The Study Of Prediction And Precaution Of Wireless Temperature Detecting Devices In Substation Based On Chaotic Time Series

With the rapid development of the national economy, people's quantity demand of electricity and dependence on electricity is increasing. Because of electrical overload and equipment aging problems, fire and explosion occur at equipments in high-voltage transformer substations by high temperatures accumulated at these spots, not only seriously affect the safety of substations production and the demand for electricity users, but also a serious threat to the city public security, and a tremendous economic loss to power suppliers. Therefore, the study of real-time temperature detecting and prewarning in substations is necessary. At present, the temperature detection substation equipment, whether it is infrared detection gun point by point temperature or optical temperature measurement, there are certain disadvantages. Infrared detection gun point by point the way temperature often occurs missing and misdeclaration. What's more, the high-voltage and intense radiation will damage the body of the workmen. However, the existing problems of Distributed optical cable temperature detecting system can't be ignored either. For example, the wiring is complicated, the circuit aging problem is serious, the circuit dismantle is difficult and it is expensive. Therefore, in this paper we take full advantage of wireless resources, the development of the wireless temperature detecting and prewarning system in substation not only saves manpower and material resources, but also cost lowly and easy to management and maintain. At the same time, it can realize the storage and analysis of the data of temperature, and prediction and precaution.In this paper, intelligent temperature sensor DS18B20, ultra-low power consumption chip MSP430 and intelligent digital transforming module CC1020 are applied to exploit low power consumption, quality performance wireless temperature detecting spots. For the first time, two or three intelligent temperature sensors will share one ultra-low power consumption chip and intelligent digital transforming module, So that it will largely decrease the number of ultra-low power consumption chip and intelligent digital transforming module, and take up much less room. Thus it developed a new low-power and high-performance wireless temperature detecting devices. At the same time, for the first time this paper analyses and forecasts the temperatures of the equipments with chaotic theory. At fist it confirms that the time series of the temperatures is chaotic, so that it can take advantage of chaotic theory and chaotic time series theory to analyze those data of the temperatures of the equipments, combining neural networks theory. Then time serial data mining method was provided to analyze the data of the temperatures of the equipments in transformer substation and forecast the temperatures in the future. It will obtain indicated precaution results from known data and guarantee prediction and precaution of equipments in time.