Research On Key Technologies Of The Low Power Communication In Smart Grid

For the complexiy of low-voltage power line carrier communicationnetwork topology in smart grid, the channel exists noise interference andmultipath fading, we need to have a research on the low-voltage powerline channel and the efficiency of network. The existing route algorithmsspend long time on searching and the efficiency is also very low, so wepropose a new dynamic routing algorithm, namely hybrid ant colonyalgorithm, and we have an experiment and simulation study on it.Firstly, by studying the noise characteristic, the impedance charact-eristic and attenuation, characteristic of the low-voltage power line carriercommunication channel. We established the multipath channel model ofpower line. We study the effects of the number of channel paths on thesignal attenuation by MATLAB simulation. Low-voltage power line carr-ier signal is modulated using OFDM technology which can well reduceinter-symbol interference and interference between subcarriers, and caneffectively combat multipath channel fading and frequency selectivefading. Secondly, in order to correct demodulate the signal, we need tobalance signal and obtain state information of channel by using the chan-nel estimation algorithm, and analysis the performance of the channelestimation algorithm. Experimental results show that LS algorithm basedon the training sequence is more suitable for power line channel estimat-ion. Through comparing the analysis of the modulation performanceamong the Chow algorithm, the gradient allocation algorithm and theequal power algorithm for allocation of the subcarriers bits of multi-userand power allocation, experimental results show that the Chow algorithmhas a good self-adaptive performance.By using power line carrier communications chip RISE3301, wehave a test on end-user of community which installed power line carrierterminal. We measured communication success rate and node communic-ation delay of the cell. By further analysis of the shortcomings of thesystem routing scheme, we propose a hybrid ant colony algorithm. Weimprove the fitness function, the path evaluation function, objective opti-mization function, genetic operators of the algorithm, and regard the node latency and load factor as evaluation factors. By comparing the improvedhybrid algorithm with ant colony algorithm on simulation comparativeanalysis, the results show that the hybrid ant colony genetic algorithmimproves11.2%than the ant colony algorithm on the part of the nodedelay; the balance performance of the load; the rate of communicationsuccess and number of hops. The improved hybrid algorithm has thefeature of short running time, the high convergence speed and good net-work performance.