on the data subsets; based on the trust status of the matching task processing result, return the matching task processing result to the corresponding container unit for storage;

[0089] The modeling engine control module is used to select a matching container unit from the data storage array based on the modeling requirement feature of the modeling engine and load it into the modeling engine;

[0090] A container change processing module, which is used to perform change processing on the matching container units based on the modeling state characteristics of the modeling engine;

[0091] An operation decision execution module, which is used to return the modeling results of the modeling engine to the operation decision task thread of the user side based on the operation state characteristics of the user side;

[0092] An instruction issuing module, which is used to send control instructions to the device side subordinate to the user side based on the execution state characteristics of the operation decision task thread.

[0093] Furthermore, the data collection and division module is used to monitor all user sides in the park, collect the full-dimensional data sets of all user sides respectively, and divide the full-dimensional data sets into several data subsets based on the data attributes of the full-dimensional data sets. Specifically:

[0094] Based on the working logs of all user sides in the park, determine the software tools in the normal state in each user side; based on the interface bandwidth of the software tools in the normal state at the user side, determine the monitoring operation parameters for the software tools in the normal state, so as to collect the full-dimensional data sets from all the software tools in the normal state in each user side;

[0095] Based on the type attributes and timeliness attributes of all data under the full-dimensional data set, mark all data, so as to divide the full-dimensional data set into several data subsets.

[0096] Furthermore, the data storage control module is used to store all data subsets into the corresponding container units subordinate to the data storage array respectively based on the storage state characteristics of the data storage array. Specifically:

[0097] Based on the data read-write states of all container units subordinate to the data storage array, determine the data read-write busy-idle attributes of all container units; based on the data read-write busy-idle attributes, identify all container units in the available state subordinate to the data storage array; based on the data type compatibility characteristics and storage capacities of all container units in the available state, store all data subsets into the corresponding container units in the available state respectively.

[0098] Furthermore, the data processing control module is used to dock the application program set of the cloud platform with the data storage array, so as to perform matching task processing on the data subsets; based on the trustworthiness state of the matching task processing results, return and store the matching task processing results to the corresponding container units. Specifically:

[0099] Verify all the applications installed in the cloud platform, determine the data processing permission information of all applications, and integrate the applications that meet the preset data processing permission requirements into an application set; based on the gateway status of the network where the cloud platform is located, dock the application set with the data storage array; and perform matching task processing on the data subset based on the running threads of all the applications under the application set.

[0100] Obtain the thread running error occurrence parameters of the matching task processing, and based on the thread running error occurrence parameters, determine whether the matching task processing result is credible.

[0101] After cleaning the container unit where the data subset is located, return and store all the credible matching task processing results to the corresponding container unit.

[0102] Furthermore, the modeling engine control module is used to select a matching container unit from the data storage array based on the modeling requirement characteristics of the modeling engine and load it into the modeling engine. Specifically:

[0103] Parse the knowledge graph of the modeling engine to determine the metadata type characteristics required by the modeling engine in the current modeling process, and use this as the modeling requirement characteristics; compare the metadata type characteristics with the container units stored in the data storage array with matching processing results to determine the container units matching the current modeling process, and load the images of the matching container units into the modeling engine.

[0104] Furthermore, the container change processing module is used to perform change processing on the matching container units based on the modeling status characteristics of the modeling engine. Specifically:

[0105] Monitor the modeling process of the modeling engine to obtain the modeling progress status characteristics of the modeling engine; based on the modeling progress status characteristics, clear the stored content of the used matching container units.

[0106] Furthermore, the operation decision execution module is used to return the modeling result of the modeling engine to the operation decision task thread of the user side based on the running status characteristics of the user side. Specifically:

[0107] Parse the operation decision running instruction received by the user side to obtain the operation decision running environment status characteristics of the user side, and use this to determine whether the user side meets the preset operation decision running environment conditions; if it meets the preset operation decision running environment conditions, return the modeling result of the modeling engine to the operation decision task thread of the user side.

[0108] Furthermore, the instruction issuing module is used to send control instructions to the device side under the user side based on the execution status characteristics of the operation decision task thread. Specifically:

[0109] Based on the execution progress status characteristics of the operation decision task thread, extract the complete operation decision result output by the operation decision task thread; based on the complete operation decision result, send a work behavior control instruction to the device end subordinate to the user end.

[0110] The operation and effect of the full-dimensional digital operation decision system of the present invention correspond to those of the above-mentioned full-dimensional digital operation decision method, and the description of the full-dimensional digital operation decision system will not be repeated here.

[0111] In an embodiment of the present invention, the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method described above is implemented.

[0112] In an embodiment of the present invention, the present invention further provides a computer device, the computer device at least includes a memory and a processor, and a computer program is stored on the memory, and when the computer program is executed by the processor, the method described above is implemented.

[0113] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of adding a necessary general hardware platform, and of course, it can also be implemented by a combination of hardware and software. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a computer product. The present invention can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them. Other embodiments can also be adopted; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A full-dimensional digital operation decision-making method, characterized in that: The method comprises the following steps: Monitor all user terminals in the park, collect full-dimensional data sets of all user terminals, and divide the full-dimensional data sets into several data subsets based on the data attributes of the full-dimensional data sets; based on the storage state characteristics of the data storage array, store all data subsets in the corresponding container units under the data storage array; Connecting the application set of the cloud platform with the data storage array to perform matching task processing on the data subset; returning the matching task processing result to the corresponding container unit based on the credibility of the matching task processing result; Based on the modeling requirement characteristics of the modeling engine, a matching container unit is selected from the data storage array and loaded into the modeling engine; based on the modeling state characteristics of the modeling engine, the matching container unit is changed, which is specifically: The knowledge graph of the modeling engine is parsed to determine the metadata type features required by the modeling engine in the current modeling process, and the features are used as the modeling requirement features; the metadata type features are compared with the container units storing the matching processing results in the data storage array to determine the container units matching the current modeling process, and the mirror image of the matching container unit is loaded into the modeling engine; Monitor the modeling process of the modeling engine to obtain the modeling progress status characteristics of the modeling engine; based on the modeling progress status characteristics, clear the storage content of the used matching container units; Based on the running state characteristics of the user terminal, the modeling result of the modeling engine is returned to the operation decision task thread of the user terminal; based on the execution state characteristics of the operation decision task thread, a control instruction is sent to the device terminal under the user terminal, which is specifically: Parsing and processing the operation decision operation instruction received by the user terminal to obtain the operation decision operation environment state characteristics of the user terminal, thereby judging whether the user terminal meets the preset operation decision operation environment conditions; if the preset operation decision operation environment conditions are met, returning the modeling result of the modeling engine to the operation decision task thread of the user terminal; Based on the execution progress status characteristics of the operation decision task thread, the complete operation decision result output by the operation decision task thread is extracted; based on the complete operation decision result, a work behavior control instruction is sent to the device terminal under the user terminal.

2. The method according to claim 1, characterized in that The monitoring of all user terminals in the park, collecting the full-dimensional data sets of all user terminals, dividing the full-dimensional data sets into a number of data subsets based on the data attributes of the full-dimensional data sets; based on the storage state characteristics of the data storage array, storing all data subsets in the corresponding container units under the data storage array, specifically: Based on the work logs of all user terminals in the park, determine the software tools in normal state in each user terminal; based on the interface bandwidth of the software tools in normal state at the user terminal, determine the monitoring operation parameters for the software tools in normal state, so as to collect a full-dimensional data set from all software tools in normal state in each user terminal; Based on the type attributes and timeliness attributes of all data under the full-dimensional data set, all data are marked, so as to divide the full-dimensional data set into a number of data subsets; Based on the data read and write status of all container units under the data storage array, determine the data read and write busy and idle attributes of all container units; based on the data read and write busy and idle attributes, identify all container units under the data storage array that are in an available state; Based on the data type compatibility characteristics and storage capacity of all container units in an available state, all data subsets are stored in corresponding container units in an available state.

3. The method according to claim 1, characterized in that The application set of the cloud platform is connected to the data storage array to perform matching task processing on the data subset; based on the credibility of the matching task processing result, the matching task processing result is returned and stored in the corresponding container unit, specifically: Verify all applications installed in the cloud platform and determine the data processing permission information of all applications, so as to integrate the applications that meet the preset data processing permission requirements into an application set; Based on the gateway status of the network where the cloud platform is located, the application set is connected to the data storage array; and based on the running threads of all applications under the application set, matching task processing is performed on the data subset; Obtaining thread running error occurrence parameters of the matching task processing, and judging whether the matching task processing result is credible based on the thread running error occurrence parameters; After cleaning the container unit where the data subset is located, all credible matching task processing results are returned and stored in the corresponding container unit.

4. A full-dimensional digital operation decision-making system, characterized in that: The system includes the following modules: A data collection and division module is used to monitor all user terminals in the park, collect the full-dimensional data sets of all user terminals, and divide the full-dimensional data sets into several data subsets based on the data attributes of the full-dimensional data sets; A data storage control module, used for storing all data subsets in corresponding container units under the data storage array based on the storage state characteristics of the data storage array; A data processing control module, used to connect the application set of the cloud platform with the data storage array, so as to perform matching task processing on the data subset; Based on the credibility status of the matching task processing result, returning the matching task processing result to the corresponding container unit for storage; The modeling engine control module is used to select a matching container unit from the data storage array and load it into the modeling engine based on the modeling requirement characteristics of the modeling engine, which is specifically: Parsing the knowledge graph of the modeling engine to determine the metadata type features required by the modeling engine in the current modeling process as the modeling requirement features; Compare the metadata type feature with the container unit storing the matching processing result in the data storage array to determine the container unit matching the current modeling process, and load the image of the matching container unit into the modeling engine; The container change processing module is used to perform change processing on the matched container unit based on the modeling state characteristics of the modeling engine, which is specifically: Monitor the modeling process of the modeling engine to obtain the modeling progress status characteristics of the modeling engine; based on the modeling progress status characteristics, clear the storage content of the used matching container units; The operation decision execution module is used to return the modeling result of the modeling engine to the operation decision task thread of the user terminal based on the operation status characteristics of the user terminal, which is specifically: Parsing and processing the operation decision operation instruction received by the user terminal to obtain the operation decision operation environment state characteristics of the user terminal, thereby judging whether the user terminal meets the preset operation decision operation environment conditions; if the preset operation decision operation environment conditions are met, returning the modeling result of the modeling engine to the operation decision task thread of the user terminal; The instruction issuing module is used to send control instructions to the device terminal under the user terminal based on the execution status characteristics of the operation decision task thread, which is specifically: Based on the execution progress status characteristics of the operation decision task thread, the complete operation decision result output by the operation decision task thread is extracted; based on the complete operation decision result, a work behavior control instruction is sent to the device terminal under the user terminal.

5. The system according to claim 4, characterized in that The data collection and division module is used to monitor all user terminals in the park, collect the full-dimensional data sets of all user terminals, and divide the full-dimensional data sets into several data subsets based on the data attributes of the full-dimensional data sets, specifically: Based on the work logs of all user terminals in the park, determine the software tools in normal state in each user terminal; based on the interface bandwidth of the software tools in normal state at the user terminal, determine the monitoring operation parameters for the software tools in normal state, so as to collect a full-dimensional data set from all software tools in normal state in each user terminal; Based on the type attributes and timeliness attributes of all data under the full-dimensional data set, all data are marked, so as to divide the full-dimensional data set into a number of data subsets; The data storage control module is used to store all data subsets in corresponding container units under the data storage array based on the storage state characteristics of the data storage array, specifically: Based on the data read and write status of all container units under the data storage array, determine the data read and write busy and idle attributes of all container units; based on the data read and write busy and idle attributes, identify all container units under the data storage array that are in an available state; Based on the data type compatibility characteristics and storage capacity of all container units in an available state, all data subsets are stored in corresponding container units in an available state.

6. The system according to claim 4, characterized in that The data processing control module is used to connect the application set of the cloud platform with the data storage array to perform matching task processing on the data subset; based on the credibility of the matching task processing result, the matching task processing result is returned and stored in the corresponding container unit, specifically: Verify all applications installed in the cloud platform and determine the data processing permission information of all applications, so as to integrate the applications that meet the preset data processing permission requirements into an application set; Based on the gateway status of the network where the cloud platform is located, the application set is connected to the data storage array; and based on the running threads of all applications under the application set, matching task processing is performed on the data subset; Obtaining thread running error occurrence parameters of the matching task processing, and judging whether the matching task processing result is credible based on the thread running error occurrence parameters; After cleaning the container unit where the data subset is located, all credible matching task processing results are returned and stored in the corresponding container unit.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 3.

Images