Brushless direct current motor driver design and adaptive control for electric vehicles

Today, interest in electric vehicles is increasing. Issues such as the drive system, engine, driver and vehicle control system to be used in electric vehicles have started to gain importance. As a result of the change of some parameters related to the vehicle with the environmental conditions, the adaptive control of the vehicle according to these changing conditions gains importance in terms of vehicle performance and energy saving. In this thesis, it is aimed to design a original new type Digital Signal Processor (DSP) based brushless direct current (BLDC) motor drive system for electric vehicles and adaptive control of this system. As a first step in the study, a passenger car was determined and its parameters were examined. The vehicle, system model has been realized, and as a result of the modeling and calculations, it has been determined that a 75kW drive power is needed. Then, a original 75kW motor drive system was designed and its practical application was realized. The original drive system design is provided by using components and DSP with security and safety features in accordance with automotive standards. All circuit elements used in control, driver and power stage have been selected in accordance with automotive standards. Circuit designs were carried out in accordance with automotive and safety standards. The control algorithm developed to generate the required PWM (Pulse Width Modulation) signals for the switching elements of the three-phase six-switch fully-controlled bridge converter was compiled and embedded by the Code Composer Studio program. This software has been successfully implemented using Texas Instrument's TMS320F28069 DSP processor to run the motor driver algorithm. In order to analyze the performance of the proposed driver, a simulation model of the driver system was created in the Matlab/Simulink environment. The efficiency of the originally designed propulsion system has been demonstrated by simulation and experimental results. In the adaptive control process, the most appropriate driving has been tried to be obtained by using data such as vehicle distance in front, road slope and vehicle weight. For this, sensors suitable for adaptation parameters were selected and used in the design of the adaptive control unit. In the application for the adaptive control system, Rasberian Linux-based operating system is installed on the Raspberry PI-III and the driver programs of the sensors are run on Linux and the sensors work. HC-SR04 Ultrasonic distance sensor is used to measure the distance between the electric vehicle and the vehicle in front of it. MPU 6050 gyro-accelometer sensor is used to measure the slope of the road where the vehicle is located. To measure the weight of the vehicle, the BMP180 pressure sensor positioned inside the vehicle tire and the Load Cell sensor for this process were also used. The data of these three parameters are processed with Raspberry PI III and used for adaptive control of the system. In practice, this upper adaptive control unit works in the background and constantly communicates with the TMS320F28069 DSP and motor driver circuit, and the data is sent and processed in real time. In the adaptive study, the FDAM drive parameters are optimized according to the value ranges of the adaptive controller data. As a result of the optimization process, first of all, the electric vehicle not only gives a warning if its distance from the vehicle in front of it falls below 3 meters, but also automatically reduces the PWM operating rate (duty cycle value), which changes the speed of the engine adaptively according to the speed of the vehicle, and prevents possible collisions. Secondly, the MPU 6050 Gyro – Accelerometer sensor measures the slope of the road where the vehicle is located and the acceleration of the vehicle. According to the optimized value ranges, the power loss of the vehicle on the ramp is prevented by adaptive control. For this, the PWM operating rate value, which changes the speed of the motor, is automatically increased. Thus, possible slowdown is prevented. Thirdly, the weight of the vehicle is measured with the BMP180 pressure sensor located inside the vehicle tire. The power required to be produced by the electric motor for the curb weight of the vehicle increases with increasing load. Although reference values are increased from the accelerator pedal for this, more effective operation is ensured by shifting the operating region of the engine with optimized parameter ranges. A load cell was also used for this process and was optimized by comparing it with the value read by the pressure sensor. As the weight of the vehicle increases, the PWM value of the engine is increased linearly according to the value of the unladen weight, thus increasing the driving comfort of the electric vehicle and providing an adaptive control. With the real-time continuous monitoring of these three parameters, adaptive control of the electric vehicle has been realized in a very efficient way. In the thesis, a systematic approach is presented for the automatic adaptive tuning of these parameters. Simulation and experimental results have shown that the approach gives successful results. In addition, with the cooling application carried out in the study, sudden heating problems of the engine for electric vehicles were prevented.