This article will quickly review some key requirements for efficient single-chip-based closed-loop control systems. Common techniques for increasing efficiency will use some examples to explain MCUS. Once you have learned some common technologies that improve efficiency, you can better find the implementation of the best shutdown, the next control cycle design.
Control ring foundation
The control loop is a key element that controls the dynamic system. A dynamic system can be a relationship between any mechanical or electrical system (usually modeling as a linear relationship between input and output). Outputs are often controlled in such a manner to maintain within the desired operation "frequency band". For example, the automatic cruise control of the car is such a system. In this system, the speed of the car is set in an expected level, even if the car encounters the hillside, the controller can keep the speed of the speed. The algorithm of the control speed uses a control ring, which applies an input (pressure on the throttle), measurement results (speed) and adjusting the input as needed to maintain the required level. A block diagram of a simple single input single output control system with a control loop, as shown in Figure 1.
Figure 1: A simple dynamic control system control loop diagram.
In the above block diagram, the input of the dynamic system generates an output. The output is measured by the feedback sensor and the measured output is compared to the reference (desired) input. If there is a difference, the system controller uses the generated error to modify the system input, so that the system output is closer to the reference input. The system controller needs enough intelligence to avoid oscillations and other issues caused by incorrect management control systems. Suppose the dynamic system is linear (output and input proportion) is not as limited as you think, because many mechanical and electrical systems are run in linear way, or it is easy to "bias", in more complex transfer functions The linear area works.
Control loop with single chip microcomputer
This is very easy to see why the MCU implements the main force of the control system. With awareness, capacity-calculation, control of various inputs and outputs, all in very high performance levels (especially in high interest rates like automotive mechanical systems) is a natural control unit MCU. In addition to the capacity of the CPU, there are many ways to make your closed loop control simple and effective.
The control loop is usually a timed relationship between each adjustment of the dynamic system. This "cycle time" determines the speed of adjustment. It is difficult to effectively control the system output if the cycle time is too long and the system dynamics (the time-transition characteristics of the control system). The misfortune of oscillation and out-of control will accumulate, so that the system is in a risk of failure, perhaps a very dramatic nature. In general, the MCU can close the loop faster (processing output sensor, determine any reference errors, and adjust the system input), better.
Effective timing and count Therefore, the best shutdown control loop required for key features and advanced peripherals of the MCU can implement the best time for control loops. For example, the Silicon Labs EFM32LG360F64G-E-CSP81 single-chip timing / counter surrounding a variety of features of control loop implementation. In addition to the main circuit model of the closed loop controller, it also requires timing and counting functions. Let's take a more detailed step diagram (Figure 2) of the EFM32LG counter / timer peripheral to see how it can help control the circuit system as common functions such as the implementation shown in Figure 1.
Design solution for closed-loop control system based on single chip microcomputer
Be
Figure 2: Timer / counter block diagram of the Silicon Labs EFM32LG microcontroller.
The timer / counter, a useful function is the ability to convert from the external source to the number of TIMN_CCN pins of the graph on the left. These inputs are helpful when observing the sensor generated when the dynamic system is measured. For example, the rotation measurement based on position measurements is often converted during each rotation. It is also noted that there is a quadrature decoder block in the middle of the map, which can be used for similar measurements. The number of records is counted when the store is reached in the Timern_ToP terminal value to trigger the interrupt, and then proactive immediately or can be stored.
The output of the timer / counter, the right TIMN_CCN pin in the figure, can use the pulse width modulation (PWM) to control the dynamic system input common scheme. In these systems, the time of signal activity is related to the voltage or current level required for control. The accurate control of the signal cycle, signal high time, and edge conversion points is critical, and can be effectively controlled in the PWM characteristics of the timer / counter. Three separate PWM outputs are also convenient for normal motor control applications, with three separate windings for changing magnetic fields associated with rotating motors.
Intelligent peripheral control
Fast and efficient loop feedback time typically requires unloading from high-power CPUs and program memory blocks using intelligent peripherals. If the peripheral device can operate independently of the CPU, this allows the CPU to perform other more complex processing tasks, and can even wait in a low power consumption until it needs to be processed. Some advanced MCUs have a special peripheral control system that can be used to connect the peripherals without intervention from the CPU. For example, a single-chip R5F52108CDFM, RX210 group of Rissa MCU, an event link controller (ELC) connection and control peripheral output as an autonomous peripheral input. A block diagram in the ELC is shown in Figure 3.
Be
Figure 3: Rissa MCU RX210 Group Event Link Controller Block Diagram.
The internal peripheral bus on the left is used to connect the peripheral device, as shown on the right side. All peripherals, including the DMA controller, Data Transmission Controller (DTC), and Interrupt Controller (ICU) can be connected to a dedicated control, autonomously based on interrupt, timer comparison result or PIN conversion activation peripheral. Up to 59 types of event signals can be connected to peripherals to initiate conversion, start timer, and start DMA or DTC transmission or any other desired peripheral devices. When the event that has been set as the trigger occurs, the operation set of the selected module will be started.
Multiple operational chains can be started to complete complex operations without CPU intervention. For example, the timer can initiate an analog-to-digital conversion and conversion value stored in the memory. The counter can track the number of conversions, and the CPU can be interrupted when the count indicates that the full data set can be processed. During the CPU processing, the clock oscillator can automatically switch to a faster mode. With ELC to play the ultimate, many of the common control circuits in sensing functions can make it easy and effectively to minimize the number of rapid cycles.
Efficient calculation
As we see, using intelligent and autonomous counters / timers and peripherals can improve loop time and reduce power consumption - this is two important aspects in control system design. Typically, each primary control system block requires calculation to process data required to detect, compare, control, and operation of control system. In fact, with the efficiency, accuracy and long service life, calculation demand has increased dramatically, has become an important system requirement. The advanced algorithm for implementing the control loop is now using a proportional integral differential algorithm, and the floating point operation usually needs to increase accuracy. If advanced calculations are not supported in hardware, it is very difficult to control loop closing at the desired frequency.
Manufacturers understand the needs of advanced processing capabilities, including numerical processing capabilities, even at low-end MCUs can accelerate complicated closed-loop control design. High-end machines typically include the most accurate control application of dedicated hardware accelerated floating point calculations. Freescale Kinetis K60 Single Chip MK61FN1M0VMD15 MK61FN1M0VM class Use 32-bit ARM Cortex-M processor and DSP instructions and single-precision floating point arithmetic units to be the most complex control algorithm, require advanced calculations. The DSP instruction includes extended single-cycle multi-cumulative (MAC) instructions for fast processing of high-precision signals, and single instruction multiple data (SIMD) instructions to process low resolution signals faster. The hardware split block only runs 2 to 12 cycles and speeds up the normal zoom operation.
To achieve higher performance, dual-core CPUs can be used to parallelize tasks. For example, a single-chip microcomputer of a Texas Musical Instrument, such as F28M35H52, both ARM Cortex-M3 32-bit CPU and 32-bit CPU floating-point capability TMS320C28X processor in Texas Wen. Figure 4 below shows a block diagram of this dual-core MCU.
Be
Figure 4: Texas Instrument F28M35X Concerto MCU block diagram.
Based on ARM's MCU, at the upper part of the figure, it can be used to manage peripheral devices, while the concerto CPU can be used to process data and manage control. Note that the PWM timer is closely linked to the resolved symbol system, so it is easy to generate the complex waveform required by the control system input. This type of dual-core CPU is important when the application has an algorithm that is easy to separate. If you need more processing capabilities for a single algorithm or high reliability, uniform dual CPU, a uniform dual CPU, such as Texas, Cortex-R4, Dictragon, RM4 ARM MCU It may be a better choice. The closed-loop control system for high reliability applications can use the built-in redundancy implemented by heterogeneous dual CPU to improve efficiency and robustness.
in conclusion
In MCU-based design, efficiently shut down the control loop does not have to fill a way to try and erroneously look for best implementations. More systematic methods, using modern advanced features, MCU can help you create more efficient, faster, lower power consumption, more efficient control system solutions.
Be
Reprinted from -Wiku Electronic Market Network
Our other product:
