Dynamic load controllers for thermostatically controlled loads should allow for accurate control of power consumption and should not disrupt the quality of service. This paper proposes an intuitive definition of nondisruptiveness for systems with second-order thermal models, based on a decomposition into fast and slow temperature modes. It enables the explicit control of the slow mode temperature using an embedded first order model; control of the fast mode is implicit. Temperature bounds are derived, and the slow mode controller is implemented using an accurate decentralised stochastic control strategy. Simulation results confirm its accuracy and nondisruptiveness.

Collectively, thermostatically controlled loads (TCLs) offer significant potential for short-term demand response. This intrinsic flexibility can be used to provide various ancillary services or to carry out energy arbitrage. This study introduces an aggregate description of the flexibility of a heterogeneous TCL as a leaky storage unit, with associated constraints that are derived from the TCL device parameters and quality of service requirements. In association with a suitable TCL control strategy this enables a straightforward embedding of TCL dynamics in optimisation frameworks. The tools developed are applied to the problem of determining an optimal multi-service portfolio for TCLs. A linear optimisation model is constructed for the optimal simultaneous allocation of frequency services and energy arbitrage. In a case study, optimal service allocations are computed for eight representative classes of cold appliances and the results are validated using simulations of individual refrigerators. Finally, it is demonstrated that clustering of appliances with similar capabilities can significantly enhance the flexibility available to the system.

Modern active distribution networks make use of intelligent switching actions to restore supply to end users after faults. This complicates the reliability analysis of such networks, as the number of possible switching actions grows exponentially with network size. This paper proposes an approximate reliability analysis method where switching actions are modelled implicitly. It can be used graphically as a model reduction method, and simulated using time-sequential or state sampling Monte Carlo methods. The method is illustrated on a simple distribution network, and reliability indices are reported both as averages and distributions. Large speedups result from the use of biased non-sequential Monte Carlo sampling - a method that is hard to combine with explicit switching models.

The large-scale integration of intermittent energy sources, the introduction of shiftable load elements and the growing interconnection that characterizes electricity systems worldwide have led to a significant increase of operational uncertainty. The construction of suitable statistical models is a fundamental step towards building Monte Carlo analysis frameworks to be used for exploring the uncertainty state-space and supporting real-time decision-making. The main contribution of the present paper is the development of novel composite modelling approaches that employ dimensionality reduction, clustering and parametric modelling techniques with a particular focus on the use of pair copula construction schemes. Large power system datasets are modelled using different combinations of the aforementioned techniques, and detailed comparisons are drawn on the basis of Kolmogorov-Smirnov tests, multivariate two-sample energy tests and visual data comparisons. The proposed methods are shown to be superior to alternative high-dimensional modelling approaches.

Thermostatically controlled loads (TCLs) such as refrigerators and air conditioners are natural candidates for short term demand response. In this paper we quantify the value associated with the TCLs' ability to provide system security and transmission constraint management services. The analysis builds on recent results that enable the aggregate control of TCLs as a leaky storage unit. We incorporate this model in a security constrained economic dispatch (SCED) that minimizes the system operational cost of a two bus-bar system, subject to frequency response and transmission constraints. Further sensitivity studies assess the impact of different penetration levels of controllable loads and transmission flow constraints.

Future power systems will have to integrate large amounts of wind and solar generation to drastically reduce CO₂ emissions. Achieving this goal comes at the cost of a reduced level of the system inertia and an increased need for fast response services. Previous research has shown the effectiveness of thermostatically controlled loads (TCLs) providing frequency response, and the ability to accurately control the aggregate power consumption of TCLs. In this paper, we explore the design space of frequency response patterns of flexible TCLs. Two distinct frequency response implementations are presented. The first makes the TCLs' power consumption a linear function of system frequency and/or its rate of change; in the second, TCLs respond to a frequency event tracking a pre-programmed reference power profile. Computer simulations illustrate strengths and weaknesses of the proposed implementations in the context of the GB 2020 Gone Green scenario.

System Protection Schemes (SPS) have the potential to greatly enhance the utilization of the network, often by automatically disconnecting generators in response to contingency events. However, malfunctions of such systems may expose the system to harmful blackouts. The operation of unreliable SPS is therefore subject to a cost-benefit balance between the benefits of increased system utilization and the risk of outages. This paper studies this trade-off in a year-round basis. The problem is firstly stated from a centralized perspective to probabilistically minimize the operational costs for a whole operating year. A case study based on a basic simple representation of the Great Britain system is considered. The results show great annual benefits from equipping the SPS with multiple generation disconnection systems, which are mainly associated with critical operating conditions. However, it is demonstrated that redundant SPS configurations do not necessarily reduce the levels of operational risk exposure.

Thermal loads such as refrigerators and electric space heaters use temperature hysteresis controllers that are insensitive to small temperature fluctuations. This results in an ability to modulate their power consumption, thus providing cost-effective frequency support, balancing services and energy arbitrage. In order to partially realise these benefits, ENTSO-E has proposed a mandatory frequency support service for thermal loads in its Network Code on Demand Connection. This is to be implemented as a proportional shift of the setpoint temperature in accordance with frequency deviations. In this paper we argue that this implementation choice results in an unpredictable response that depends strongly on controller details. Furthermore, it restricts the flexibility to implement advanced controllers that deliver multiple services simultaneously. We present a case study that demonstrates very different frequency response patterns from three controllers that are each compatible with the proposed Code. Alternative implementations of the code and controllers are presented to illustrate the scope for improvement.

Thermostatically controlled loads (TCLs), such as refrigerators, air-conditioners and space heaters, offer significant potential for short-term modulation of their aggregate power consumption. This ability can be used in principle to provide frequency response services, but controlling a multitude of devices to provide a measured collective response has proven to be challenging. Many controller implementations struggle to manage simultaneously the short-term response and the long-term payback, whereas others rely on a real-time command-and-control infrastructure to resolve this issue. In this paper, we propose a novel approach to the control of TCLs that allows for accurate modulation of the aggregate power consumption of a large collection of appliances through stochastic control. By construction, the control scheme is well suited for decentralized implementation, and allows each appliance to enforce strict temperature limits. We also present a particular implementation that results in analytically tractable solutions both for the global response and for the device-level control actions. Computer simulations demonstrate the ability of the controller to modulate the power consumption of a population of heterogeneous appliances according to a reference power profile. Finally, envelope constraints are established for the collective demand response flexibility of a heterogeneous set of TCLs.

Generating capacity adequacy studies play a significant role in long term capacity planning. Risks of capacity deficits are usually reported in the form of one or more average quantities, which cannot fully convey the nature of the risks being faced. Chronological Monte Carlo simulations may be used to construct comprehensive multi-dimensional risk profiles, but such profiles tend to be difficult to interpret. This paper proposes the use of a clustering method to partition the risk profile into clusters of similar outcomes with associated probabilities. The results are presented in accessible tabular form, and prototypical scenarios can be analysed in detail to provide further insight.

Power systems are facing complex challenges in order to achieve environmental targets, such as a strong abatement of CO₂ emissions. Hence, an increasing share of renewable energy sources (RES) will change the generation portfolio. This is likely to push network operation into a risky territory, due to a lack of system inertia. This paper quantifies the impact of providing 'effective inertia' by controlling domestic thermostatic loads. This inertial support mechanism would permit full absorption of available wind power even for those scenarios characterized by low demand and high wind penetration.

An increasing share of nuclear and renewable generation is reducing the ability of the power system to withstand the sudden loss of a generating unit. Hence, there is a need for additional response and reserve services. Previous research has shown that frequency-responsive thermostatic loads would be able to support the system, but the proposed control methods typically suffer from undesirable side-effects at the device or system level. We propose a hybrid controller that addresses the observed shortcomings and illustrate its effectiveness using simulations in a model system. Quantitative comparisons are made with two control schemes from the literature.