Ebook: Mechatronics and Automation Technology

With the development of science and technology, mechatronics and automation technology has become an important symbol of information and modern industrial production, and has a huge impact in many fields such as manufacturing industry, robots, automation, automobiles, biomedicine. It has changed the face of the entire traditional machinery manufacturing industry.

This book contains the proceedings of the 2022 International Conference on Mechatronics and Automation Technology (ICMAT 2022), which was held virtually due to the COVID-19 pandemic on October 29, 2022. The Conference is an ideal platform for bringing together researchers, practitioners, scholars, professors and engineers from all around the world to exchange the newest research results and stimulate the scientific innovations. The conference received 117 submissions, and accepted 82 papers for publication after a rigorous process of peer-review. The topics covered mechanical manufacturing and equipment, robots, information technology, automation technology, automotive systems and other related fields.

The book provides an overview of technologies and applications in mechatronics and automation technology, as well as current research and development, and will be of interest to researchers, engineers, and educators working in the field.

The Editor

To improve the management of collaborative manufacturing tasks in MES and promote the digital transformation of discrete manufacturing enterprises, the collaborative manufacturing of discrete manufacturing based on MES has been studied. First, the business flow of collaborative manufacturing in an MES was analyzed systematically. Second, the algorithm of detailed scheduling for collaborative tasks was studied, and a model of external collaborative resources was designed. Then, a solution of collaborative scheduling was presented based on a hybrid genetic algorithm. Finally, a solution for the execution process management of collaborative manufacturing was promoted based on the integration of MES and ERP. The technologies researched can improve the general management of collaborative manufacturing and guide practical production management.

This research has developed a simple and efficient algorithm for Bangla speech recognition using an artificial neural network. This supervised learning model consists of four major phases: collecting audio samples, conversion into frequency domain & filtering, normalization, and applying the back propagation algorithm to train the artificial neural network. Proper filtering has granted higher noise immunity and has decreased the number of elements in the input layer of the neural network. Thus, the overall processing speed is increased. The audio samples for this research have been collected through a digital microphone, and the complete model is illustrated using GNU Octave. Although this research focuses on the Bangla language, this model can be applied to any other language.

In this paper, a kind of temperature sensor for fermentation bed in large scale chicken farm was studied. The temperature sensing equipment includes a temperature sensor housing, a circuit board and a temperature probe detection sensor. The circuit board is provided with a microprocessor and a wireless Bluetooth communication module. The temperature probe detecting sensor is fixedly connected with the sensor housing, and at least two temperature detecting elements are arranged inside the temperature probe detecting elements, which are spaced inside the probe tube. The microprocessor is connected with the temperature detecting element and wireless Bluetooth module respectively. The temperature sensor is used to detect the wall temperature of the probe tube and generate temperature signals. The microprocessor is used to receive the detection signals and forward them to the wireless Bluetooth module. The wireless Bluetooth module is used to convert the temperature detection signals into wireless signals and send them out.

With the rapid development of image recognition technology, it has its applications in medical, security and other fields, but this technology has many areas to be improved, such as the accuracy of recognition, real-time and other issues, which have always been a hot research topic in this field. This article involves the development process of image recognition (traditional image processing, machine learning, deep learning), and focuses on the advantages and disadvantages of the popular YOLO algorithm in image recognition. Finally, it describes the technical problems faced by the current target detection technology and the corresponding solutions.

In the Saudi Master Gas System Expansion project, the portal frame of the overhead line of the high-voltage substation was reformed. The safety and stability of the laying of the high-voltage cable was guaranteed through the optimized design of the supporting structure and the production of the high-voltage cable head. Good practical results have been obtained, the construction quality and period have been guaranteed. It provides a guiding and operational statement for the improved technology of portal frame of overhead line in high-voltage substation.

Titanium alloy is a kind of difficult-to-machine material, which needs to seek an efficient processing method. In this paper, the deposition samples of TB6 titanium alloy were prepared by laser deposition manufacturing. The internal microstructure of the as-deposited was analyzed by Optical Microscope(OM) and Scanning Electron Microscope(SEM). The formation mechanism of interlaminar bands and the variation law of microhardness were studied. The microstructure of TB6 titanium alloy prepared by laser deposition is composed of most equiaxed and a small part of original β grains elongated along the deposition direction and approximately ellipsoidal. The original β grains are mainly composed of primary α phase(α_p), grain boundary α phase(α_GB) and matrix β phase. The temperature fluctuation between the deposition layers in the laser deposition process leads to the change of α_p phase size, which forms the layer band distribution with different contrasts.

The TB6 titanium alloy prepared by laser deposition is prone to uneven microstructure. In this paper, heat treatment was used to regulate the microstructure. Through the observation and analysis of microstructure and microhardness under different heat treatment states, the variation law of microhardness with the change of heat treatment process was explored. The results show that due to the distribution of layer bands in the as-depositeds of TB6 titanium alloy fabricated by laser deposition manufacturing. The microhardness decreases gradually with the increase of deposition height in a single deposited layer. The average microhardness at the dark band position is 377.4 HV_0.2, while the microhardness at the top bright band position decreases by 7.3 %. Low temperature annealing treatment has little effect on the microhardness. After high temperature annealing treatment, the microhardness decreases by about 4.6 % compared with the deposition state. The secondary α phase (α_s) precipitated after solution and aging treatment has obvious strengthening effect on the β matrix. The microhardness is greatly improved, and increases with the increase of solution temperature. The microhardness after solution and aging treatment at low and high temperatures is 10.0 % and 19.2 % higher than that of the deposited state, respectively.

Aiming at the problem that the staff do not wear safety helmets or even wear safety helmets in the construction and maintenance of new energy charging stations, an improved YOLOv3 safety helmet detection method is designed in this paper. Firstly, the 128*128 feature map output is added based on the three feature map output of YOLOv3 algorithm, and then the FI module is added to fuse the four scale feature information to improve the detection accuracy of small targets. Firstly, on the basis of the 3 feature map outputs of the YOLOv3 algorithm, 128*128 feature map outputs are added, and then the FI module is added to fuse the information of the 4 scale feature information to improve the detection accuracy of small targets; Secondly, the DIoU function is used to optimize the boundary frame loss function, so that the boundary frame regression is more accurate. Compared with the original algorithm YOLOv3, the proposed algorithm has higher detection accuracy, shorter time consuming and meets the requirements of real-time detection. Compared with other advanced algorithms, the robustness, detection accuracy and detection speed of the proposed algorithm are better, which can provide technical reference for avoiding staff safety hazards.

With the development of new energy vehicles, the detection and fault diagnosis of high voltage system of new energy vehicles are becoming more and more important. The leakage of high-voltage system of new energy vehicles will lead to the failure of power on and normal operation of vehicles. At the same time, it is very important for the safety protection of the whole vehicle. Taking the leakage detection of byd-qin hybrid high-voltage system as an example, this paper analyzes the fault generation mechanism and puts forward the detection technology of new energy vehicles, so as to help maintenance personnel better grasp the diagnosis technology of new energy vehicles.

Face quality evaluation can filter out low quality face image to save computational resources and improve the system performance, labeling the face image quality score by manual consume too much manpower. To solve this problem, an unsupervised face image evaluation based on face recognition is proposed. We use the face recognition model to calculate the features of faces and label the images quality score. The face recognition model is compressed by knowledge distillation method to obtain efficient quality assessment model. Experimental results show that this method can effectively evaluate the quality of face image and improve the performance of face recognition.

In modern society, due to the diversification of information and the acceleration of network communication, the barriers of communication between teachers, parents and college students are increasing, so that teachers or parents can’t understand the psychological state of contemporary students in time. When college students have psychological problems, they are more willing to ask for help from classmates and friends than from teachers or psychologists. Therefore, this system is devoted to solving the problem that college students are not willing to talk to their teachers and parents, and providing the characteristic psychological analysis function for college students to evaluate their psychological state. Firstly, This paper describes the research background and significance, and then, it introduces Wechat Mini Program, Springboot framework, database and behavior recognition technology. On the basis of the above, the overall architecture, functional structure and database of the system are designed. Finally, the functional modules of the system are coded. The system interface is beautiful, easy to operate, and set practicality and interest in one, can let college students more understand their psychological state, face mental health problems.

By analyzing the characteristics of Archimedes spiral equation, this equation takes the included angle as the variable, and the algorithm of equal included angle straight line approximation can be used to calculate the interpolation points. Using the advantage of parametric programming by FANUC system macro program, a G instruction(G102) is customized to realize the machining of Archimedes spiral surface. This G instruction is used in the same way as the system’s inherent G code, similar to a custom fixed loop code. It is different from the program number that G code must indicate when calling macro program. The G102 code can complete the machining of Archimedean spiral surface only by assigning values to each letter according to the format specified in the text. After processing verification on GLU28x40 CNC, the use method of customized G code is as flexible and convenient as the fixed cycle G code provided by CNC. It can process Archimedes spiral profile that meets the requirements, and use the equal angle straight line approximation algorithm for interpolation. The calculation speed is fast, which effectively improves the processing efficiency of the profile.

Under the background of the general plan for general survey of Germplasm Resources issued by the State Forestry and grassland administration, in order to establish the forest and grass germplasm resources management system, it is urgent to identify the forest vegetation types. In the past, manual identification was often used, which was inefficient and had a high error rate. Considering the small number of samples in the plant leaf database, in order to improve the accuracy of plant leaf identification, In the framework of tensorflow, a convolution neural network method for plant leaf image recognition based on transfer learning is proposed. Firstly, the plant leaf image is preprocessed, and the plant leaf image data set is expanded through the horizontal transformation, random clipping, translation transformation, color and illumination transformation of the original image, and is divided into training set and test set in the ratio of 7:3, the concept V3 model is applied to image data processing by migration learning. The trained models ResNet50 and Concept V3 are migrated and trained on the plant leaf image data set, and the full connection layer is replaced in the pre training model, so that it can adapt to the recognition of plant leaf images. The accuracy of the test set obtained from the pre training model of this method is 95.22% and 95.45%, reaching the excellent level required by the task.

In order to improve the overall economy of fuel cell vehicle, this paper proposes an energy management strategy based on the Car-Following Model of Long Short-Term Memory Neural Network (LSTM-CFM), which classified vehicles according to the Vehicle_ID, Frame_ID, Global_Time and other core data based on the vehicle information collected from the Next Generation Simulation (NGSIM) data set on the US 101 highway in south California. The two adjacent vehicles were identified and the data were extracted and processed into the LSTM-CFM for training. In the simulation stage, UDDS (a standard driving condition) is taken as the driving condition of the followed vehicle and 50m is taken as the initial distance between the following vehicle and the target vehicle. Under these initial conditions, we can get the speed prediction data of target vehicle based on the Car-Following model. The energy management strategy is designed based on the principle of minimum equivalent fuel consumption, and the equivalent factor can be solved by using the speed prediction. The simulation results show that the economy of the fuel cell vehicle under the LSTM-CFM energy management strategy is better than that of the rule-based strategy, which verifies the effectiveness of the LSTM-CFM energy management strategy.

This paper investigates the 5G power application status in China, and compares the mainstream communication technologies of the existing power system, such as wired, 4G, wifi, 5G and so on. Effectively combined the advantages of 5G technology with power applications, the paper clarifies the application scenarios of 5G in the power system. The research focuses on typical power applications under the background of new power system construction, including intelligent inspection, distribution network protection and control, distributed energy grid connection and dispatching stability control, and analyzes the technical indicators and economy of various typical applications. Finally, this paper analyzes the challenges faced by 5G power application, and proposes solutions, as well as prospects for 5G power application.

This paper analyses the TSN technology and its basic standard components. It also analyses the architectures of 5G System and 5G RAN. Base on the two analyses, three technology directions of 5G and TSN integration are studied. Finally, to realize the integration, key technologies are discussed.

This paper studies the electro-hydraulic proportional position control system of a hydraulic rock drilling jumbo. First of all, the composition of the system is introduced and the transfer function of the system components is established, and the mathematical model of the whole system is finally obtained through calculation and simplification. Then, the PID and the fuzzy PID control strategy were designed by MATLAB software, and the system model was built in simulink. Finally, the results indicate that the fuzzy PID controller has better anti-interference ability, and the response speed is more advantageous than that of PID.

A multi-exposures with dose modulation method is presented to fabricate SU-8 mold which is used for microfluidic channels. The method used a maskless digital lithography device with a 405 nm LED source. Digital micromirror device (DMD) maskless lithograph can reach higher precision to meet the requirements of microfluidic chips. For a thick SU-8 layer, multi-exposures with low dose method can effectively suppress the T-shaped structure formed due to top overexposed and bottom underexposed. Two SU-8 molds were fabricated with 55 μm and 25 μm in height in this work. The number of exposures is 62 and 55, respectively. The actual contour of SU-8 structures closely matches the design contour. The PDMS microchannel structure was fabricated using the SU-8 mold. The minimum width of a single microchannel achieved 10 μm. This method provides a more flexible method for fabricating SU-8 molds with higher precision and is well suited for on-chip cell isolation.

As the expansion of edge devices, such as local edge servers or even IoT (Internet of Things) devices, edge computing reforms data processing flow, engendering improvement in faster insights, improved response times and better bandwidth availability. However, industry encounters computational resource constraints of machine learning on edge devices, e.g., limitation of memory and processing power. To tackle this situation, we attempt to enhance flexibility of neural network models for variable hardware constraints. In this paper, we propose a paradigm of neural network combination implementation as a primitive solution; further more, we manage to build up a general neural network deployment framework for edge devices.

This paper proposes a simple and efficient VRM format display optimization scheme. The optimization effect is significant. The display of avatars has played a pivotal role in the development of all types of interactive media. VRM is an easy-to-create, share, and copyright-protect virtual image file format. VRM is an easy-to-create, share, and copyright-protected avatar format, but its common browser-based display solution, three-vrm, suffers from serious performance problems. In this project, we propose several performance optimization schemes, each of which is experimentally analyzed and verified, and finally propose a simple and efficient optimization scheme for the display of the VRM format, which achieves significant results. This technical report contains four main sections: the first provides an introduction to the background of the project and a brief comparison of the virtual image implementation options that exist on the Internet today to discover the direction and significance of the project. The technical and system features and innovations used in the project are also covered. The second part describes the research process, intermediate results, and phased conclusions of the project. In the third part, the principles of the implementation and the core code of the final solution in the project are explained and sorted out in detail. The fourth part discusses and summarises the limitations of the system and areas for improvement.

The construction of the new laser-driven proton therapy facility (CLAPA-II), designed by Peking University, has begun in Beijing. This paper describes the design of a cavity beam position monitor (CBPM) for the CLAPA-II. The proton beam which is accelerated to 100 MeV by 2-PW Laser has parameters of 108∼1010 particles range, 1 Hz repetition rate, 20∼40 mm beam length range. A non-interceptor BPM is needed to characterize the beam motion state in real time. The simulation has been performed to study CBPM properties and cross-talk.

With the development of economy and the improvement of people’s living standards, the garbage problem has become increasingly prominent, and people’s concept of environmental protection has been continuously enhanced. In response to the popularization of garbage classification and environmental pollution of garbage, the industry has seen innovative designs for traditional garbage bins. The intelligent waste distribution bin studied in this paper adopts micro-control chip, infrared detection device, mechanical transmission device and linkage mechanism, which can solve the difficulty in separation from the source, cultivate the public’s awareness of waste separation and reduce the cost of waste disposal, It protects user privacy and improves security through blind processing, and thus has practical research significance.

As for the challenges of local motion planning subject to constrained state for quadrotor in evaluating the health status of photovoltaic power generation in complex environment, including fast replanning, limited onboard computational resources, a practical framework for constructing motion trajectory from the bounded states is proposed in this work. The references and nominal inputs are made by the scheme of receding horizon fashion of model predictive control (MPC) by means of gradient-based method in the continuous domain. In order to avoid the complex calculation of the optimal gradient, the ability of approximation of a neural network make it possible for the fast evaluation and online optimization. The simulations with variety types of trajectories are made by means of a quadrotor to demonstrate the good efficiency and real-time of the proposed approach.

In order to avoid the instability flutter of aero-engine blade and prevent the failure of thin film thermocouple caused by the blade in complex environment, the structural strength of the thermocouple on the actual blade surface was studied in this paper. Firstly, based on the 3D scanning of the real blade, the blade model is established after the point cloud graphics. The surface thermocouple model of the blade is established in Solidworks for simulation, and the modal and harmonic response analysis of the blade is simulated. Secondly, the vibration of the curved thin film thermocouple and the stress distribution of the blade under the acceleration shock were studied in Comsol, and the failure weakness was analyzed to avoid the failure of the thin film thermocouple caused by the bending, fracture or shedding of the thin film thermocouple on the blade due to vibration and shock, providing the most suitable thermocouple selection for the engine blade.