Planning of safe working space for the hot-line working robot ICBot

Abstract

With the popularization of intelligent substations, hot-line working robots that can replace the manual operation without shutting down the power of the corresponding equipments are increasingly widely applied in the maintenance and monitoring of substations. Robotic operation can ensure the safety of the operators and reliability of operation while improving the economic benefits of substation. However, because of the compact structure of the intelligent substations, the operating space is small. Further, as the main material of hot-line working robots is metal, which greatly changes the strength and distribution of the electric field around its working position when entering the working position and in the process of operation. In serious cases, interphases or phase to ground air breakdown occurs. Hence the working space of hot-line working robots is limited not only by the narrow space between devices in the substation but also by the electrical safety clearance of phase to phase and phase to ground. Therefore, it is necessary to find a safe working space (“SWS”) for the robot which should not be too small for the robot to accomplish the task effectively. The purpose of this paper is to provide: (i) a design reference for hot-line working robots which can not only guarantee the electrical safety distance but also minimize the effect to the electric field and (ii) a method to find the SWS of the robot in which the robot can accomplish the task effectively with the example of the hot-line working robot for isolated circuit breaker (“ICBot”). We analyze the influence of different installation positions and postures of the manipulators on the electric field in this paper and present a calculation method of the SWS of the manipulator in a very non-uniform electric field caused by the intervention of manipulators, which is significant to the research on hot-line working robots. The SWS has been tested in simulated and energized experimental system, and the experiment results show that the robot can work safely and efficiently.

Keywords

Hot-line robot electric field distortion air breakdown substation insulation

1. Introduction

There is increasing needs for robots capable of performing complex and dangerous maintenance tasks. An autonomous hot-line working robot could replace workers in some risky tasks involving high voltage equipments, which can significantly reduce the time cost of maintenance. Many hot-line working robots were designed for different operation tasks. Since 1984, Kyushu Electric Power Company (KYUDEN) has introduced hot-line maintenance techniques [1] which realized semi-automatic operation for distribution lines in various outdoor facility conditions in the natural environment. A high-performance robot presented by LineScout Technology [2] is designed to undertake detailed and comprehensive inspections of transmission lines. The flying robot could track and inspect a desired transmission lines automatically [3,4]. There is no need to analyze the SWS for these robots because the distance between phases of transmission lines are much larger than the size of the robots and the influence caused by distortion of the electric field around the transmission lines caused by the intervention of the robots on the grid can be neglected.

There are also types of hot-line working robots designed to operate in the power distribution system. Wang Gang, Zhang Hongwei [5] introduced a dual arm robot with master-slave made for maintenance of high voltage power line. Xiao Lv and Lu Shouyin [6] designed a set of master-slave manipulator which can implement the hot-line working operation in substations. The posture of slave manipulator can be adjusted by the operator controlling the master manipulator. The most widely used robots in substation are insulator cleaning robots for suspension insulator strings [7–10].

The manipulators of some of these robots that contact the hot-line directly are made of insulating materials. Others are working mid-potentially or earth potentially. Therefore the analysis of the SWS of these robots are also unnecessary.

But the ICBot is made of metallic material and it will work equipotentially with hot-line in a compact substation. Therefore, the top priority in the design of such hot-line working robot is to guarantee the safety of the equipments in the substation. To achieve such purchase: (i) the robot must not contact any object except the device that need to be maintained; (ii) the robot must keep sufficient electrical safety phase to phase and phase to ground distance so that no air break down will occur when the metal robot causes the distortion of electric field. By analyzing the structure of the substation and transforming the coordinates of each object to the coordinate system of the robot, the robot will know clearly where the obstacles is. Keeping sufficient electrical safety phase to phase and phase to ground distance is relatively difficult due to the uncertainty and invisibility of the electric field distortion. During the operation of the robot, the electric field around the equipment under maintenance will change distinctly. The electric field strength of certain points on the robot or the equipments are so large that may cause corona or air breakdown and lead to short circuit. Therefore, we need to analyze the electric field under the conditions of different attitude of the robot during the operation to make sure that the field strength of any point in the working space is within limits. Meanwhile, the working space should not be too small for the robot to finish the task.

In this paper, we study how the electric field is influenced when the ICBot approaches the working position and operates the connect fittings in 110 kv substation. We present a fast method to calculate the SWS in a non-uniform electric field, which is verified in simulated energized experimental system voltage. The experiment results show that the robot can work safely and efficiently in the SWS we have calculated.

2. Electric field analysis

In traditional substations, when an isolation circuit breaker (ICB) need to be separated from the bus bar, operators need to manually screw the bolts by using a long insulating rod attached with a wrench. In such circumstance, the operation may not be reliable and the operators may not be safe. An appropriate robotic assembly can undoubtedly improve the reliability of operation and safety of operators. As shown in Fig. 1, a set of fittings includes two parts: an L-type Shaft (LS) and a cable clamp (CC) which are installed on porcelain insulator at the height of 5.2 m. The CC is fixed on a porcelain insulator connected to the bus bar and the LS is connected with the ICB through an aluminium conductor steel reinforced (ACSR). With four bolts screwed on the CC, the ICB can be easily connected with the bus bar by fastening the LS or disconnected by pulling out the LS.

Fig. 1.

The structure of connect fittings.

Taking the disassemble of connect fittings for example, to accomplish the operation, manipulators are needed for carrying the end-effectors. One good layout of the manipulators is that one screws the bolts from the top of the CC and the other installed on the side of the operation platform pulls out the LS as shown in Fig. 2b. Such dual manipulator system contains two KUKA industrial manipulators with six DOF. Manipulator 1 (A1) carries a camera, a light source, two screw fixtured nutrunners as its missions are positioning and screwing bolts; manipulator 2 (A2) is responsible for carrying removable part of connect fittings. There are a clamping jaw and a force sensor on the end-effector of the A2. Since the platform has to work equipotentially (hot-line working), all the above mentioned devices are powered by batteries and inverters which are also equipped on the platform. As to the electric elements which can be integrated into the metallic control cabinet such as batteries, inverter, controller of manipulators and motors, PLCs and switches, we believe they are safe from the risk of induced potential difference. For the devices that cannot be put into the cabinet such as the fixtured nutrunners, motors, force sensor and cables, we try to make a “Faraday cage” for each of them by wrapping them with copper foil to avoid potential difference inside these devices. All the shells, copper foil and cabinet must be in conduction to avoid potential difference.

Fig. 2.

Two main processes during the operation. (a) ascending process. (b) operation process.

Almost all the equipments are made of metal. The other main part of the ICBot is an autonomous mobile carrier which can carry and lift the platform to the operation spot as shown by the yellow arrow in Fig. 2a. Four 1.5 m length composite insulators which can guarantee the phase to ground safe clearance are equipped between the mobile carrier and the operation platform. The robot is composed of three main parts: the operating platform with a dual manipulator system, the insulating lift and the autonomous mobile carrier which can carry and lift the platform to the operation spot. The precision of the autonomous navigation system will affect the SWS of the dual manipulator system. The composite insulators of the insulating lift can guarantee the electrical safety clearance of phase to ground.

Based on the structure of ICBot, the distortion of the electric field can be analyzed. As shown in Fig. 2a, the platform is lifted by the autonomous mobile carrier before the operation. During this process, the platform go through the following three stages: earth potential, intermediate potential and 110 kv potential. The phase-ground safe clearance can be guaranteed at earth or 110 kv potential by the composite insulators. We focus on the distorted electric field caused by the platform in intermediate potential. Considering the electric field is mainly affected by metal, we remove the non-metal objects such as the insulator in Fig. 2a and 2b and establish the simulation models in ANSOFT MAXWELL respectively as shown in Fig. 3.

Fig. 3.

The simulation models of two main processes during the operation. (a) ascending process. (b) operation process.

2.1. Analysis of electric field during ascending process

In order to analyze the influence of the platform’s shape (different postures of the manipulator) on the electric field, we have conducted simulations with different postures of the platform model, the sliding table on which does not stretch forward. The material of each object in the simulation model is as shown in Table 1.

Table 1
The material of each object in the simulation model

Platform ACSR GIS pipe Environmental

Steel Aluminum Steel Air at STP

Platform	ACSR	GIS pipe	Environmental
Steel	Aluminum	Steel	Air at STP

The ACSRs on top of the robot are charged with 110 kV three-phase AC voltage, and for the purpose of the simulation we select the most dangerous moment in its period, which is when the voltage of the ACSR right above the robot reaches the maximum (in 110 kV substation, the maximum instant phase to ground voltage is $110\times \sqrt{2}/\sqrt{3}$ kV). According to the three-phase voltage law, at this point of time, the adjacent phases voltage is half the maximum value and in the opposite direction. Therefore we apply −45 kv, 90 kv and −45 kv voltage excitation on A, B and C phase, respectively. The voltage excitation on the GIS pipe is 0V because it is grounded. The simulation result is shown in Fig. 4 and Table 2:

Fig. 4.

Electric field simulations with different postures of the manipulators during the ascending process. (a) The platform is close to the ACSR (110 kV energized). (b) The platform is close to the GIS pipe (grounded). (c) The platform is far away from the ACSR and the GIS pipe.

Table 2

Simulation results during the ascending process

Pos	D ₁ (mm)	D ₂ (mm)	D ₁ + D ₂ (mm)	E _P (kV/cm)	E _A (kV/cm)
(a)	788	808	1596	1.45	4.29
(b)	1325	305	1630	0.46	3.89
(c)	1330	798	2128	0.44	3.89

D ₁: The minimum distance between the platform and the ACSRs; D ₂: The minimum distance between the platform and the GIS pipe. E _P: Maximum electric field strength on the operation platform, E _A: Maximum electric field on the surface of ACSRs;

As shown in Table 2, among postures (a), (b) and (c) in the platform’s ascending process, the smallest values of E _P, E _A and U _R are obtained when the platform is in the state as shown in Fig. 4c when the value of D ₁ + D ₂ is the biggest. Therefore, in the design of equipotential operation platform, as long as the combined safe clearance of phase to ground (D ₁ + D ₂) can be guaranteed, the operation of the robot will be safe when the platform rises. If there is a telescopic mechanism or moveable device on the platform, it is necessary to keep them at a posture that makes the operation platform as compact as possible to ensure the combined safe clearance.

2.2. Analysis of electric field during operation process

When the robot operates on the connect fittings, the whole platform is connected to and equipotential with the corresponding ACSR. Therefore, the platform and the ACSR can be considered as a whole and be applied with voltage excitation together in the simulation.

When the robot is working on the connect fittings of phase B, it influences the electric field around phase A and phase C at the same time. Therefore, this paper focuses on the robot’s the influence on the electric field during its operation on phase B. The robot operates on phase A to C in sequence. While it operates on phase B, the connect fitting and transmission line attached to phase A have been removed, and phase A can be regarded as one point in the simulation.

As shown in Fig. 2b, A2 is located between the phase B and C, which shortens the safe distance between phase B and C. Therefore, we selected the most dangerous situation to conduct the simulation, which is when the potential difference between phase B and C reaches the maximum value. We apply voltage excitation of 0 V, $55\sqrt{2}$ kV and $-55\sqrt{2}$ kV on phase A, B and C, respectively (see section 3.1.3 below for details).

We conducted the following two sets of simulations to find out the influence of changes of the posture of A2 on the strength and distribution of the electric field.

In simulation 1 as shown in Fig. 5, the position of the elbow joint of A2 is fixed, and we move the end-effector towards the adjacent ACSR. In simulation 2 as shown in Fig. 6, the position of the end-effector of A2 is fixed, and we move the elbow joint toward the adjacent ACSR. The simulation results are as shown in Fig. 5, Fig. 6, Table 3 and Table 4:

Fig. 5.

Simulations with different distances between the end-effector and ACSR.

Fig. 6.

Simulations with different distances between the elbow joint and ACSR.

Table 3

Simulations with different distance between the end-effector and ACSR (joint fixed)

Pos.	D _j (mm)	D _e (mm)	E _max (kV/cm)	E _j (kV/cm)	E _e (kV/cm)	E _A (kV/cm)
(a)	1005	1166	8.39	4.45	5.65	8.39
(b)	1005	1005	9.41	4.07	5.31	9.41
(c)	1005	642	10.52	2.75	8.20	10.52

Table 4

Results of simulations with different distance between the elbow joint and ACSR (end-effector fixed)

Pos.	D _j (mm)	D _e (mm)	E _max (kV/cm)	E _j (kV/cm)	E _e (kV/cm)	E _A (kV/cm)
(a)	1360	1220	8.17	2.42	6.82	8.17
(b)	1005	1166	8.39	4.45	5.65	8.39
(c)	971	1180	9.65	6.25	4.43	9.65

The simulation results shows the following rules:

1. The electric field strength on the surface of the manipulator increases as its distance from the ACSR shortens.

2. The maximum electric field strength in the environment always appears on the ACSR.

3. With the shortest distance between the arm and the cable fixed, electric field strength increases as the average distance between each point of the manipulator and the ACSR shortens.

4. The electric field strength is bigger around the surface of objects with less radius of curvature.

Therefore, a hot-line working robot needs to be small, and sharp corners, angles and edges should be avoided on its components. When the operation platform ascends (intermediate potential), it should be as compact as possible, which means all the scalable components thereof should retract as much as possible to maximize the safe clearance. During the operation process, the robot is equipotential with the corresponding ACSR. The interphase clearance apart from the adjacent ACSR changes with the vacation of the size of the operation platform and the posture of the extending parts thereon. Therefore, any movement of the components of the platform should be strictly limited within a certain space (SWS) to ensure the safety of the operation.

Hot-line working robots with different functions have different structures and different SWS. In this paper, we calculate the SWS of ICBot. Based on the simulation results, it is clear that the safety clearance can be ensured as long as all components are retracted during ascending process (as shown in Fig. 4c). But in the operation process, the distance between A2 and the adjacent ACSR has great influence on the maximum strength of the electric field. Therefore the key point in the planning of the SWS is the calculation of the shortest safe distance between A2 and the adjacent ACSR.

3. Risk evaluation indexes

3.1. Air breakdown model

Normally air medium is widely used as an insulating medium in different electrical power equipments and overhead lines. When air molecules become ionized in a very high electric field, the air changes from an insulator to a conductor and provides a path whereby charges can flow between electrodes. In the past several decades, extensive amount of research work has been done to understand the fundamental characteristics of the electrical breakdown. Townsend theory and Streamer theory explain the mechanism of air breakdown under different conditions such as temperature and pressure, etc. [11,12]. In this paper, we analyze the air breakdown model based on the following three elements: gap length, uniformity of electric field and excitation voltage.

3.1.1. Gap length

The electric field conditions, in which an electrical discharge can develop and propagate in air gap, change by orders of magnitude with the gap length: in small gaps (millimeter range) the average electric field needed for breakdown is around 3 × 10⁶ V/m [13–15]; Its exact value varies with the shape and size of the electrodes and increases with the pressure of the air. In larger gaps (up to one meter) it becomes five time lower, around 5 × 10⁵ V/m, while it decreases to 1 × 10⁵ V/m for a ten meters gap [16]. In this paper, the distance between the ACSRs of the two phases is 1.7 meters, which will be reduced to 1.4 to 0.5 meters as the posture of the A2 changes during the operation. Therefore, we regard it a long gap air breakdown model.

3.1.2. Uniformity of electric field

Different shapes of electrodes lead to different nonuniformity of electric field. Plate-plate and rod-plate are two typical models of electrodes, which lead to uniform and non-uniform electric field, respectively. The experiments [17–19] show that, with the same gap length, the breakdown voltage in non-uniform electric field is much lower than that in uniform electric field. The electric field discussed in this paper is non-uniform because the two electrodes are the manipulator and the ACSR of adjacent phase.

3.1.3. Excitation voltage

Air breakdown voltage also varies with the waveform of the excitation voltage. DC, power frequency (PF) and impulse are three typical waveforms. In uniform electric field, the breakdown voltage (BDV) of DC, PF and impulse is nearly the same. But in non-uniform electric field such as that caused by rod-plate electrodes, the BDV of PF is equal to the BDV of DC when the rod is charged with positive voltage, which is lower than the BDV of DC when the rod is charged with negative voltage due to the polarity effect [20,21]. In the situation of small rod-plate air gaps, the BDV of impulse voltage is relatively higher than that of dc voltage [22]. The BDV of impulse may be lower than the BDV of PF when the wavefront timeT1 is within a certain range.

In 110 kV substations, the effective value of voltage between two phases of incoming line is 110 kv. This means the maximum voltage drop U _m between the manipulator and the ACSR of the adjacent phase is $110\sqrt{2}$ kV during the operation of the substation, which will not be affected by operating overvoltage and thunder overvoltage because of the protection by the lightning arrester. The above mentioned voltage can be simplified as a $110\sqrt{2}$ kV DC voltage in the simulation.

Based on the above, the air breakdown model presented in this paper has the following features: (i) long gap air breakdown, (ii) non-uniform electric field, (iii) the voltage drop between two electrodes is $110\sqrt{2}$ kV DC voltage. Assuming U _m is between phase B and C, the voltage on phase A, B and C will be 0 V, $55\sqrt{2}$ kV and $-55\sqrt{2}$ kV, respectively as decided by the features of three phase electricity. The influence by phase A and the ground is small enough to be neglected in this model.

3.2. Long gap air breakdown

Although the critical distance dmin between the manipulator and the ACSR which leads to the air breakdown can be calculated and simulated precisely according to the their structure and the excitation voltage. However, it cannot be ensured that air breakdown will not occur as long as the manipulator keeps a distance larger than dmin. Experiments show that air breakdown occurs randomly under same conditions. Therefore, we need to find a dmin bearing no risk of air breakdown to guarantee the absolute safety of the operation. The breakdown in the gap is initiated by the breakdown of micro-projections on the cathode [23]. The physical mechanism of the arc needs to be studied from the microscopic level, in order to analyze the motion characteristics of the arc more accurately [24]. An air breakdown in long gap and non-uniform electric field go through four subsequent phases: (i) first corona inception and development, (ii) stem formation, (iii) leader head propagation, (iv) leader channel evolution [25]. According to G. Carrara and L. Thione’s experiments [26] as shown in Fig. 7, nothing is observed before the applied voltage is up to U _i (corona inception voltage), when a corona burst is seen on the energized electrode tip. They believed that the 50% breakdown voltage is considered as the sum of three components: U ₁ (leader inception voltage), ΔU _l (increase of voltage during leader propagation), ΔU _s (statistical increase of voltage). Each of them is related to different stages of the discharge phenomenon. Consequently, air breakdown will not occur before corona inception. Therefore, we believe the operation of the robot is safe if the maximum electric field strength on the surface of both electrodes is smaller than the corona inception electric field strength of them. Different from that of the ACSR, the structure of the manipulator electrode can be modified to have a higher corona inception electric field and lower surface field strength. This means the first corona will appear on the ACSR while the surface electric field strength of it increases.

Fig. 7.

Image converter picture of a discharge on a gap of length d. stage a: absence of phenomena, T: time U: voltage; stage c: corona, primary dark periods, T _i, U _i: corona inception; stage i: leader stem, secondary dark periods, T ₁, U ₁: leader inception; stage l: leader continuous propagation, T _c, U _c: continuous leader inception; stage f: final jump, T _f, U _f: final jump inception, T _b, U _b: Breakdown.

3.3. Surface electric field of ACSR

3.3.1. Corona inception electric field of ACSR

Empirical methods are generally used [27] to determine the corona onset gradients of practical transmission-line conductors. The generic empirical formula for the corona onset gradient in kV/cm of cylindrical conductors is of the form: $\begin{eqnarray}\displaystyle E_{c}=mE_{0}{\sigma}\left(1+\frac{K}{\sqrt{{\sigma}r_{c}}}\right) & & \displaystyle\end{eqnarray}$ (1) where:

E _c and K: empirical constants depending on the nature of the applied voltage;

r _c: the conductor radius (in centimeters);

m: the conductor surface roughness factor;

σ: relative air density, given as $\begin{eqnarray}\displaystyle {\sigma}=\left(\frac{273+t_{0}}{273+t}\right)\cdot \left(\frac{p}{p_{0}}\right) & & \displaystyle\end{eqnarray}$ (2) where:

t temperature of ambient air (in °C)

p pressure of ambient air (in Torr);

t ₀ reference values, usually 20 °C

p ₀ 760 Torr.

In reality, the power transmission line is not a perfect cylindrical conductor. It is made up of a group of stranded wires [28]. Figure 8 shows the cross section of ACRS. According to Tiebing Lu’s research [29], the electric field strength of the surface of real wires is bigger than that of the simplified model. In order to calculate the electric field strength of the surface of real wires, rough coefficient is used which is the ratio of the maximum electric field of the surface of cylindrical conductors and that of stranded wires. The rough coefficient of ACSR is about 0.71. It can be seen from Eq. (1) above that the conductor surface roughness factor has a predominant influence on E _c. The following values of the empirical constants are assumed for both positive and negative corona: E ₀ = 30 kV/cm; K = 0.3 [30].

Fig. 8.

Cross section of ACSR.

We set the range of t to −20 °C to 40 °C and p to 788 Torr (corresponding to altitude of −300 m) to 674 Torr (corresponding to altitude of 1000 m), respectively. As a result, the minimum corona onset gradient E _c of the ACRS is 22.46 kV/cm.

3.3.2. Simplification of the electrodes

The ACSR electrode can be simplified to a cylinder (C ₁) that has same diameter with the ACSR. The surface electric field of ACSR is influenced by three main elements: the excitation voltage of both electrodes, the gap distance and the structure of the other electrode which is the manipulator. We simplify the complicated structure of the manipulator for the purpose of the calculation and simulation.

As shown in Figs 5b and 6b, in the case of equal minimum distance between two electrodes, the maximum electric field strength on the surface of the ACSR E _max is stronger when the average distance between the manipulator and ACSR is shorter. This can also be proved by using charge simulation method to calculate the electric field strength [31,32]. Based on this theory, the manipulator can be simplified to a cylindrical electrode (C ₂) as shown in Fig. 9. Such cylindrical electrode is parallel to the C ₁ and has equal diameter (slightly smaller than 100 mm) with the manipulator A2. The simplified model leads to a stronger E _max, which means the safe clearance calculated based on this model can ensure the safety on the real electrodes.

Fig. 9.

Simplification of the electrodes.

Considering the distance between the axes of two cylinders is much larger than the radius of each cylinder, the model can be simplified to two cylindrical electrodes with same radiuses r (12.6 mm) and distance in between as d. Then the maximum electric field strength on the surface of C ₁ is [33]: $\begin{eqnarray}\displaystyle E_{max}=\frac{V}{d}\cdot \frac{\sqrt{\left(\frac{d}{2\cdot r}\right)^{2}-\left(\frac{d}{2\cdot r}\right)}}{\ln \left(1+\frac{d}{2\cdot r}+\sqrt{\left(\frac{d}{2\cdot r}\right)^{2}-\left(\frac{d}{2\cdot r}\right)}\right)}. & & \displaystyle\end{eqnarray}$ (3) To verify the relationship of d and E _max, the same model is simulated in ANSOFT MAXWELL. The results are recorded as E _{max_s}. Since the electrodes in the model are cylinders with smooth surface (the rough coefficient of which is 1), the real maximum electric field strength on the surface of the ACSR is: $\begin{eqnarray}\displaystyle E_{\max \_\text{ACSR}}=E_{max}/m. & & \displaystyle\end{eqnarray}$ (4) The results are listed in Table 5.

Table 5

Maximum electric field strength of different gap distance

d (mm)	E _max (kV/cm)	E _{max _s} (kV/cm)	E _{max_ACSR} (kV/cm)	E _{max_ACSR_S} (kV/cm)
400	18.27	19.68	26.11	28.11
500	16.74	17.25	23.91	24.64
600	16.05	16.23	22.92	23.18
700	15.5	15.05	22.14	21.5
800	15.02	14.69	21.45	20.98
900	14.65	14.33	20.92	20.47

Based on the results listed in Table 5 and the corona onset gradients E _c (22.46 kV/cm), we can find that as long as the distance between the axes of two cylinders is larger than 700 mm, the maximum electric field strength on the surface of the ACSR will be less than E _c. There will be no corona on each cylinder. Consequently, there will be no air breakdown between them. Since the manipulator A2 causes weaker E _max than cylinder C ₂ does, the manipulator A2 will not cause corona nor air breakdown. Due to the dimension of the electrodes, under such circumstances, the distance between the surface of electrodes is 600 mm. Therefore, the SWS of the ICBot is where the distance between it and adjacent ACSR is greater than 600 mm.

4. Experiment

We have carried out multiple experiments in the equal proportion simulated experiment system. We simulate the porcelain insulators of bus bar with a set of composite insulators. A ICBot shown in Fig. 10 includes the components mentioned above: two KUKA manipulators (KR 10 R1100 SIXX), two screw fixtured nutrunners (Ingersoll Rand ETD ST), one camera (AVT Manta G-125B/C) and one clamping jaw (IAI).

Fig. 10.

The experiment system.

The diameter of the ACSR is 25.2 mm. The distance between the connect fittings of phase B and C is 1.7 meters. We charge the ACSRs of phase B and C as we did to the simulation model in Section 2.2. The temperature and air pressure is 8 °C and 766 Torr.

During the experiments, we change the posture of A2 to vary the distance d between it and the ACSR of phase C until it reaches 700 mm. No corona is observed on the ACSR of phase C. After that we implement an assembly operation of connect fittings of phase B using trajectory planning of A2 within the SWS. Neither corona nor air breakdown occurs after the LS contacts the CC (the inductive discharge between LS and CC is unavoidable as discussed in chapter 1).

5. Conclusion

In this paper we have analyzed electric field distortion caused by the intervention of the hot-line working robots and proposed a design reference for hot-line working robots which can not only guarantee the electrical safety distance but also minimize the robots’ influence on the electric field. Then we focused on manipulators A2’s influence on the electric field under different postures, and presented a method to calculate the minimum gap distance between A2 and the adjacent ACSR. Based on such distance, we have found the SWS for the robot and are able to plan the trajectory of A2.

This system is currently an exploratory study related to robotic hot-line working in substations. It will be tested in substation field condition soon after. We have conducted multiple experiments in the simulated energized experiment system whose size is in equal proportion to the actual 110 kV intelligent substation. The results show that: (i) neither corona nor air breakdown occurs during the operation; (ii) the robot can complete the task effectively in the SWS.

References

Maruyama

, Robotic applications for hot-line maintenance, Industrial Robot: An International Journal 27(5) (2000), 357–365.

Pouliot

and Montambault

, Field-oriented developments for linescout technology and its deployment on large water crossing transmission lines, Journal of Field Robotics 29(1) (2012), 25–46.

and Chen

, Aerial surveillance system for overhead power line inspection, Center for Self-Organizing and Intelligent Systems (CSOIS), Utah State Univ., Logan, Tech. Rep. USU-CSOIS-TR-04-08 Sep. 2000.

Hrabar

Merz

and Frousheger

, Development of an autonomous helicopter for aerial power line inspections, in: International Conference on Applied Robotics for the Power Industry, 2010, pp. 1–6.

Wang

Zhang

H. W.

and Qi

, Dual arm robot for maintenance of high voltage power line, Machine and Hydraulic (2003), NO.5 -049 -3.

and Lu

, Design and implementation of hot-line working robot master-slave manipulator, in: International Conference on Robotics and Biomimetics, Guangzhou, China, 11–14, Dec., 2012.

IEEE Power Engineering Society , IEEE Guide for Cleaning Insulators, New York, USA, 1995, pp. 10–18.

Paris

, Equipment to carry out maintenance operations, particularly washing, on insulator chains of high voltage electric lines, U.S. Patent 5 119 851, Jun. 9, 1992.

and Chen

, The research of the automatic washing-brushing robot of 500 kV DC insulator string, in: Proc. Sixth International Conference on Transmission and Distribution Construction and Live Line Maintenance, Las Vegas, 1993, pp. 411–424.

10.

Hirose

and Aoki

, Development of the insulator washer robot, in: Proc. 1995 IEEE International Conference on Robotics and Automation, Nagoya, 1995, pp. 1783–1789.

11.

Kuffel

Zaengl

and Kuffel

, High Voltage Engineering Fundamentals, Newnes, Oxford, 2000.

12.

Naidu

N. S.

, High Voltage Engineering, McGraw Hill, New York, 1996.

13.

Paul

, College Physics, Worth, 1987, p. 467.

14.

Macmillan

S. J.

, Encyclopedia of Physics, Simon & Schuster, 1996, p. 353.

15.

Riley

A. L.

, Dielectrics, 1999--2000.

16.

Gallimberti

Bacchiega

Bondiou-Clergerie

and Lalande

, Fundamental processes in long air gap discharges, Applied Physics (2002), 1335–1359.

17.

Kojima

Hotta

Kitamura

and Hayakawa

, Classification of impulse breakdown mechanisms under non-uniform electric field in air, IEEE Transactions on Dielectrics and Electrical Insulation 23(1) (2016), 194–201.

18.

Drapeau

J. F.

Olivier

and Gervais

, Non-uniform field breakdown of gases, in: Conference on Electrical Insulation & Dielectric Phenomena, Report, 1988.

19.

Önal

, Breakdown characteristics of gases in non-uniform fields, Journal of Electrical & Electronics Engineering 4 (2004), 1177–1182.

20.

Khalifa

, High Voltage Engineering, Theory and Practice, Marcel Dekker Inc., New York, USA, 1990.

21.

Hua

Zhonggan

and Li

, The influence of ions on electrical breakdown in air, IEEE Proc. Dielectr. Materials 1 (1988), 168–171, S.

22.

Leandros

Frangiskos

and Topalis

, Method of controlling corona effects and breakdown voltage of small air gaps stressed by impulse voltages, Eurocon (2013), 1207–1214.

23.

Cong

Shu

and Li

, Research portfolio and prospect on physical characteristics and modelling methods of secondary arcs related to high-voltage power transmission lines, International Journal of Applied Electromagnetics and Mechanics 56(2) (2018), 195–209, IOS Press.

24.

Reddy

C. S.

Sharma

and Mittal

K. C.

, Recovery analysis of spark gap by anode temperature decay method for different materials, International Journal of Applied Electromagnetics and Mechanics 47(3) (2015), 643–651, IOS Press.

25.

Zangeneh

Gholami

and Zamani

, A new method for calculation of corona inception voltage in stranded conductors of overhead transmission lines, in: Proc. IEEE Int. Power and Energy Conf., Nov., 2006, pp. 571–575.

26.

Carrara

and Thione

, Switching surge strength of large air gaps: A physical approach, IEEE Transactions on Power Apparatus & Systems 95(2) (1976), 512–524.

27.

Drapeau

J. F.

Olivier

and Gervais

, Non-uniform field breakdown of gases, in: Conference on Electrical Insulation & Dielectric Phenomena, Report, 1988.

28.

Laney

, Analysis and application of transmission line conductors, Dissertations & Theses - Gradworks, 2013.

29.

Xiong

Cui

Rao

and Wang

, Analysis of corona onset electric field considering the effect of space charges, IEEE Transactions on Magnetics 47(5) (2011), 1390–1393.

30.

Sarma Maruvada

, Corona in Transmission Systems: Theory, Design and Performance, Eskom Holdings, Johannesburg, South Africa, 2011.

31.

Yializis

, An optimised charge simulation method for the calculation of hige voltage field, IEEE Trans. PAS 97(6) (1978), 2433–2440.

32.

Malik

N. H.

, A review of the charge simulation method and its applications, IEEE Trans. EI 24(1) (1989), 3–20.

33.