Optimum placement of the capacitor with the aim of improving the voltage profile and minimizing the power losses of the real distribution network by modeling different loads

Dissertation for receiving a Master's degree (M.Sc.)

Direction: Power Electronics

Abstract:

Distribution networks have various loads that consume different amounts of reactive power. they do Considering the effect of the type of load on the location and size of capacitor banks in distribution systems, in this thesis, a novel method for modeling different network loads in order to improve the voltage profile and reduce power losses in the presence of shunt capacitors is carried out. For this purpose, two models are presented: a) commercial-household-agricultural-public-industrial and b) constant impedance - constant current - constant power. In fact, the main advantage of this thesis is to bring the study closer to the real world, because almost all the studies conducted in the field of optimal placement of the capacitor have basically ignored the effect of the type of load on the location and optimal capacity of the installed capacitor, while this thesis will prove that considering a different load model will have an effect on the location/capacitance of the capacitor. Another feature of this thesis is the formulation of the objective function as a multi-objective problem. For this purpose, voltage deviation has been added to the single-objective function. The above problem will be solved using the particle swarm optimization algorithm [1] (PSO). In order to prove the effect of the proposed load modeling on the answers to the problem of optimal placement of the capacitor, the following scenarios are used: load modeling and without it with and without the presence of the capacitor. style="direction: rtl;"> 

1-1 Preface

The analysis of distribution networks is one of the main concerns of network operators. A system engineer must know information about the number, size, location and type of network elements in order to analyze the distribution network. Since the majority of distribution systems are practically radial, in order to model and analyze the network, various challenges must be overcome, which include [1]:

System

The distribution network has a wide variety of components that have complexity and large dimensions. For example, most connected loads are fed from only one of the three phases.

Load distribution

Load distribution in feeders and branches is usually not the same and therefore the distribution system is unbalanced. Unbalanced systems are inappropriate.

Data

           The entire distribution system operates with minimal monitoring and control. Therefore, the available real system data

           is very limited for their modeling and analysis.                                                                      

           Designers need detailed information to model and design alternatives for complex systems, to

           in order to overcome various problems associated with the design, operation and control of distribution systems.   

           The answer to some types of problems such as conductor selection, voltage regulation, capacitor placement requires models and

                                                                                                                                                                                                                   .   

            Some reasons for the complexity of distribution system design aspects are presented below.

Commercial aspects of distribution systems

Due to the internal connection of various distribution network equipment, one component affects the technical and economic performance of other components in the system, which makes evaluating the technical-economic solution a complex process.

Size

Many components make up a distribution system. This means that identifying and analyzing alternatives is very difficult.It is very difficult and sometimes impossible to evaluate all the possible alternatives.

Uncertainty

In long-term planning, it is necessary to predict the improvement in the future. Any uncertainty affects the distribution network design. Sometimes, the best design, load and economic predictions do not include the uncertainty about future system loading for improper design of the distribution network. Therefore, it is necessary to design the distribution system in order to satisfy future demand by avoiding uncertainty in load forecasting and other aspects of distribution networks.

Other aspects of safety

In the design process, power system designers should consider other aspects of safety including legal authorities, social groups, intervention of business leaders and other services in calculations.

1-2 A study of the effect of capacitor on the network

The experience of distribution network designers in the form of an expert system is expressed in the form of a number of rules with the title of basic law [1]. Many concepts in the design of distribution networks are in agreement with demand. The total power and production demand is not simply "sum of consumers". The demand of a consumer (which is referred to in the term "load") is not exactly satisfied by the producer (which is referred to as "useful power"). Useful power is the active power required to produce useful work, it is the power consumed in the resistive part of the circuit. Power tools are not ideally resistive and have reactance, which leads to the consumption of reactive power in the circuit [2]. Generally, a source should supply the mixed or apparent power of all consumers. The apparent power of a point load (consumer) is obtained from the following relationship:

(1-1)

where,

Si: apparent power in node i in kVA

Pi: active power required in Node i in kW

Qi: reactive power required in node i in kVAr

Electrical companies have found that it is not economically possible to produce reactive power in the power plant and supply all consumers far from the substation. Installation of capacitor banks is very economical in the absence of loads or reactive load centers of the power system. System reactive power can be sent in two ways. One method is to eliminate the reason or need for it, which is done by managing part of the load impedance. This work can be done by compensating the induced reactance by the post using series capacitors as in figures (1-1) and (1-2). This method is called series compensation. Abstract: Distribution networks have various loads which the loads consume different levels of reactive power. By considering the impact of type of load on location and size of capacitor banks in distribution systems, in this thesis, a novel technique has been proposed for modeling various loads to improve voltage profile and reduce power loss of distribution network in the presence of shunt capacitor. For this, two models have been presented; Model A: Commercial - Domestic - agricultural - General - Industrial, Model B: Constant impedance - Constant Current - Constant Power. Indeed, the main challenge of this work is present a practical study, because all studies of optimal capacitor placement have negligible impact of load model on location and capacity of capacitor banks. While this thesis will prove the location and size of installed capacitors change by considering the load model. Another contribution of this work is formulating objective function as a multiobjective problem. For this, the voltage deviation has been added to the single objective function of optimal capacitor placement. Particle Swarm Optimization (PSO) algorithm has been used to solve the problem.