PIDparameters.h

#pragma once

#ifndef PIDPARAMETERS_H
#define PIDPARAMETERS_H

struct PIDparameters
{
	/* The gain K is set to a positive value if the measured value goes in the same
	 * direction as the control signal, otherwise negative.
	 * An increase in the gain K means that the gain curve is raised for all frequencies while the
	 * phase curve remains unchanged. The increased gain at low frequencies means that any
	 * stationary errors are reduced. Increasing the gain will increase the cutoff frequency ωc of
	 * the compensated system. The increased cutting frequency means that the speed of the closed
	 * system increases. Because now the phase margin must be read at a higher frequency
	 * the gain increase will also mean that the phase margin decreases and
	 * the robustness becomes worse.
	*/
	float proportionalGain;

	/* The integration time in seconds.
	 * A reduction in Ti means an increase in integral action. A reduction in Ti causes the gain to increase
	 * for low frequencies, while the phase decreases. The increased gain at low
	 * frequencies mean that any stationary errors are reduced.
	 * The increased gain causes the speed of the closed system to increase.
	 * Because the phase margin must now be read at a higher frequency, at the same time as the phase
	 * decreased, the robustness of the closed system becomes worse.
	*/
	float integrationTimeSecond;

	/* The derivative time in seconds.
	 * An increase in Td means that the gain increases for high frequencies,
	 * while the phase increases. The increase in gain means that the cutoff frequency
	 * ωc increases.
	 * The increased cutting frequency means that the system becomes faster. The fact that
	 * the phase margin must now be read at a higher frequency gives a poorer robustness, but
	 * since we have simultaneously raised the phase, we cannot draw this conclusion. There actually is
	 * a possibility that one can achieve both an increase in speed and an improved robustness by raising Td.
	 * However, this only applies within certain limits. Then gives an increase in Td impaired robustness.
	*/
	float derivateTimeSecond;

	/* Time constant in seconds that controls how quickly the integral part should
	 * be restored after a limitation of the control signal. Tt is referred to as "tracking time
	 * constant". The time constant Tt should be greater than Td but less than Ti.
	 * A rule of thumb is to choose Tt according to Tt = sqrt(Ti * Td) for a PID controller and
	 * for a PI controller Tt = 0.5 * Ti.
	*/
	float trackingTimeConstantSecond;

	/* The time constant of the analogue measured value filter in e.g. seconds.
	 * To avoid the alias effect, the filter must eliminate any frequencies higher than half the
	 * sampling rate. The time constant Tanalogue f shall be chosen so that the
	 * frequencies above half the sampling frequency are attenuated, e.g. 16 times.
	*/
	float filterTimeConstantSecond;

	/* The time constant of the derivative filter in seconds. This filter filters the measured value
	 * before the D-part. The filter is necessary for the prediction of the D-part to be reliable even if the
	 * measurement signal contains noise.
	 * Rule of thumb for the D filter Tf = Td / 10 is a good rule.
	 * Tf can be chosen according to Tf = Td/N, where typical
	 * values of N are between 2 and 20.
	 * The disadvantage of filtering is that the filter causes a delay
	 * or, more precisely, a phase shift of the signal. A delay negatively affects the D-part
	 * prediction function, which is not desirable. It is important to find a balance between
	 * how hard you want to filter and how much delay you can allow.
	*/
	float derivateFilterTimeConstantSecond;

	/* Setpoint weighting factor, value 0-1.
	*/
	float setpointWeightingFactor;

	/*
	* The sampling time is the time in e.g. seconds between each execution of the
	* regulator algorithm. Sampling rate = 1/h. A rule of thumb is to choose the sampling
	* time shorter than one-fifth of the process's time constant T.
	*/
	float sampleTimeSecond;
};
#endif // PIDPARAMETERS_H