Research On The Technology Of Active Noise Control For Power Transformer

With the development of society and the improvement of living standards, an impact on people gradually enhances by noise pollution, especially radiation of the low frequency noise from power transformer. The noise from the substation surrounding residents is offen complained. Because urbanization is accelerating, substations in the suburbs gradually go into the residential area and the city. Therefore, related departments focus on problems how to effectively reduce the power transformer noise.According to the active noise control (ANC) system, a lot of achievements for the adaptive control strategy are made. However the existing algorithms still have obvious defect. Therefore, it is a potential and important work to develop a new control algorithm. At the same time, optimal arrangement of adaptive control system must be further studied in order to more effectively apply the adaptive active noise control technology to engineering. So these problems were studied both of theoretically and experimentally.The main content of this dissertation is as follows:(1) Generation mechanism and characteristics of the power transformer noise are analysed. And the mechanism of ANC is studied. Key problems about active noise control for power transformer noise are analyzed, i.e. the adaptive active noise control algorithms, the impact of the secondary channel upon adaptive active noise control system, the target function and the system optimization way.(2) The principle and performance of NLMS, BLMS and FxLMS algorithm is studied, Generalized FIR Filter Feedforward Control (GFWC) for active noise control for power transformer noise is given. It is one of its important characteristics and advantages for GFWC algorithm that dynamic characteristic of basis function can be brought in generalized FIR filter. Without the dynamic characteristic of expectation, generalized FIR filter, which has less parameter and better approximation characteristics than conventional filter, is simplified as conventional FIR filter by creating basic function. Therefore, the algorithm has good convergence performance. The control system based on this algorithm has the advantage of good stability and easy realization.(3) Optimal objective function must be provided when ANC system for the power transformer is optimized. Therefore, theory models of power transformer noise are deduced according to three cases, i.e., pipeline noise, free field noise and local mute area noise for power transformers. Theoretical formulas are constructed in the three cases.(4) For the adaptive control technique is perfectly applied to power transformer noise control, based on particle swarm optimization (pso) algorithm, comprehensive optimization position of the secondary source and error sensor is studied. Firstly, this dissertation briefly introduces the basic theory and the algorithm flow of the classical particle swarm algorithm. Secondly, integrated optimization arrangement models are built with optimization variables,i.e., position and number of secondary source and error sensor and with the biggest noise reduction of system in the whole space and mute area. Finally, optimal arrangement of secondary sources and error sensor in the system is studied through the particle swarm optimization algorithm by the simulation experiment.(5) Adaptive active noise control system for power transformers is designed based on GFWC algorithm. The control strategy design based on GFWC algorithm mainly includes feedforward compensation analysis, optimal arrangement of second source&error sensor, the construction of generalized FIR filter, and the construction&realization of basic function. Finally, simulation research is implemented according to the optimization, identification and realization of system.(6) The adaptive active control system for low-frequency noise of power transformers is developed and debugged. Intelligent controller of the system uses DSP as the core unit and GFWC algorithm as the core algorithm. This dissertation offers power transformer noise as example. The noise is simulated by100Hz sine signal. The feasibility and control effect of the system is analyzed through the real-time control eperiment.