Industrial Internet data storage method, system, computer equipment and storage medium

Abstract

The present disclosure provides an industrial Internet data storage method, system, computer equipment and storage medium, wherein the method includes: classifying the data storage locations of the industrial Internet to form data storage node sets of different levels, wherein the data storage node sets of the same level include several data storage nodes of the same level; constructing a data storage model, which includes data storage nodes of various levels, the connection relationship between nodes, and the benefits brought by the data generated by the local device nodes at the production site when the data is transmitted to other data storage nodes; based on the data storage model, using a preset data storage strategy to seek the optimal data storage node that meets the optimal benefits of the system; transmitting the data currently generated by the local device node to the optimal data storage node for storage. The technical solution provided by the present disclosure can realize data sharing while reducing latency and transmission consumption, and is more suitable for industrial Internet data storage.

Description

Technical Field

[0001] The present disclosure relates to the technical field of data storage, and in particular to an industrial Internet data storage method, an industrial Internet data storage system, a computer device, and a computer-readable storage medium. Background Art

[0002] Through the comprehensive interconnection of people, machines, and things, the Industrial Internet has achieved ubiquitous interconnection of all factors in the industrial chain links such as design, R&D, production, management, and services, including topics such as people, machines, workshops, and enterprises. It has built an advanced manufacturing system and modern service industry system that are fully connected with all factors, the entire industrial chain, and the entire value chain. It is the infrastructure for realizing the digitalization, networking, and intelligent development of industry, an important cornerstone of the Fourth Industrial Revolution, and a key support for promoting the deep integration of the digital economy and the real economy.

[0003] A large number of industrial equipment in the Industrial Internet will generate a large amount of data during the production process, and a data storage solution based on on-site data servers has been generated accordingly. However, due to the limitation of industrial equipment resources (such as storage and processing), the capabilities of industrial equipment are restricted, the cost of industrial equipment is increased, and the sharing of data is also affected, which is not conducive to the development of the full-factor Industrial Internet architecture. In order to solve this problem, a data storage solution based on private cloud or public cloud has been generated, which saves data to cloud servers with strong data storage and shared access capabilities. This solution will significantly enhance the capabilities of industrial equipment and effectively reduce costs. However, transferring data to cloud servers will face higher latency and transmission consumption, which will bring many problems to some equipment, such as industrial equipment that needs to strictly implement latency requirements. Summary of the invention

[0004] The present disclosure is completed in order to at least partially solve the technical problems existing in the prior art such as high latency and transmission consumption when transferring data generated by the industrial Internet to a cloud server.

[0005] According to one aspect of the present disclosure, a method for storing industrial Internet data is provided, the method comprising:

[0006] Classify the data storage locations of the Industrial Internet to form data storage node sets of different levels, where a data storage node set of the same level contains several data storage nodes of the same level;

[0007] Construct a data storage model that includes data storage nodes at all levels, the connection relationship between nodes, and the benefits of data generated by local device nodes at the production site when they are transferred to other data storage nodes;

[0008] Based on the data storage model, a preset data storage strategy is used to seek an optimal data storage node that satisfies the optimal benefit of the system, wherein the optimal benefit of the system refers to the maximum value of the total benefit brought by data transmission from the local device node to other data storage nodes;

[0009] The data currently generated by the local device node is transmitted to the optimal data storage node for storage.

[0010] According to another aspect of the present disclosure, an industrial Internet data storage system is provided, the system comprising:

[0011] A classification module, which is configured to classify the data storage locations of the industrial Internet to form data storage node sets of different levels, wherein a data storage node set of the same level includes a number of data storage nodes of the same level;

[0012] A construction module configured to construct a data storage model, which includes data storage nodes at all levels, connection relationships between nodes, and benefits brought by data generated by local device nodes at the production site being transmitted to other data storage nodes;

[0013] a target optimization module, which is configured to seek an optimal data storage node that satisfies the optimal system benefit by adopting a preset data storage strategy based on the data storage model, wherein the optimal system benefit refers to the maximum value of the total benefit brought by data being transmitted from the local device node to other data storage nodes; and

[0014] A transmission module is configured to transmit the data currently generated by the local device node to the optimal data storage node for storage.

[0015] According to another aspect of the present disclosure, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor runs the computer program stored in the memory, the processor executes the aforementioned industrial Internet data storage method.

[0016] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the processor executes the aforementioned industrial Internet data storage method.

[0017] The technical solution provided by the present disclosure may have the following beneficial effects:

[0018] The industrial Internet data storage method provided by the present invention first divides the data storage location of the industrial Internet into data storage nodes of different levels, then constructs a data storage model, and then finds the optimal data storage node in the data storage model based on a preset data storage strategy, and transmits the data currently generated by the local device node to the optimal data storage node for storage. Compared with the data storage solutions based on private cloud or public cloud in the prior art, while being able to realize data sharing, it also reduces latency and transmission consumption, and is more suitable for industrial Internet data storage.

[0019] Other features and advantages of the present disclosure will be described in the following description, and partly become apparent from the description, or be understood by practicing the present disclosure. The purpose and other advantages of the present disclosure can be realized and obtained by the structures particularly pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to provide further understanding of the technical solution of the present disclosure and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the technical solution of the present disclosure and do not constitute a limitation on the technical solution of the present disclosure.

[0021] Figure 1 A schematic diagram of a process flow of an industrial Internet data storage method provided by an embodiment of the present disclosure;

[0022] Figure 2 A schematic diagram of the structure of the industrial Internet data storage system provided by an embodiment of the present disclosure;

[0023] Figure 3 A schematic diagram of the structure of a computer device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0024] To make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the specific implementation of the present disclosure is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present disclosure, and is not used to limit the present disclosure.

[0025] The integrated and innovative development of the fifth generation mobile communication technology (5G) and the industrial Internet will drive the manufacturing industry to transform from single-point, local information technology applications to digitalization, networking and intelligence, and also open up a broader market space for 5G.

[0026] With the deployment of 5G networks, MEC (Mobile Edge Computing) in 5G networks has migrated data storage and processing capabilities to the scope of access networks. By providing IT service environments and cloud computing capabilities at the edge of mobile networks (such as base stations), it can provide low-latency, high-bandwidth, and real-time network services compared to existing cloud computing. Communication operators can make full use of the data storage and processing capabilities within the 5G network to provide rich data storage services for all parties in the Industrial Internet. Since MEC has high computing and data storage capabilities, this service also provides more abundant options for the large amount of data in the Industrial Internet.

[0027] However, while MEC in 5G networks provides more abundant data storage and data processing services, it also makes data storage optimization strategies more complicated. Therefore, how to store various types of data generated on industrial sites to different data storage nodes becomes the key to the problem. In order to solve this problem, the present disclosure provides a data storage solution for industrial Internet, which will be described in detail below.

[0028] It should be noted that, in the subsequent description, the use of suffixes such as "module", "component" or "unit" to represent elements is only to facilitate the description of the present disclosure, and has no specific meaning in itself. Therefore, "module", "component" or "unit" can be used in a mixed manner.

[0029] Figure 1 A flow chart of the industrial Internet data storage method provided by the embodiment of the present disclosure. Figure 1 As shown, the method includes the following steps S101 to S104.

[0030] S101. Classify the data storage locations of the Industrial Internet to form data storage node sets of different levels, wherein the data storage node sets of the same level include several data storage nodes of the same level;

[0031] S102. Constructing a data storage model, which includes data storage nodes at all levels, connection relationships between nodes, and the benefits of data generated by local device nodes at the production site when data is transmitted to other data storage nodes;

[0032] S103. Based on the data storage model, a preset data storage strategy is used to seek an optimal data storage node that satisfies the optimal system benefit, wherein the optimal system benefit refers to the maximum value of the total benefit brought by data transmission from the local device node to other data storage nodes;

[0033] S104. Transmit the data currently generated by the local device node to the optimal data storage node for storage.

[0034] In this embodiment, the data storage location of the industrial Internet is first divided into data storage nodes of different levels, and then a data storage model is constructed. Then, based on a preset data storage strategy, the optimal data storage node is found in the data storage model, and the data currently generated by the local device node is transmitted to the optimal data storage node for storage. Compared with the data storage solutions based on private clouds or public clouds in the prior art, while being able to realize data sharing, it also reduces latency and transmission consumption, and is more suitable for industrial Internet data storage.

[0035] In a specific implementation, the data storage nodes are divided into five levels, namely local device nodes, local data nodes, user-side data nodes, regional data nodes and cloud data nodes. Moreover, each level of data storage node set includes a number of data storage nodes of the same level.

[0036] In the Industrial Internet, a large amount of industrial field data needs to be stored and processed for use by local device nodes, other nodes (such as nodes in other factories for collaborative production), local applications, and other applications (such as production quality monitoring systems). Data storage is limited by the capacity of hardware devices and data processing capabilities. Storing data in different locations will have a significant impact on overall performance.

[0037] The rich network architecture of the 5G network enables it to have a rich variety of data storage node types, providing a variety of options for data storage. This embodiment makes full use of this feature of the 5G network to classify the data storage locations of the industrial Internet to form data storage nodes of different levels. Specifically, according to the characteristics of data storage nodes at different locations of the network, the data storage nodes are divided into local device nodes L1, local data nodes L2, 5G user-side data nodes L3, 5G regional data nodes L4 and cloud data nodes L5, so as to quickly find the optimal data storage node according to the data storage requirements.

[0038] Among them, local device node L1: a local device node located at the production site, which generates data for local control, monitoring, automatic production program processing and use. Its characteristics are that there is no transmission cost for program call data and the reading and writing speed is fast. However, due to cost, volume, and environmental constraints, the device storage and data speed are limited.

[0039] Local data node L2: A data storage and processing node deployed by a production enterprise at a production site for mutual invocation between devices and for use by processing programs that require the collection of multiple devices, such as production control programs that coordinate multiple devices. It is characterized by large data storage capacity, strong processing capabilities, and a relatively superior environment. However, due to network limitations, it mainly serves local fixed devices.

[0040] 5G user-side data node L3: Deployed in the MEC environment of the base station, it can be directly connected to the device nodes within the coverage of the base station, serving all nodes within the coverage area, with large data storage capacity, strong processing capacity, and a relatively superior environment. Since the base station and the equipment all use high-speed wireless links, the layout is flexible and suitable for regional production scenarios with mixed fixed and mobile equipment, such as production scenarios with automatic production equipment and handling robots. Since the MEC data node of the base station is close to the device node, it has the advantage of low latency and is suitable for processing programs with strict corresponding speed requirements. In addition, since the data node is inside the 5G network, it is convenient to coordinate production with device nodes in other regions. However, due to the limitations of the MEC hardware of the base station, it is not suitable for data storage and processing with very large amounts of data.

[0041] 5G regional data node L4: deployed in the MEC of the regional aggregation point of the 5G network, it can cover all device nodes in the entire area. Its data storage and data processing capabilities are more powerful than the MEC of the base station. Because it is located inside the 5G network, it can obtain network basic resources, which is convenient for data node systems in different locations to work. However, due to the long distance from the device nodes, it has greater latency and data transmission costs.

[0042] Cloud data node L5: Deployed on the cloud computing platform, it has powerful data storage and processing capabilities and is suitable for sharing data among various nodes within a large coverage area. Since it is located on a public network platform, data transmission needs to pass through the 5G network, resulting in large delays and high data transmission costs.

[0043] After dividing the data storage location of the industrial Internet into data storage nodes of different levels, in order to facilitate analysis and optimize calculations, the data storage is abstracted into a computing model G = (V, E), where V = {v1, v2, ..., v x} represents a set of x points, E = {e1, e2, ..., e y In this embodiment, points are used to represent data storage nodes at various levels, and lines are used to represent data paths between nodes, so as to realize data communication between nodes.

[0044] For any two nodes in model G, the comprehensive weight is To indicate that for Details are as follows.

[0045] In a specific implementation, the benefits brought by the data generated by the local device node at the production site in step S102 when it is transmitted to other data storage nodes are calculated using the following formula:

[0046]

[0047] in, Be1 represents the benefit brought by storing data from node i to node j, Be2 represents the benefit brought by the data storage capacity of node j (for example, data stored in L2-level nodes has a longer data storage period than data stored in L1-level nodes), Be3 represents the benefit brought by the processing capacity of node j (for example, data stored in L2-level nodes has the ability to share data with L1-level warehouse scheduling nodes, supporting warehouse scheduling nodes to arrange raw material supply according to production real-time data), and Be1 represents the benefit brought by the data storage capacity of node j (for example, data stored in L2-level nodes can use the data processing capacity of the node to process data, and the benefit brought by processing large amounts of real-time data into small amounts of feature data); i belongs to the local device node L1 of the production site, and j belongs to other data storage nodes L2-L5 except the local device nodes.

[0048] In a specific implementation, the data storage model constructed in step S102 satisfies preset delay constraints and reliability constraints, which are as follows:

[0049] in represents the delay constraint (also called time cost constraint) for data to be stored from node i to node j, τ i,j represents the delay constraint condition of data storage between node i and node j set by the system;

[0050] in represents the reliability constraint (also called quality constraint) of data stored from node i to node j, Q i,j Represents the reliability constraints of data storage between node i and node j set by the system.

[0051] In a specific embodiment, It is calculated using the following formula:

[0052] Among them, t1 represents the transmission time of data from node i to node j, t2 represents the time from the output of the first byte of data from node i to the completion of the transmission of the last byte, t3 represents the time from the first byte of data entering node j to the last byte entering node j, t4 represents the waiting time of data in the read queue of node i, and t5 represents the waiting time of data in the write queue of node j.

[0053] It is calculated using the following formula:

[0054] Among them, R1 represents the reliability of data stored in node j, specifically the reliability of the data storage device of node j (such as hardware redundancy, power supply, ambient temperature, humidity, etc.), R2 represents the reliability of the communication link between node i and node j, and R3 represents the security of data stored in node j (such as security protection level, security level of data access, etc.).

[0055] After adopting the data storage model described in the embodiment of the present disclosure, the optimization problem becomes seeking the optimal benefit of the entire system under the conditions of satisfying the delay constraint and reliability constraint when the data generated by the local device node at the production site is transmitted to other data storage nodes:

[0056] and

[0057] where i∈{v1,v2,...v m}, m is the local device node L1 of all production sites, which refers to the final device for data generation and consumption; j∈{v1,v2,...v n}, n is all the data storage nodes L2-L5 except the local device node L1; the connection relationship between node i and node j (i,j)∈{e1,e2,...,e k}; st means being subject to the following conditions (subject to).

[0058] For the large amount of data generated by L1 local device nodes, these data can be stored in any storage location in the industrial Internet. As for the specific storage location that makes the performance of the entire system the best, it is the optimization goal, and the system optimal benefit G m,n This is the target formula for optimization, but optimization is a complex process. Since there are too many factors to choose from, direct calculation is unsolvable and belongs to the NP problem (Nondeterministic Polynomially problem). It is impossible to obtain the optimal solution, or it takes too long to obtain the optimal solution. Therefore, it is necessary to find a suitable algorithm that can find the optimal solution within the specified time so that the system can achieve better performance. In order to obtain the optimal benefit G of the system m,n In order to find the optimal solution, the embodiment of the present disclosure adopts a particle swarm optimization algorithm to obtain a globally optimized data storage strategy, and then uses the data storage strategy to seek the optimal data storage node that satisfies the optimal benefit of the system.

[0059] The particle swarm optimization algorithm is an optimization algorithm that simulates the predation process of birds and fish. The particles in the initial state gather into multiple small groups through self-organization and move in the same direction at the same speed. Then the small groups gather into large groups, and the movement process of the large group may be dispersed into multiple small groups. Individual particles benefit from their own discoveries and historical experiences, as well as the discoveries and experiences of other particles. When many particles move in the same direction, they have good optimization performance.

[0060] After applying the particle swarm optimization algorithm in the embodiment of the present disclosure, in the target search space D, a particle swarm is composed of w particles from other data storage nodes above the node at the same level, where the attributes of the i-th particle at time t are composed of two vectors:

[0061] 1) Speed: in And v min and v max Represent the minimum and maximum values ​​of the velocity of the i-th particle respectively;

[0062] 2) Location: in And l D and u D Represent the lower and upper limits of the search space of the i-th particle respectively.

[0063] Two optimal positions are recorded in each iteration of the particle swarm optimization algorithm:

[0064] 1) Individual optimal position:

[0065] 2) Optimal position of population:

[0066] Among them, 1≤i≤w, 1≤d≤D, and D is the target search space.

[0067] The embodiment of the present disclosure uses a particle swarm optimization algorithm to obtain the preset data storage strategy in step S103, and then uses the strategy to obtain the optimal data storage node, specifically as follows:

[0068] In a specific implementation, step S103 specifically includes the following steps S103a to S103g.

[0069] S103a. Abstracting all data storage nodes in the data storage model into individual particles in a particle swarm;

[0070] S103b updates the speed and position of each particle according to the preset particle speed update formula and particle position update formula;

[0071] S103c. Calculate the gain of each particle after the speed and position are updated (i.e. );

[0072] S103d. Compare the benefit of each particle obtained in step S103c with its historical optimal benefit. If it exceeds its historical optimal benefit, the current position of the particle is used as the optimal position of the particle. Otherwise, return to step S103b to continue updating the speed and position of the particle.

[0073] In this step, each time a particle updates its speed and position, the particle's gain is calculated and recorded. Based on the historical records, the particle's historical optimal gain can be obtained, and the speed and position corresponding to the particle's historical optimal gain can also be obtained.

[0074] S103e. Compare the benefit corresponding to the optimal position of each particle with the optimal benefit of the population. If it exceeds the optimal benefit of the population, update the optimal position and the best fitness value of the population.

[0075] S103f. Determine whether a global optimal solution is obtained or a preset maximum number of iterations is reached. If not (neither a global optimal solution is obtained nor the maximum number of iterations is reached), return to step S103b and perform the next iteration; if yes, execute step S103f;

[0076] The population's optimal fitness value is used to evaluate the quality of the optimal solution found through iteration; the global optimal solution refers to an acceptable satisfactory solution; the maximum number of iterations can be set and adjusted by those skilled in the art according to actual needs;

[0077] S103g. The data storage node corresponding to the optimal position of the particle obtained after the iteration is completed is used as the optimal data storage node.

[0078] Furthermore, before step S103b, the following step may be included: initializing the population.

[0079] In this embodiment, after classifying the data storage locations of the industrial Internet, a data storage strategy is formed according to the delay, reliability, and benefit grading model. Starting from the local device node, selection can be made from a large number of data storage nodes at each level in the order of L1 to L5, thereby selecting nodes step by step until the optimal data storage node is found. This is consistent with the pattern of a large number of particles moving in one direction in the particle swarm optimization algorithm. Therefore, this embodiment uses the particle swarm optimization algorithm to form a data storage strategy for optimizing data storage.

[0080] In the classic particle swarm optimization algorithm, the speed of each particle is composed of three parts: the first is its own speed in the previous iteration, which is the inheritance of the particle's previous speed and reflects the inertia of particle motion; the second is "self-cognition", which indicates the influence of the particle's previous flight experience on the subsequent flight direction; the third is "social cognition", also known as social experience, which indicates the flight experience of all particles in the population. During the evolution of the particle swarm algorithm, each particle has been tracking two optimal values: one is the individual's historical optimal position, and the other is the population's historical optimal position.

[0081] However, the inventors found that the classical particle swarm optimization algorithm has the following problems when applied:

[0082] 1) A large number of particles will inquire about the status of each data storage node one by one, including common parameters such as current storage capacity, processing capacity, and security features. If each particle inquires, it will cause a large number of repeated query data packets, affecting the convergence speed of optimization and increasing system overhead;

[0083] 2) Each local data node generates different data features. For example, the quality monitoring data of production equipment needs to be stored in a higher-level data storage node so that it can be read by the monitoring system. The intermediate data of the production process has high latency requirements, but a small sharing range. It is only read by nodes around the node at this level to coordinate actions, so it is saved in a lower-level data storage node. Therefore, the data features of each particle are different. If they move in the same way, it will affect the convergence speed of the optimization.

[0084] 3) After a particle finds an optimal data storage node, a large number of other particles tend to the same data storage node, which will cause the burden of the data storage node to increase sharply and affect the performance of the system.

[0085] In view of the above problems, the present disclosure makes the following improvements to the particle swarm optimization algorithm:

[0086] In a specific implementation, step S103 further includes:

[0087] S103h. Each data storage node regularly broadcasts the current public status of the node to surrounding nodes.

[0088] In this embodiment, the feature broadcast function of the data storage node is added. Specifically, each data storage node regularly broadcasts the current public state of the node to the surrounding nodes. After such setting, each particle can know the public state of the next hop node of the node at the current node, which greatly reduces the number of queries for public feature information.

[0089] In a specific implementation, step S103 further includes:

[0090] S103i. Number each data storage node according to the level as (a, b), where a represents the number of the data storage node level (one of L1-L5), and b represents the number of other data storage nodes at the same level as the data storage node;

[0091] Accordingly, a vector parameter about the particle data characteristic value is added to the particle velocity part in the particle velocity update formula in step S103b to control whether the particle is transferred to the next node in the direction of level number a or in the direction of number b of other data storage nodes at the same level.

[0092] Since the data characteristics of the particles will affect the direction of the particle velocity v, this embodiment adds a vector parameter j to the particle velocity update formula. i (h), where h is the data feature value of the i-th particle. According to the size of its shared feature, parameter j i (h) It is possible to control whether the particle is transferred to the next node in the direction of level number a or in the direction of number b of other data storage nodes at the same level. For example, for data with a larger shared eigenvalue, it is transferred in the direction of a first, otherwise it is transferred in the direction of b first. Particles move according to the level number (a, b), which improves the optimization convergence speed of different types of data and reduces the probability of particles moving from high-level data storage nodes to low-level data storage nodes.

[0093] In a specific implementation, the social cognition part in the particle speed update formula in step S103b adds a data storage node index parameter, which affects the attraction of the optimal data storage node to the surrounding particles.

[0094] In this embodiment, a data storage node indicator parameter is added to the social cognition part (population history) in the particle speed update formula

[0095]

[0096] in, represents the data storage capacity of the i-th data storage node at time t, It represents the length of the data read queue of the i-th data storage node at time t, Represents the computing power of the i-th data storage node at time t.

[0097] parameter Influence the attraction of the optimal data storage node of the population to particles. When the optimal data storage node of the population is lightly loaded, the attraction to particles is increased. Conversely, when the load is heavy, the attraction to particles is reduced, thereby avoiding a large number of particles from selecting the same optimal data storage node at the same time, causing the performance of the node to drop sharply and affecting the performance of the entire system.

[0098] Combining the above improved methods, the particle velocity update formula is as follows:

[0099]

[0100] in, represents the velocity of the i-th particle at time t+1; j i (h) is a vector parameter, h is the data characteristic value of the i-th particle; represents the velocity of the ith particle at time t, represents the individual optimal position of the i-th particle at time t, represents the optimal position of the population at time t, represents the position of the i-th particle at time t; c1 and c2 are learning factors, generally c1=c2=2; r1 and r2 are random numbers between (0,1); is the data storage node indicator parameter; 1≤i≤w,1≤d≤D, D is the target search space.

[0101] The particle position update formula is as follows:

[0102]

[0103] in, represents the position of the i particles at time t+1; represents the position of the i-th particle at time t; represents the velocity of the i-th particle at time t+1.

[0104] Based on the foregoing content, it can be known that in order to save different types of data generated by different local device nodes L1 to the optimal data storage node according to the real-time status of the network, the present invention abstracts the on-site network of the industrial Internet into an appropriate data storage model based on the 5G network, including benefits, delay constraints, reliability constraints and other contents, and then, based on this model, uses a particle swarm optimization algorithm to obtain the globally optimized data storage strategy, and then uses this data storage strategy to seek the optimal data storage node that meets the system's optimal benefits.

[0105] It should be noted that the order of the above steps is only a specific example provided to illustrate the embodiments of the present disclosure. The present disclosure does not limit the order of the above steps, and those skilled in the art may adjust them as needed in practical applications.

[0106] The industrial Internet data storage method provided by the embodiment of the present disclosure first divides the data storage location of the industrial Internet into data storage nodes of different levels, and then adopts an improved particle swarm optimization algorithm to obtain a globally optimized data storage strategy. Then, based on the acquired data storage strategy, different types of data generated by different local device nodes are saved to the optimal data storage node according to the real-time status of the network. This can not only realize data sharing, but also reduce latency and transmission consumption, thereby improving the performance of the entire industrial Internet data storage.

[0107] Figure 2 This is a schematic diagram of the structure of the industrial Internet data storage system provided by the embodiment of the present disclosure. Figure 2 As shown, the system 2 includes: a classification module 201, a construction module 202, a target optimization module 203 and a transmission module 204.

[0108] Among them, the classification module 201 is configured to classify the data storage locations of the industrial Internet to form data storage node sets of different levels, wherein the data storage node sets of the same level include several data storage nodes of the same level; the construction module 202 is configured to construct a data storage model, which includes data storage nodes at various levels, the connection relationship between the nodes, and the benefits brought by the data generated by the local device nodes at the production site when they are transmitted to other data storage nodes; the target optimization module 203 is configured to use a preset data storage strategy to seek the optimal data storage node that meets the system's optimal benefits based on the data storage model, and the system's optimal benefits refer to the maximum value of the total benefits brought by the data being transmitted from the local device node to other data storage nodes; the transmission module 204 is configured to transmit the data currently generated by the local device node to the optimal data storage node for storage.

[0109] In a specific implementation, the data storage nodes formed by the classification module 201 are divided into five levels, namely: local device nodes, local data nodes, user-side data nodes, regional data nodes and cloud data nodes.

[0110] In a specific implementation, in the data storage model constructed by the construction module 202, the benefits brought by the data generated by the local device nodes at the production site when transmitted to other data storage nodes are calculated using the following formula:

[0111]

[0112] in, represents the benefit brought by data storage from node i to node j, Be1 represents the benefit brought by the data storage capacity of node j, Be2 represents the shared benefit of data stored in node j, and Be3 represents the benefit brought by the processing capacity of node j; i belongs to the local device node of the production site, and j belongs to other data storage nodes except the local device node.

[0113] In a specific implementation, the data storage model constructed by the construction module 202 satisfies the preset delay constraint condition and reliability constraint condition, which are as follows:

[0114] in represents the delay constraint for data to be stored from node i to node j, τ i,j represents the delay constraint condition of data storage between node i and node j set by the system;

[0115] in represents the reliability constraint of data stored from node i to node j, Q i,j Represents the reliability constraints of data storage between node i and node j set by the system.

[0116] In a specific embodiment, It is calculated using the following formula:

[0117] Wherein, t1 represents the transmission time of data from node i to node j, t2 represents the time from when the first byte of data is output from node i to when the last byte is transmitted, t3 represents the time from when the first byte of data enters node j to when the last byte enters node j, t4 represents the waiting time of data in the read queue of node i, and t5 represents the waiting time of data in the write queue of node j;

[0118] It is calculated using the following formula:

[0119] Among them, R1 represents the reliability of data stored in node j, R2 represents the reliability of the communication link between node i and node j, and R3 represents the security of data stored in node j.

[0120] In a specific implementation, the target optimization module 203 includes: a processing unit, an updating unit, a calculating unit, a first comparing unit, a second comparing unit and a determining unit.

[0121] Among them, the processing unit is configured to abstract all data storage nodes in the data storage model into individual particles in the particle swarm; the updating unit is configured to update the speed and position of each particle according to a preset particle speed update formula and a particle position update formula; the calculating unit is configured to calculate the benefit of each particle after the speed and position are updated; the first comparing unit is configured to compare the benefit of each particle with its historical optimal benefit, and if it exceeds its historical optimal benefit, its current position is used as the optimal position of the particle; the second comparing unit is configured to compare the benefit corresponding to the optimal position of each particle with the optimal benefit of the population to which it belongs, and if it exceeds the optimal benefit of the population to which it belongs, the optimal position and the best fitness value of the population are updated; then the updating unit, the calculating unit, the first comparing unit and the second comparing unit cooperate with each other to iterate multiple times until a global optimal solution is obtained or a preset maximum number of iterations is reached; the determining unit is configured to use the data storage node corresponding to the optimal position of the particle obtained after the iteration is completed as the optimal data storage node.

[0122] In a specific implementation, the processing unit is further configured to enable each data storage node to periodically broadcast the current public state of the node to surrounding nodes.

[0123] In a specific implementation, the processing unit is further configured to number each data storage node according to the level as (a, b), where a represents the number of the data storage node level, and b represents the number of other data storage nodes at the same level as the data storage node; and,

[0124] In the particle velocity update formula, a vector parameter about the particle data characteristic value is added to the particle velocity part to control whether the particle is transferred to the next node in the direction of level number a or in the direction of number b of other data storage nodes at the same level.

[0125] In a specific implementation, the processing unit is further configured to add a data storage node index parameter to the social cognition part of the particle velocity update formula, and the parameter affects the attraction of the optimal data storage node to the surrounding particles.

[0126] The industrial Internet data storage system provided by the embodiment of the present disclosure first divides the data storage location of the industrial Internet into data storage nodes of different levels, and then adopts an improved particle swarm optimization algorithm to obtain a globally optimized data storage strategy. Then, based on the acquired data storage strategy, different types of data generated by different local device nodes are saved to the optimal data storage node according to the real-time status of the network, which can not only realize data sharing, but also reduce latency and transmission consumption, thereby improving the performance of the entire industrial Internet data storage.

[0127] Based on the same technical concept, the embodiment of the present disclosure also provides a computer device, such as Figure 3 As shown, the computer device 3 includes a memory 31 and a processor 32. The memory 31 stores a computer program. When the processor 32 runs the computer program stored in the memory 31, the processor 32 executes the aforementioned industrial Internet data storage method.

[0128] Based on the same technical concept, the embodiment of the present disclosure also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the processor executes the aforementioned industrial Internet data storage method.

[0129] The following problems exist in the data storage solutions currently used in the Industrial Internet:

[0130] 1) Industrial sites based on the Industrial Internet will generate a large amount of data, but due to cost and site conditions, the data storage and data processing capabilities of a large number of on-site equipment nodes are limited; the data of on-site equipment nodes needs to be shared with other nodes or other devices, as well as local and remote production control, raw material supply, warehouse management, quality control, logistics and other software; in addition, the Industrial Internet based on 5G realizes the ubiquitous interconnection of all elements of production elements such as machines, workshops, enterprises, and design, R&D, production, management, and services in the industrial chain, which makes the requirements for data storage more complex. The existing data storage strategy does not take into account the needs of data storage and application elements under the 5G network;

[0131] 2) 5G networks significantly improve the performance of the industrial Internet. Existing data storage solutions do not consider the overall optimization plan of data storage nodes including 5G, nor do they consider the properties of data storage nodes at each level. Nodes are generally selected and judged one by one. Since the state of storage nodes in the network is changing rapidly, a faster optimization convergence speed of data storage strategies is required. Existing data storage strategies cannot meet the requirements of fast restrictions.

[0132] The industrial Internet data storage method, system, computer device and storage medium provided by the embodiments of the present disclosure make full use of the rich data storage node types of 5G network, and divide the industrial Internet data storage nodes into L1 to L5 categories according to the characteristics of data storage nodes at different locations of the network. At the same time, according to the characteristics and various needs of data generation and storage of the industrial Internet under the 5G network, the overall fast convergence optimization storage strategy is studied based on the improved particle swarm optimization algorithm, and then the optimal data storage node is quickly found according to the data storage requirements and storage strategy, which has a positive significance for improving the overall performance of the system.

[0133] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0134] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present disclosure.

Claims

1. An industrial Internet data storage method, characterized in that: include: Classify the data storage locations of the Industrial Internet to form data storage node sets of different levels, where a data storage node set of the same level contains several data storage nodes of the same level; Construct a data storage model that includes data storage nodes at all levels, the connection relationship between nodes, and the benefits of data generated by local device nodes at the production site when they are transferred to other data storage nodes; Based on the data storage model, a preset data storage strategy is used to seek an optimal data storage node that satisfies the optimal benefit of the system, wherein the optimal benefit of the system refers to the maximum value of the total benefit brought by data transmission from the local device node to other data storage nodes; Transmitting the data currently generated by the local device node to the optimal data storage node for storage; The benefits of transmitting data generated by local device nodes at the production site to other data storage nodes are calculated using the following formula: in, represents the benefits brought by data storage from node i to node j, Be1 represents the benefits brought by the data storage capacity of node j, Be2 represents the shared benefits brought by data storage in node j, and Be3 represents the benefits brought by the processing capacity of node j; i belongs to the local device node of the production site, and j belongs to other data storage nodes except the local device node; The data storage model meets the preset delay constraints and reliability constraints, which are as follows: in represents the delay constraint for data to be stored from node i to node j, τ i,j represents the delay constraint condition of data storage between node i and node j set by the system; in represents the reliability constraint of data stored from node i to node j, Q i,j Represents the reliability constraint of data storage between node i and node j set by the system; It is calculated using the following formula: Wherein, t1 represents the transmission time of data from node i to node j, t2 represents the time from when the first byte of data is output from node i to when the last byte is transmitted, t3 represents the time from when the first byte of data enters node j to when the last byte enters node j, t4 represents the waiting time of data in the read queue of node i, and t5 represents the waiting time of data in the write queue of node j; It is calculated using the following formula: Among them, R1 represents the reliability of data storage in node j, which is specifically evaluated by hardware redundancy, power supply, ambient temperature, and ambient humidity. R2 represents the reliability of the communication link between node i and node j. R3 represents the security of data storage in node j, which is specifically evaluated by the security protection level and the security level of data access. Based on the data storage model, a preset data storage strategy is used to seek the optimal data storage node that satisfies the optimal benefit of the system, including: Abstracting all data storage nodes in the data storage model into individual particles in a particle swarm, and updating the speed and position of each particle according to a preset particle speed update formula and a particle position update formula; Calculate the gain of each particle after velocity and position update; Compare the profit of each particle with its historical optimal profit. If it exceeds its historical optimal profit, its current position is taken as the optimal position of the particle. Compare the benefit corresponding to the optimal position of each particle with the optimal benefit of the population. If it exceeds the optimal benefit of the population, update the optimal position and the best fitness value of the population. Repeat the above steps until a global optimal solution is obtained or a preset maximum number of iterations is reached after multiple iterations, and the data storage node corresponding to the optimal position of the particle obtained after the iteration is completed is used as the optimal data storage node.

2. The method according to claim 1, characterized in that The data storage nodes are divided into five levels, namely: local device nodes, local data nodes, user-side data nodes, regional data nodes and cloud data nodes.

3. The method according to claim 1, characterized in that Also includes: Each data storage node regularly broadcasts the current public status of the node to the surrounding nodes.

4. The method according to claim 1, characterized in that: Also includes: Number each data storage node as (a, b) according to the level, where a represents the level number of the data storage node, and b represents the numbers of other data storage nodes at the same level as the data storage node; In the particle velocity update formula, a vector parameter about the particle data characteristic value is added to the particle velocity part to control whether the particle is transferred to the next node in the direction of level number a or in the direction of number b of other data storage nodes at the same level.

5. The method according to claim 1, characterized in that Also includes: In the social cognition part of the particle velocity update formula, a data storage node indicator parameter is added, which affects the attraction of the optimal data storage node to the surrounding particles.

6. An industrial Internet data storage system, characterized in that: include: A classification module, which is configured to classify the data storage locations of the industrial Internet to form data storage node sets of different levels, wherein a data storage node set of the same level includes a number of data storage nodes of the same level; A construction module configured to construct a data storage model, which includes data storage nodes at all levels, connection relationships between nodes, and benefits brought by data generated by local device nodes at the production site being transmitted to other data storage nodes; A target optimization module is configured to use a preset data storage strategy to seek an optimal data storage node that satisfies the optimal system benefit based on the data storage model, wherein the optimal system benefit refers to the maximum value of the total benefit brought by data transmission from the local device node to other data storage nodes; as well as, A transmission module, which is configured to transmit the data currently generated by the local device node to the optimal data storage node for storage; In the data storage model constructed by the building module, the benefits brought by the data generated by the local device nodes at the production site when transmitted to other data storage nodes are calculated using the following formula: in, represents the benefits brought by data storage from node i to node j, Be1 represents the benefits brought by the data storage capacity of node j, Be2 represents the shared benefits brought by data storage in node j, and Be3 represents the benefits brought by the processing capacity of node j; i belongs to the local device node of the production site, and j belongs to other data storage nodes except the local device node; The data storage model constructed by the construction module satisfies the preset delay constraints and reliability constraints, which are as follows: in represents the delay constraint for data to be stored from node i to node j, τ i,j represents the delay constraint condition of data storage between node i and node j set by the system; in represents the reliability constraint of data stored from node i to node j, Q i,j Represents the reliability constraint of data storage between node i and node j set by the system; It is calculated using the following formula: Wherein, t1 represents the transmission time of data from node i to node j, t2 represents the time from when the first byte of data is output from node i to when the last byte is transmitted, t3 represents the time from when the first byte of data enters node j to when the last byte enters node j, t4 represents the waiting time of data in the read queue of node i, and t5 represents the waiting time of data in the write queue of node j; It is calculated using the following formula: Among them, R1 represents the reliability of data storage in node j, which is specifically evaluated by hardware redundancy, power supply, ambient temperature, and ambient humidity. R2 represents the reliability of the communication link between node i and node j. R3 represents the security of data storage in node j, which is specifically evaluated by the security protection level and the security level of data access. The target optimization module includes: a processing unit, an updating unit, a calculating unit, a first comparing unit, a second comparing unit and a determining unit; Among them, the processing unit is configured to abstract all data storage nodes in the data storage model into individual particles in the particle swarm; the updating unit is configured to update the speed and position of each particle according to a preset particle speed update formula and a particle position update formula; the calculating unit is configured to calculate the benefit of each particle after the speed and position are updated; the first comparing unit is configured to compare the benefit of each particle with its historical optimal benefit, and if it exceeds its historical optimal benefit, its current position is used as the optimal position of the particle; the second comparing unit is configured to compare the benefit corresponding to the optimal position of each particle with the optimal benefit of the population to which it belongs, and if it exceeds the optimal benefit of the population to which it belongs, the optimal position and the best fitness value of the population are updated; then the updating unit, the calculating unit, the first comparing unit and the second comparing unit cooperate with each other to iterate multiple times until a global optimal solution is obtained or a preset maximum number of iterations is reached; the determining unit is configured to use the data storage node corresponding to the optimal position of the particle obtained after the iteration is completed as the optimal data storage node.

7. A computer device, characterized in that: It includes a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the industrial Internet data storage method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the processor executes the industrial Internet data storage method according to any one of claims 1 to 5.

Images