We present an energy management scheme for the power-over-fiber system used in the electric power industry. The scheme dynamically adjusts output power of the system by tuning the driven current of the laser, according to the power consumption of the external load, which is deduced by the output current and voltage of the photovoltaic cell. In addition, an auxiliary power is employed to increase the output power and improve the dynamic response of the system. Experiments were made in which the static response, dynamic response, and the stability of the scheme were tested. The results show that our scheme can automatically control the output power of the system corresponding to the external load, which leads to the alleviation of the working temperature and potentially increases the reliability of the system and even prolongs the lifetime.
The optical power supply technology has been presented several years and used in Electric current transformer (ECT). However, the reliability of the technology is proved to be low in actual application, which limit its widely used in electric power industry. In this paper, we propose a novel optical power supply system for Ultra-High Voltage (UHV) Grid, which utilizes 810-nm laser diode in low-voltage (LV) environment and GaAs photocell for power supply in highvoltage (HV) environment. A Lithium battery is further employed in HV environment which makes the power supply scheme more flexible and reliable. In addition, two VCSELs are also used for information transmission between the LV and the HV environment. And the parameters of the system, such as load power, laser driving current, laser temperature, semiconductor chilling plate driving current, ambient temperature and ambient humidity, are collected, transmitted via optical fiber and processed by the power management scheme. The scheme is able to dynamically control the output power of the laser diode according to the load power variance and make the Lithium battery working when the optical power supply is unavailable or burst heavy power supply demand is emerged. In this way, the output power variance of the laser can be kept as low as possible, which effectively increase the reliability of the system.
In this paper, we analyze the performance of the electro-optic hybrid optical current transformer (HOCT) proposed by
ourselves for high-voltage metering and protective relaying application. The transformer makes use of a fast variable
optical attenuator (FVOA) to modulate the lightwave according to the voltage from the primary current sensor, such as
low-power current transformer (LPCT). In order to improve the performance of the transformer, we use an optic-electro
feedback loop with the PID control algorithm to compensate the nonlinearity of the FVOA. The linearity and accuracy of
the transformer were analyzed and tested. The results indicate that the nonlinearity of the FVOA is completely
compensated by the loop and the ratio and phase errors are under 0.07% and 5 minutes respectively, under the working
power of less than 1 mW power. The transformer can be immune to the polarization and wavelength drift, and also
robust against the environmental interference.
In this paper, a novel architecture for fiber Bragg grating (FBG) sensor network with self-healing function is proposed, which is made of by a set of sector sub-networks including FBG sensors, main node and some remote nodes. The main node is responsible for sending the lightwave from the source to the sensor parts in the network. The remote nodes are built by using of optical switches and couplers so as to check the breakpoint and reconfigure the sensor network with different route if any link fails. The simulation and discussion results show that the networks consisting of different nodes could provide different performances due to the change in insertion losses at the nodes and reflecting spectrum from FBG sensors, and successively, offer different levels of survivability. More importantly, reconfiguring the sensor network in case of the failure for certain links, the new route composed by the different remote nodes may influence the network performances. This sensor network can be also expanded to large scale by combining three or more sector sub-networks. In order to meet the remand for survivability, remote nodes must be redesigned carefully when certain links fail. The improved performances are verified by the simulation. The results indicate the proposed architecture can facilitate a reliable sensor network with large scale and multipoint smart structure.