Impact of electrical equipment on the power factor

A large proportion of electricity consumers alter the electricity quality through the equipment they use. The alteration is caused by a drop in the power factor, which leads to additional losses in the power grid and additional reactive power costs. In order to identify the consumers’ electrical equipment that alter the power factor, we monitor their electricity use in a production hall looking at the variation of the effective average values of voltages, currents, powers and power factor. The analysis of these records gives us the difference between the power factor and cos φ (the phase difference between phase voltage and phase current in single-phase operation). Our monitoring shows, at the same time, the necessity of an individual analysis of each electrical equipment in the production hall under observation, as well the necessity to locally compensate the power factor.


Introduction
The quality of electrical energy is becoming more important, lately, especially since more and more processes are automated.Any wide deviations of the electrical energy parameters (voltage, frequency, total harmonic distortion THD, etc.) from their values prescribed by the industrial standards, is considered to be an anomaly negatively influencing the functioning of other electrical equipment in the system, causing additional losses in the power grid.Various methods are used to identify these anomalies, such as waveform analysis, analysis of the mode of variation of the quantities of interest, proximity-based machine learning techniques [1], and so on.
To determine the electricity quality, it has to be monitored for each space of use, such as hospitals [2], educational institutions [3], companies and production premises [4], for group of household equipment [5].
Where production facilities are in discussion, the introduction of automated material processing equipment has brought the benefit of increased productivity by higher working speed, lower number of people involved in the process and higher quality products.The automation of industrial and domestic processes, though, has also led to an increase in electricity consumption [6], and, at the same time, introduced alterations of power quality through the flicker phenomena [7], voltage and current harmonics, power factor decrease, waveform distortion.The electrical power installation of these electric equipment is negatively influenced by electronic components affecting the quality of the power supply.
In order to reduce electricity consumption during the manufacturing process, research has shown that the appropriate choice of processing parameters is decisive to reaching this objective, contributing also to quality improvements of the final product [8].
Solutions to reduce harmonics and increase the power factor include the use of compensation batteries, automatic correction units or compensation batteries controlled by microcontrollers or PLCs [9][10].
Despite the advantages of modernizing and automating manufacturing processes, various disadvantages due to plant operation have been identified.For example, the laser cutting process generated vapors which are disturbing for the workers involved in the process.Moreover, when using high power lasers without adequate protective measures (masks, local and forced ventilation) workers are exposed to toxicity values exceeding the safe values defined in industrial standards [11].
The aim of this paper is to identify the sources that cause an alteration of the electrical energy quality for an industrial consumer, with the aim to find local solutions to increase the power factor.To this end we monitor the energy consumption of a production hall of an industrial consumer.The monitoring used an grid analyzer which recorded the electricity parameters for 24 consecutive days, at every 200 ms.Analyzing the collected data we identified that, during the monitored time period, the electrical equipment were operating at a power factor lower that the limit (neutral) one.This causes additional costs for reactive energy, which necessitates a local compensation.Analyzing the variation modes of the power factors in the sinusoidal and the distorting modes, which has very different values for the non operating times, we see the need to investigate all the equipment in the industrial hall individually.In this way we can identify the electric equipment that introduce distorting power into the power system, and we can recommend the optimal solutions to reduce the inductive or capacitive reactive energy.

The power factor
According to industry regulations, the cost of active and reactive electrical energy is paid by the consumer.Reactive electrical energy can be [12]: -Inductive, when its direction coincides with the direction of active energy, at the point of billing, from the supplier to the consumer; -capacitive, when its direction is opposite to the direction of the active energy, at the billing point; The consumer takes active energy from the grid and feeds reactive energy into it.The limit power factor (previously called neutral), set at 0.9, is the value up to which inductive or capacitive reactive energy must be paid.For the inductive reactive energy, a second threshold is set at 0.65, and below this limit the energy price is three times the price when the reactive power is in the range 0.65 ÷ 0.9.For the capacitive reactive energy there are no thresholds set.When the active power consumption happens at a lower power factor than the set limit, additional energy losses occur in the power grids.To reduce these losses, the energy supplier charges additionally for the reactive energy at a power factor lower than 0.9.
To reduce the additional costs caused when the power factor is below the set thresholds, consumers have the possibility to use different methods of reactive power compensation.
In practice, the difference between cos φ (see further below in this section) and the power factor (PF) is not considered.When no distortion energy is present, these two values coincide.When distortion energy is present, PF is lower than cos φ.As most of the electrical devices and equipment contain electronic components that introduce distorting energy, it is necessary that these two parameters are carefully analyzed.
By definition, the angle φ is the phase difference between the phase voltage and the phase current in a monophase sinusoidal regime.cos φ in the absence of distorting power, depends only on the active power, P, reactive power, Q, and apparent power, S: where S is: The power factor, λ, is defined only for the circuits where the distorting power, D, occurs: where: Thus, for non-sinusoidal systems, we use the power parallelepiped is used for power factor analysis, presented in Figure 1  In this figure, φ is the phase shift angle between the active and the apparent power for systems in which these vary sinusoidally, and Φ is the phase shift angle between active and apparent power for systems where deforming power occurs.
Current regulations do not provide for separate billing of distorting energy, but since: the distorted power occurrence causes a lower cos φ which causes additional costs for the reactive energy.
To determine the difference between cos φ and PF can also be done theoretically using trigonometric functions [14].
Since in our monitoring of the industrial consumer revealed a large amount of reactive energy to be paid for, we looked to identify the sources of inductive and capacitive reactive energy, and we decided to analyze the electricity consumption in each production hall.The paper presents the analysis carried out in the mechanical processing hall.
The industrial consumer takes electricity from the medium voltage supply grid via a step-down transformer.At the same time, a 40 kW photovoltaic system is installed on a production hall to provide part of the electricity needs.
Analyzing the parameters of the supplied and received electrical energy, we found that there are deformations of the current waveforms and alternating values of the reactive power, being sometimes capacitive (negative values) and sometimes inductive (positive values) [15].The variation curve of the reactive energy shown in Figure 2   The main equipment in the hall we analyzed are two mechanical laser processing installations with a power of 2 kW and 1 kW respectively.In addition to these equipment in the hall there are: a compressor for supplying air to the laser cutting installations, a 40 kW Uninterrupted Power Source (UPS), 1 abkant (a bending installation for sheet metal and metal profiles), a travelling crane, a sliding door, an LED lighting system and several monophase (computer, radio) and three-phase (fans for exhausting noxious gases) consumers.Fig. 3 shows the block power schema of these equipment.
The analysis was carried out between 17.03.2022and 30.03.2022.Fig. 4 shows the variation of the average effective values of the phase voltages (first 3 records), the currents on the three phases (records 4-6) and the null current (record 7).
We analyzed the variation of reactive power, shown in Fig. 5, and the variation of cos φ (red line in Fig. 6) and of the PF (blue line in Fig. 6).From this record, two areas can be distinguished: the one corresponding to working hours and the one to the to rest hours.In all cases, cos φ > PF, but during working hours both values are lower, which is explainable.
To detect the source of deforming power, we analyzed how cos φ and PF vary on a working day (Fig. 7).We observe that during working hours the share of deforming power is small (cos φ and PF are almost equal), while outside the working hours the deforming power is considerable.

Results
An analysis of the variation of the voltages on the 3 phases (Fig. 4) shows that they vary between 223 V and 247 V, except on 23.03.2022 at 08:35 when there was a voltage drop of about 29 minutes.The analysis of the variation of the currents on the 3 phases (Fig. 4) shows that they vary between 10.5 A and 63 A on weekdays and between 0.2 A and 2.6 A on weekends and at night.
The analysis of the reactive power variation (Fig. 5) shows that it is inductive, except for the turning on/off moments of the equipment in the hall, when small peaks of capacitive reactive power occur.We detected inductive reactive power peaks whose value is 20.5 kVAr.From our recording it does not appear that there is a distorting reactive power, therefore it was necessary to analyze how cos φ and PF vary.
Looking at the variation of cos φ (Fig. 6) we see that it varies between 0.55 and 0.99 being lower than the limit during working hours.The power factor PF (Fig. 6), which also takes into account the distorting power, has values between 0.34 and 0.9.
From the analysis of the variation of cos φ and PF (Fig. 7) we see that during working hours the share of distorting power is small (cos φ and PF almost equal).However, during non-working hours there is a large difference between cos φ and PF, hence the need to find the source of occurrence of distorting power.
The source of the large distorting power during non working time may be caused by an equipment that has many electronic components it its switching.Hence the need for further investigations.

Conclusion
Analyzing the voltages variations, it is found that they fall within the allowed limits.On further analysis, voltage pulsations have been identified on the T-phase, the source of which needs to be identified.
Since during working hours the value of reactive power is considerable, solutions must be found to compensate for this.
The cos φ variation shows that during operating hours it is lower than the limit.This makes it necessary to increase it and find solutions to reduce the distorting power, especially during nonworking hours.
This requires further research to identify the electric receptors that introduce distorting power and to find solutions for local compensation.

Funding
The publication of this article was supported by the 2023 Development Fund of the UBB.

Figure 2 .Figure 3 .
Figure 2. Variation of the reactive power