Experimental Study On PID Control Algorithm Free Essay Example

Results and Experimental Process

Chapter 1 of this thesis has covered the Introduction of the proportional valve controlled hydraulic cylinder system and established its background over the years until the present day. Chapter 2 explained the mathematical model of the hydraulic proportional valve position control of the entire system, and analysis on the simulation of the proportional valve controlled hydraulic cylinder system, it established the factors that happen to affect the control of the system, further explanation of dead-time composition and algorithm of PID were established as to know which dead one to use for particular system.

We can write
your paper for you

100% original

24/7 service

50+ subjects

Order Now

Chapter 3 design of the hardware and construction of the experimental platform. Chapter 4 establishes the software’s used to accomplish the requirements of the thesis, and the link with micro-controller and the interface. In this chapter, the hardware’s, software’s, theoretical of proportional valve controlled hydraulic cylinder, and dead zone characteristic are applied to the hydraulic proportional position control system and analyze the results according to the command signal (mainly step signal).

Experimental study on PID control algorithm

PID is an easy algorithm to implement, it’s a simple algorithm and has strong adaptability. The PID control algorithm has been used in many occasions where the need for accuracy, speed, the fast response needed, but is some cases it is difficult for a PID control algorithm to find a better solution between stability and accuracy. In essence, if the system needs an improve in accuracy and reduce the errors, it automatically needs to increase the control mode, and in doing so, the enhancement might cause a reduction in stability.

If the focus changes to ensuring the system have the stability, the system accuracy becomes hard to control and the desired effect won’t be met. But if the PID algorithm is mixed with other algorithms like dead-time compensation, the system stands a chance of having fast response, accuracy, and stability.

• ADCVoltage=Get_Adc_Average(ADC_Channel_1,10);
• ReferenceDistance=setpoint;
• error=ReferenceDistance*4095/250-ADCVoltage;
• sum_error+=error;
• derror=error-last_error;
• last_error=error;
• KP=kp;
• KI=ki;
• KD=kd;
• PID_OUT=((KP*error*0.01+KI*sum_error*0.000001+KD*derror*0.1)+2047.5);
• ADCVoltage=Get_Adc_Average(ADC_Channel_1,10);
• ReferenceDistance=setpoint;
• error=ReferenceDistance*4095/250-ADCVoltage;
• sum_error+=error;
• derror=error-last_error;
• last_error=error;
• KP=kp;
• KI=ki;
• KD=kd;
• PID_OUT=((KP*error*0.01+KI*sum_error*0.000001+KD*derror*0.1)+2047.5);

Comparing the experimental results

According to the experimental rule, the initial error value of the Hydraulic cylinder of the proportional position controlled should be taken into account for the reciprocating motion of the process, because it will have an influence on the forward and backward speed of the system. But during the experiment, the PID control parameters should be set reasonably so the PID control effect can act on the system and can give the results more accuracy. The algorithm used in this experiment is PID algorithm for a closed-loop system, and is optimized to control and track error in the system.

System with no-load experiment

The analysis of figure 5.2 indicates that the system is stable, and has a fast response with an overshoot of 4.62%, the rise time of 23ms and steady-state error of 0.13 (nearly to zero). Figure 5.3 analysis indicates that the system is partially stable, and has a fast response with an overshoot of 9.29%, the rise time of 33ms and steady-state error of 0.13 (nearly to zero). The system results had a feedback from the proportional valve and gave out results with an overshoot of 4.51%, the rise time of 28ms and steady-state error of 0.13.

System with load experiment

The analysis of the results with load are given in the following figures. When the load of 10kg was added to the system experiment, the results show a great stability throughout the experiment. The above figure (5.5) had an overshoot of 10.02% (it is within the overshoot requirements of 10%), steady-state error of 0.13 and the rise time of 35ms. With great stability, the system had a fast response and accuracy. The system established that adding a load to the system has a great influence on the system because the system has good stability, accuracy and fast response. Here the steady-state was 0.13, the rise time of 20ms and the overshoot of 3.86%. figure 5.7 20 kg load results of P, I and D

Increasing load cause the PID parameters to change extremely, as the load increase the Ki and Kd are cause to increase significantly. This had an improvement in the accuracy rapidly fast, and stability not so much. But after tuning the PID, the expected results we acquired. The overshoot was 9.87%, the rise time was 37ms and the Steady-state error was 0.13.

Problems during experiment

During the experiment, there were some problems meet with the signal conditioning. The signal condition that was designed had a voltage drop at the input of D/A, when you input 3300 (3.3V) at it, on the first output of LM324N was dropped to 2.19V and that was a problem in getting 10V output for D/A. The A/D part of the signal condition was also placed wrong at the input which resulted as a problem in getting the right signal from the feedback signal. The voltage regulator pin also had pin broken inside it and it caused me to use the supply voltage of 15V and had to solder the wire to the signal conditioning.

Later, during the experiment, the MCU (STM32) chip overheated and cause the chip to not work at all. The chip overheated and it was late to buy a new one, as it would take 1-7 days of delivery to Beijing delivery services.

Solutions for the problems encountered

As a result of the signal conditioner having problems, the signal conditioner was fixed but there was no time to print it again. The experiment was done using the other senior student signal conditioner and MCU (STM32). But beside the signal conditioner, the code and interface screen used at the experiment were not senior’s student during the experiment. The experiment was done it time and results were found for the final thesis analysis.

Summary of this chapter

From the results, one can say that the steady-state error and overshoot of the system will increase with the increase of the load. Why the increase of overshoot? Because the direction of the load force is directly proportional to that of the hydraulic cylinder in the forward motion of the system. For future purposes, the system needs further research on how to improve the influence of the load on the proportional valve controlled hydraulic cylinder system.

This chapter concludes the by introducing the algorithms to the hydraulic proportional valve system experimentally so based on PC humane interface control and the experimental platform of proportional valve controlled hydraulic cylinder system, and also the dead-zone was introduced to compensate the linearity. The PID algorithm and the dead-time compensation made the system track stability and accuracy, and with the Steady-state error of less than 1m, hence improving the system.

Chapter 6 Summary6.1Conclusion and Main workIn this thesis, the main work of the experimental platform of the proportional valve controlled hydraulic system is built to realize the control effect on the proportional control position of the system. During the research, the discovery of mathematical model of proportional valve controlled hydraulic cylinder is discovered to explain the system mathematically so, and the aspect like non-linearity, dead zone, and the variable load was being analyzed and studied for further on how to eliminate them during the experiment. According to the distinctiveness of the experimental platform, the displacement tracking position and the dead zone are allocated together to get accurate control effect on the system. The main work done in this thesis is summarized below:

The hardware designs, mathematical models and working principle of the proportional valve controlled hydraulic cylinder system are demonstrated in a structural understanding of the system. The establishment of the transfer function of the system is designed according to the data analyzed from the mathematical models of the proportional valve controlled hydraulic cylinder system and simulation of the system was created.

The elements that had an influence on the system were analyzed and studied on how to detect them so they don’t have a huge impact on the proportional valve controlled hydraulic cylinder system. After analyzing the elements, the PID algorithm and dead-time compensation were designed for the system to have better control results. Use of Matlab Simulink created a link of comparison of the simulation and real-time experiment, and to further study the characteristics that influence the system. According to the proportional valve controlled hydraulic cylinder system experimental platform, the characteristics of the dead-time compensation was studied and analyzed the input of different signals, the motion of the system and the load changes how they affect the final results.

The software’s construction was studied and the PC control system was used as an interface screen by using LabVIEW software and C language from the code to create a link of communication, and processing of data. The written code of the PID algorithm and dead-time compensation were applied on the interface screen to display the real-time results on the PC. The signal conditioner was design by Altium Designer software for the requirements of this thesis and tested on until the choose parameters of the signal condition weren’t accurate enough to deliver the right signals to the experimental platform.

Future work

The focus will be on sine tracking and introduce other algorithm to improve the stability of the system. Hopefully, the system will be introduced to fuzzy logic algorithm with dead-time compensation, to get accuracy with a good sine tracking. The focus must also be on improving the steady-state error and introduce the dead-zone on the Matlab Simulink. In masters thesis this will be the focus.

AcknowledgementI owe my gratitude to God and all the people who made it possible for me to study in china and get a good quality degree from one of the biggest country in the world.

Secondly, I want to thank my advisor, Professor Li, for taking me as one of her students, and also for giving me an opportunity to prove to myself I can finish lot of work load under stress and always making herself available for advising me whenever I am at the dead end and believing in me that I can accomplish the task that was at hand. It has been a pleasure to work with and learn from such an extraordinary individual.

Thirdly, I also want to thank my senior friend, Mao Wei. Without his extraordinary theoretical ideas about the proportional valve controlled hydraulic cylinder system and computational expertise on particular software, this thesis would have been a distant dream if it wasn’t of his help and advices. My colleagues at the systems engineering laboratory (room 115) have enriched my graduate life in many ways.

I would like to thank my family for their constant support and unwavering confidence in me, and have pulled me through against impossible odds at times. Words cannot express how grateful I am for their love and support, I owe them my gratitude. It is impossible to remember all, and I apologize to those I’ve inadvertently left out.

Lastly, thank you all and thank God!

My Future PlansMy plans for the future definitely include becoming a Professional Engineer. This means I plan to get licensed with Engineering Council of South Africa (ECSA). To become licensed, I must first get evaluated on whether my qualification quality is good or are there any things to do like internship to fulfill the gap. I then will have to work under a Professional Engineer for at least 4 years. After that I have to take two exams and pass the test on knowledge an engineer should know.  If all goes well according to the evaluation, I will then be eligible for my individual professional license.

As soon as I apply and get my license then I will be a registered licensed professional engineer. It is important to become a registered licensed professional engineer because of four main reasons. One reason is that a client will see that I have the credentials of an expert and that I can be trusted to do the work. The second reason is that employers will see that I can take on a higher level of responsibility and that I am a valuable asset to them. The third reason is that by being licensed co-workers will trust and value my opinions and will look up to me. The fourth reason is that I will be happy with myself because the license shows my hard work ethics and my desire to succeed.Licensed engineers are the reason the world has been able to make the huge leaps forward in technology that we as humans have made. It is the trust that people have in the engineers that brings the world forward. To help continue to bring the world even farther forward than we already have it is important to become licensed.

With this license people will trust the engineer and the engineer can invent new revolutionary things that in a few years will seem common because of the speed at which the world is progressing. Another future plan is to also further my studies and get my master’s degree while I work.