Methods of analysis of the power system security

Increasing the intelligent level of the PS control systems caused by the implementation of Smart technologies changes the structure and the properties of PS and increases the importance of system reliability analysis. System reliability analysis includes two components – the balance analysis and the regime analysis. On the one hand, there are a large number of studies that assess the reliability of the power system examining various aspects and methods of solving this problem. In practice, the security analysis is limited by the calculations of power flows, static and dynamic stability for a number of forecast periods for the normal and repair circuits considering the most severe disturbances. The existing approach allows defining the requirements and adjusting emergency control systems, but does not allow evaluating and comparing solutions for power grid constructions. The authors propose a new method for power system reliability evaluation, which is suitable for planning development and operation of power systems. The method includes a general description of the algorithm which allows to compare various development scenarios, as well as to assess the reliability level of their implementation.


Introduction
The rise of intelligent level of the power system (PS) control due to the implementation of Smart technologies and conception of intelligent power system (IPS) of active-adaptive grid (AAG) changes the structure and characteristics of PS [1][2], simultaneously increasing the importance of assessment power system security, as aspect of system reliability. System reliability (PS reliability) includes balance and regime components.
In Russia, in accordance with [3], power system security is understood as a property of the system to maintain the prescribed modes of operation when conditions change, element failures and sudden perturbations occur. In other countries, the term security has a similar meaning, which is understood as the ability of the system to withstand sudden disturbances [4][5][6]. There is distinguish between static and dynamic power system regime security. In assessing static regime security, dynamic transient processes in PS are not taken into account, the model of the system is limited to the equations of steady-state. Dynamic regime security is related to the dynamics of PS behavior, when a nonlinear model of transient processes in PS or its linearized version can be considered.
There is a great number of studies on assessment of power system reliability (in particular [7][8]) dealing with various aspects and methods for solution of this problem. In practice, the security analysis is limited by calculations of static and dynamic stability for a number of forecast periods for normal and repair circuits considering the most severe disturbances [9]. Such an approach allows us to define requirements and adjust emergency control (EC). However, it cannot assess and compare solutions for power grid constructions.
The authors propose a new method to determine application of various aspects of security at different stages of development planning and during operation power system. A technique and an algorithm are proposed that allow to determine the magnitude of possible under-supply of electricity for various emergency situations, taking into account the emergency control and the expected actions of the dispatcher. This approach allows us to determine the mathematical expectation of annual damage from all accidents in PS, resulting in a shortage of electricity.
The authors propose the new method of steady state calculation of PS taking into account the discrete and interval characteristics of the parameters of the regime. Using this method it is possible to perform the analysis of system reliability assessing the control of PS and finding different ways of automatic or operational control allowing PS to move from pre-emergency to acceptable post-emergency state in the event of emergency disturbances.

Reliability criteria for power system expansion planning
The most important criterion for decision-making in operation and in planning of power system development is system reliability and PS viability. Moreover, both balance and regime aspects of system reliability should be provided [5][6][7][8]. Planning is carried out at various levels with different time interval, therefore, various aspects of system reliability are considered for different types of planning. When choosing the sufficiency of measures of power grid construction in the development process of power systems, the algorithm given in Fig. 1 is used.  We distinguish two components in assessment of PS security: Assessment of PS security in the aspect in dynamics process of its transition to postemergency state after emergency disturbances: evaluation of disturbance consequences (including static and dynamic stability analysis), operation of relay protection (RP), emergency control (EC), automatic voltage regulators (AVR), automatic frequency and power control (AFPC), sources of reactive power and on-load tap-changers (OLTC). Changes of generator power are recorded with primary regulation within the primary reserve.
Assessment PS security in the aspect of its recovery process after the potential emergencies and transition from post-emergency to normal state: changes of generators power are recorded with primary, secondary and operational regulation within primary and secondary reserves.

Methods and assessment criteria of PS security in the aspect of its recovery process after the potential emergencies and transition from post-emergency to normal state
At the moment of emergency disturbance by RP and EC operation, partial load shedding (power consumers) is possible as well as output of regime parameters (current, voltage and frequency) beyond permissible values. After RP and EC response, automatic regulators and regime automatics (RA) together with the operational-dispatcher personnel mobilize hot reserve at power plants. Regime parameters are inserted into permissible limits and power consumers disconnected during the emergency are connected. According to the rules [10,11], it takes 20 minutes to carry out these measures. However, if this time is not enough, the forced regime of power system is introduced and other 20 minutes (totally 40 minutes from the beginning of the emergency) are spent for normalization of the post-emergency regime.

Original regime
Search for standard disturbances (1, 2 and 3 groups) or search for disturbances from criteria "n-1" and "n-2" Calculation of post-emergency steady regime with RP and EC modelling Analysis of post-emergency steady regime and transient process (permissibility of regime parameters, preservation of static and dynamic stability, etc.) Summary statistics and identification of security parameters Input of post-emergency regime in the feasible area (imitation of action of operational-dispatcher personnel) using reserves of primary and secondary regulation The authors propose to use as one of the PS security criteria the ability of the PS to recover to normal state after an emergency disturbance for 20 (40) minutes, taking into account the operation of the systems of RS, PA, RA and automatic regulators, and further implementation of operational activities by operational dispatch personnel. It is necessary to check if the electricity supply to consumers of the 2nd and higher categories of reliability comply with the requirements for reliability. If it is provided for particular pre-emergency scheme-regime conditions and particular emergency disturbance, it is considered as a sufficient PS security, i.e. it does not require additional power grid construction for improved reliability. If after 20 (40) minutes the restrictions for consumers remain, the volume and probability undersupply of electrical energy are determined, taking into account the subsequent mobilization of all types of reserves.
The authors propose [12] the algorithm for determining the probabilistic characteristics of electrical energy undersupply by iterating over all emergency disturbances and the integration of the probabilistic volume of undersupply ( • in some cases, if the calculating model does not switch on distribution systems, besides changes of the grid topology and generation, the load redistribution between the nodes is specified (thus, changes of the distribution grid topology is considered including those of the grids of internal power supply of consumers with several supply centers).

Enumeration of emergency disturbances caused by long-term grid tripping is specified. Every disturbance is specified by:
• enumeration of tripped grid elements (branches of the calculating model, e.g. TL with tapping is modelled with several nodes and branches); • probability of concrete emergency disturbance; • expected time of recovery. 9. Search for emergency disturbances (Item 8) is specified for the original (normal) scheme and for all specified control schemes (Item 7) by turn: • pre-emergency regime is assessed and changes are overlaid in the control schemes (Item 7, tripping/switching on of branches and changes of generation and load); • emergency disturbances are searched by turn (Fig. 2) overlaying tripping specified for a concrete disturbance (Item 8); • EC operation is modelled during estimation of the steady regime (Item 5, in case of its actuation for a concrete disturbance); • if the iterative process is non-convergence during estimation of the steady regime, it is necessary to model EC (Item 5, using optimization method and method of directed or complete search) to reach post-emergency steady regime; Where Y 0.RF and Y 0.Reg is the energy intensity of the gross domestic product (GDP) for the country as a whole, and the energy intensity of the gross regional product (GRP) for specific subjects of the Federation. For large or particularly important industrial enterprises, the amount of damage is determined individually, but not below the energy intensity.
The expectation value of the annual loss from all emergencies in the power system associated with the electrical power undersupply will be determined from the arithmetical sum of expectation values of annual losses from concrete emergencies (ML ann. emerg. No. xxx ): ..
1    n ann emerg ann emerg №i i

ML ML
Besides the loss of the expected electrical power undersupply, it is necessary to take into account a probable loss from unforeseen tripping of consumers or generation. It is suggested to consider this loss only in those cases and volumes which are stipulated in the agreements (agreements of power supply or technological connection). If these aspects are not stipulated in the agreement, the loss is not considered. The maximum allowed time of electrical power undersupply for concrete consumers should correspond to agreement conditions and category of consumers (according to the decrees of the Russian Federation Government).

PS steady state calculation with taking into account the discrete and interval characteristics of the regime parameters
The solution of the problem of paragraph 9 is proposed to perform by PS steady state calculation calculating taking into account the discrete and interval characteristics of the regime parameters [13][14].
In general, the balanced steady state parameters as well as their functions should be as close as possible to  The steady state of PS is usually described by the equations of balance of currents or powers in the grid nodes and in vector form: here, the vector