现代炼钢电弧炉的基本功能是将尽可能多的电功率输入到熔池内，以获得高的生产率和低的物料、能量消耗以及好的环保指标。炼钢电弧炉按其吨钢平均变压器额定容量, 或单位炉膛面积平均变压器额定容量分为普通功率(RP)、高功率(HP)和超高功率(UHP)三种。超高功率电弧炉概念自70年代提出，目标在于极大地提高电弧炉炼钢的生产率和降低成本，开创了电弧炉炼钢技术发展新纪元。但由于生产时对电网影响与干扰是多方面的，实践中也发现了涉及到电能质量的所有方面。由于超高功率电弧炉的变压器功率水平高，变压器容量高达数十兆伏安，在炼钢过程中对电网造成严重的冲击和干扰，这些“公害”必须加以控制和治理。

1对电网的干扰

1.1功率因数低

电弧炉从电网获得电能，其中一部分转化为有用的热能，而另一部分则为无功能量。为了使电弧能稳定燃烧，电弧炉的功率因数不能取得太高。因电弧炉负载是高感性的，电弧炉的接入使供电电网的功率因数恶化。超高功率电弧炉运行在熔化期时，功率因数甚至低到0.1，这样引起母线电压严重降低。电压降低又相应降低电弧炉的有功功率，使熔化期延长，生产率下降。

1.2电压闪烁和波动

超高功率电弧炉是供电电网的很大的负载，而且在运行中经常产生突然的、强烈的电压冲击，导致电网电压的快速波动，频率为0.1～30Hz。频率在1～10Hz之间的电压波动会引起照明白炽灯和电视画面的闪烁，使人们感到烦躁，这类干扰称之为“闪烁”或“闪变”。强烈的闪烁会造成电机转动不稳定，电子装置误动作甚至损坏，也会使电网供电的用户(包括电弧炉本身)的实际功率减少，闪烁是对电网的一种公害。

对于实际的、有限容量的电网，电弧炉负载引起的电网电压波动百分数为

ΔU%=ΔQ/Sk×100% (1)

式中ΔQ——无功功率的冲击量，Mvar;

Sk——本电弧炉供电母线小短路容量，MVA。

考虑到电弧炉的正常运行状态与短路状态之间的无功功率变动大致与炉子变压器的额定容量相当，即ΔQmax≈SF，故可以用下式来估算电网上大波动百分数

ΔUmax%=SF/Sk×100% (2)

国标《电能质量˙电压允许波动和闪变》(GB12326—2000)规定了电力系统公共连接点的电压波动和闪变电压允许值。超高功率电弧炉在其运行过程中产生的波动和闪变往往都超过这个规定值，必须加以抑制。

1.3三相电压与电流不对称

引起电网供电质量变差的另一个重要问题是电弧炉三相负载不对称引起的电网三相电压及三相电流的不对称。其后果是相关电网的所有用户(包括电弧炉)的经济性和生产率降低。分析表明，在公共连接处电网短路容量Sk比电弧炉变压器额定容量SF高50倍以上时，电弧炉三相负载不对称所造成的电网电压的不对称性未超过国标《电能质量˙三相电压允许不平衡度》(GB/T15543—1995)中规定的允许值。

当前电弧炉大型化、更超高功率化，结果供电电网的容量往往只有电弧炉变压器容量的20～40倍或更低，故必须采用补偿。

1.4高次谐波

交流电弧炉在炼钢过程中其电流会产生非正弦畸变和各次谐波，对电网造成干扰。其主要原因有：

(1)电弧的电阻值不恒定，并且在交流电弧的半个周期中电弧电阻也在变动，这造成电弧电流的非正弦畸变。

(2)交流电的正负半周换相，石墨电极和钢交替作阴极和阳极，因不同材料的发射电子能力不一样，故使电流的正负两个半周的波形不对称，造成偶次谐波。

(3)三相电弧不均衡，导致三次谐波。

(4)供电系统连接的各种谐波源导致各种谐波的形成，如静补装置中的整流器等。

电弧炉的谐波电流成份主要为2～7次，其中2、3次大，其平均值可达基波分量的5%～10%，谐波电流流入电网，使电压波形发生畸变，引起电气设备发热、振动以及保护误动作等。国标《电能质量˙公用电网谐波》(GB/T14549—93)对综合电压畸变率、谐波电流注入量均作了具体规定，为抑制电弧炉产生的谐波提供了依据和标准。

2抑制电弧炉对电网和自身影响的途径

抑制超高功率电弧炉干扰的途径总的来讲有二：一是提高供电电源的电压等级，以提高与电网公共连接点的短路容量，使其对电网和自身的影响在允许范围内;二是采用SVC装置，使其多项指标限制在允许范围内。两种途径相比，途径一是治标的方法，因为电炉对电网和自身影响的各种量值并未消除，而是送到更高电压级的电网去扩散，随着电炉不断建设发展，这些量值在电网中增加积累，泛滥成灾，将会形成电网所不能接受的程度，而增加了对广大用户的影响，因此，使用范围越来越小。途径二是治本的办法，它使电炉对电网和自身影响的各种量值大部分就地消除了，故其使用范围越来越大，前途广阔。

2.1SVC装置

近些年来发展起来的SVC装置是一种快速调节无功功率的装置，已成功地用于电力、冶金、采矿和电气化铁道等冲击性负荷的补偿上，它可使所需无功功率作随机调整，从而保持在电弧炉等冲击性负荷连接点的系统水平的恒定

Qi=QD+QL-QC (3)

式(3)中Qi、QD、QL、QC分别为系统公共连接点的无功功率、负荷所需的无功功率、可调(可控)电抗器吸收的无功功率、电容器补偿装置发出的无功功率，单位均为kvar。

当负荷产生冲击无功ΔQD时，将引起

ΔQi=ΔQD+ΔQL-ΔQC (4)

式中ΔQC=0，欲保持Qi不变，即ΔQi=0，则ΔQD=-ΔQL，即SVC装置中感性无功功率随冲击负荷无功功率作随机调整，此时电压水平能保持恒定不变。

SVC由可控支路和固定(或可变)电容器支路并联而成，主要有四种型式。

(1)可控硅阀控制空芯电抗器型(称TCR型)它用可控硅阀控制线性电抗器实现快速连续的无功功率调节，它具有反应时间快(5～20ms)、运行可靠、无级补偿、分相调节、能平衡有功、适用范围广、价格便宜等优点。TCR装置还能实现分相控制，有较好的抑制不对称负荷的能力，因而在电弧炉系统中采用广泛，但这种装置采用了先进的电子和光导纤维技术，对维护人员要专门培训提高维护水平。

(2)可控硅阀控制高阻抗变压器型(TCT型)优点与TCR型差不多，但高阻抗变压器制造复杂，谐波分量也略大一些。由于有油，要求一级放火，只宜布置在一层平面或户外，容量在30Mvar以上时价格较贵，而不能得到广泛采用。

(3)可控硅开关控制电容器型(TSC型)分相调节，直接补偿，装置本身不产生谐波，损耗小，但是它是有级调节，综合价格比较高。

(4)自饱和电抗器型(SSR型)维护较简单，运行可靠，过载能力强，响应速度快，降低闪变效果好，但其噪声大，原材料消耗大，补偿不对称电炉负荷自身产生较大谐波电流，无平衡有功负荷能力。

2.2无源滤波装置

该装置由电容器、电抗器，有时还包括电阻器等无源元件组成，以对某次谐波或以上次谐波形成低阻抗通路，以达到抑制高次谐波的作用。由于SVC的调节范围要由感性区扩大到容性区，所以滤波器与动态控制的电抗器一起并联，这样既满足无功补偿、改善功率因数，又能消除高次谐波的影响。

国际上用于大型炼钢电弧炉的滤波器种类有：各阶次单调谐滤波器、双调谐滤波器、二阶宽频带与三阶宽频带高通滤波器等。

(1)单调谐滤波器一阶单调谐滤波器的优点是滤波效果好，结构简单，缺点是电能损耗比较大，但随着品质因数的提高而减少，同时又随谐波次数的减少而增加，而电炉正好是低次谐波，主要是2～7次，因此，基波损耗较大。二阶单调谐滤波器当品质因数在50以下时，基波损耗可减少20%～50%，属节能型，滤波效果等效。三阶单调谐滤波器是损耗小的滤波器，但组成复杂些，投资也高些，用于电弧炉系统中，2次滤波器选用三阶滤波器为好，其他次选用二阶单调谐滤波器。

(2)高通(宽频带)滤波器一般用于某次及以上次的谐波抑制。当在电弧炉系统中采用时，对5次以上起滤波作用时，通过参数调整，可形成该滤波器回路对5次及以上次谐波形成低阻抗通路。

用于大型电炉的滤波器组合基本的有两类：一是用3～5组单调谐滤波器组成，二是由2～4组单调谐滤波器和一组二阶宽频带滤波器组成。[敏感词]类组合对高次谐波滤波效果要差一些，但电能损耗低些;第二类组合对高次数滤波效果好，分工也明确，设计也简单容易些。两者组合各有优缺点，总的发展趋势是在滤波效果好的前提下减少组数，以节省占地和投资，又要尽可能优化组合以节省电能损耗。

3有源滤波器

虽然无源滤波器具有投资少、效率高、结构简单及维护方便等优点，在现阶段广泛用于配电网中，但由于滤波特性受系统参数影响大，只能消除特定的几次谐波，而对某些次谐波会产生放大作用，甚至谐振现象等因素，随着电力电子技术的发展，人们将滤波研究方向逐步转向有源滤波器(APF)。

APF即利用可控的功率半导体器件向电网注入与谐波源电流幅值相等、相位相反的电流，使电源的总谐波电流为零，达到实时补偿谐波电流的目的。它与无源滤波器相比，有以下特点：

a.不仅能补偿各次谐波，还可抑制闪变，补偿无功，有一机多能的特点，在性价比上较为合理;

b.滤波特性不受系统阻抗等的影响，可消除与系统阻抗发生谐振的危险;

c.具有自适应功能，可自动跟踪补偿变化着的谐波，即具有高度可控性和快速响应性等特点。

The basic function of modern steelmaking electric arc furnaces is to input as much electrical power as possible into the molten pool to achieve high productivity, low material and energy consumption, and good environmental indicators. Electric arc furnaces for steelmaking are classified into three types based on their average transformer rated capacity per ton of steel or per unit furnace area: ordinary power (RP), high power (HP), and ultra-high power (UHP). The concept of ultra-high power electric arc furnace was proposed in the 1970s, with the goal of greatly improving the productivity and reducing costs of electric arc furnace steelmaking, ushering in a new era of development in electric arc furnace steelmaking technology. However, due to the multifaceted impact and interference on the power grid during production, all aspects related to power quality have also been found in practice. Due to the high power level and capacity of transformers in ultra-high power electric arc furnaces, which can reach tens of megavolt amperes, they cause serious impact and interference to the power grid during the steelmaking process. These "public hazards" must be controlled and treated.

1. Interference with the power grid

1.1 Low power factor

Electric arc furnaces obtain electrical energy from the power grid, with a portion converted into useful thermal energy and another portion being reactive energy. In order to ensure stable combustion of the arc, the power factor of the arc furnace cannot be too high. Due to the high inductance of the electric arc furnace load, the connection of the electric arc furnace deteriorates the power factor of the power supply grid. During the melting period of ultra-high power electric arc furnaces, the power factor can even be as low as 0.1, causing a severe decrease in bus voltage. The decrease in voltage correspondingly reduces the active power of the electric arc furnace, prolonging the melting period and decreasing productivity.

1.2 Voltage flicker and fluctuation

Ultra high power electric arc furnaces are a significant load on the power grid and often generate sudden and strong voltage surges during operation, resulting in rapid fluctuations in grid voltage with frequencies ranging from 0.1 to 30Hz. Voltage fluctuations with frequencies between 1 and 10Hz can cause flickering in incandescent lamps and television screens, causing people to feel irritated. This type of interference is called "flickering" or "flicker". Intense flickering can cause unstable motor rotation, electronic device misoperation or even damage, and can also reduce the actual power of users (including arc furnaces themselves) supplied by the power grid. Flickering is a public hazard to the power grid.

For actual, limited capacity power grids, the percentage of voltage fluctuations caused by electric arc furnace loads is

ΔU%=ΔQ/Sk×100% (1)

In the formula, Δ Q represents the impact of reactive power, Mvar;

Sk - Small short-circuit capacity of the power supply busbar for this electric arc furnace, MVA.

Considering that the reactive power variation between the normal operating state and short-circuit state of the electric arc furnace is roughly equivalent to the rated capacity of the furnace transformer, i.e. Δ Qmax ≈ SF, the following equation can be used to estimate the percentage of large fluctuations on the power grid

ΔUmax%=SF/Sk×100% (2)

The national standard "Electricity Quality - Allowable Voltage Fluctuations and Flicker" (GB12326-2000) specifies the allowable values of voltage fluctuations and flicker at the common connection points of the power system. The fluctuations and flicker generated during the operation of ultra-high power electric arc furnaces often exceed this specified value and must be suppressed.

1.3 Asymmetric three-phase voltage and current

Another important issue causing the deterioration of power supply quality in the power grid is the asymmetry of three-phase voltage and current caused by the asymmetry of three-phase loads in electric arc furnaces. The consequence is a decrease in the economy and productivity of all users of the relevant power grid, including electric arc furnaces. Analysis shows that when the short-circuit capacity Sk of the power grid at the public connection is more than 50 times higher than the rated capacity SF of the arc furnace transformer, the asymmetry of the power grid voltage caused by the three-phase load asymmetry of the arc furnace does not exceed the allowable value specified in the national standard "Electric Energy Quality - Allowable Unbalance of Three phase Voltage" (GB/T15543-1995).

Due to the current large-scale and higher power generation of electric arc furnaces, the capacity of the power grid is often only 20-40 times or lower than that of the arc furnace transformer, so compensation must be adopted.

1.4 High order harmonics

During the steelmaking process, AC electric arc furnaces generate non sinusoidal distortion and various harmonics in their current, causing interference to the power grid. The main reasons are:

(1) The resistance value of the arc is not constant, and the arc resistance also varies during the half cycle of the AC arc, which causes non sinusoidal distortion of the arc current.

(2) The positive and negative half cycles of alternating current are reversed, with graphite electrodes and steel alternating as cathodes and anodes. Due to the varying electron emission capabilities of different materials, the waveforms of the positive and negative half cycles of the current are asymmetric, resulting in even harmonics.

(3) The three-phase arc is unbalanced, resulting in third harmonic.

(4) The various harmonic sources connected to the power supply system lead to the formation of various harmonics, such as rectifiers in static compensation devices.

The harmonic current components of electric arc furnaces are mainly 2-7 times, with the 2nd and 3rd times being larger, with an average value of 5% -10% of the fundamental component. Harmonic currents flowing into the power grid cause voltage waveform distortion, leading to electrical equipment heating, vibration, and protection misoperation. The national standard "Electric Energy Quality - Harmonics in Public Power Grids" (GB/T14549-93) provides specific regulations on the comprehensive voltage distortion rate and harmonic current injection amount, providing a basis and standard for suppressing harmonics generated by electric arc furnaces.

Ways to suppress the impact of electric arc furnaces on the power grid and themselves

There are generally two ways to suppress interference from ultra-high power electric arc furnaces: firstly, to increase the voltage level of the power supply to improve the short-circuit capacity of the common connection point with the power grid, so that its impact on the power grid and itself is within the allowable range; The second is to use SVC devices to limit multiple indicators within the allowable range. Compared with the two approaches, the first approach is a temporary solution, because the various values of the impact of electric furnaces on the power grid and themselves have not been eliminated, but are sent to higher voltage level power grids for diffusion. With the continuous construction and development of electric furnaces, these values accumulate in the power grid and become rampant, which will form an unacceptable level for the power grid and increase the impact on the majority of users. Therefore, the scope of use is becoming smaller and smaller. The second approach is a fundamental solution, which eliminates most of the various values of the impact of electric furnaces on the power grid and themselves on site. Therefore, their use is becoming increasingly widespread and their prospects are broad.

2.1SVC device

The SVC device developed in recent years is a device that quickly adjusts reactive power and has been successfully used in compensating for impact loads such as electricity, metallurgy, mining, and electrified railways. It can randomly adjust the required reactive power to maintain a constant system level at the connection point of impact loads such as electric arc furnaces

Qi=QD QL-QC (3)

In equation (3), Qi, QD, QL, and QC represent the reactive power at the system's common connection point, the reactive power required by the load, the reactive power absorbed by the adjustable (controllable) reactor, and the reactive power generated by the capacitor compensation device, all in kvar.

When the load generates impact reactive power Δ QD, it will cause

ΔQi=ΔQD ΔQL-ΔQC (4)

In the equation, Δ QC=0. To keep Qi constant, i.e. Δ Qi=0, then Δ QD=- Δ QL, i.e. the inductive reactive power in the SVC device is randomly adjusted with the reactive power of the impulse load, and the voltage level can remain constant.

SVC is composed of controllable branches and fixed (or variable) capacitor branches connected in parallel, and there are mainly four types.

(1) The thyristor valve controlled air core reactor type (TCR type) uses a thyristor valve to control a linear reactor to achieve fast and continuous reactive power regulation. It has the advantages of fast response time (5-20ms), reliable operation, stepless compensation, phase separation regulation, ability to balance active power, wide applicability, and low price. TCR devices can also achieve phase separation control and have good ability to suppress asymmetric loads, so they are widely used in electric arc furnace systems. However, this device adopts advanced electronic and optical fiber technology, and maintenance personnel need to be specially trained to improve their maintenance level.

(2) The advantages of thyristor controlled high impedance transformer type (TCT type) are similar to TCR type, but the manufacturing of high impedance transformers is complex and the harmonic components are slightly larger. Due to the presence of oil, it is required to set fire to the first level and should only be installed on the first floor or outdoors. When the capacity is above 30Mvar, the price is relatively expensive and cannot be widely adopted.

(3) The thyristor switch controlled capacitor type (TSC type) has phase separation regulation and direct compensation. The device itself does not generate harmonics and has low losses. However, it has stage regulation and a relatively high overall price.

(4) The self saturating reactor type (SSR type) has simple maintenance, reliable operation, strong overload capacity, fast response speed, and good flicker reduction effect. However, it has high noise, high raw material consumption, and compensates for the large harmonic current generated by asymmetric electric furnace loads, without the ability to balance active loads.

2.2 Passive filtering device

The device is composed of passive components such as capacitors, reactors, and sometimes resistors to form a low impedance path for a certain harmonic or higher harmonics, in order to suppress high-order harmonics. Due to the expansion of the regulation range of SVC from the inductive region to the capacitive region, the filter is connected in parallel with the dynamically controlled reactor, which not only satisfies reactive power compensation and improves power factor, but also eliminates the influence of high-order harmonics.

The types of filters used internationally for large-scale steelmaking electric arc furnaces include single tuned filters of various orders, double tuned filters, second-order wideband and third-order wideband high pass filters, etc.

(1) The advantages of a first-order single tuned filter are good filtering effect and simple structure. The disadvantage is that the power loss is relatively large, but it decreases with the improvement of quality factor and increases with the decrease of harmonic order. The electric furnace happens to be a low order harmonic, mainly 2-7th order, so the fundamental loss is relatively large. When the quality factor of the second-order single tuned filter is below 50, the fundamental loss can be reduced by 20% to 50%, which is energy-saving and has equivalent filtering effect. A third-order single tuned filter is a filter with low loss, but its composition is more complex and the investment is higher. It is better to use a third-order filter for secondary filters and a second-order single tuned filter for other filters in electric arc furnace systems.

(2) High pass (wideband) filters are generally used for harmonic suppression of one or more harmonics. When used in an electric arc furnace system, the filter circuit can form a low impedance path for harmonics of the 5th order and above by adjusting the parameters.

There are two basic types of filter combinations used for large electric furnaces: one is composed of 3-5 sets of single tuned filters, and the other is composed of 2-4 sets of single tuned filters and one set of second-order wideband filters. The first type of combination has poorer filtering effect on high-order harmonics, but lower power loss; The second type of combination has better filtering effect for high order filtering, clear division of labor, and simpler and easier design. The combination of the two has its own advantages and disadvantages, and the overall development trend is to reduce the number of groups while achieving good filtering effects, in order to save land and investment, while optimizing the combination as much as possible to save energy loss.

3 Active Filters

Although passive filters have the advantages of low investment, high efficiency, simple structure, and easy maintenance, and are widely used in distribution networks at present, due to the significant influence of system parameters on filtering characteristics, they can only eliminate specific harmonics, and may have amplification effects or even resonance phenomena on certain harmonics. With the development of power electronics technology, people have gradually shifted their research direction towards active filters (APF).

APF uses controllable power semiconductor devices to inject current into the power grid that is equal in amplitude and opposite in phase to the harmonic source current, so that the total harmonic current of the power supply is zero, achieving the goal of real-time compensation of harmonic current. Compared with passive filters, it has the following characteristics:

a. Not only can it compensate for various harmonics, but it can also suppress flicker, compensate for reactive power, and has the characteristic of multi energy in one machine, which is relatively reasonable in terms of cost-effectiveness;

b. The filtering characteristics are not affected by system impedance and can eliminate the risk of resonance with system impedance;

c. It has adaptive function and can automatically track and compensate for changing harmonics, with high controllability and fast response characteristics.