CN108156226A - The industrial Internet of Things cognition energy management system and computational methods of a kind of cloud and mist fusion - Google Patents

Abstract

The invention discloses a cloud-fog fusion industrial internet of things cognitive energy management system and a calculation method thereof, the structure of which includes an industrial equipment layer, an industrial cloud and fog cognitive computing layer and an energy management layer. The industrial cloud and fog cognitive computing layer includes industrial cloud computing and industrial fog computing; as the top layer of the architecture, the energy management layer is responsible for providing diversified energy management applications, including energy perception modules, energy analysis modules, energy prediction modules and energy optimization modules. The present invention uses cloud computing to provide unlimited resources to alleviate the problem of limited fog computing resources, uses the edge information processing capability of fog computing to alleviate the problems of high delay, network congestion, and low reliability caused by cloud computing, and builds a penetration model between fog computing and cloud computing To achieve cloud-fog integration, under the guidance of minimizing the consumption of cloud and fog resources, energy management services can be reasonably allocated between clouds and fog, and cloud and fog resources can be used reasonably and efficiently, so as to realize efficient IIoT energy management.

Description

A cloud-fog fusion industrial internet of things cognitive energy management system and calculation method

technical field

The invention belongs to the application technical field of the Internet of Things in the field of energy, and specifically relates to a cloud-mist fusion industrial Internet of Things cognitive energy management system and a calculation method.

Background technique

The transformation of the domestic manufacturing industry means that the strong promotion of industrial informatization plays an important role in promoting the era of the industrial Internet. The Industrial Internet is an industrial revolution + network revolution, not industry + Internet. The Industrial Internet of Things belongs to the Industrial Internet. Through the existing Internet of Things (IoT) technology and Big Data (Big Data) technology, analysis, prediction and automation algorithms are used in the field of industrial manufacturing Connect traditional machine equipment to achieve more stable human-computer interaction.

However, with the continuous development of the Industrial Internet of Things (IIoT), the continuously increasing energy consumption and serious environmental pollution issues have aroused widespread concern from all walks of life. Therefore, there is an urgent need for an efficient and green energy management method to reduce energy consumption and reduce environmental pollution. In view of the characteristics of IIoT, such as large scale, high accuracy requirements, and time delay sensitivity, IIoT proposes the following points for energy management Requirements: ①High-speed and reliable network transmission capability, ②Massive data storage and computing capability, ③Intelligent analysis and decision-making ability, ④Information security guarantee, ⑤Intelligent interaction effect.

At present, research on IIoT energy management at home and abroad is mainly focused on integrating cloud computing (Cloud Computing) technology into energy management to solve the problems of limited resources and difficult expansion of traditional energy management. However, while cloud computing brings convenience to IIoT energy management, it also brings great challenges. As the IIoT continues to mature, massive energy data information will inevitably be generated. If all the data is moved to the cloud for storage and computing, it will inevitably cause an input/output bottleneck between the cloud center and industrial equipment, which will greatly reduce the entire IIoT transmission rate. At the same time, it brings serious network congestion, and all data is stored in the cloud, which also has great security risks. On the one hand, the industrial Internet energy management architecture in the existing technology should no longer stick to cloud computing. Fog computing (Fog Computing) has the ability to store and calculate data on industrial equipment (or between equipment, on the network). Consider incorporating fog computing technology into energy management as well. On the other hand, the cloud-based energy management model only processes energy data information, lacks the ability to process network edge data, and does not have the ability to alleviate IIoT congestion and ensure data security.

In the existing patent literature, the patent application number 201505000110.X discloses a visual modeling-based industrial Internet of Things energy management method, based on the technical specifications of an industrial Internet of Things system and functional components, to achieve the usage perspective model and The activities and functions derived from the functional perspective model are mapped from "activities" to "functional components" to "implementation components", and key system features (edge layer, platform layer, and enterprise layer). In this way, a complete industrial Internet of Things energy management system is completed. Its main process is shown in Figure 1. It can realize that the edge layer collects data from the industrial control system and transmits it to the platform layer; and receives control commands for the industrial control system from the platform layer, and the platform layer receives, processes, and forwards control commands from the enterprise layer to the edge layer; The layer aggregates, processes, and forwards data to the enterprise layer, which implements domain-specific applications, decision support systems, and provides application interfaces to end users. This document is a basic patent for modeling the energy management system of the Internet of Things, and a relatively high-level technical system has been constructed for the information security of the Industrial Internet of Things. However, the shortcoming of this technology is that it does not address the theoretical problems of the Industrial Internet of Things, including intelligence issues, digital issues, reliability issues, controllability issues, and security issues, especially for the three systems of physical security, information security, and system self-healing The characteristics do not give a complete solution to the supporting technical system.

The patent document with the application number 201423274662.X discloses a cloud-integrated industrial Internet of Things network support modeling method, which proposes the structural system of the Industrial Internet of Things from three perspectives of communication, service and information: ubiquitous network architecture , application layer overlay network architecture, and service-oriented architecture. This approach incorporates the Industrial Internet of Things as a technical problem to be solved for smart manufacturing. Its main process is shown in Figure 2. In the service-oriented layer, through the Internet gateway or intermediate server, authorized users can access the device information directly extracted by the object network. At this time, the server acts as a receiver in the object network and performs data collection from each object; the ubiquitous network layer It acts as an interface between the Internet cloud and the object network, including heterogeneous access networks, 3G networks and wireless LANs, with interoperability; the application layer overlay network can be in the form of wireless sensor networks or factory ad hoc networks. Different forms of networks enable better coordination and integration in multi-access and multi-operator environments, while high-quality communication channels provide new opportunities for service applications. This patent document utilizes cloud computing's data perception, collection, storage and computing capabilities to integrate with the Industrial Internet of Things. However, this method does not have the optimal information flow and fast transmission capabilities between the cloud center and industrial equipment, and there are also major security risks.

Contents of the invention

Aiming at the deficiencies of the above-mentioned existing technologies, the purpose of the present invention is to propose a cloud-fog fusion industrial Internet of Things cognitive energy management solution based on cloud computing to provide unlimited resource technology and fog computing edge information processing technology. Construct a permeable cognitive model between clouds and fog to achieve cloud fusion. Under the guidance of minimizing the consumption of cloud and fog resources, energy management services can be reasonably allocated among clouds and fog, and cloud and fog resources can be used reasonably and efficiently. Network congestion, low reliability issues.

In order to achieve the above purpose, the technical solution adopted by the present invention is a cloud-fog fusion industrial Internet of Things cognitive energy management system, whose structure includes an industrial equipment layer, an industrial cloud and fog cognitive computing layer, and an energy management layer, wherein the industrial equipment layer includes data Acquisition equipment, communication equipment, and central equipment use these physical equipment to perform data collection, transformation, upper-layer transmission, and local or remote control; the industrial cloud and fog cognitive computing layer includes industrial cloud computing and industrial fog computing, and industrial cloud as a centralized computing The center provides abundant storage and computing resources for energy management, and plays a centralized control role for the entire IIoT energy management. Industrial fog provides real-time storage and computing resources for energy management in a distributed manner. The model improves resource utilization; as the top layer of the architecture, the energy management layer is responsible for providing diversified energy management applications, including energy perception modules, energy analysis modules, energy prediction modules, and energy optimization modules. On the one hand, it is used to convey energy management instructions to the lower layer. On the other hand, it provides a good human-computer interaction environment for IIoT users.

Further, the above-mentioned energy sensing module is responsible for sensing disordered, scattered, and unsystematic raw energy data information, and classifying and summarizing them according to certain standards, so as to simplify, visualize, and systematize the raw data.

The above-mentioned energy analysis module analyzes the data information through the method of statistical calculation, reflecting the trend, degree of dispersion and correlation strength of the original energy data information.

The above-mentioned energy prediction module performs energy consumption prediction and energy supply prediction.

The above-mentioned energy optimization module establishes a relationship model between the execution performance and energy consumption of industrial equipment based on real-time data and historical data, and applies a multi-objective optimization control algorithm to find the optimal energy management solution to reduce energy consumption while maintaining the excellent performance of industrial equipment.

The present invention further proposes a calculation method used by the above-mentioned cloud-mist fusion industrial Internet of Things cognitive energy management system, comprising the following steps:

Step 1: Decompose the complex instruction D issued by the upper energy management module into multiple energy management services

Step 2: the industrial cloud layer is responsible for classifying these services _si ;

Step 3: The solute is composed of the number SM, storage capacity SS, and computing capacity SC of industrial cloud servers, and dynamically adjusts and distributes these services by using the principle of osmosis;

Step 4: Using the energy management service s _i as a solvent, carry out cognitive movement according to the difference of industrial cloud and fog resources on both sides of the semi-permeable membrane, so as to balance the concentration at both ends of the membrane and realize the reasonable distribution of services;

Step 5: Consider the following factors to configure the semi-permeable membrane: energy management service balance L _balance , processing delay interval D ⁱ (i=Cloud, Fog, Avg), and cloud computing upper and lower limits Ensure its intelligence, and then control the flow of energy management services;

Step 6: During the infiltration process of services between industrial clouds, adjust and configure the number SM, storage capacity SS, and computing capacity SC of industrial cloud servers, and determine the migration direction of services between industrial clouds according to the configuration difference f.

Compared with prior art, the beneficial effect of the present invention is:

1. The present invention uses cloud computing to provide unlimited resources to alleviate the problem of limited resources in fog computing, and uses the edge information processing capability of fog computing to alleviate the problems of high delay, network congestion, and low reliability caused by cloud computing.

2. By building a penetration model between fog computing and cloud computing to realize cloud and fog fusion, under the guidance of minimizing the consumption of cloud and fog resources, energy management services can be reasonably allocated among clouds and fog, and cloud and fog resources can be used reasonably and efficiently, so as to realize IIoT Energy efficient management.

Description of drawings

Figure 1 is an industrial Internet energy management model based on visual modeling.

Figure 2 is a cloud-based integrated industrial IoT network support model.

Fig. 3 is the cognitive energy management model of the cloud-fog fusion industrial Internet of Things of the present invention.

Detailed ways

The specific embodiment of the present invention will be further described in conjunction with the accompanying drawings.

The system structure of the cloud-fog fusion industrial Internet of Things cognitive energy management proposed by the present invention is shown in FIG. 3 , including an industrial equipment layer, an industrial cloud-fog cognitive computing layer, and an energy management layer. in:

Industrial equipment layer: use industrial physical equipment for data acquisition, transformation, upper layer transmission and local or remote control.

Industrial cloud and fog cognitive computing layer: on the one hand, it is used to store and analyze the underlying energy data information to provide data protection for energy management; on the other hand, it provides corresponding energy management services for upper-level energy management and controls the underlying physical equipment, such as configuring industrial equipment, Virtual clusters move in and out.

Energy management layer: On the one hand, it is used to convey energy management instructions to the lower layer, and on the other hand, it provides a good human-computer interaction environment for IIoT users.

Brief description: In the IIoT environment, industrial equipment is constantly engaged in production activities and consumes a lot of energy. The energy management application obtains abundant computing resources from the industrial cloud cognitive computing layer, decomposes the instructions issued by different energy management modules into energy management services, and then distributes these services to different industrial clouds for execution, controls industrial equipment, and finally reports the execution results Then return to the corresponding energy management module.

For the above three levels, the specific research content is described as follows:

(1) Industrial equipment layer: the industrial physical equipment layer, the main equipment includes data acquisition equipment, communication equipment and central equipment.

Data acquisition equipment solves the problem of data transformation between the human world and the industrial world. They are responsible for collecting energy data information of industrial equipment and transmitting the data to the upper layer through communication equipment to obtain more potential energy information.

The central equipment is the equipment for industrial production in IIoT. It is the main consumer of energy and the main control object of energy management. They can be controlled locally or remotely.

These devices can be regarded as nodes in IIoT. Nodes can be divided into different sub-networks according to their functions, locations and scopes to form virtual clusters. Each virtual cluster has a one-to-one mapping relationship with the industrial fog in the upper layer. At the same time, nodes (devices) can freely leave or join any virtual cluster according to changes in the environment, time, and their own state, and disconnect or establish a connection with the corresponding industrial fog on the upper layer. Industrial fog can adaptively adjust the load of these nodes (devices) according to its own resources.

(2) Industrial cloud and fog cognitive computing layer: This layer includes industrial cloud computing and industrial fog computing. As a centralized computing center, the industrial cloud provides abundant storage and computing resources for energy management, and plays a role in centralized control of the entire IIoT energy management; as a distributed method, the industrial fog provides real-time storage and computing resources for energy management, alleviating the burden of the industrial cloud. Latency, congestion and security issues caused. Moreover, the cognitive model is penetrated between the industrial fog and the industrial cloud to improve resource utilization.

(3) Energy management layer: As the top layer of the architecture, this layer is responsible for providing diversified energy management applications, including energy perception module, energy analysis module, energy prediction module and energy optimization module.

Energy perception module: This module is responsible for sensing disordered, scattered and unsystematic raw energy data information, and classifying and summarizing them according to certain standards, so as to simplify, visualize and systematize the raw data.

Energy analysis module: analyze the data information through the method of statistical operation, reflect the trend, degree of dispersion and correlation strength of the original energy data information. For example, through the statistical analysis module, you can know the highest and lowest energy consumption per unit time of each workshop and its occurrence time.

Energy Forecasting Module: This module has two main aspects, energy consumption forecasting on the one hand and energy supply forecasting on the other. Energy consumption refers to the energy consumption of various IIoT energy-consuming equipment within a certain period of time, including raw coal, crude oil and their products, natural gas, and electricity. Energy production refers to the supply of IIoT various energy sources within a certain period of time, including raw coal, crude oil, natural gas, hydropower, nuclear power generation, biomass energy, solar energy, etc.

Energy optimization module: Its function is to establish a relationship model between industrial equipment execution performance and energy consumption based on real-time data and historical data, apply multi-objective optimization control algorithms, and find the optimal energy management plan, while maintaining the excellent performance of industrial equipment Reduce energy consumption.

The calculation method of the industrial Internet of Things cognitive energy management system based on the above-mentioned cloud-fog fusion mainly uses the osmotic cognitive mechanism, similar to the osmotic in chemistry, and uses the semi-permeable membrane to recognize and balance the solution concentration on both sides of the membrane. Including the following steps:

Step 1: First, decompose the complex instruction D issued by the upper energy management module into multiple energy management services s _i (i=1,2,3,...,n);

Step 2: the industrial cloud layer is responsible for classifying these services _si ;

Step 3: The solute is composed of the number SM, storage capacity SS, and computing capacity SC of industrial cloud servers, and dynamically adjusts and distributes these services by using the principle of osmosis;

Step 4: At this time, the energy management service s _i is used as a solvent to carry out cognitive movement according to the difference of industrial cloud and fog resources on both sides of the semi-permeable membrane, so as to balance the concentration at both ends of the membrane and realize the reasonable distribution of services;

Step 5: The semi-permeable membrane must consider many factors, such as the energy management service balance L _balance , the processing delay interval D ⁱ (i=Cloud, Fog, Avg), and the upper and lower limits of cloud computing Ensure its intelligence, and then control the flow of energy management services;

Step 6: In addition, during the infiltration process of services between industrial clouds and fogs, the resources can be adjusted and configured, and the migration direction of services between industrial clouds and fogs can be determined according to the configuration difference f.

In an industrial production scenario that includes various physical devices, a cloud data center, multiple fog data centers, industrial wireless networks, intelligent control equipment, intelligent production, packaging, transportation equipment, etc., the industrial equipment in this scenario is all intelligent Equipment, integrated with smart sensors, can collect energy data information in industrial production in real time, and these devices also have networking capabilities to share this information.

In addition, these smart industrial devices have certain computing power. Multiple smart devices are combined through the network to form industrial fog. Industrial fog can provide local energy management services to control these devices locally. Multiple industrial fogs can communicate with each other or with the industrial cloud.

Therefore, the industrial cloud understands the operating status and energy consumption of the entire industrial scene, and performs global energy management on the entire scene, and supervises and controls the local energy management that occurs in the industrial fog. Users participate in the energy management system through the intelligent computer in the industrial scene, not only can visually view the energy consumption status of the scene, but also issue energy management instructions to complete specific energy management functions.

In such an industrial scenario, compare the energy management effect of the traditional energy management architecture and the energy management architecture of the industrial Internet of Things based on cloud and fog integration:

1. Comparing the hourly energy consumption cost of industrial equipment, the energy consumption cost of industrial equipment in each period is less under the energy management architecture based on cloud-fog integration, and compared with the traditional architecture, the fluctuation of energy consumption cost in each period is small, and the entire industry The energy consumption of the scene is stable. Specifically, the total energy consumption cost of industrial equipment under this architecture is about 13,200 yuan, far less than the total cost of 21,538 yuan under the traditional architecture, which is more efficient and has better energy management effects.

2. Comparing the hourly pollutant discharge of industrial equipment, the energy management framework based on cloud and fog integration has less pollutant discharge in most periods of time, and the total discharge of 1208kg is less than 1759kg under the traditional framework, which is greener and more environmentally friendly.

It should be noted that the above-mentioned embodiments provided by the present invention are only illustrative, and do not limit the scope of the specific implementation of the present invention. The protection scope of the present invention shall include those changes or substitutions obvious to those skilled in the art.

Claims (6)

1. A cognitive energy management system for the industrial Internet of Things that combines cloud and fog, characterized in that its structure includes an industrial equipment layer, an industrial cloud and fog cognitive computing layer, and an energy management layer, wherein the industrial equipment layer includes data acquisition equipment, communication equipment, and a center Equipment, using these physical devices for data collection, transformation, upper-layer transmission, and local or remote control; the industrial cloud and fog cognitive computing layer includes industrial cloud computing and industrial fog computing. The storage and computing resources play a role in centralized control of the entire IIoT energy management. Industrial fog provides real-time storage and computing resources for energy management in a distributed manner. The penetration of cognitive models between industrial fog and industrial cloud improves resource utilization; energy As the top layer of the architecture, the management layer is responsible for providing a variety of energy management applications, including energy perception modules, energy analysis modules, energy prediction modules, and energy optimization modules. Good human-computer interaction environment. 2. The cloud-fog fusion industrial IoT cognitive energy management system according to claim 1, characterized in that the energy sensing module is responsible for sensing disordered, scattered, and unsystematic raw energy data information, and conducts according to certain standards. Classify and summarize, so that the original data are simplified, visualized and systematized. 3. The cognitive energy management system of the Industrial Internet of Things based on cloud and fog fusion according to claim 1, characterized in that the energy analysis module analyzes data information by means of statistical calculations, reflecting the trend, degree of dispersion and relative strength. 4. The cognitive energy management system of the industrial internet of things with cloud and fog fusion according to claim 1, characterized in that the energy forecasting module performs energy consumption forecasting and energy supply forecasting. 5. The cloud-fog fusion industrial IoT cognitive energy management system according to claim 1, characterized in that the energy optimization module establishes a relationship model between industrial equipment execution performance and energy consumption based on real-time data and historical data, and applies Multi-objective optimization control algorithm, looking for the optimal energy management scheme, reducing energy consumption while maintaining the excellent performance of industrial equipment. 6. a computing method that the industrial internet of things cognitive energy management system that cloud and mist merges according to claim 1 uses, is characterized in that comprising the steps: Step 1: Decompose the complex instruction D issued by the upper energy management module into multiple energy management services s _i (i=1,2,3,...,n); Step 2: the industrial cloud layer is responsible for classifying these services _si ; Step 3: The solute is composed of the number SM, storage capacity SS, and computing capacity SC of industrial cloud servers, and dynamically adjusts and distributes these services by using the principle of osmosis; Step 4: Using the energy management service s _i as a solvent, carry out cognitive movement according to the difference of industrial cloud and fog resources on both sides of the semi-permeable membrane, so as to balance the concentration at both ends of the membrane and realize the reasonable distribution of services; Step 5: Consider the following factors to configure the semi-permeable membrane: energy management service balance L _balance , processing delay interval D ⁱ (i=Cloud, Fog, Avg), and cloud computing upper and lower limits (i=min,max), to ensure its intelligence, and then control the flow of energy management services; Step 6: During the infiltration process of services between industrial clouds, adjust and configure the number SM, storage capacity SS, and computing capacity SC of industrial cloud servers, and determine the migration direction of services between industrial clouds according to the configuration difference f.