Crowd-sourcing cooperative control method for tobacco production host

Abstract

The invention discloses a tobacco production host crowd-sourcing cooperative control method, which comprises a plurality of control hosts for respective production processes of different tobacco ingredients, and comprises the following steps of: 1, setting an intelligent node on each control host, and establishing a distributed communication network through the intelligent nodes, the intelligent node comprises a sensor, a controller, a communication module and a storage unit; 2, starting all intelligent nodes, initializing a communication network, inputting production tasks and equipment parameters, setting priorities of the production tasks, and starting cooperative control; and step 3, dynamically adjusting operation parameters of each host according to the priority of the production task, the equipment state and the resource use condition, and realizing task allocation cooperative control. Through cooperative control of the intelligent nodes, operation of host equipment is more efficient, and production task distribution is more reasonable; the system stability is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of tobacco processing, and in particular to a method for collective intelligence collaborative control of tobacco production hosts. Background Art

[0002] In the tobacco production process, the operating efficiency and synergy of main equipment (such as cigarette making machines, cutters, etc.) have a significant impact on overall production quality and output. However, in existing technologies, the control of main equipment mostly adopts independent operation or simple centralized control methods, which has the following problems:

[0003] Inefficiency: The lack of real-time coordination between host devices leads to uneven distribution of production tasks. Some devices may be idle while others may be overloaded.

[0004] Slow equipment response: When a host computer is ahead of or behind in its tasks, other equipment cannot quickly adjust their operating status, resulting in a decrease in the efficiency of the entire production line.

[0005] Resource waste: Due to the lack of intelligent task allocation mechanism, resource utilization is low.

[0006] Low degree of intelligence: Traditional control methods are difficult to adapt to complex and changing production environments and cannot flexibly respond to emergencies. Summary of the Invention

[0007] Based on the above problems, developing a method that can realize the intelligent collaborative control of host groups is of great significance for improving the intelligent level and overall efficiency of tobacco production. Therefore, the purpose of the present invention is to provide a method for the intelligent collaborative control of host groups in tobacco production.

[0008] In order to achieve the above object, the scheme of the present invention is:

[0009] A method for intelligent collaborative control of tobacco production hosts, including multiple control hosts for production processes of different tobacco ingredients, the intelligent collaborative control method comprising:

[0010] The first step is to set up an intelligent node on each control host and establish a distributed communication network through the intelligent node. The intelligent node includes a sensor, a controller, a communication module and a storage unit;

[0011] in:

[0012] The sensor is used to collect the operating status of the intelligent node device in real time;

[0013] The controller is used to coordinate control and adjust the equipment according to the collected operating status and operating parameters;

[0014] The communication module is used to realize real-time communication between intelligent nodes using wireless or wired communication methods;

[0015] The storage unit is used to store equipment operation history data and collaborative control operation parameters;

[0016] Step 2: Start all intelligent nodes, initialize the communication network, input production tasks and operating parameters, set the priority of production tasks, and start collaborative control;

[0017] The third step is to dynamically adjust the operating parameters of each host according to the priority of the production task, equipment status and resource usage to achieve collaborative control of task allocation.

[0018] The solution is further that: the method further includes calibrating the intelligent node, and the calibration includes: accurate calibration of collected data and calibration of stability and real-time performance of communication network data transmission.

[0019] The solution is further: the method further includes establishing a fault detection module, and monitoring the intelligent node in real time through the fault detection module. The real-time monitoring includes monitoring the operating status of the equipment, and dynamically adjusting the operating parameters based on the monitoring, and dynamically adjusting the resource allocation strategy according to changes in production tasks.

[0020] The solution is further: the method further includes regular maintenance, regular maintenance of the communication network and intelligent nodes, and when the operating status of a device changes, adjusting the operating parameters of other devices to share the tasks.

[0021] The solution further includes: the collaborative control is implemented by a collaborative control algorithm, and the collaborative control algorithm includes:

[0022] 1. Build a collective intelligence collaborative control model and set A i is the intelligent agent, representing the intelligent agent corresponding to each device, and E is the factory manufacturing environment;

[0023] 2. Obtain a model of a single agent’s ability to perceive the environment.

[0024] F Ai =f(E,T1,T2,T3...,T n )

[0025] Where: T n is the perception mode of agent i; E is the factory manufacturing environment;

[0026] 3. Construct a swarm agent behavior model. Substitute the acquired single agent perception capability model into the swarm agent behavior model to obtain swarm intelligent behavior emergence. The specific formula is as follows:

[0027] A i =S(FAi )

[0028] Among them, A i represents the emergence of group agent behavior; S represents the process of agent strategy formation;

[0029] 4. Selection of swarm agent behavior strategies: different behavior strategies are selected for different types of devices; the behavior strategies include: motion behavior strategy, collaborative behavior strategy, and cognitive behavior strategy;

[0030] The motion behavior strategy is to use a motion model based on position prediction, which uses a Markov model to predict the next position of the agent:

[0031]

[0032] Among them, x t represents the position of the agent at time t, P(x t+1 |x t ) is the agent’s position x t Move to position x t+1 probability;

[0033] The collaborative behavior strategy is the collaboration between multiple agents. In collaborative spectrum sensing, the voting fusion criterion is expressed as:

[0034]

[0035] Where M is the number of agents involved in perception, and p is the detection probability or false alarm probability of a single agent.

[0036] The cognitive behavior strategy is the learning and decision-making process of the agent to adapt to environmental changes and task requirements. In the collective intelligence behavior strategy, the utility function of the agent is expressed as:

[0037]

[0038] Where s is the state of the agent, a is the action of the agent, R(s, a) is the immediate reward, γ is the discount factor, and s′ is the next state.

[0039] The beneficial effects of the present invention are:

[0040] Improve production efficiency. Through the collaborative control of intelligent nodes, the operation of host equipment is more efficient and the distribution of production tasks is more reasonable. Enhance system stability. The collaborative processing mechanism can quickly respond to abnormal equipment conditions and reduce downtime. Reduce resource waste. With a high degree of intelligence, the group intelligence collaborative control algorithm can adapt to complex and changing production environments and flexibly respond to emergencies.

[0041] The invention will be further explained in detail below with reference to the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic diagram of the control flow of the present invention;

[0043] Figure 2 This is a schematic diagram of the collective intelligence collaborative control model architecture. DETAILED DESCRIPTION

[0044] A method for intelligent collaborative control of tobacco production hosts, including multiple control hosts for production processes of different tobacco ingredients, includes:

[0045] The first step is to set up an intelligent node on each control host and establish a distributed communication network through the intelligent node. The intelligent node includes a sensor, a controller, a communication module and a storage unit;

[0046] in:

[0047] The sensor is used to collect the operating status of the intelligent node equipment in real time, such as temperature, pressure, speed, etc.

[0048] The controller is used to coordinate control and adjust the equipment according to the collected operating status and operating parameters;

[0049] The communication module is used to realize real-time communication between intelligent nodes using wireless or wired communication methods;

[0050] The storage unit is used to store the equipment operation history data and collaborative control operation parameters, such as collaborative control algorithm, to adjust the equipment operation parameters;

[0051] Step 2: Start all intelligent nodes, initialize the communication network, input production tasks and operating parameters, set the priority of production tasks, and start collaborative control;

[0052] The third step is to dynamically adjust the operating parameters of each host according to the priority of the production task, equipment status and resource usage to achieve collaborative control of task allocation.

[0053] In the embodiment, the method further includes calibrating the intelligent node, wherein the calibration includes accurately calibrating the collected data and calibrating the stability and real-time performance of the communication network data transmission.

[0054] In the embodiment: the method further includes establishing a fault detection module, and monitoring the intelligent node in real time through the fault detection module. The real-time monitoring includes monitoring the operating status of the equipment, and dynamically adjusting the operating parameters according to the monitoring, and dynamically adjusting the resource allocation strategy according to the changes in production tasks.

[0055] In the embodiment: the method further includes regular maintenance, which includes regular maintenance of the communication network and the intelligent nodes. When the operating status of a certain device changes, the operating parameters of other devices are adjusted to share the tasks.

[0056] In the embodiment, the collaborative control is implemented by a collaborative control algorithm, and the collaborative control algorithm includes:

[0057] 1. Build a collective intelligence collaborative control model and set A i is the intelligent agent, representing the intelligent agent corresponding to each device, and E is the factory manufacturing environment;

[0058] 2. Obtain a model of a single agent’s ability to perceive the environment.

[0059] F Ai =f(E, T1, T2, T3…, T n )

[0060] Where: T n is the perception mode of agent i; E is the factory manufacturing environment;

[0061] 3. Construct a swarm agent behavior model. Substitute the acquired single agent perception capability model into the swarm agent behavior model to obtain swarm intelligent behavior emergence. The specific formula is as follows:

[0062] A i =S(F Ai )

[0063] Among them, A i represents the emergence of group agent behavior; S represents the process of agent strategy formation;

[0064] 4. Selection of swarm agent behavior strategies: different behavior strategies are selected for different types of devices; the behavior strategies include: motion behavior strategy, collaborative behavior strategy, and cognitive behavior strategy;

[0065] The motion behavior strategy is to use a motion model based on position prediction, which uses a Markov model to predict the next position of the agent:

[0066]

[0067] Among them, x t represents the position of the agent at time t, P(x t+1 |x t ) is the agent’s position x t Move to position x t+1 probability;

[0068] The collaborative behavior strategy is the collaboration between multiple agents. In collaborative spectrum sensing, the voting fusion criterion is expressed as:

[0069]

[0070]

[0071] Where M is the number of agents involved in perception, and p is the detection probability or false alarm probability of a single agent.

[0072] The cognitive behavior strategy is the learning and decision-making process of the agent to adapt to environmental changes and task requirements. In the collective intelligence behavior strategy, the utility function of the agent is expressed as:

[0073]

[0074] Where s is the state of the agent, a is the action of the agent, R(s, a) is the immediate reward, γ is the discount factor, and s′ is the next state.

[0075] The following is based on Figure 1 and Figure 2 The above method is described in further detail, wherein:

[0076] S01: Smart Node Installation: Install a smart node on each host device. The smart node consists of sensors, controllers, communication modules, and storage units. Sensors collect real-time data about the device's operating status (such as temperature, pressure, and speed). The controller adjusts the device's operating parameters based on the collected data and the collaborative control algorithm. The communication module enables real-time communication between smart nodes, using either wireless or wired communication. The storage unit stores historical device operating data and collaborative control algorithm parameters.

[0077] S02: Communication network construction: Establish a distributed communication network to ensure the communication reliability and redundancy between intelligent nodes. Use redundant design to avoid single point failure affecting the entire system.

[0078] S03: Development of swarm intelligence collaborative control algorithm. Design a collaborative control algorithm based on swarm intelligence, so that intelligent nodes can dynamically adjust the operating parameters of each host according to the priority of production tasks, equipment status and resource usage. Implement a task allocation mechanism, dynamically assign tasks to each host device according to task priority and equipment status; implement a collaborative adjustment mechanism, when the operating status of a device changes (such as a fault or overload), other devices can quickly adjust the operating parameters and share the task; implement a fault collaborative processing mechanism, develop a fault detection module, and intelligent nodes monitor the operating status of the equipment in real time. When an abnormality is detected, it will immediately notify other nodes; implement a task reallocation mechanism, through the swarm intelligence algorithm, reallocate the tasks of the faulty device to other normal devices to ensure the continuity of the production process; implement a fault recovery mechanism, when the faulty device is repaired, the intelligent node can automatically reintegrate it into the collaborative control network.

[0079] S04: Smart Node Startup: Start all smart nodes, initialize the communication network, and ensure normal communication between nodes. Input production tasks and equipment parameters, and start the collaborative control algorithm.

[0080] S05: System calibration: Calibrate the smart nodes to ensure the accuracy of the data collected by the sensors. Test the communication network to ensure the stability and real-time performance of data transmission.

[0081] S06: Real-time monitoring and adjustment: Intelligent nodes monitor equipment status in real time and dynamically adjust operating parameters. Resource allocation strategies are dynamically adjusted based on changes in production tasks.

[0082] S07: Regular maintenance: Regularly maintain the communication network and intelligent nodes to ensure the stable operation of the system; regularly update the collaborative control algorithm and return to step S03 to optimize system performance.

[0083] Principle of swarm intelligence collaborative control algorithm:

[0084] S01: Build a collective intelligence collaborative control model, agent A i Represents the intelligent agent corresponding to each device, E is the factory manufacturing environment, and its specific structure is shown in the attached figure. Figure 2 .

[0085] S02: Obtain a model of a single agent's ability to perceive the environment. A single agent is an abstraction of the device's actual perception capabilities at the software level. The device's capabilities depend on the number of corresponding sensors and collection points deployed.

[0086] F Ai =f(E,T1,T2,T3...,T n )

[0087] Where: T nis the perception mode of agent i; E is the manufacturing environment.

[0088] S03: Swarm agent behavior model. Substitute the acquired single agent perception capability model into the swarm agent behavior model to obtain swarm intelligent behavior emergence. The specific formula is as follows;

[0089] A i =S(F Ai )

[0090] Among them, Ai represents the emergence of group intelligent agent behavior; S represents the process of intelligent agent strategy formation.

[0091] S04: Selection of swarm agent behavior strategies. Due to the different types of devices, different behavior strategies should be selected for different types of devices. The behavior strategies can be divided into three categories: motion behavior strategy, collaborative behavior strategy, and cognitive behavior strategy.

[0092] 1. Exercise behavior strategy

[0093] Motion behavior strategies typically involve the movement and path planning of an agent to optimize task execution efficiency. Common motion models include Markov models and Bayesian inference models. A motion model based on position prediction can be used to predict the next position of an agent:

[0094]

[0095] Among them, x t represents the position of the agent at time t, P(x t+1 |x t ) is the agent’s position x t Move to position x t+1 probability.

[0096] 2. Collaborative Behavior Strategy

[0097] Collaborative behavior strategy involves the collaboration between multiple agents to improve the efficiency and quality of task completion. In collaborative spectrum sensing, the voting fusion criterion can be expressed as:

[0098]

[0099] Where M is the number of agents involved in perception, and p is the detection probability or false alarm probability of a single agent.

[0100] 3. Cognitive behavioral strategies

[0101] Cognitive behavioral strategies involve the learning and decision-making process of intelligent agents to adapt to environmental changes and task requirements. In collective intelligence behavioral strategies, the utility function of the intelligent agent can be expressed as:

[0102]

[0103] Where s is the state of the agent, a is the action of the agent, R(s, a) is the immediate reward, γ is the discount factor, and s′ is the next state.

Claims

1. A method for collaborative control of tobacco production hosts using collective intelligence, including multiple control hosts for different tobacco production processes, characterized in that: The intelligent collaborative control method includes: The first step is to set up an intelligent node on each control host and establish a distributed communication network through the intelligent node. The intelligent node includes a sensor, a controller, a communication module and a storage unit; in: The sensor is used to collect the operating status of the intelligent node device in real time; The controller is used to coordinate control and adjust the equipment according to the collected operating status and operating parameters; The communication module is used to realize real-time communication between intelligent nodes using wireless or wired communication methods; The storage unit is used to store equipment operation history data and collaborative control operation parameters; Step 2: Start all intelligent nodes, initialize the communication network, input production tasks and operating parameters, set the priority of production tasks, and start collaborative control; The third step is to dynamically adjust the operating parameters of each host according to the priority of the production task, equipment status and resource usage to achieve collaborative control of task allocation.

2. The collaborative control method according to claim 1, characterized in that: The method further includes calibrating the intelligent node, wherein the calibration includes: accurately calibrating the collected data and calibrating the stability and real-time performance of data transmission on the communication network.

3. The collaborative control method according to claim 1, characterized in that: The method further includes establishing a fault detection module, and monitoring the intelligent node in real time through the fault detection module. The real-time monitoring includes monitoring the operating status of the equipment, and dynamically adjusting the operating parameters according to the monitoring, and dynamically adjusting the resource allocation strategy according to the changes in production tasks.

4. The collaborative control method according to claim 1, characterized in that: The method further includes regular maintenance, which includes regular maintenance of the communication network and the intelligent nodes. When the operating status of a device changes, the operating parameters of other devices are adjusted to share the tasks.

5. The collaborative control method according to claim 1, characterized in that: The collaborative control is implemented by a collaborative control algorithm, which includes:

1. Build a collective intelligence collaborative control model and set A i is the intelligent agent, representing the intelligent agent corresponding to each device, and E is the factory manufacturing environment; 2. Obtain a model of a single agent’s ability to perceive the environment. F Ai =f(E,T1,T2,T3...,T n ) Where: T n is the perception mode of agent i; E is the factory manufacturing environment; 3. Construct a swarm agent behavior model. Substitute the acquired single agent perception capability model into the swarm agent behavior model to obtain swarm intelligent behavior emergence. The specific formula is as follows: A i =S(F Ai ) Among them, A i represents the emergence of group agent behavior; S represents the process of agent strategy formation; 4. Selection of swarm agent behavior strategies: different behavior strategies are selected for different types of devices; the behavior strategies include: motion behavior strategy, collaborative behavior strategy, and cognitive behavior strategy; The motion behavior strategy is to use a motion model based on position prediction, which uses a Markov model to predict the next position of the agent: Among them, x t represents the position of the agent at time t, P(x t+1 |x t ) is the agent’s position x t Move to position x t+1 probability; The collaborative behavior strategy is the collaboration between multiple agents. In collaborative spectrum sensing, the voting fusion criterion is expressed as: Detection probability False alarm probability Where M is the number of agents involved in perception, and p is the detection probability or false alarm probability of a single agent. The cognitive behavior strategy is the learning and decision-making process of the agent to adapt to environmental changes and task requirements. In the collective intelligence behavior strategy, the utility function of the agent is expressed as: Where s is the state of the agent, a is the action of the agent, R(s,a) is the immediate reward, γ is the discount factor, and s′ is the next state.

Images