Skip to main content

Design of a 2-DOF-PID controller using an improved sine–cosine algorithm for load frequency control of a three-area system with nonlinearities

Abstract

This paper proposes an improved sine–cosine algorithm (ISCA) based 2-DOF-PID controller for load frequency control. A three-area test system is built for study, while some physical constraints (nonlinearities) are considered for the investigation of a realistic power system. The proposed method is used as the parameter optimizer of the LFC controller in different scenarios. The 2-DOF-PID controllers are used because of their capability of fast disturbance rejection without significant increase of overshoot in set-point tracking. The 2-DOF-PID controllers’ efficacy is observed by examining the responses with the outcomes obtained with PID and FOPID controllers. The simulation results with the suggested scheme are correlated with some of the existing algorithms, such as SCA, SSA, ALO, and PSO in three different scenarios, i.e., a disturbance in two areas, in three areas, and in the presence of physical constraints. In addition, the study is extended to a four-area power system. Statistical analysis is performed using the Wilcoxon Sign Rank Test (WSRT) on 20 independent runs. This confirms the supremacy of the proposed method.

Introduction

Literature survey

The pivotal aspiration of the modern power system is to provide a reliable power supply without any interruption. This can be achieved when equilibrium between demand and generation of power is maintained. Frequency is one of the vital parameters which indicates the balance between demand and supply or generation. As frequency and load are inversely proportional to each other, frequency increases when the load is less than the generation, while frequency reduces when the load on the system is more than the supply. Maintaining the frequency at its standard value is an essential task and can be achieved by a technique known as load frequency control (LFC) [1]. LFC has the significant responsibility to maintain the drift in frequency within the permissible limit. In addition, it also maintains the drift in tie-line power between multi-area systems to its permissible value. The mechanical input to the power generator is provided to have a balance between generation and demand of power and thereby, LFC controls the input to the generator according to the requirements. LFC basically perform the following tasks:

  • Nullifies the steady-state error in frequency that is due to step load changes and thereby minimizes transient response and time error.

  • Reduces the static change in tie-line power to zero due to step load retribution.

  • Supplies the power to any area during emergency from the rest areas.

A detailed review of LFC using PID controller based on soft computing, IMC techniques, and robust control schemes is presented in [2]. In [3], a PID controller with a sliding mode control scheme based on the ALO method for a four-area system is discussed, while a comparative analysis of a backtracking search algorithm and fruit fly optimizer-based PID controller for a two-area system considering nonlinearities is explained in [4]. Reference [5] uses a PID controller based on the ALO method for a two-area and a three-area system with a non-reheated thermal power system for analyzing different performance indices. In [6], a differential evolution (DE) method based on a PID controller is provided for a two-area thermal system with GRC and a two-area thermal system with diverse generating units such as thermal, hydro, and diesel. Reference [7] uses bacterial foraging optimization for LFC of an unequal three-area system, whereas a flower pollination algorithm-based PID controller is designed in [8] for frequency control of a two-area system considering GDB. A fuzzy PID controller based on a sine cosine algorithm for LFC of a hybrid renewable system is presented in [9], and [10] uses a sine cosine algorithm-based fuzzy PID controller for the LFC of a three-area system having nonlinearities. An imperialist competitive algorithm-based fuzzy PI controller for LFC of two-area systems is presented in [11], while [12] carries out a detailed review on the different types of fractional order controllers used for LFC where various controllers are used, including FOPI, FOPID, PIFOD, TID, FOPIDN, PFOID, etc. It has been observed from different literature surveys that FOPID is the most commonly used controller for LFC. The DE-based FOPID controller is used for LFC of a three-area system in [13], while a sine cosine algorithm-based TID controller is implemented for a two-area hybrid source system in [14]. FOPID based on gases Brownian motion optimization [15] is proposed for LFC of a two-area system having GDB, while a FOPID controller is designed in [16] for a single area system using Kharitonov’s theorem. A Salp Swarm Algorithm (SSA) based TID controller is suggested in [17] for LFC for the systems incorporating FACTS devices, while a Grey wolf optimized multi-degree of freedom PID controller is proposed for the LFC of a two-area system in [18]. A 2-DOF-PID controller based on the quasi-oppositional Jaya algorithm is applied for LFC of multi-source three-area system in [19]. The 2-DOF-PID controller based on MFO is implemented for LFC of a two-area system [20], and it is also suggested for a two-area system with some nonlinearities in [21]. The SSA-based 2-DOF-PID controller is used in [22] for LFC of a two-area multi-source system. Some hybrid algorithms such as the hybrid gravitational search and pattern search algorithm [23], and the hybrid firefly-pattern search algorithm [24] have been proposed for LFC. Recently, ALO-based adaptive neuro fuzzy interference system (ANFIS) [25], PSO [26, 27], two-dimensional sine-logistic mapped-based SCA [28], and SSA [29, 30] have been used for LFC.

It is observed from the literature survey that in the study of LFC, the researchers have mainly focused on three things, viz. designing new controllers, proposing new optimization techniques, and modeling different types of power system. SCA is a recently developed technique that has been implemented to solve various engineering problems [31]. However, SCA suffers from slow convergence, and getting trapped in local optima. For better performance, SCA is improved in this work, and the Improved SCA (ISCA) method is applied to tune the controller of LFC. The 2-DOF-PID controller based on ISCA is designed and implemented for LFC of a three-area system with and without nonlinearities, and then further applied to a four-area system. The objective function used is ITAE, and by minimizing the ITAE, the different performance parameters such as peak undershoot, settling time of frequency and tie-line power are improved.

To further confirm the superiority of the proposed method, a statistical analysis is performed. For that analysis, each method is simulated for 20 independent runs, and the Wilcoxon Sign Rank Test (WSRT) is used. This is a significance test where \(+, -, \approx\) show ‘superior’, ‘inferior’, or ‘equivalent’ with respect to the compared ones.

Contribution and organization of the paper

The main contributions of the work are as follows:

  • An improved form of SCA is proposed, and the superiority of the proposed ISCA method is justified using unimodal and multimodal benchmark functions.

  • The proposed ISCA is used to optimize the LFC controller variables, and then the performance is compared with some other recently developed algorithms such as SCA, ALO, SSA, and PSO.

  • A three-area test system and a four-area test system are modeled for the case study.

  • Three types of controllers, viz. PID, 2-DOF-PID, and FOPID are implemented to test the systems, and the effectiveness of the 2-DOF-PID controller is validated.

  • For the first time, statistical analysis and WSRT are performed in LFC study to draw conclusions.

  • The impact of the presence of physical constraints on system performance is investigated.

The work is organized as follows: the proposed power system is discussed in Sect. 2, while Sect. 3 describes the proposed controller. The proposed optimization strategy is detailed in Sect. 4, and the problem formulation is defined in Sect. 5. The results are discussed in Sect. 6, and the conclusions are summarized in Sect. 7.

Proposed power system

A three-area system is considered for the case study. The system is an unequal system that consists of three thermal generators having different parameter values. In each area, there are governor, turbine, generation and load sections. The complete system is shown in Fig. 1 with mathematical models in transfer function forms. As shown in Fig. 1, the time constants of each area are denoted as: (1) Tg1, Tg2, Tg3 for the governors; (2) Tt1, Tt2, Tt3 for the turbines; and (3) Tps1, Tps2, Tps3 for the generation and load section. The gains for generation and load sections are denoted as Kp1, Kp2, Kp3 for each area. The other parameters of the system are B1, B2, B3 for the frequency biased, R1, R2, R3 for the droops, and T12, T23, T31 for the synchronizing coefficients. Drifts in frequency for each area are represented by \(\Delta F_{1}\), \(\Delta F_{2}\), \(\Delta F_{3}\), while \(\Delta P_{tie12}\), \(\Delta P_{tie23}\), \(\Delta P_{tie31}\) are the incremental changes in tie-line power. The numerical values of the parameters are shown in “Appendix 1”.

Fig. 1
figure 1

Block diagram of a three-area power system

The proposed controller

Numerous variants of PID controller have been used for many years, because of its simplicity and ability to provide reliable results. In this paper, a variation of PID controller known as 2-DOF- PID controller is used as the LFC controller because of its capability of fast disturbance rejection without significant increase of overshoot in set-point tracking. DOF stands for the degree of freedom, which means the extent of closed-loop transfer function that can be handled distinctly in a control system. The basic arrangement of this controller is shown in Fig. 2, where two separated loops can be seen. Two inputs are applied to the controller, of which one is a reference, and the other is the output of the system. The error signal generated because of difference in these two signals is used by the controller for the generation of the controller output signal which consists of the proportional, integral, and derivate portions according to their weight. The mathematical expression for the 2-DOF-PID is:

$$\begin{aligned} u & = K_{p} \left( {\left( {PW} \right)r - y} \right) + \frac{{K_{i} }}{s}\left( {r - y} \right) \\ & \quad + \frac{{K_{d} s}}{Ns + 1}\left( {\left( {DW} \right)r - y} \right) \\ \end{aligned}$$
(1)

where r and y are two input signals, r is a reference and y is the output of the system. Kp, Ki, Kd are the proportional, integral, and derivative weights, respectively. N is the filter coefficient, and u is the controller output. PW and DW are the set-point weights on proportional and derivative sections, respectively.

Fig. 2
figure 2

Structure of the 2-DOF-PID controller

The projected optimization techniques

Sine–cosine algorithm (SCA)

The SCA algorithm is a stochastic population-based optimization technique inspired by the mathematical functions of sine and cosine. It was recently developed in [32]. Because of the use of the sine and cosine mathematical functions, this algorithm provides cyclic space for exploitation, in which search agents can update their position as per a position changing equation, as:

$$Y_{j}^{n + 1} = Y_{j}^{n} + a_{1} \sin \left( {a_{2} } \right) \times \left| {a_{3} P_{j}^{n} - Y_{j}^{n} } \right|$$
(2)
$$Y_{j}^{n + 1} = Y_{j}^{n} + a_{1} \cos \left( {a_{2} } \right) \times \left| {a_{3} P_{j}^{n} - Y_{j}^{n} } \right|$$
(3)

where \(a_{1}\), \(a_{2}\), \(a_{3}\) are the arbitrary numbers and are the main parameters of this algorithm, and \(Y_{j}^{n + 1}\) and \(Y_{j}^{n}\) are the next and current positions of the solution at the time of the nth iteration in the jth dimension, respectively. \(P_{j}^{n}\) is the terminus point in the jth dimension. Equations (2) and (3) can be combined by another parameter \(a_{4}\). Depending upon the value of \(a_{4}\) which is an arbitrary numeral in the array of [0, 1], the algorithm will choose the equation for renovating the position of the investigating agent, given as:

$${ }Y_{j}^{n + 1} = \left\{ {\begin{array}{*{20}c} {Y_{j}^{n} + a_{1} \sin \left( {a_{2} } \right) \times \left| {a_{3} P_{j}^{n} - Y_{j}^{n} } \right|\;\;if\;\; a_{4} < 0.5 } \\ {Y_{j}^{n} + a_{1} \cos \left( {a_{2} } \right) \times \left| {a_{3} P_{j}^{n} - Y_{j}^{n} } \right|\;\,if\;\; a_{4} \ge 0.5} \\ \end{array} } \right.$$
(4)

The parameter \(a_{1}\) will decide the next location province, which can be between the destination and another location. It has the objective of harmonizing the exploitation and exploration of this optimizer, and its value can be given by:

$$a_{1} = b - n\frac{b}{N}$$
(5)

where N is the maximum number of iterations, b is a constant, and n represents the current iteration.The direction of movement of the search agent, whether towards the global optima or away from it, is decided by \(a_{2}\). Better results are obtained by considering the range of \(a_{2}\) between [− 2 to 2], while sine and cosine functions are between 0 to 2π. The objective of \(a_{3}\) is to emphasize the destination and is implemented by choosing a random value. If it is greater than 1 it will stochastically emphasize the destination while it will deemphasize if less than 1.

Improved sine–cosine algorithm (ISCA)

Nevertheless, SCA is very capable of handling the real-time problem, though there is scope to improve the algorithm to improve the rate of convergence, the ability not to be trapped in nearby optima, and to maintain a balance between exploration and exploitation. The above limitations of traditional SCA are due to the updating scheme of its search agents. In SCA, most of the search agents run towards the global optima, but sometimes can get trapped in local optima and thus converge to that premature local optima. To overcome this, a new scheme for updating the location of the search agent is introduced in this paper. This scheme mainly consists of the SCA/best-target shown in (6) and (7), and the SCA/rand-target shown in (8) and (9). The best-target search agent of the SCA assists the search agents in moving towards the best position obtained so far and searching locally around the best search agent, which results in the intensification of the solution. On the other hand, the rand-target search agent of SCA moves the search agents towards the arbitrary position, which results in more search space exploration. In the next step, the exploring capability of both schemes is combined by taking the mean as shown in (10), and the resultant is set as the new search agent. The characteristics of the proposed ISCA are as follows:

  1. 1.

    It maintains balance between exploration and exploitation.

  2. 2.

    It has fewer parameters, i.e., the number of parameters of the proposed ISCA is 3 while it is 4 in the original SCA.

  3. 3.

    It has a better convergence rate than the SCA.

  4. 4.

    It avoids getting trapped in local optima.

The values of the three parameters are decided in accordance with (11), (12), and (13), respectively.

$$Y_{1} = Y_{best}^{n} + a_{1} \sin \left( {a_{2} } \right) \times \left| {a_{3} Y_{rand}^{n} - Y_{j}^{n} } \right|$$
(6)
$$Y_{2} = Y_{best}^{n} + a_{1} \cos \left( {a_{2} } \right) \times \left| {a_{3} Y_{rand}^{n} - Y_{j}^{n} } \right|$$
(7)
$$Y_{3} = Y_{rand}^{n} + a_{1} \sin \left( {a_{2} } \right) \times \left| {a_{3} Y_{best}^{n} - Y_{j}^{n} } \right|$$
(8)
$$Y_{4} = Y_{rand}^{n} + a_{1} \cos \left( {a_{2} } \right) \times \left| {a_{3} Y_{best}^{n} - Y_{j}^{n} } \right|$$
(9)
$$Y_{j}^{n + 1} = Mean\left( {Y_{1} ,Y_{2} ,Y_{3} ,Y_{4} } \right)$$
(10)
$$a_{1} = b\left( {1 - \frac{b}{N}} \right)$$
(11)
$$a_{2} = 2 \times \pi \times rand\left( {0, 1} \right)$$
(12)
$$a_{3} = 2 \times rand\left( {0, 1} \right)$$
(13)

where b is a constant which is set to 2, N is the maximum number of iterations, n is the current iteration, and \(rand\left( {0, 1} \right)\) denotes a random number that will be generated in the range of 0–1.

The flow chart of the ISCA is shown in Fig. 3. The algorithm mainly has three steps, i.e., initialization, iteration, and termination. In the first step, the algorithm initializes the parameters, such as the maximum number of iterations (N), search agent number (c), number of variables to be tuned (d) with their upper (ub) and lower (lb) bound, first set of search agents (solution). In the second step, it generates a single new search agent by taking the average of four search agents which are being generated by the proposed search schemes. In the last step, the best agent so far obtained is selected as the solution to the optimization problem.

Fig. 3
figure 3

Flow chart of ISCA

Performance estimation of the suggested method

The superiority of the proposed technique is tested against ALO, SCA, SSA, and PSO using the 13 standard unimodal and multimodal benchmark functions. Every single algorithm is run 20 times for each benchmark function. The average and standard deviations of different benchmark functions for the ISCA, SCA, ALO, SSA, and PSO algorithms are shown in Table 1. The statistical Wilcoxon signed-rank test is carried out on the results. This is shown in Table 2 to confirm the superiority of the ISCA. From Tables 1 and 2, it is found that the ISCA outperforms other methods in eight functions \(f_{1} ,f_{2} ,f_{3} ,f_{4} ,f_{5} ,f_{7} ,f_{10} ,f_{11}\), while PSO outperforms other methods for \(f_{6}\), \(f_{8} , f_{12}\) functions, and SCA and ALO outperform other methods for \(f_{9}\) and \(f_{13}\) functions, respectively. Hence, the proposed method achieves better performance than the existing methods.

Table 1 Performance evaluation of ISCA over SCA, SSA, ALO, and PSO with the results of unimodal and multimodal benchmark test function
Table 2 Wilcoxon signed-rank test results on unimodal and multi-model functions indicating the inferior (-), superior ( +), or equivalent (≈), method in comparison to the proposed method

Problem formulation

Whenever there is a disturbance in the system, the prime objectives of LFC are: (1) nullifying the drift in frequency; and (2) keeping the exchange power of the tie-line at its schedule value. For any optimization problem in LFC, the objective function needs to be defined to achieve the objectives. Various criteria have been included in the literature to accumulate the frequency deviation and tie-line power deviation in the objective function. From the literature survey, it is found that ITAE (integral of time multiplied absolute error) is a promising criterion and is generally preferred over others like IAE, ISE, and ISTE. Hence, ITAE is used as the objective function for the test system shown in Fig. 1. In LFC, ITAE of the drift in each area frequency and incremental change in tie-line power is taken as the objective function:

$$J = ITAE = \mathop \sum \limits_{i = 1}^{NA} \mathop \smallint \limits_{0}^{{t_{sim} }} \left( {\left| {\Delta f_{i} } \right| + \mathop \sum \limits_{{\begin{array}{*{20}c} {j = 1} \\ {j \ne i} \\ \end{array} }}^{NA} \left| {\Delta P_{{tie_{i - j} }} } \right|} \right).t.dt$$
(14)

where \(\Delta f_{i}\) is the drift in each area frequency, \(\Delta P_{{tie_{i - j} }}\) is the incremental change in tie-line power, and \(t_{sim}\) is the simulation time period. The controller parameter boundary is the problem constraint. As a result, the design challenge can be expressed as an optimization problem:

$${\text{Minimize}}\;{\text{J}}$$

For the PID controller

$$K_{{p_{min} }} \le K_{p} \le K_{{p_{max} }} ;\;K_{{i_{min} }} \le K_{i} \le K_{{i_{max} }} ;\;K_{{d_{min} }} \le K_{d} \le K_{{d_{max} }} ;\;N_{min} \le N \le N_{max} ,$$

For the FOPID controller:

$$\begin{aligned} & K_{{p_{min} }} \le K_{p} \le K_{{p_{max} }} ;\;K_{{i_{min} }} \le K_{i} \le K_{{i_{max} }} ;\;K_{{d_{min} }} \le K_{d} \le K_{{d_{max} }} ; \\ & \mu_{min} \le \mu \le \mu_{max} ;\;{\uplambda }_{min} \le {\uplambda } \le {\uplambda }_{max} \\ \end{aligned}$$

For the 2-DOF-PID controller:

$$\begin{aligned} & K_{{p_{min} }} \le K_{p} \le K_{{p_{max} }} ;\;K_{{i_{min} }} \le K_{i} \le K_{{i_{max} }} ;\;K_{{d_{min} }} \le K_{d} \le K_{{d_{max} }} ; \\ & N_{min} \le N \le N_{max} ;\;b_{min} \le b \le b_{max} ;\;c_{min} \le c \le c_{max} \\ \end{aligned}$$

Results and discussions

On the test system of Fig. 1, many simulation studies have been carried out in order to determine the optimal combination of the suggested algorithm and the controller in order to reach a better outcome. For this purpose, various types of studies have been considered. In Sect. 6.1, the performance of various controllers is compared to identify the better one for further study, whereas in Sect. 6.2, disturbances are applied in two areas, while various algorithms are compared and statistical analysis is carried out. Similarly, in Sect. 6.3, disturbances are applied to all three areas, various algorithms are compared, and statistical investigation is carried out to find the better algorithm. In Sect. 6.4, nonlinearities of GRC, GDB, and commutation delay are considered in the test system, with various algorithms applied to find the controller parameters, while statistical analysis is carried out to find the better algorithm. For statistical assessment of these techniques, WSRT is performed to demonstrate the superior (\(+\)), equivalent (\(\approx\)) or inferior (\(-\)) schemes in contrast to the proposed ISCA method. The study is then extended to a four-area power system shown in Fig. 8 in Sect. 6.5 and again various algorithms are compared for this scenario.

Examination of controllers

From the literature survey, it is found that some variants of PID controllers have been considered to implement the proposed algorithm. The most common controllers are the PID controller with filter (PIDF), 2-DOF-PID controller (2-DOF-PID), and Fractional order PID controller (FOPID). In this work, these controllers are initially compared for the considered test system (Fig. 1) to identify the better one for the present study. A disturbance of 2% is applied in area-1 and area-2. Figure 4 illustrates the convergence curve of the different controllers, the drift in frequency and tie-line power of each area for this case, while Table 3 shows the performance parameter of LFC for these controllers. From Table 3 and Fig. 4, it can be seen that the 2-DOF–PID controller outperforms the other two as it converges faster and attains the least objective function value. It also has a lower settling time and undershoot for the frequency deviation as well as for the tie-line power. Given these advantages of the 2-DOF- PID controller, it will be used as the LFC controller for further study as discussed in the following subsections.

Fig. 4
figure 4

Transient responses of frequency and tie-line power of all areas and convergence curves of the controllers for 2% SLD in area-1 and area-2 for 2-DOF PID, FOPIDF, and PID controllers

Table 3 Performance parameters of LFC with 2-DOF PID, FOPIDF, and PID controllers

Examination of optimization techniques when disturbances are in two areas

In this part, to validate the effectiveness of the proposed method, the ISCA is applied for the tuning of the LFC controller parameters. The 2-DOF-PID controlled test system is simulated when disturbances of 2% are applied in area-1 and area-2. The comparison is carried out with recently used algorithms of PSO, SSA, SCA, and ALO. Figure 5 illustrates the convergence curves and the transient responses of the test system. The convergence curves show that the ISCA achieves the lowest objective function (ITAE value) and converges faster than the other methods. Table 4 gives the performance parameters of the different algorithms for this case. From Fig. 5 and Tables 4, it can be seen that the ISCA performs much better than the other optimization techniques.

Fig. 5
figure 5

Transient responses of frequency and tie-line power of all areas and convergence curves of the various algorithms for SLD of 2% in area-1 and area-2 with the proposed and other existing algorithms

Table 4 Performance parameters of LFC with ISCA, SSA, ALO, SCA and PSO, in the case of SLD of 2% in area-1 and area-2

To further highlight the efficacy of the ISCA in optimizing the LFC controller parameters, simulation is carried out for 20 independent times by ISCA, PSO, SSA, ALO, and SCA. Table 5 presents the mean and standard deviation of each technique. The ISCA gives the minimum mean value of the objective function. Further, for statistical assessment of these techniques, WSRT is performed on the test system for this scenario, and the results are shown in Table 5.

Table 5 Comparison of fitness values and WRST results of the three-area system with SLD of 2% in area-1 and area-2 using Different Optimization Techniques (Values in bold show best value)

Examination of optimization techniques when the disturbance is in each area

In this section, the ISCA is used as the optimizer for the controller parameters. The 2-DOF-PID controlled test system is simulated when a disturbance of 2% is applied in each area of the test system, and the results are compared with the PSO, SSA, SCA, and ALO. Figure 6 compares the convergence curves and the transient responses of the test system. As seen, the convergence curve of the ISCA shows that the ISCA achieves the lowest objective function (ITAE value) and converges faster than the other methods. Table 6 gives the performance parameters of different algorithms for this case.

Fig. 6
figure 6

Transient responses of frequency and tie-line power of all areas and convergence curves of the various algorithms for SLD of 2% in each area with the proposed and other algorithms

Table 6 Performance parameters of LFC with ISCA, SSA, ALO, SCA and PSO in the case of SLD of 2% in each area

To further show the efficacy of the ISCA in optimizing the LFC controller parameters, simulation is done for 20 independent times by ISCA, PSO, SSA, ALO, and SCA. Table 7 presents the mean and standard deviation of each technique. The ISCA gives the minimum mean value of the objective function. Further, for statistical assessment of these techniques, WSRT is performed, and the results are given in Table 7. It is evident that the proposed method is superior.

Table 7 Comparison of fitness values and WRST results of the three-area system with SLD of 2% in each area using Different Optimization Techniques (Values in bold shows best value)

Examination of optimization techniques considering physical constraints of GRC, GDB, and communication delay

In this section, the test system is modified and some physical constraints of GRC, GDB, communication delay, and reheated turbine are taken into consideration. Since practical power systems have nonlinearity as mentioned above, the modified system with consideration of these constraints gives a better representation of the practical systems. The communication delay of 40 ms, GDB of 0.036 pu, and GRC of 3% pu are considered in each area of the system, while a disturbance of 1% is applied to area-1. The proposed and other algorithms are compared for the 2-DOF-PID controlled system. The dynamic responses and convergence curves for this case are given in Fig. 7, and the performance parameters are shown in Table 8. It is concluded from the convergence curve, dynamic response, and performance parameters that the proposed ISCA method has much better tuning efficiency than other optimization methods.

Fig. 7
figure 7

Transient responses of frequency and tie-line power of all areas and convergence curves of the various algorithms for the three-area system with physical constraint and SLD of 1% in area-1 with the proposed and other algorithms

Table 8 Performance parameters of LFC with ISCA, SCA, ALO, SSA, and PSO for the three-area system with physical constraints with SLD of 1% in area-1

To further highlight the efficacy of the ISCA in optimizing the LFC controller parameters, simulation is done for 20 independent times by ISCA, PSO, SSA, ALO, and SCA. Table 9 shows the mean and standard deviation of each technique. The ISCA gives the minimum mean value of the objective function. Further, for statistical assessment of these techniques, WSRT is performed and the results are given in Table 9.

Table 9 Comparison of fitness values and WRST results of the three-area system with physical constraints with SLD of 1% in area-1 using Different Optimization Techniques (Values in bold shows best value)

Extension to four-area power system

To further examine the capability of the proposed algorithm the study is extended to a four-area power system. The schematic diagram of this system is shown in Fig. 8 while the values of relevant parameters are given in “Appendix 1”. The 2-DOF-PID controlled system is simulated for the 2% disturbance in area-1. Figure 9 shows the comparative analysis with various techniques for this test system while Table 10 gives the corresponding performance parameters. From Fig. 9 and Table 10, it can be seen that the proposed ISCA algorithm performs better than other techniques.

Fig. 8
figure 8

Schematic diagram of the four-area system

Fig. 9
figure 9

Transient responses of frequency and tie-line power of all areas and convergence curves of the various algorithms for the four-area system with SLD of 2% in area-1 with the proposed and other algorithms

Table 10 Performance parameters of LFC with ISCA, SSA, ALO, SCA and PSO in the case of SLD of 2% in area-1 of the four-area system

To further demonstrate the efficacy of ISCA in optimizing the LFC controller parameters, simulation is done for 20 independent times by ISCA, PSO, SSA, ALO, and SCA. Table 11 shows the mean and standard deviation of each technique. The ISCA again gives the minimum mean value of the objective function. Again, WSRT is performed and the results are given in Table 11.

Table 11 Comparison of fitness values and WRST results of the four-area system with SLD of 2% in area-1 using Different Optimization Techniques (Values in bold shows best value)

Conclusion

In this work, an improved version of the SCA, i.e. the ISCA, is proposed as the tuning tool for the load frequency controller of a multi-area unequal system. The balance between exploitation and exploration is maintained as the proposed method combines the exploitation proficiency of the SCA/best-target search agent and the exploration proficiency of the SCA/rand-target search agent. First, some controllers are compared for the considered test system, and the 2-DOF-PID controller emerges as the best. This is then further used as the load frequency controller for different analyses. The performance of the ISCA tuned LFC controller is examined for the test system for various schemes, such as disturbances in two areas, disturbance in each area, disturbance in the presence of various nonlinearities in the test system, and in an extended four-area power system. The performance of the ISCA is compared with some of the existing algorithms, while the effectiveness of the proposed method is tested by statistical analysis for different scenarios. After considering all the analyses, it is observed that the proposed method is better than the other techniques, such as PSO, SCA, SSA, and ALO, in terms of better convergence rate, least objective function value, minimum settling time, and undershoot of deviation in frequency and tie line power. Therefore, it can be concluded that the proposed method can be applied to solve real-world engineering problems.

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Abbreviations

ISCA:

Improved sine–cosine algorithm

ALO:

Ant-lion optimization

LFC:

Load frequency control

DOF:

Degree of freedom

Tps :

Time constant of the power system

FOPID:

Fractional order proportional-integral-derivative

SCA:

Sine–cosine algorithm

Tsg :

Time constant of speed governor

Ri :

Speed regulation constant

Kps :

Gain of the power system

PID:

Proportional-integral-derivative

β:

Frequency bias constant

SSA:

Salp swarm algorithm

PW, DW:

Proportional and derivative set-point weights respectively

Tt :

Time constant of the turbine

ΔPtie :

Tie-line power deviation

Tij :

Synchronization time constant

\(\Delta P_{ref}\) :

Reference input to the controller

PSO:

Particle swarm optimization

\(\Delta P_{D}\) :

Load disturbance

Δf:

Frequency deviation

WRST:

Wilcoxon sign rank test

SLD:

Step load disturbance

GRC:

Generation Rate Constraint

GDB:

Governor dead band

References

  1. Kundur, P. (2009). Power system stability and control (8th ed.). Tata McGraw-Hill.

    MATH  Google Scholar 

  2. Hote, Y. V., & Jain, S. (2018). PID controller design for load frequency control: Past, present and future challenges. IFAC-PapersOnLine, 51(4), 604–609. https://doi.org/10.1016/j.ifacol.2018.06.162

    Article  Google Scholar 

  3. Ah, G. H., & Li, Y. Y. (2020). Ant lion optimized hybrid intelligent PID-based sliding mode controller for frequency regulation of interconnected multi-area power systems. Transactions of the Institute of Measurement and Control, 42(9), 1594–1617. https://doi.org/10.1177/0142331219892728

    Article  Google Scholar 

  4. Dinesh Madasu, S., Sai Kumar, M. L. S., & Singh, A. K. (2017). Comparable investigation of backtracking search algorithm in automatic generation control for two area reheat interconnected thermal power system. Applied Soft Computing, 55, 197–210. https://doi.org/10.1016/j.asoc.2017.01.018

    Article  Google Scholar 

  5. Jagatheesan, K., Anand, B., Dey, K. N., Ashour, A. S., & Satapathy, S. C. (2018). Performance evaluation of objective functions in automatic generation control of thermal power system using ant colony optimization technique-designed proportional–integral–derivative controller. Electrical Engineering, 100(2), 895–911. https://doi.org/10.1007/S00202-017-0555-X

    Article  Google Scholar 

  6. Mohanty, B., Panda, S., & Hota, P. K. (2014). Differential evolution algorithm based automatic generation control for interconnected power systems with non-linearity. Alexandria Engineering Journal, 53(3), 537–552. https://doi.org/10.1016/j.aej.2014.06.006

    Article  Google Scholar 

  7. Nanda, J., Mishra, S., & Saikia, L. C. (2009). Maiden application of bacterial foraging-based optimization technique in multiarea automatic generation control. IEEE Transactions on Power Systems, 24(2), 602–609. https://doi.org/10.1109/TPWRS.2009.2016588

    Article  Google Scholar 

  8. Madasu, S. D., Sai Kumar, M. L. S., & Singh, A. K. (2018). A flower pollination algorithm based automatic generation control of interconnected power system. Ain Shams Engineering Journal. Retrieved June 05, 2021, from https://reader.elsevier.com/reader/sd/pii/S2090447916300892?token=888DDE26161EBD8536B32321A023F501D4402644778263B171C2669901F5B191A3543FB2C83D59DD636863E6AEAA35DF&originRegion=eu-west-1&originCreation=20210605180519

  9. Rajesh, K. S., & Dash, S. S. (2019). Load frequency control of autonomous power system using adaptive fuzzy based PID controller optimized on improved sine cosine algorithm. Journal of Ambient Intelligence and Humanized Computing, 10(6), 2361–2373. https://doi.org/10.1007/s12652-018-0834-z

    Article  Google Scholar 

  10. Sahu, P. C., Prusty, R. C., & Sahoo, B. K. (2020). Modified sine cosine algorithm-based fuzzy-aided PID controller for automatic generation control of multiarea power systems. Soft Computing, 24(17), 12919–12936. https://doi.org/10.1007/s00500-020-04716-y

    Article  Google Scholar 

  11. Arya, Y. (2018). Automatic generation control of two-area electrical power systems via optimal fuzzy classical controller. Journal of the Franklin Institute, 355(5), 2662–2688. https://doi.org/10.1016/j.jfranklin.2018.02.004

    MathSciNet  Article  MATH  Google Scholar 

  12. Latif, A., Hussain, S. M. S., Das, D. C., Ustun, T. S., & Iqbal, A. (2021). A review on fractional order (FO) controllers’ optimization for load frequency stabilization in power networks. Energy Reports, 7, 4009–4021. https://doi.org/10.1016/J.EGYR.2021.06.088

    Article  Google Scholar 

  13. Delassi, A., Arif, S., & Mokrani, L. (2018). Load frequency control problem in interconnected power systems using robust fractional PIλD controller. Ain Shams Engineering Journal, 9(1), 77–88. https://doi.org/10.1016/j.asej.2015.10.004

    Article  Google Scholar 

  14. Khadanga, R. K., Kumar, A., & Panda, S. (2021). A novel sine augmented scaled sine cosine algorithm for frequency control issues of a hybrid distributed two-area power system. Neural Computing and Applications. https://doi.org/10.1007/s00521-021-05923-w

    Article  Google Scholar 

  15. Zamani, A., Barakati, S. M., & Yousofi-Darmian, S. (2016). Design of a fractional order PID controller using GBMO algorithm for load–frequency control with governor saturation consideration. ISA Transactions, 64, 56–66. https://doi.org/10.1016/j.isatra.2016.04.021

    Article  Google Scholar 

  16. Sondhi, S., & Hote, Y. V. (2016). Fractional order PID controller for perturbed load frequency control using Kharitonov’s theorem. International Journal of Electrical Power & Energy Systems, 78, 884–896. https://doi.org/10.1016/j.ijepes.2015.11.103

    Article  Google Scholar 

  17. Sharma, M., Prakash, S., Saxena, S., & Dhundhara, S. (2021). Optimal fractional-order tilted-integral-derivative controller for frequency stabilization in hybrid power system using salp swarm algorithm. Electrical Power Components and Systems, 48(18), 1912–1931. https://doi.org/10.1080/15325008.2021.1906792

    Article  Google Scholar 

  18. Charan Patel, N., Kumar Debnath, M., Prasad Bagarty, D., & Das, P. (2018). GWO tuned multi degree of freedom PID controller for load frequency control. International Journal of Engineering & Technology, 7(3.3), 548. https://doi.org/10.14419/ijet.v7i2.33.14831

    Article  Google Scholar 

  19. Guha, D., Roy, P. K., & Banerjee, S. (2020). Quasi-oppositional JAYA optimized 2-degree-of-freedom PID controller for load-frequency control of interconnected power systems. International Journal of Modelling and Simulation, 00(00), 1–23. https://doi.org/10.1080/02286203.2020.1829444

    Article  Google Scholar 

  20. Singh, R., Kesarwani, S. K., Gupta, N. K., & Ashfaq, H. (2020). Design of 2-Dof Pid controller for load frequency control of two area power system using Mfo algorithm. International Journal of Engineering and Advanced Technology, 9(3), 158–161. https://doi.org/10.35940/ijeat.c5009.029320

    Article  Google Scholar 

  21. Sahu, R. K., Panda, S., & Rout, U. K. (2013). DE optimized parallel 2-DOF PID controller for load frequency control of power system with governor dead-band nonlinearity. International Journal of Electrical Power & Energy Systems, 49(1), 19–33. https://doi.org/10.1016/j.ijepes.2012.12.009

    Article  Google Scholar 

  22. Mohapatra, T. K., Dey, A. K., & Sahu, B. K. (2020). Employment of quasi oppositional SSA-based two-degree-of-freedom fractional order PID controller for AGC of assorted source of generations. IET Generation, Transmission and Distribution, 14(17), 3365–3376. https://doi.org/10.1049/iet-gtd.2019.0284

    Article  Google Scholar 

  23. Sahu, R. K., Panda, S., & Padhan, S. (2015). A novel hybrid gravitational search and pattern search algorithm for load frequency control of nonlinear power system. Applied Soft Computing Journal, 29, 310–327. https://doi.org/10.1016/j.asoc.2015.01.020

    Article  Google Scholar 

  24. Sahu, R. K., Panda, S., & Pradhan, P. C. (2015). Design and analysis of hybrid firefly algorithm-pattern search based fuzzy PID controller for LFC of multi area power systems. International Journal of Electrical Power & Energy Systems, 69, 200–212. https://doi.org/10.1016/j.ijepes.2015.01.019

    Article  Google Scholar 

  25. Fathy, A., & Kassem, A. M. (2019). Antlion optimizer-ANFIS load frequency control for multi-interconnected plants comprising photovoltaic and wind turbine. ISA Transactions, 87, 282–296. https://doi.org/10.1016/j.isatra.2018.11.035

    Article  Google Scholar 

  26. Pain, S., & Acharjee, P. (2014). Multiobjective optimization of load frequency control using PSO. International Journal of Emerging Technology and Advanced Engineering, 4(7), 16–22.

    Google Scholar 

  27. Jaber, A. S., Ahmad, A. Z., & Abdalla, A. N. (2013). An investigation of scaled-FLC using PSO for multi-area power system load frequency control. Energy and Power Engineering, 05(04), 458–462. https://doi.org/10.4236/epe.2013.54b088

    Article  Google Scholar 

  28. Khokhar, B., Dahiya, S., & Parmar, K. P. S. (2021). Load frequency control of a microgrid employing a 2D sine logistic map based chaotic sine cosine algorithm. Applied Soft Computing, 109, 107564. https://doi.org/10.1016/j.asoc.2021.107564

    Article  Google Scholar 

  29. Sharma, M., Prakash, S., & Saxena, S. (2021). Robust load frequency control using fractional-order TID-PD approach via salp swarm algorithm. IETE Journal of Research. https://doi.org/10.1080/03772063.2021.1905084

    Article  Google Scholar 

  30. Khokhar, B., Dahiya, S., & Singh-Parmar, K. P. (2020). A robust cascade controller for load frequency control of a standalone microgrid incorporating electric vehicles. Electric Power Components and Systems, 48(6–7), 711–726. https://doi.org/10.1080/15325008.2020.1797936

    Article  Google Scholar 

  31. Mirjalili, S. M., Mirjalili, S. Z., Saremi, S., & Mirjalili, S. (2020). Sine cosine algorithm: Theory, literature review, and application in designing bend photonic crystal waveguides (Vol. 811). Springer.

    MATH  Google Scholar 

  32. Mirjalili, S. (2016). SCA: A sine cosine algorithm for solving optimization problems. Knowledge-Based Systems, 96, 120–133. https://doi.org/10.1016/j.knosys.2015.12.022

    Article  Google Scholar 

Download references

Acknowledgements

Not applicable.

Funding

No funding has been received for this work.

Author information

Authors and Affiliations

Authors

Contributions

NKG: Formulation of the method, Simulation and Result, Paper writing. MKK: Formulation of the method. AKS: Supervisions and Paper correction. All authors read and approved the final manuscript.

Authors’ information

Neelesh Kumar Gupta born in 1991. He received B. Tech degree form UPTU Lucknow, Utter Pradesh, India in electrical and electronics engineering in 2014, M. Tech in Power System Engineering from NIT Jamshedpur, Jharkhand, India, in 2018. Currently he is the Research Scholar in Electrical Engineering department of NIT Jamshedpur Jharkhand, India.

Manoj Kumar Kar born in 1988. He received B. Tech degree from B.P.U.T, Odisha India, in 2010, M. Tech (Power Electronics and Control Drives) from VSSUT, Burla, Odisha, India, in 2012. He is having five years of teaching experience. Currently, he is the Research Scholar in the Electrical Engineering Department, NIT Jamshedpur Jharkhand, India.

Arun Kumar Singh born in 1956. He received his B. Sc. (Engg.) degree from Kurkshetra University, India in 1982, M Tech from IIT BHU, India in 1985, Ph.D. from IIT Kharagpr, India in 1994. He is having more than thirty years of teaching experience. Currently, he is working as Professor in the Electrical Engineering Department, NIT Jamshedpur Jharkhand, India.

Corresponding author

Correspondence to Neelesh Kumar Gupta.

Ethics declarations

Competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix 1

Appendix 1

Three area unequal power system

Tg1 = 0.08; Tg2 = 0.06; Tg3 = 0.07; Tt1 = 0.4 Tt2 = 0.44; Tt3 = 0.3; Kp1 = 105; Kp2 = 100; Kp3 = 120; Tps1 = 20; TPS2 = 22; Tps3 = 20; B1 = 0.3483; B2 = 0.3827; B3 = 0.3692; R1 = 3; R2 = 2.73; R3 = 2.82; T12 = 0.2 T23 = 0.44 T31 = 0.3.

Four area unequal power system

Tg1 = 0.08; Tg2 = 0.06; Tg3 = 0.07, Tg4 = 0.06; Tt1 = 0.4 Tt2 = 0.44; Tt3 = 0.3, Tt4 = 0.44; Kp1 = 105; Kp2 = 100; Kp3 = 120, Kp4 = 100; Tps1 = 20; Tps2 = 22; Tps3 = 20; Tps1 = 22; B1 = 0.3483; B2 = 0.3827; B3 = 0.3692; B4 = 0.3827; R1 = 3; R2 = 2.73; R3 = 2.82; R4 = 2.73; T12 = 0.2 T23 = 0.44 T31 = 0.3; T14 = 0.2.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Gupta, N.K., Kar, M.K. & Singh, A.K. Design of a 2-DOF-PID controller using an improved sine–cosine algorithm for load frequency control of a three-area system with nonlinearities. Prot Control Mod Power Syst 7, 33 (2022). https://doi.org/10.1186/s41601-022-00255-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s41601-022-00255-w

Keywords

  • 2-DOF-PID
  • FOPID
  • Load frequency control
  • WSRT
  • Statistical analysis
  • Improved sine–cosine algorithm