Search

This paper deals with the Automatic Generation Control (AGC) strategy for multi-source interconnected power system in deregulated environment. Physical constraints like Generation Rate Constraints (GRC), Governor Dead Band (GDB) and Boiler Dynamics (BD) are incorporated in the system for the realistic approach. In this paper, an accurate model of HVDC tie-link is considered for the proposed AGC mechanism. Inertia Emulation based Control (INEC) technique is adopted to utilize the stored energy of HVDC tie-link of the interconnected power system. Distributed Generation (DG) system is also considered in the proposed AGC structure. The DG system comprises of Solar Photovoltaic (SPV), Wind Turbine System (WTS), Diesel Engine Generator (DEG), Aqua Electrolyzer (AE) and Fuel Cell (FC). Finally, a novel application of Volleyball Premier League (VPL) algorithm has been introduced to obtain the optimized gains of the proposed controller. The proposed controller includes the cascade structure of 2-Degree of Freedom Proportional-Integral controller and Fractional Order Proportional-Derivative controller with filter (2-DOF-PI-FOPDN). The robustness and effectiveness of VPL algorithm based proposed controller has been tested by comparing with the published literatures on same platform.

This paper presents a Semi-supervised Learning approach for anomaly detection in the Automatic Generation Control loop of the power systems. The proposed technique is an ensemble method, which employs Gaussian One-class Support Vector Machine, K-Means clustering, and a novel Region Expanding Oversampling technique. Unlike previous research in this field, the proposed method learns from only healthy class data that are readily available in the historical database or can be obtained by offline simulation. The algorithm is proposed keeping in mind that the defenders may not predict the attack sequences ahead in time; hence, the attack instances are not utilised during the training. To validate the proposed method's performance in detecting unseen attacks, apart from general False Data Injection attacks, we have formulated two varieties of intelligently crafted attacks. At first, a coordinated stealthy attack is designed that has the potential to degrade system stability. Secondly, a new financially orientated market operation attack is formulated that induces higher financial settlements for the attacker-preferred generation companies. Finally, the performance of the proposed ensemble method is compared with existing one-class classifiers, which reveals the superiority of the proposed ensemble framework in terms of average geometric accuracy (98.79%), True Positive Rate (98.38%), and True Negative Rate (99.22%).

The paper discusses practical implementation and computational aspects of an AGC scheme based on model predictive control (MPC) for a real-world power system, namely the all-island Irish transmission system. This system is particularly interesting from the frequency regulation point of view due to the high penetration of wind power generation. This leads to a significant reduction in system inertia, which in turn impacts on the ability of the system operator to regulate the system frequency. The paper compares the performance of the MPC-based AGC with a conventional AGC based on PI controllers and draws relevant conclusions. The case study is a high-resolution simulation of a 1479-bus model of the Irish electrical grid with significant wind penetration, stochastic loads, and controller delays. European Commission Horizon 2020 Science Foundation Ireland

From the perspective of power balancing, we consider in this paper the timescale of real-time dispatch in power systems with high penetration of intermittent power sources. Due to the integration of intermittent power sources, real-time dispatch must be employed in the process of power balancing. The hierarchy of power system operation is typically comprised of the short-term scheduling, the real-time dispatch, and the automatic generation control (AGC). Therein, the AGC is the last-level defense of the systemwide power balancing, which has limited power adjustment capability. Consequently, the real-time dispatch must eliminate as much power imbalance as possible, and its capability of doing so depends on its timescale. Based on the statistical forecast uncertainty functions, a quantified relation between the uncertainty level of the power system and the critical timescale for real-time dispatch is established in this paper. Simulations demonstrate the effectiveness of the proposed formula.

With the increased share of power generation based on wind energy, the complexity of the unit commitment (UC) and economic dispatch (ED) problems increases due to the stochastic nature of wind power. Therefore, an accurate and fast optimization method is needed when the generation process involves large quantities of wind sources to effectively manage generation mix and load requirements. In this paper, model predictive control (MPC) is used to solve power system UC/ED problems with the presence of wind energies. Because the UC/ED problem is nonlinear, non-convex and mixed integer problem, the MPC must be used as nonlinear. To produce a nonlinear MPC (NMPC), MPC must be integrated with a fast optimization methodology. This paper presents a generic mathematical formula for a NMPC, integrated with swarm optimization technique to describe the nonlinear behavior in the mathematical formulation. This new formulation will be called swarm model predictive control (SMPC) optimization. The control model will be able to address the effect of the system disturbances and fluctuations using a controlled autoregressive integrated moving average (CARIMA). A general form of future predictions can be expressed as a function of input and output past data, and a future control sequence, and the degree of freedom in the SMPC problem. Also, the prediction part improves the swarm technique, because it identifies the size of search space in a better way. In this paper, UC schedule is designed using the swarm technique offline, while ED is solved using the proposed SMPC optimization method on real-time basis and fed into the automatic generation control system. There is no load deficit in real time ED results in less spinning reserve requirements compared with swarm optimization and dynamic program techniques used discretized load. The system under study is a standard IEEE 30 bus modelled in MATLAB environment, with all data from the city of Florida, USA.

The power grid is a dynamic system encompassing numerous control devices with varying degrees of time constants, posing a challenge for dynamic simulators to efficiently step through time. We introduce a robust event-driven simulator to mimic dynamic simulation that exploits the inherent temporal sparsity by solving for the quasi-steady state of the grid using frequency-dependent power flow at each time-step. In this methodology, the effect of fast transient control actions is captured within the quasi-steady state while a time-dependent outer loop handles slow transients that dynamically change over time. The event-driven approach offers a fast simulation framework capable of scaling to large systems by taking time-steps based on proceeding events and uses a robust power flow simulator, Simulation with Unified Grid Analyses and Renewables (SUGAR), to solve for the quasi-steady state. We demonstrate the efficacy of this approach by simulating the effects of different control actions including automatic generation control (AGC), automatic voltage (AVR) and over excitation limits on large systems.

In traditional power system operations real-time control policies are based on simple control policies such as automatic generation control and/or local voltage control. Although these simple policies have worked well in the past, the increased variability associated with higher penetrations of renewable energy strengthens the case for more general approaches. In this paper we analyze the design of real-time control policies through the lens of stochastic AC-OPF. Using a chance-constrained approach in combination with a reformulation based on polynomial chaos expansion, we show that the solution corresponds to a general polynomial control policy. By restricting this general policy, we are able to recover the traditional (affine) control policies as a special case. This allows to quantify the value of implementing non-affine policies, and to analyze their impact on the system behavior, as demonstrated in the numerical example.

This paper presents an analysis and systemization of automatic generation control (AGC) in distribution networks (DNs) with high penetration of distributed resources, including electric vehicles (EVs). A methodology is developed that allows designing the AGC service at the distribution level, and an optimization model is proposed to assess the potential of AGC provision from EVs according to an objective of optimal economic management. A realistic case study is considered to analyze the proposed approach, and to illustrate both the potential of the methodology and the effectiveness of the optimization model. Results show that the proposed methodology represents a flexible tool that any system operator could use for the operational planning and the management of ancillary services such as AGC with EVs.

Existing reliability indices for power systems provide measures of long-term stability; in this work, we propose a new reliability criterion that estimates the probability of a spinning-reserve shortage occurring, thus indicating the short-term stability of the grid frequency. In the formulation, load, wind power, and vehicle-to-grid (V2G) power are represented as random variables, and conventional generators are incorporated in a deterministic manner. Using the equilibrium of demand and supply and the physical constraints of automatic generation control (AGC), two inequalities are derived, from which the probability of successful frequency regulation is obtained. The inverted probability, referred to as the failure rate for frequency regulation (FRFR), is employed as a metric in the short-term reliability criterion. Then, the developed criterion was applied in several case studies. First, the impact of wind-power deployment was estimated in terms of the required spinning reserve. The acceptable penetration level of wind power was then investigated in the case that V2G power is also present. The variation of FRFR with respect to the commitment level of thermal (non-renewable) power plants was also investigated under the renewable portfolio standard (RPS) to illustrate the recursive effect of the policy. Finally, FRFR was calculated for the IEEE reliability test system and compared with conventional reliability indices.

This paper presents a design of a sliding mode controller (SMC) with chattering reduction feature applied to interconnected automatic generation control (AGC). After formulating the design of SMC as an optimization problem, the proposed method utilizes the genetic algorithms (GA) to find the optimal feedback gains and switching vector values of the controller. In order to guarantee the enhancement of the system dynamical performance and a reduction in the SMC chattering, two objective functions were investigated in the optimization process. The tested two-interconnected AGC model incorporates nonlinearities in terms of generation rate constraint (GRC) and a limiter on the integral control value. Comparison with previous AGC methods reported in literature validates the significance of the proposed SMC design.

In this paper, a comparative transient performance of two types of single machine power system is analyzed. The two types of system configurations are viz. (i) automatic voltage regulator (AVR) loop with single input conventional power system stabilizer (CPSS) combined with automatic generation control (AGC) loop, (ii) AVR with CPSS combined with capacitive energy storage (CES) unit-based AGC loop. For AGC loop both thermal unit and hydro unit are individually considered. The thermal unit is considered with either single or double stage reheat turbine, hydro unit is considered with mechanical or electric governor. Integral controller is provided in the AGC loop. It is shown that the CES-based integral controlled AGC loop along with CPSS assists in the best transient performance of the power system in all cases under different operating conditions. Parameters of the CPSS installed in AVR loop, gains of the integral controller in AGC loop are optimized with the help of a novel particle swarm optimization, developed by the authors, called as craziness-based particle swarm optimization (CRPSO). Transient performance is carried out with 1% step perturbation in either reference voltage or load torque.

This paper proposes an approach of optimal sensitivity applied in the tertiary loop of the automatic generation control. The approach is based on the theorem of non-linear perturbation. From an optimal operation point obtained by an optimal power flow a new optimal operation point is directly determined after a perturbation, i.e., without the necessity of an iterative process. This new optimal operation point satisfies the constraints of the problem for small perturbation in the loads. The participation factors and the voltage set point of the automatic voltage regulators (AVR) of the generators are determined by the technique of optimal sensitivity, considering the effects of the active power losses minimization and the network constraints. The participation factors and voltage set point of the generators are supplied directly to a computational program of dynamic simulation of the automatic generation control, named by power sensitivity mode. Test results are presented to show the good performance of this approach.

Fixed gain controllers for automatic generation control are designed at nominal operating conditions and fail to provide best control performance over a wide range of operating conditions. So, to keep system performance near its optimum, it is desirable to track the operating conditions and use updated parameters to compute control gains. A control scheme based on artificial neuro-fuzzy inference system (ANFIS), which is trained by the results of off-line studies obtained using particle swarm optimization, is proposed in this paper to optimize and update control gains in real-time according to load variations. Also, frequency relaxation is implemented using ANFIS. The efficiency of the proposed method is demonstrated via simulations. Compliance of the proposed method with NERC control performance standard is verified.

Under a deregulated environment, ancillary services (AS) are defined as the set of products separated from the energy production, which are related to security and reliability of a power system. Ancillary services can be classified in: (a) active power ancillary services (load–frequency control (LFC)—including primary control, automatic generation control (AGC), tertiary control, balancing service and black start provision) and (b) reactive power ancillary services (voltage control). The system operator is the entity responsible for the secure operation of the power system and in this way, the management of all the AS is considered a specific function of the system operator. Under a deregulated framework, ancillary services are separated from the energy production, and can be mandatory or remunerated under market-driven mechanisms. This paper overviews the management of ancillary services in the Spanish power system, including a detailed technical description of the services and the organization of the ancillary services markets. In addition, a comprehensive review of different optimization algorithms and tools used by the agents and system operator within the Spanish electricity business is presented.

The paper presents a new technique of automatic generation control of an isolated small-hydro power plant. Presently surplus generation due to decrease in load is dissipated in a dump load. But surplus water can be used for irrigation, which is the primary requirement of the local community for their survival at large number of locations. It is being wasted in the dump load and the new scheme proposed in this paper eliminates the dump load, while the input power of the hydro plant is controlled by an on/off and a servo controlled valves. The on/off control linearly raises or lowers the generation by 50%, while the servo motor controls the generation to the required extent. It is shown that the rate of opening or closing of the on/off control valve has considerable effect on the transient response of the system. Finally, the dynamic responses due to step disturbance for different nominal loadings of the system, gain settings, rate of closing/opening of the on/off control valve are also presented.

The conventional model of a power system control area used for automatic generation control (AGC) is crucially extended to include individual power producers. Taking into account the influence of the power network, a systematic method of obtaining a detailed model of a control area is presented. It is proved that within the framework of the model developed, both frequency control as well as regulation of the power of individual power stations becomes possible. To this end, new advanced optimal control methods are used for the design of suitable proportional–integral (PI) controllers. Using the proposed model and the new control strategy, the application of automatic generation control and its effectiveness is extensively studied and evaluated on a 6-bus illustrative power system example; furthermore, the method is tested on the New England power system, which is commonly used as a large test system.

This paper presents a load frequency control design for interconnected electric power systems using a set of fuzzy logic rules. The design objectives are (i) to comply with the North American Electric Reliability Council's (NERC) control performance standards, CPS1 and CPS2, (ii) to reduce wear and tear of generating unit's equipments, and (iii) to design a feasible control structure. A test system with multiple generation and distribution companies that takes into account regulation and load following services is used to demonstrate the effectiveness of the proposed methodologies.

In this paper, various novel heuristic stochastic search techniques have been proposed for optimization of proportional–integral–derivative gains used in Sugeno fuzzy logic based automatic generation control of multi-area thermal generating plants. The techniques are classical particle swarm optimization, hybrid particle swarm optimizations and hybrid genetic algorithm simulated annealing. Numerical results show that all optimization techniques are more or less equally very effective in yielding optimal transient responses of area frequency and tie-line power flow deviations. The gains obtained by particle swarm optimization are more optimal than those obtained by GA/hybrid GA-simulated annealing. Particle swarm optimizations take the least time to achieve the same optimal gains. These gains are for nominal system parameters. For varying off-nominal on-line system parameters, fast acting Sugeno fuzzy logic manipulates the nominal gains adaptively to determine transient responses.

A continuous and near instantaneous balance between generation supply and load demand must be maintained in power system operation. As a kind of regulation reserve, automatic generation control (AGC) is widely employed for maintaining system security and reliability, especially for controlling frequency deviation and supporting commercial transactions. In a competitive electricity market, AGC services are either contracted via negotiations between the market operator and generation companies concerned, or procured via a certain form of auction. A new framework to determine AGC capacity requirement and to deal with regulation dispatch under uncertain demand in a competitive electricity market is presented in this paper. First, the stochastic characteristics of AGC capacity requirement and services are analyzed. A stochastic optimization decision-making model for determining optimal AGC capacity requirement is then developed, followed by a pricing analysis which reveals that a trade-off between economic efficiency and system security must be made by the market operator. Finally, an optimal AGC dispatch algorithm is suggested. A numerical study is given to further demonstrate the essential features of the proposed model and method.

Optimal integral gains (for integral gain control) and proportional-integral-derivative gains (for PID control) are computed by genetic algorithm (GA) and then hybrid genetic algorithm-simulated annealing (GA-SA) techniques for nominal values of area input parameters and optimal transient responses of area frequency deviations in terms of settling times, undershoots, overshoots and d f /d t as output with incremental increase of area load for interconnected three equal generating areas. Though it is well known that the normal PID control is usually superior to integral control because of the advantages of each of the three individual control actions (proportional, integral and derivative), the author’s contribution in the paper is optimizing these individual PID gains through GA or GA-SA methods to obtain an optimal PID controller, which would be further better than an optimal integral controller. These optimal PID gains are tested by plotting transient responses analytically by MATLAB based software program and then by “SIMULINK of MATLAB software.” Both methods yield same results and prove that optimal PID gains are superior to suboptimal, arbitrary PID gains and optimal integral gains as well with respect to transient responses. The author’s next contribution is to show optimal PID gains as determined by hybrid GA-SA technique to be more globally optimal than those determined by GA method. For off-nominal input parameters, transient responses as determined by fast acting Sugeno fuzzy logic technique reflect the same superiority of GA-SA based optimized gains, specially for PID control, the same has also been verified by “MATLAB–SIMULINK” software.

This paper is devoted to the development of a neural network structure which implements the line,power flow state estimator algorithm for solving the set of nonlinear equations of power system automatic generation control analysis. The principal context is that of on-line network analysis in energy management systems with particular reference to the automatic generation control function. The author shows that the complete line power flow state estimator formulation maps into an array of two-layer neural networks. The development starts from a formulation for solving as a minimization the equation system to which the state estimator sequence leads at the each iteration. A neural network structure is given which implements the steepest descent method for minimizing the objective function. It is shown that some of the input values of neural networks are formed from power flow measurements as on-line. A principal feature of the extensive parallel processing capability of the architecture developed is that the computing time of state estimator analysis is independent of the number of nodes in a power network for which analysis is carried out. (C) 2003 Elsevier B.V. All rights reserved.

The evolution of many voltage collapse scenarios on large power systems can be alleviated by controlled load shedding associated with rearrangement of generator outputs. Dynamic simulation of power system mid-term voltage response is required to adequately determine the impact of load shedding actions and to design appropriate control systems. The dynamic simulation must take into account load-voltage behaviour, transformer tap changer controls and limits, generator over-excitation limiters, automatic generation control and system protective devices. This paper demonstrates the impact of load shedding as a corrective action through simulation of the system dynamic response to a disturbance. Whilst load shedding can stabilise a system, rigorous techniques are required to identify the magnitude and location of shedding. The application of voltage modal analysis in combination with the determination of reactive power margins in resolving this problem is demonstrated.

This paper presents a new nonlinear convex network flow programming (NLCNFP) model and algorithm for solving the on-line economic power dispatch and automatic generation control (AGC) with security and economy. Two AGC functions—load frequency control (LFC) and economic dispatch control (EDC) are modeled on the same timescale as EDC. This new AGC modeling and solution algorithm meets not only the control requirement of frequency and tie-line power as in the conventional load frequency control, but also the line security and system economy. The proposed approach is tested on an IEEE 30-bus system with satisfactory results.

This paper deals with the application of genetic algorithms for optimizing the parameters of conventional automatic generation control (AGC) systems. A two-area nonreheat thermal system is considered to exemplify the optimum parameter search. A digital simulation is used in conjunction with the genetic algorithm optimization process. Several integral performance indices, or cost functions, are considered in the search for the optimal AGC parameters. The work is further extended to include the use of more elaborate feedback control strategies, such as the proportional-plus-integral type, within the decentralized frame. The results reported in this paper have not been obtained before and they demonstrate the effectiveness of the genetic algorithms in the tuning of the AGC parameters.

This paper describes a control scheme based on a generalized minimum variance strategy for the load frequency and tie-line power control of a multi-area interconnected thermal power system. Each control area in the interconnection is represented by a reduced-order stochastic AR-MAX model with parameters estimated using an extended least squares technique. The area generator corrective control is computed by a self-tuning algorithm derived from the minimization of a generalized cost function, which is defined in terms of the system output and the weighting on the control effort. The control algorithm is further coupled with a simple load accommodation routine to account for the non-zero mean component of the load disturbances. Simulation studies show that, compared with conventional methods, the proposed control strategies are more effective, especially when system non-linearities are taken into account.

A describing-function method is applied to the study of the existence and behavior of limit cycles observed in power systems. A parameter sensitivity study was conducted on a system model of an isolated area. Results on the qualitative behavior of parameter changes are presented.

A decentralized scheme is proposed for the design of automatic generation control (AGC) for interconnected power systems. This scheme uses the natural structural properties of interconnected power systems and gives a new method for decentralized AGC regulator design, using the existing theory of proportional-plus-integral linear regulators applied to each individual power area. It is shown that a decentralized design, based upon the expansion-contraction theory of large-scale systems with overlapping and modified linear regulators with incomplete subsystem feedback, is suitable to handle the design of AGC regulators for present-day interconnected power systems. The resulting AGC retains the autonomous area control concept and load distribution property of conventional AGC regulators, while at the same time allows for improved transients and stability margins for interconnected areas. The application of this decentralized AGC scheme is illustrated on a 17th-order model of a two-area interconnected power system.

In this paper, a systematic method that predicts limit cycle oscillations in a hydroelectric system with governor deadband on the parameter plane is presented. A new technique for constructing limit cycle loci on this plane is also proposed. This approach is based on the characteristic equation as well as the describing function approximation and avoids the use of derivation and separation of stability equations. For each family of solution loci in the parameter plane, only two integrations are needed. The stability of the limit cycle can also be determined easily from the plane using the gradient approach. Finally, the proposed method is employed to investigate the influence that several variations in the modeling of a hydro system may have on the existence of limit cycles. The results from this study will shed some light on the selection of automatic generation control (AGC) parameters as a remedy for this phenomenon.

This paper presents a new approach to the automatic generation control of power systems. It makes use of the theory of variable structured systems to investigated the automatic generation control of a steam electric power plant. Improvements in both the dynamic and static performances can be achieved by means of the sliding mode provided by a variable structure controller. The main feature of the sliding mode is that the system willhave low sensitivity; certain desirable properties can be obtained by minimizing a quadratic performance index specified in the sliding mode. For comparison, results obtained using a convetional controller and a linear optimal controller are also included.