Real and reactive power
The AGC comes into play as the demand for real and reactive power is never steady as they continually change along a trend line. Therefore generators which could quickly react for these kinds of responses (e.g. gas turbines, hydro-generators) must be actively fed with the necessary to either increase or decrease depending upon the requirement of the plant. Failing to do so will result in the change of frequency of the whole power network. Since the fact in this era, all the power networks are interconnected to each other, and therefore it's not so easy to easy to respond to these frequency changes in quick time; hence automatic generation control of voltage was installed in each generating unit so that the mainframe computers actively react to these frequency changes in quick time.
This assignment deals with the power system a two area power system connected with a single tie line was simulated using MATLAB. The load frequency controller devices were attached to either of the area. In order to actively monitor the frequency deviation an integral controller was modelled in both the areas so as to give a zero steady state error in the tie line when the LFC is activated.
On addition of both the equations implies
Therefore on steady state conditions or else there will be no load frequency control governing the action of the generators. Eventually this turns out to give;
Therefore the steady-state frequency error = 0 and the power flow on the tie lines ( is also equal to zero when the LFC is activated in both the areas. On making the steady state - frequency to zero, the area 1 take the 400MW load on its own and then the makes the Area 2 to come back to its original position without any load increment.The figure 2 below is the SIMULINK model, of the two areas with a single tie line, the output of the LFC controls are expressed graphically.
The data file shown in figure 3 that sets the parameters to this SIMULINK model that actually sets whether or not the LFC is activated.
% Data for EE5523_21_3
On setting the constants k1 and k2 = 0, one is actually switching off the LFC control for both the areas, this constant on attaining greater than zero will activate the LFC control measures as the integral loop of the system is activated, thereby making the steady state frequency error go to zero. But for this task it is made to switch off and thereby simulating the model gives a steady state frequency error due to load increase of 400MW in area 1. The steady-state frequency, generation change in area 1 and are 2, and ∆Ptie12 shown in the following figures match the calculated values thus proving the correctness of the calculations.
On setting up the data file in such away so that the both K1 = 0 and K2 = 0.0001, now area 2 is activated for LFC and hence the data file was recompiles and then simulation was run so that the parameters values get updated. The generation in the both the areas and the ∆Ptie12 are illustrated in the figures to come.
The calculated values did match up with the simulated results, but for the area 2 change in generation proved that the dynamic response was not similar to plots that were shown in task 4.
The transient response of this plot is disturbed mainly due to the value of the constant k2 which activates the LFC of the area 2. Hence this turn on the integral loop of the SIMULINK model thereby causing this dynamic response. However the effect of this dynamic response is minimised by equalising the to B2 which give adequate performance of the system. Furthermore to further stabilise the value of k2 was set to be minimal, if not this would result results in an instability of the area 2 thus causing the frequency to deviate from its statutory frequency limit between 0.5 Hz of 50 Hz. This was further investigated by increasing k2 value to 0.1 shown in figure.
Now on both the areas the LFC is activated as both the k values are set to 0.001, hence the data file was updated with and the simulation of the generation on both the areas, frequency deviation and the tile line error graphs are illustrated below.
The theoretical proof matched the simulation as the LFC is making errors to minimise to zero thus stabilising the frequency hence maintain it at 50 Hz for a load increase of 400MW.
The area 1 will pick up the 400 MW (which is shown in figure 13) increase of generation as the steady - state frequency error returns to zero, and area 2 returns to its operating condition of having zero error.
The initial fluctuations on all of the graph show that the integral loop as is can be seen in the SIMULINK model along with the k and the B constants is trying to stabilise the system in running on a closed loop circle.
The action of LFC on governing generators in a two area system interconnected by a single tie line was well understood. Tasks 1, 2 and 3 initially helped in the realisation of this model in theory and then with the help of MATLAB SIMULINK this scenario was well understood graphically.
In real life manual monitoring of frequency is completely irrelevant these day as power systems are getting more and more complex as the interconnections are getting bigger so the LFC mode of control would be in the forefront of examining small signal instability as this mode of controlling the frequency acts more slowly to a response, this was quite evident in figure 12, 13, 14 and 15 where the damping of the signals takes a while to come to a stable position. It can also be envisaged in task 4 where it takes more time either to increase or lower the signal.