CN-122013453-A - Full-automatic overedger based on thing networking

CN122013453ACN 122013453 ACN122013453 ACN 122013453ACN-122013453-A

Abstract

The invention discloses a full-automatic overedger based on the Internet of things, which comprises a fuzzy PID controller based on SPSO algorithm, wherein network parameters of a convolutional neural network are optimized through SPSO algorithm, the optimized neural network is utilized to perform online self-tuning on the parameters of fuzzy PID control, an overedger information acquisition module is used for acquiring state parameter information of the overedger, and an overedger fault prediction system of a wireless sensor network is used for evaluating and controlling the running quality of the overedger, predicting faults and performing intelligent fault diagnosis. The equipment has ultralow running noise, achieves the effects of energy conservation, environmental protection and noise reduction, reduces pollution to the working environment and harm to human bodies, and improves the safety guarantee of operators.

Inventors

CAI JINGXIAN
ZHENG JUN
ZHOU HAOXIANG
YIN SHAOJIE
WANG MENG
WANG WEI
ZHU YOUYOU

Assignees

浙江大学台州研究院

Dates

Publication Date: 20260512
Application Date: 20260209

Claims (10)

1. A full-automatic overedger based on the Internet of things is characterized by comprising The fuzzy PID controller based on SPSO algorithm optimizes the network parameters of the convolutional neural network by SPSO algorithm, and performs on-line self-tuning on the parameters of the fuzzy PID control by using the optimized neural network; The overedger information acquisition module is used for acquiring state parameter information of the overedger; The overedger fault prediction system of the wireless sensor network is used for evaluating and controlling the running quality of the overedger, predicting faults and intelligently diagnosing faults.
2. The full-automatic overedger based on the Internet of things of claim 1, wherein the SPSO algorithm mutation rule is used for maintaining the diversity of particles.
3. The full-automatic overedger based on the Internet of things, as set forth in claim 2, wherein the fuzzy PID controller takes an error e and an error change rate ec as inputs, meets the requirements of different moments e and ec on PID parameter self-tuning, determines a fuzzy subset and formulates a fuzzy rule two-dimensional table to obtain quantized output, defuzzifies the quantized output, obtains a control quantity through the PID controller, and controls a controlled object.
4. The full-automatic overedger based on the Internet of things of claim 1, wherein the information acquisition module of the overedger comprises an STM32F407ZGT6 microcontroller, an MP2359 chip, a linear voltage stabilizing chip AMS1117, an A/D converter and an acquisition system.
5. The full-automatic overedger based on the Internet of things of claim 4, wherein the STM32F407ZGT6 microcontroller has a working voltage of 3.3V, the MP2359 chip is used for performing buck conversion to reduce the power supply voltage to 5V, and the linear voltage-stabilizing chip AMS1117 is used for converting the analog quantity information acquired by the bottom layer sensing module into a digital quantity signal by a small voltage-drop voltage-stabilizing A/D converter of 5V to 3.3V, so that the STM32 system is convenient for analyzing and processing the digital quantity.
6. The full-automatic overedger based on the Internet of things according to claim 4, characterized in that the acquisition system adopts an AD7606 chip to realize the conversion of the acquired information quantity, the STM32F407ZGT6 microcontroller and the AD7606 chip realize the information interaction through an SPI bus, the control signal pin of the A/D conversion chip is connected with the GPIO port corresponding to the STM32F407ZGT6 microcontroller, and the information transmission is realized through the parallel port DB 0-15.
7. The full-automatic overedger based on the Internet of things of claim 1, wherein the overedger fault prediction system of the wireless sensor network comprises a sensing layer, a transmission layer and an application layer, and the three layers respectively correspond to real-time acquisition, forwarding and application processes of operation parameters of a numerical control machine tool.
8. The full-automatic overedger based on the Internet of things of claim 7, wherein the sensing layer is a wireless sensor network formed by each sensor node, and the real-time information of equipment operation parameters such as temperature, vibration, pressure, noise and the like is collected through the wireless sensor network and a data source is provided for subsequent applications.
9. The full-automatic overedger based on the Internet of things of claim 7, wherein the transmission layer is composed of gateway hardware, standard transmission protocols and LAN/WAN networks, and completes data transmission of the whole system. The standard network transmission protocol and LAN/WAN network used therein send the data information collected from the sensing layer to the application layer system through the gateway.
10. The full-automatic overedger based on the Internet of things of claim 7, wherein the application layer is positioned at the top layer of the system, supports Internet application services, performs rough processing, proper processing and analysis and prediction on data sent by the transmission layer, is convenient for maintenance manager of the overedger to monitor equipment state in real time, and is ready for early warning and alarming.

Description

Full-automatic overedger based on thing networking Technical Field The invention belongs to the field of overedgers, and relates to a full-automatic overedger based on the Internet of things. Background The industrial overedger is a special sewing machine which is commonly applied to mass production in textile and clothing factories, can be used for hemming clothing, and can also be used for sewing fabrics such as underwear, sportswear, T-shirts, knitting and the like. With the rapid development of domestic technology, some products of domestic sewing machine enterprises realize a certain technological breakthrough, but still stay in semi-automatic production, and advanced computer technology, control technology, microelectronic technology and power electronic technology are not combined, so that the overedger cannot be applied to large-scale full-automatic industrial production. Disclosure of Invention The invention provides a full-automatic overedger based on the Internet of things, which aims to overcome at least one defect of the prior art. In order to achieve the aim, the invention adopts the following technical scheme that the full-automatic overedger based on the Internet of things comprises The fuzzy PID controller based on SPSO algorithm optimizes the network parameters of the convolutional neural network by SPSO algorithm, and performs on-line self-tuning on the parameters of the fuzzy PID control by using the optimized neural network; The overedger information acquisition module is used for acquiring state parameter information of the overedger; The overedger fault prediction system of the wireless sensor network is used for evaluating and controlling the running quality of the overedger, predicting faults and intelligently diagnosing faults. Further, the mutation rule of SPSO algorithm is used for maintaining the diversity of particles. Furthermore, the fuzzy PID controller takes the error e and the error change rate ec as inputs, meets the self-setting requirement of PID parameters at different moments e and ec, determines a fuzzy subset, formulates a fuzzy rule two-dimensional table to obtain quantized output, defuzzifies the quantized output, obtains control quantity through the PID controller and controls a controlled object. Further, the information acquisition module of the overedger comprises an STM32F407ZGT6 microcontroller, an MP2359 chip, a linear voltage stabilizing chip AMS1117, an A/D converter and an acquisition system. Furthermore, the working voltage of the STM32F407ZGT6 microcontroller is 3.3V, the MP2359 chip is used for performing buck conversion, the power supply voltage is reduced to 5V, the linear voltage stabilizing chip AMS1117 is used for converting the analog quantity information acquired by the bottom layer sensing module into a digital quantity signal through the small voltage drop voltage stabilizing A/D converter from 5V to 3.3V, and the STM32 system is convenient for analyzing and processing the digital quantity. Furthermore, the acquisition system adopts an AD7606 chip to realize the conversion of the acquired information quantity, the STM32F407ZGT6 microcontroller and the AD7606 chip realize the information interaction through an SPI bus, the control signal pin of the A/D conversion chip is connected with the GPIO port corresponding to the STM32F407ZGT6 microcontroller, and the information transmission is realized through the parallel port DB 0-15. Furthermore, the overedger fault prediction system of the wireless sensor network comprises a sensing layer, a transmission layer and an application layer, and the three layers respectively correspond to real-time acquisition, forwarding and application processes of the operation parameters of the numerical control machine tool. Furthermore, the sensing layer is a wireless sensor network formed by each sensor node, and the wireless sensor network is used for collecting real-time information of equipment operation parameters such as temperature, vibration, pressure, noise and the like and providing a data source for subsequent application. Furthermore, the transmission layer is composed of gateway hardware, standard transmission protocol and LAN/WAN network, and completes the data transmission of the whole system. The standard network transmission protocol and LAN/WAN network used therein send the data information collected from the sensing layer to the application layer system through the gateway. Furthermore, the application layer is positioned at the top layer of the system, supports Internet application service, performs rough machining, proper processing and analysis prediction on data sent by the transmission layer, is convenient for a maintenance manager of the overedger to monitor the equipment state in real time, and performs early warning and alarm preparation. In summary, the invention has the following advantages: Compared with the traditional overedger, the invention increases the motor control technology a