APRÈN. Portal de la producció docent del professorat de la UPC

This is a detailed paper on the design and implementation of a smart farming system using STM32 microcontrollers and some sensors such as for temperature and humidity, CO2 concentration, soil moisture and light sensors. It focuses on improving crop growing efficiency through automation and intelligent management of greenhouse environment. The system collects environmental parameters and displays them in real time on a display screen or a mobile app for automatic regulation and remote control. This innovative approach aims to improve agricultural productivity by integrating advanced sensor technologies, data collection and control mechanisms. The software used for program writing and designing is Keil 5. The hardware and software design aspects of the system were continuously improved after referring to a large number of literary sources, including circuit design, programming. I conducted software and hardware tests,the experimental results were in accordance with the design requirements. The significance of this work is that it revolutionizes the traditional farming methods by introducing precision farming and smart greenhouse management. With the development of science ,technology and intelligence, in recent years, more and more intelligent systems have been applied to daily life. Agricultural intelligence is an emerging industry, how to effectively improve the cultivation efficiency of crops, efficient use of crop growth period, greenhouse environment detection and regulation has become the most critical part of the market there are many intelligent machines that can help improve the efficiency of plant cultivation. However, most of the key elements such as environmental monitoring, automatic processing, timely alarms and remote control are not perfectly combined together, and can only achieve simple functions. Therefore, it is particularly important to design an intelligent agricultural system that combines single-chip microcomputer technology, sensor technology and remote-control technology. The system can realize the collection and real-time display of crop growth environmental parameters. It can be combined with timely alarm and automatic processing, it can also carry out staff information security management and remote adjustment, which to a large extent liberates the manual, realizes data monitoring, and improves production efficiency. This paper mainly combines STM32 microcontroller, DHT11 temperature and humidity detection module, rainfall detection module, CO2 concentration module, soil humidity module, oled display screen and wifi module to form the whole system, which can display and send the collected data to the mobile phone app for viewing relevant environmental parameters and controlling related peripherals. The embedded system uses the STM32 chip to collect the data of the environment in the agricultural greenhouse with the help of various sensors. By comparing with the preset threshold of the system, it can complete automatic adjustment and remote-control adjustment functions. At the same time, the WIFI module can realize the function of data cloud. Remote control is mainly carried out in the upper computer app design through Wifi, the upper computer and the lower computer are connected to complete information interaction, real-time monitoring data and video screen in the greenhouse can be displayed in the mobile app terminal, which is convenient for farmers to view the relevant environmental parameters in the farm and remote-control related peripherals in time. In the process of design, on the one hand, the circuit diagram design and software program design are completed for the subject; on the other hand, the simulation model is made through the construction, and the object function is further debugged through the program debugging, functional testing and other steps, and the target function is repeatedly debugged, and the subject goal is finally realized.

The main research content of this paper is based on motor theory, power electronics technology, motor control theory, brushless DC motor research object, based on magnetic field oriented control technology to do further design and research of motor control algorithm.The work completed in this paper is summarized in the following parts: (1) By referring to a large number of Chinese and English references, this paper summarizes the cutting-edge technology and research hotspots of the motor control system, conducts a comprehensive research and analysis on the origin of the motor, the development history of the controller and the general situation of the motor control strategy. In order to achieve the servo control of the motor, the FOC algorithm is used in this project to output sine wave to drive the motor control system. The angle position information of motor is obtained by encoder. (2) The core components of the brushless DC motor are introduced, the rotor and stator structure and material selection characteristics of the motor body are introduced, the rotor position detection and the working principle of the electronic commutation circuit are introduced, and the corresponding connecting circuit is built according to the working principle of BLDCM. (3) The principle and control flow of FOC algorithm are introduced. The coordinate transformation and the analysis and judgment of the sector are explained in detail. In particular, park transform and Clark transform are analyzed in detail, including the related formulas and calculation process are described. (4) In terms of software design, the features and advantages of the STM32CubeMX integrated development environment are first introduced. The hardware abstraction layer of CubeMX can conveniently select the required pins and automatically generate the initialization program, which greatly simplifies our subsequent code writing work. In addition, the program structure diagram is designed and the realization process is described in detail. At the same time, the main algorithm program function and some relatively important program code in the main function are given, including: duty cycle calculation, electrical angle calibration, speed measurement, etc. The pulse width modulation principle of motor control is introduced, and the program code of PWM pulse output is given. Finally, the realization of lower closed-loop control and PID algorithm is introduced, and then all the designed hardware circuits and hardware devices are connected according to the sequence of signals, and the design and welding of the entire BLDCM hardware circuit is completed, including: The circuit module of IGCMF15F60GA minimum system, the circuit module of power supply circuit and the encoder module of real-time detection of rotor position information play the role of core control. After the experimental platform was built, the KEIL MDK V5 debugger was used to download the compiled program code to the control chip through STLINK V2, and finally the motor was tested to be able to follow up control according to the plan. The test results show that it is feasible to use the FOC algorithm as the brushless DC motor servo control scheme, the control precision of the brushless DC motor is improved, and the motor fluctuation and noise intensity are suppressed to a certain extent. The brushless DC motor designed and realized in this paper is driven by FOC algorithm to achieve a servo control system, which is completed smoothly. Under the control of the algorithm, the output of the brushless DC motor can be started smoothly, and the speed of the motor can be adjusted without pole or the angle of the electric machine can be locked while the servo control is completed, and the automatic reset can be achieved when external forces are applied. In addition, in the experiment, the noise generated by the motor is very small, and the one-to-many control makes it more suitable for a wide range of applications in factory production and residential life. The brushless direct current motor has the good speed regulation performance of the DC motor, and can achieve uniform and smooth stepless speed regulation under heavy load. At the same time, compared with the traditional brush DC motor, there is no motor contact caused by brush friction commutation, no power generation and other problems. Therefore, it has been widely concerned by industry workers and scholars. Under normal circumstances, the follow-up control for brushless DC motor is generally square wave control. However, when the square wave control is started, the fluctuation is large and the noise is large, which seriously affects the industrial application environment, which causes a great challenge to the practical application of BLDCM. On the other hand, FOC adopts sine wave control mode, which can start smoothly and mute the whole process, and fundamentally solve the noise problem caused by square wave control. Moreover, this control mode divides the stator current of the motor into excitation current and torque current according to the set relationship. The accuracy and rapidity of speed control are greatly improved. Based on the above situation, the study of brushless DC motor FOC algorithm servo control has important theoretical value and practical significance. This thesis focuses on the follow-up control system of DC brushless motor based on FOC algorithm, and the specific research content is summarized as follows: Firstly, the thesis introduces the research background and significance of brushless DC motor and its drive control system, focuses on their research status and development trend, and introduces the structure and working principle of brushless DC motor. Then, this thesis introduces the basic control principle of FOC algorithm, expounds the corresponding control theory and coordinate transformation theory in detail, and gives the sector calculation process. The servo control system of BLDCM is designed based on FOC algorithm, and the specific control flow is given. Finally, this thesis uses IGCMF15F60GA as the main controller chip of the control system to build the corresponding hardware experiment platform, and carries out joint debugging and testing of the BLDCM system based on FOC algorithm. The experimental results show that the brushless DC motor using the FOC control algorithm can achieve stable operation, and at the same time, it can also achieve the stepless adjustment of the motor speed or the lock of the electromechanical angle while completing the follow-up control function, and can realize the automatic reset under the application of external force. The noise generated by the motor in the experiment is very small.

Automatic obstacle avoidance technology is widely used in the development of intelligent vehicles and is indispensable for the implementation of autonomous driving. Intelligent vehicles with obstacle avoidance capabilities are used in many industries and play an important role, not only for civilians but also for military purposes. Research on automatic obstacle avoidance can provide relevant control strategies for various intelligent vehicles as well as reference hardware devices. This project is based on the open source Arduino platform, I designed a three- wheeled smart car equipped with ultrasonic sensors, infrared sensors, a Bluetooth module and other modules. By using multiple sensors, acquiring information from the driving environment of the intelligent vehicle, and PID control strategies, the study of automatic obstacle avoidance technology is carried out. The intelligent vehicle in this project uses two infrared sensors at the front and ultrasonic sensors on both sides to detect the distribution of obstacles around it, obtain information such as distance, and apply corresponding obstacle avoidance strategies. On the other hand, to ensure that the smart car can move in a straight line forward and backward without deviation due to the instability of the motor drive module, a PID closed-loop control is added to ensure that the speeds of the left and right motor are the same. After testing, the smart car is stable, the hardware circuit is responsive, and can correctly realize obstacle avoidance and drive in a straight line, meeting the automatic obstacle avoidance requirements of the smart car. This project solves the problems of low efficiency and inability to collect real-time multi- directional information that exists in the current automatic obstacle avoidance system for the intelligent car.

The aim of this project is to create a temperature control lab for teaching applications. The project consists of the design of a printed circuit board (PCB) that, together with an Arduino MEGA 2560 Rev3, conform a temperature plant that allows the user to control the temperature of each BJT transistor. In order to control the temperature of these, different control algorithms will be designed with the help of numerical programs such as Matlab and Arduino. The temperature plant is composed of the PCB, the Arduino MEGA 2560 Rev3, two BJT transistors, two TMP36 sensors, one LED, an OLED display, a 4x4 keypad and a fan. In the project, the utility of this components is explained. Eventually, simulations and analysis of the PCB will be performed and comparisons between different controllers will be made.

This project shows the modeling, design and implementation of a GPC (Generalized Predective Control), MIMO (Multiple Input Multiple Output) controller, on a study board called TCLAB, which is made up of an Arduino Leonardo, an LED, two transistors that act as heat sources and two temperature sensors. Starting from the non-linear modeling of the system, its linearization is carried out, following mathematical procedures. With these, the respective transfer functions are obtained that will be used for the design of the GPC predictive controller. Taking into account that it is a multiple input and multiple output system, a predictive controller is designed, which takes into account the physical interaction between the inputs and outputs of the system. This linear model is used for the design of the GPC controller, which is simulated and then implemented on the TCLAB plant. Based on the same linear equations of the system, a PI (Proportional Integral) controller is regulated, with which the respective simulation is made, followed by its implementation on the plant. Finally, tests are carried out to make a comparison between these two controllers and it is found that the GPC controller is more robust and accurate than the PI controller, for different temperature ranges.

This project deals with the design and assembly of a model for the development and implementation of different controls with an educational and / or research nature. The functionality of the model is the levitation of an expanded polystyrene sphere inside a methacrylate tube through the action of one or two fans, located at the ends of the tube, to later be able to carry out position and speed controls with it. This project is situated in the current context, given the large number of educational models for the design of controls that are currently on the market, and the few for levitation by air flow. The goal is to offer a comprehensive and versatile learning and testing platform. Thus, it begins by designing and assembling all the Hardware related to the circuitry and all the elements necessary for the operation of the model. Subsequently, the 3D pieces that are the container structure of all the electronic components, the chassis of the model, are designed and printed. Once all the assembly is finished and the model is operational, an empirical model is obtained on which different controls are designed, using only one fan or using both. These controls are programmed on a microcontroller and applied to the model to define the robustness and stability of the educational platform designed and assembled. Next, the results obtained from the simulations are compared with those obtained on the already assembled model, and it is concluded which controls are the ones that give the best results, for the system with one or two fans. Finally, different conclusions are drawn about the project and future lines of work are proposed to continue with the improvement and enhance its use.

Even though new systems like 3D printing start to get more viable, casting remains an important manufacturing technology. During a casting process, some segments at different spots of the material at the microscopic level are solidifying. Those are called grains, and they greatly influence the mechanical properties of the material. To increase the material properties the material science part (CDAL) of the UPC started with the development of a viscosity analysis prototype. This prototype analysis the viscosity, but the measured data is not accurate enough and the casting process doesn´t stop automatically. For this reason, the results from the previous group were analyzed and the existing inputs and outputs were described. With this information the main tasks were started, that include designing a case for the prototype, developing a viscosity model, doing research on viscosity and developing a software solution for the existing prototype. The results of those tasks lead to a safety increase through the case, an accurate viscosity analysis and an automated closed loop system to control the viscosity analysis. This system is now able to read the torque and speed values via Modbus, calculate the viscosity for various materials, include safety features, provide an interface and to enable customization through a maintenance window. Only the problems with the Festo motor control couldn`t be solved, due to the high complexity and communication problems between the devices. Further development could be to integrate the two motors into the program. To simplify that, replacing the motor and its control system is recommended.