Full-automatic overedger based on Internet of Things

Abstract

The invention discloses a full-automatic overedger based on Internet of Things. The full-automatic overedger comprises a fuzzy PID controller based on an SPSO algorithm: network parameters of a convolutional neural network are optimized through the SPSO algorithm, and the optimized neural network is utilized to perform online self-tuning on parameters of fuzzy PID control; the overedger information acquisition module is used for acquiring state parameter information of the overedger; the overedger fault prediction system based on the wireless sensor network is used for evaluating and controlling the operation quality of an overedger, predicting faults and intelligently diagnosing the faults, compared with a traditional overedger, the motor control technology and the Internet of Things technology are added, the production efficiency and the product quality are improved, and the production cost is reduced. Meanwhile, the overedger is convenient and accurate to control and high in response speed. The equipment operation noise is ultralow, the energy-saving, environment-friendly and noise-reducing effects are achieved, the pollution to the working environment and the harm to human bodies are reduced, and the safety guarantee of operators is improved.

Description

Technical Field

[0001] This invention belongs to the field of overlock sewing machines and relates to a fully automatic overlock sewing machine based on the Internet of Things. Background Technology

[0002] Industrial overlock sewing machines are special sewing machines commonly used in large-scale production in textile and garment factories. They can be used to hem garments and sew together fabrics such as underwear, sportswear, T-shirts, and knitwear.

[0003] With the rapid development of science and technology in China, some domestic sewing machine companies have achieved certain technological breakthroughs in their products, but they are still stuck in semi-automatic production and have not combined advanced computer technology, control technology, microelectronics technology and power electronics technology, which makes it impossible for overlock sewing machines to be applied to large-scale fully automated factory production. Summary of the Invention

[0004] In order to overcome at least one deficiency of the prior art, the present invention provides a fully automatic overlock sewing machine based on the Internet of Things.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a fully automatic overlock sewing machine based on the Internet of Things, comprising... Fuzzy PID Controller Based on SPSO Algorithm: The network parameters of the convolutional neural network are optimized by SPSO algorithm, and the optimized neural network is used to perform online self-tuning of the parameters of fuzzy PID control. Overlock sewing machine information acquisition module: used to collect status parameter information of the overlock sewing machine; Wireless sensor network-based fault prediction system for overlock sewing machines: used for the assessment and control of overlock sewing machine operating quality, fault prediction, and intelligent fault diagnosis.

[0006] Furthermore, the mutation rules of the SPSO algorithm are used to maintain particle diversity.

[0007] Furthermore, the fuzzy PID controller takes the error e and the error change rate ec as inputs, satisfies the self-tuning requirements of e and ec for PID parameters at different times, determines the fuzzy subset and formulates a two-dimensional table of fuzzy rules to obtain the quantized output, then defuzzifies the quantized output, and obtains the control quantity through the PID controller to control the controlled object.

[0008] Furthermore, the overlock sewing machine information acquisition module includes an STM32F407ZGT6 microcontroller, an MP2359 chip, an AMS1117 linear voltage regulator chip, an A / D converter, and an acquisition system.

[0009] Furthermore, the STM32F407ZGT6 microcontroller operates at 3.3V. The MP2359 chip performs a step-down conversion, reducing the power supply voltage to 5V. The AMS1117 linear regulator chip then converts the 5V to 3.3V small-dropout regulated A / D converter into a digital signal, facilitating the STM32 system's analysis and processing of the digital data.

[0010] Furthermore, the acquisition system uses an AD7606 chip to convert the acquired information. The STM32F407ZGT6 microcontroller and the AD7606 chip interact via the SPI bus. The control signal pins of the A / D conversion chip are connected to the corresponding GPIO ports of the STM32F407ZGT6 microcontroller, and information is transmitted via parallel port DB:0-15.

[0011] Furthermore, the wireless sensor network-based overlock sewing machine fault prediction system includes a sensing layer, a transmission layer, and an application layer. These three layers correspond to the real-time acquisition, forwarding, and application of CNC machine tool operating parameters, respectively.

[0012] Furthermore, the perception layer is a wireless sensor network composed of various sensor nodes. The wireless sensor network collects real-time information on equipment operating parameters such as temperature, vibration, pressure, and noise, and provides a data source for subsequent applications.

[0013] Furthermore, the transport layer consists of gateway hardware, standard transmission protocols, and a LAN / WAN network, completing the data transmission for the entire system. The standard network transmission protocols and LAN / WAN network used therein send the data collected from the perception layer to the application layer system via the gateway.

[0014] Furthermore, the application layer is located at the top of the system and supports Internet application services. The application layer performs rough processing, appropriate processing, analysis and prediction on the data sent from the transport layer, which facilitates the maintenance and management personnel of the overlock sewing machine to monitor the equipment status in real time and prepare for early warning and alarm.

[0015] In summary, the advantages of this invention are: Compared to traditional overlock sewing machines, this invention incorporates motor control technology and Internet of Things (IoT) technology, improving production efficiency and product quality. Furthermore, the machine offers convenient, precise, and fast control. Its ultra-low operating noise achieves energy-saving, environmentally friendly, and noise-reducing effects, minimizing pollution to the working environment and harm to personnel, and enhancing operator safety. Detailed Implementation

[0016] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0017] Example: A fully automatic overlock sewing machine based on the Internet of Things, including Fuzzy PID Controller Based on SPSO Algorithm: The network parameters of the convolutional neural network are optimized by SPSO algorithm, and the optimized neural network is used to perform online self-tuning of the parameters of fuzzy PID control. Overlock sewing machine information acquisition module: used to collect status parameter information of the overlock sewing machine; Wireless sensor network-based fault prediction system for overlock sewing machines: used for the assessment and control of overlock sewing machine operating quality, fault prediction, and intelligent fault diagnosis.

[0018] In the particle optimization process, a similarity particle swarm optimization (SPSO) algorithm is added to measure the similarity between particles. The mutation rules of the SPSO algorithm are used to maintain the diversity of particles, thereby improving the defect of the particle swarm optimization algorithm that is prone to premature convergence.

[0019] In the similarity particle swarm optimization algorithm, the similarity sim(i,j) between two particles i and j must satisfy the following criterion: Rule 1: sim(i,j) = 1; Rule 2: When dim (i,j) →∞, sim (i,j) = 0.

[0020] Criterion 3: For any two particles i and j, sim (i,j)∈[0,1].

[0021] The first criterion defines the particle as being most similar to itself, at which point the similarity value is 1, meaning the maximum similarity between particles is 1. Criterion 2 defines that if the distance between two particles is infinite, that is, they are basically dissimilar, the similarity value between the two particles is 0, that is, the minimum similarity value between particles is 0. Criterion 3 defines the similarity range between two particles. Criterion 1 and Criterion 2 can be considered as two extreme cases of Criterion 3.

[0022] Similarity measures the degree of particle clustering. The closer two particles are, the more similar they are, and the higher the similarity value. According to the iteration of the particle swarm optimization algorithm, any particle i will converge to the global optimal particle gbest, i.e., xi(t) → gbest. That is, the similarity s(i,g) between particle i and the global optimal particle gbest will become smaller and smaller, eventually approaching 0. In order to maintain the diversity of the population, when the particles cluster near the optimal position gbest and reach the set mean, the position xi of particle i is randomly mutated according to the population clustering degree C(t) and the similarity s(i,g) between the optimal particle gbest.

[0023] The greater the aggregation degree C(t) of the particle swarm or the greater the similarity s(i,g) between a particle i and the optimal particle g, the greater the probability of random particle mutation and the greater the opportunity to increase particle diversity. To balance global search capability and local search capability, the mutation coefficient am decreases linearly with the increase of iteration number. This is beneficial for improving global search capability in the early stage of algorithm iteration and maintaining local detection capability in the later stage of iteration.

[0024] Fuzzy adaptive PID control, using the error e and the rate of change of error ec as inputs, can meet the self-tuning requirements of PID parameters at different times for e and ec. The structure of fuzzy PID control is achieved by modifying the PID parameters using fuzzy rules.

[0025] The process involves determining a fuzzy subset and creating a two-dimensional table of fuzzy rules to quantize the output. The quantized output is then defuzzified to obtain the new PID parameters. These parameters are then processed by a PID controller to generate the control input, which in turn controls the controlled object. Fuzzy rules are crucial for ensuring good control characteristics of the control system and consist of three fuzzy rule tables.

[0026] The overlock sewing machine information acquisition module includes an STM32F407ZGT6 microcontroller, an MP2359 chip, an AMS1117 linear regulator chip, an A / D converter, and an acquisition system. The STM32F407ZGT6 microcontroller operates at 3.3V. Considering the power supply of the external interface circuits, it is necessary to ensure that the voltage and current are at their rated values. The overlock sewing machine uses the MP2359 chip, which has high efficiency and low heat generation, to step down the power supply voltage to 5V. Then, the AMS1117 linear regulator chip, which has low cost and low ripple, is used to achieve a small voltage drop from 5V to 3.3V. Through two voltage reduction steps, the energy loss and reduced conversion efficiency caused by high heat generation due to large voltage reduction are reduced.

[0027] The A / D converter transforms the analog information acquired by the underlying sensing module into digital signals, facilitating the STM32 system to analyze and process the digital signals; the high-precision A / D converter ensures the efficiency and accuracy of the acquisition system.

[0028] The acquisition system uses the AD7606 chip to convert the acquired information. This chip supports ±5V and ±10V bipolar analog signal input. The STM32F407ZGT6 microcontroller interacts with the AD7606 chip via the SPI bus. The control signal pins of the A / D conversion chip are connected to the corresponding GPIO ports of the STM32F407ZGT6 microcontroller, and information is transmitted via parallel port DB:0-15.

[0029] The STM32F407ZGT6 microcontroller achieves information interaction with the outside world through communication interface circuits. For RS233 and RS485 interfaces, MAX232 and MAX485, which are compatible with 5V power supply, are selected respectively to complete the level conversion and realize information interaction.

[0030] The wireless sensor network-based fault prediction system for overlock sewing machines consists of a sensing layer, a transmission layer, and an application layer. These three layers correspond to the real-time acquisition, forwarding, and application of CNC machine tool operating parameters, respectively.

[0031] The perception layer is a wireless sensor network composed of various sensor nodes. It collects real-time information on equipment operating parameters such as temperature, vibration, pressure, and noise through the wireless sensor network and provides a data source for subsequent applications.

[0032] The transport layer, composed of gateway hardware, standard transmission protocols, and a LAN / WAN network, completes the data transmission for the entire system. The standard network transmission protocols and LAN / WAN network used rapidly transmit data collected from the perception layer to the application layer system via the gateway. During data transmission, potential transmission errors are effectively avoided, ensuring data integrity and reliability.

[0033] The application layer, located at the top of the system, supports internet application services for user convenience. It performs preliminary processing, appropriate handling, analysis, and prediction on the data sent from the transport layer, facilitating real-time monitoring of equipment status by overlock machine maintenance personnel and enabling them to prepare for early warnings and alarms. The wireless sensor network-based overlock machine fault prediction system continuously acquires a large amount of product operation data, providing an information foundation for the assessment and control of overlock machine operation quality, fault prediction, and intelligent fault diagnosis. Through big data analysis, deep learning-based equipment fault anomaly analysis technology, and intelligent diagnostic technology, potential faults can be detected early, preventing their occurrence and further deterioration, reducing downtime losses caused by frequent maintenance, and improving equipment utilization.

[0034] Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort should fall within the scope of protection of the present invention.

Claims

1. A fully automatic overlock sewing machine based on the Internet of Things, characterized in that: include Fuzzy PID Controller Based on SPSO Algorithm: The network parameters of the convolutional neural network are optimized by SPSO algorithm, and the optimized neural network is used to perform online self-tuning of the parameters of fuzzy PID control. Overlock sewing machine information acquisition module: used to collect status parameter information of the overlock sewing machine; Wireless sensor network-based fault prediction system for overlock sewing machines: used for the assessment and control of overlock sewing machine operating quality, fault prediction, and intelligent fault diagnosis.

2. The fully automatic overlock sewing machine based on the Internet of Things according to claim 1, characterized in that: The mutation rules of the SPSO algorithm are used to maintain particle diversity.

3. The fully automatic overlock sewing machine based on the Internet of Things according to claim 2, characterized in that: The fuzzy PID controller takes the error e and the error change rate ec as inputs, meets the self-tuning requirements of e and ec for PID parameters at different times, determines the fuzzy subset and formulates a two-dimensional table of fuzzy rules to obtain the quantized output, then defuzzifies the quantized output, and obtains the control quantity through the PID controller to control the controlled object.

4. The fully automatic overlock sewing machine based on the Internet of Things according to claim 1, characterized in that: The overlock sewing machine information acquisition module includes an STM32F407ZGT6 microcontroller, an MP2359 chip, an AMS1117 linear regulator chip, an A / D converter, and an acquisition system.

5. The fully automatic overlock sewing machine based on the Internet of Things according to claim 4, characterized in that: The STM32F407ZGT6 microcontroller operates at 3.3V. The MP2359 chip performs a step-down conversion, reducing the power supply voltage to 5V. The AMS1117 linear regulator chip then converts the 5V to 3.3V small-dropout regulated A / D converter into a digital signal, facilitating the STM32 system's analysis and processing of the digital data.

6. The fully automatic overlock sewing machine based on the Internet of Things according to claim 4, characterized in that: The acquisition system uses the AD7606 chip to convert the acquired information. The STM32F407ZGT6 microcontroller interacts with the AD7606 chip via the SPI bus. The control signal pins of the A / D conversion chip are connected to the corresponding GPIO ports of the STM32F407ZGT6 microcontroller, and information is transmitted via parallel port DB:0-15.

7. The fully automatic overlock sewing machine based on the Internet of Things according to claim 1, characterized in that: The wireless sensor network-based fault prediction system for overlock sewing machines includes a sensing layer, a transmission layer, and an application layer. These three layers correspond to the real-time acquisition, forwarding, and application of CNC machine tool operating parameters, respectively.

8. A fully automatic overlock sewing machine based on the Internet of Things according to claim 7, characterized in that: The perception layer consists of a wireless sensor network composed of various sensor nodes. This wireless sensor network collects real-time information on equipment operating parameters such as temperature, vibration, pressure, and noise, and provides a data source for subsequent applications.

9. A fully automatic overlock sewing machine based on the Internet of Things according to claim 7, characterized in that: The transport layer, consisting of gateway hardware, standard transmission protocols, and a LAN / WAN network, completes the data transmission for the entire system. The standard network transmission protocols and LAN / WAN network used therein send data collected from the perception layer to the application layer system via the gateway.

10. A fully automatic overlock sewing machine based on the Internet of Things according to claim 7, characterized in that: The application layer is located at the top of the system and supports Internet application services. The application layer performs rough processing, appropriate processing, analysis and prediction on the data sent from the transport layer, which facilitates the maintenance and management personnel of the overlock sewing machine to monitor the equipment status in real time and prepare for early warning and alarm.