【Daily Question】20090311 What issues should be considered when adjusting regulator parameters on-site?

Generally, the following points should be noted: 1. The set parameter values obtained through various methods fall within a certain range, and they must be adjusted on-site according to the actual production conditions. 2. It is best to adjust the regulator parameters when the operating conditions during production are relatively stable. Apart from introducing appropriate artificial disturbances, it is preferable to modify the regulator parameters by observing the natural disturbances that occur during production over an extended period of time (it is advisable to use a fast-recording device for this purpose). 3. Artificial disturbances generally include internal disturbances, external disturbances, and prescribed value disturbances. Given value perturbations have a significant impact on the production process; it is essential to control their magnitude. Attention should also be paid to the extent of these perturbations, whether they are internal or external. After applying a disturbance, it is necessary to wait for a while to observe the changes in the parameter being adjusted; do not rush to apply a second disturbance until the situation is clear. 4. For PID regulators, the mutual interference between parameters must be taken into account. The regulator parameter values obtained through the tuning method are not the actual scale values of the regulator parameters; they must be corrected using the inter-regulator interference coefficients to become the actual scale values. 5. When setting the parameters, the nonlinear factors of the controlled object and the control mechanism must be taken into account. If the nonlinearity is not severe, the tuning parameters should be set more conservatively. 6. When adjusting, the operating conditions of the production process must be taken into account. Generally speaking, the adaptation range of regulator parameters is for the normal operating conditions. When the operating conditions change significantly, the regulator parameters become inadequate. It is not due to poorly tuned regulator parameters, but rather to a limited range of parameter adaptation; to address this issue, adaptive control must be employed. By the way, the PID parameter tuning method for DCS involves determining the regulator’s proportional band PB, integral time Ti, and derivative time Td. It can generally be determined through theoretical calculations, but the error is too large. Currently, the most commonly used methods are engineering tuning methods: such as the empirical method, the attenuation curve method, the critical proportional band method, and the response curve method. The general process of various methods is as follows: (1) The empirical method, also known as the on-site trial method, involves first determining the parameter values PB and Ti of the regulator. A disturbance is introduced into the control system by changing the setpoint, and the shape of the control curve is observed and evaluated on-site. If the curve is not ideal, PB or Ti can be changed, and the control process curve can be drawn again. Repeated trials are carried out until the control system meets the requirements for dynamic performance; at that point, PB and Ti represent the optimal values. If the regulator is a PID three-way type, differential action must be added on top of the tuned PB and Ti values. Since the differential action has the ability to counteract deviations, once a Td value is determined, the set PB and Ti values can be reduced slightly and further adjustments can be made on-site until the optimal values for PB, Ti, and Td are achieved. Obviously, the parameters tuned using the empirical method are accurate. But it takes more time. To shorten the tuning time, the following points should be noted: ① Determine the initial parameter values PB, Ti, and Td based on the characteristics of the control object. The parameter values of similar control systems in actual operation can be referenced, or the values given in Table 3-4-1 can be used, so as to make the determined initial parameters as close as possible to the ideal set values. This can **reduce the number of trial adjustments on site. ②During the trial process, if it is observed that the controlled quantity changes slowly and does not reach a stable value promptly, this is caused by either an excessively high PB or an excessively long Ti. However, there is a difference between the two: when PB is too high, the curve fluctuates significantly and its changes are irregular; when Ti is too long, the curve contains oscillatory components, and it approaches the desired value very slowly. In this way, PB or Ti can be changed according to the curve shape. ③Too small a PB, too short a Ti, and too long a Td can all cause the oscillations to decay slowly or not at all; the difference is that with too small a PB, the oscillation period is shorter ； Ti is too short, resulting in a longer oscillation period ； Td is too long, resulting in the shortest oscillation period. ④If amplitude-controlled oscillations occur during the tuning process, and this phenomenon cannot be eliminated by adjusting the parameters of the regulator, it may be due to an inaccurate calibration of the valve positioner, gaps in the actuation mechanism of the control valve (or an oversized control valve), or disturbances in the controlled process that result in amplitude-controlled oscillations. At this point, it is not sufficient to focus only on tuning the regulator parameters; rather, other instruments and components also need to be checked and adjusted. (2) The attenuation curve method requires a 4:1 attenuation ratio for tuning purposes. First, the integral and derivative actions of the regulator are disabled, and the proportional band PB of the pure proportional control is tuned using trial and error (which is much simpler than adjusting two or three parameters at once), so as to meet the 4:1 attenuation requirement. The proportional band PBs and oscillation period Ts at this point are then recorded. If integral and differential effects are taken into account, calculations can be performed using the empirical formula given in Table 3-4-2. If the parameters set in this way are appropriately adjusted. For some control objects, the control process proceeds rapidly, making it difficult to determine the attenuation ratio from the recorded curve. At this point, the system reaches a stable state after just 2 fluctuations of the quantity in question; this can be approximated as an attenuation process with a ratio of 4:1, and the time taken for one fluctuation is Ts. (3) Critical proportional band method: When adjusting the parameters of the regulator using this method, it is first necessary to disable the integral and derivative actions, allowing the control system to operate under pure proportional control with a larger proportional band. Then, the PB value is gradually reduced; after each reduction, the process curve must be carefully observed until amplitude-controlled oscillations occur. At that point, the proportional band PBk (referred to as the critical proportional band) and the oscillation period Tk are recorded. The parameter values of the regulator are then determined using the empirical formulas provided in Table 3-4-3. After calculating the parameter values using this table, set the ratio band at a value slightly higher than the calculated value, set Ti and Td at the calculated value, and conduct on-site observations; if the ratio band can be reduced, then set PB at the calculated value. This method is simple and widely applied. However, it is not suitable for control systems with a very small PBk. (4) The response curve method: The first three methods for tuning regulator parameters are all carried out without prior knowledge of the characteristics of the control object. If the characteristic parameters of the controlled object are known, namely the time constant T, the time delay ξ, and the gain coefficient K, the parameters of the regulator can be calculated using empirical formulas. The tuning result using this method can meet the requirement of an attenuation rate of φ=0.75. This post was last edited by wopale3 on 2009-3-11 11:37]

When tuning PID parameters, it would be ideal to have a theoretical method for determining them, but in practical applications, the PID parameters are more often determined through trial and error. Increasing the proportional coefficient P generally speeds up the system’s response and helps to reduce the steady-state error when such an error exists. However, an excessively large proportional coefficient leads to significant overshoot and oscillations, thereby deteriorating the stability of the system. Increasing the integration time I helps to reduce overshoot and oscillations, thereby improving the stability of the system, but it prolongs the time required for the system to reach steady state. Increasing the differential time D helps to speed up the system’s response time, reduces overshoot, and improves stability; however, it weakens the system’s ability to suppress disturbances. During trial adjustments, one can refer to the impact trends of the above parameters on the system control process, and follow a tuning procedure that first uses proportional control, then integral control, and finally derivative control. First, tune the proportional part. Increase the proportional parameter from small to large and observe the corresponding system response until a response curve with fast response and low overshoot is obtained. If the system has no steady-state error, or the steady-state error is already within an acceptable range and the response curve is satisfactory, then only a proportional controller is needed. If the static error of the system, based on proportional control, does not meet the design requirements, an integral term must be added. During tuning, start by setting the integration time to a relatively large value; then slightly reduce the already adjusted proportional coefficient (usually to 0.8 of its original value), and subsequently decrease the integration time so that the steady-state error is eliminated while the system maintains good dynamic performance. During this process, the proportional coefficient and integral time can be adjusted repeatedly based on the quality of the system’s response curve, in order to achieve a satisfactory control process and optimal tuning parameters. If satisfactory results cannot be achieved by repeatedly adjusting the dynamic behavior of the system during the aforementioned adjustments, a differential element can be added. First, set the differentiation time D to 0. On this basis, gradually increase the differentiation time while simultaneously adjusting the proportionality coefficient and integration time, and trial and error until a satisfactory tuning effect is achieved.

Although there are theoretical methods for calculating PID parameters, due to the many factors that affect closed-loop regulation, it is not possible to describe them all precisely mathematically; as a result, the values calculated often have little practical significance. Therefore, aside from the parameters obtained through actual debugging, there are no available empirical parameter values. Even for two systems that appear to be identical, completely different parameter values can be obtained through actual debugging. One can try to use some relatively conservative PID parameters; for example, the gain should not be too high, and the integration time should not be too short, in order to avoid oscillations. On this basis, it can be put into operation directly to observe the waveform changes resulting from feedback. Giving a step input and observing the system’s response is the best way. If, after the feedback reaches a given value, it takes multiple oscillations to stabilize or remains unstable altogether, it should be considered whether the gain is too high or the integration time is too short ； If the feedback cannot be provided in a timely manner and the rise rate is very slow, it is necessary to consider whether the gain is too low or the integration time is too long... In short, tuning PID parameters is a comprehensive process with interrelated elements, and making multiple attempts during the actual tuning process is a very important and essential step.

Tuning the parameters of a regulator is a very delicate task; it is necessary to understand the impact of these parameters on the production process, while also continuously monitoring how the production process is performing, in order to ensure that production is not disrupted despite the need to adjust the regulator parameters properly. Generally, the following points should be noted: 1. The set parameter values obtained through various methods fall within a certain range, and they must be adjusted on-site according to the actual production conditions. 2. It is best to adjust the regulator parameters when the operating conditions during production are relatively stable. Apart from introducing appropriate artificial disturbances, it is preferable to modify the regulator parameters by observing the natural disturbances that occur during production over an extended period of time (it is advisable to use a fast-recording device for this purpose). 3. Artificial disturbances generally include internal disturbances, external disturbances, and prescribed value disturbances. Given value perturbations have a significant impact on the production process; it is essential to control their magnitude. Attention should also be paid to the extent of these perturbations, whether they are internal or external. After applying a disturbance, it is necessary to wait for a while to observe the changes in the parameter being adjusted; do not rush to apply a second disturbance until the situation is clear. 4. For PID regulators, the mutual interference between parameters must be taken into account. The regulator parameter values obtained through the tuning method are not the actual scale values of the regulator parameters; they must be corrected using the inter-regulator interference coefficients to become the actual scale values. 5. When setting the parameters, the nonlinear factors of the controlled object and the control mechanism must be taken into account. If the nonlinearity is not severe, the tuning parameters should be set more conservatively. 6. When adjusting, the operating conditions of the production process must be taken into account. Generally speaking, the adaptation range of regulator parameters is for the normal operating conditions. When the operating conditions change significantly, the regulator parameters become inadequate. It is not due to poorly tuned regulator parameters, but rather to a limited range of parameter adaptation; to address this issue, adaptive control must be employed. Source: China Power Plant Centralized Control Operation Technology Network :) :)

Although there are theoretical methods for calculating PID parameters, due to the many factors that affect closed-loop regulation, it is not possible to describe them all precisely mathematically; as a result, the values calculated often have little practical significance. Therefore, aside from the parameters obtained through actual debugging, there are no available empirical parameter values. Even for two systems that appear to be identical, completely different parameter values can be obtained through actual debugging. One can try to use some relatively conservative PID parameters; for example, the gain should not be too high, and the integration time should not be too short, in order to avoid oscillations. On this basis, it can be put into operation directly to observe the waveform changes resulting from feedback. Giving a step input and observing the system’s response is the best way. If, after the feedback reaches a given value, it takes multiple oscillations to stabilize or remains unstable altogether, it should be considered whether the gain is too high or the integration time is too short ； If the feedback cannot be provided in a timely manner and the rise rate is very slow, it is necessary to consider whether the gain is too low or the integration time is too long... In short, tuning PID parameters is a comprehensive process with interrelated elements, and making multiple attempts during the actual tuning process is a very important and essential step.

1. The set parameter values obtained through various methods fall within a certain range, and they must be adjusted on-site according to the actual production conditions. 2. It is best to adjust the regulator parameters when the operating conditions during production are relatively stable. Apart from introducing appropriate artificial disturbances, it is preferable to modify the regulator parameters by observing the natural disturbances that occur during production over an extended period of time (it is advisable to use a fast-recording device for this purpose). 3. Artificial disturbances generally include internal disturbances, external disturbances, and prescribed value disturbances. Given value perturbations have a significant impact on the production process; it is essential to control their magnitude. Attention should also be paid to the extent of these perturbations, whether they are internal or external. After applying a disturbance, it is necessary to wait for a while to observe the changes in the parameter being adjusted; do not rush to apply a second disturbance until the situation is clear. 4. For PID regulators, the mutual interference between parameters must be taken into account. The regulator parameter values obtained through the tuning method are not the actual scale values of the regulator parameters; they must be corrected using the inter-regulator interference coefficients to become the actual scale values. 5. When setting the parameters, the nonlinear factors of the controlled object and the control mechanism must be taken into account. If the nonlinearity is not severe, the tuning parameters should be set more conservatively. 6. When adjusting, the operating conditions of the production process must be taken into account. Generally speaking, the adaptation range of regulator parameters is for the normal operating conditions. When the operating conditions change significantly, the regulator parameters become inadequate. It is not due to poorly tuned regulator parameters, but rather to a limited range of parameter adaptation; to address this issue, adaptive control must be employed.

Generally, the following points should be noted: 1. The set parameter values obtained through various methods fall within a certain range, and they must be adjusted on-site according to the actual production conditions. 2. It is best to adjust the regulator parameters when the operating conditions during production are relatively stable. Apart from introducing appropriate artificial disturbances, it is preferable to modify the regulator parameters by observing the natural disturbances that occur during production over an extended period of time (it is advisable to use a fast-recording device for this purpose). 3. Artificial disturbances generally include internal disturbances, external disturbances, and prescribed value disturbances. Given value perturbations have a significant impact on the production process; it is essential to control their magnitude. Attention should also be paid to the extent of these perturbations, whether they are internal or external. After applying a disturbance, it is necessary to wait for a while to observe the changes in the parameter being adjusted; do not rush to apply a second disturbance until the situation is clear. 4. For PID regulators, the mutual interference between parameters must be taken into account. The regulator parameter values obtained through the tuning method are not the actual scale values of the regulator parameters; they must be corrected using the inter-regulator interference coefficients to become the actual scale values. 5. When setting the parameters, the nonlinear factors of the controlled object and the control mechanism must be taken into account. If the nonlinearity is not severe, the tuning parameters should be set more conservatively. 6. When adjusting, the operating conditions of the production process must be taken into account. Generally speaking, the adaptation range of regulator parameters is for the normal operating conditions. When the operating conditions change significantly, the regulator parameters become inadequate. It is not due to poorly tuned regulator parameters, but rather to a limited range of parameter adaptation; to address this issue, adaptive control must be employed.