城市污泥堆肥温度的空间变异性研究

Spatial Variability of Temperature During the Sewage Sludge Composting Process

利用半变异函数对城市污泥堆肥温度的空间变异特性进行了研究 ,对堆体温度进行了克里格法 ( KRIGING)插值。采用通风静态垛堆肥工艺 ,试验了 0 .79、2 .0 3m3 / ( min·m3 )两种通风量。沿着堆肥池长度方向设定 2个纵剖面 ,每个纵剖面的面积为 6 .0 m× 1 .0 m,按 0 .5 m× 0 .1 m布设网格。结果表明 ,在水平方向上堆肥温度的半变异函数用球状模型进行拟合效果较好 ,而在垂直方向上的半变异函数用线性模型进行拟合效果较好 ;在水平方向上两个剖面的温度变程 ( range)分别为 0 .90 m、1 .2 5 m,在垂直方向上的变程分别为 0 .75 m、1 .0 0 m;利用克里格法进行最优内插估值得到的温度等值线图表明 ,高温区域一般位于堆体中层 0 .4～ 0 .6 m,低温区域一般位于堆体下层 0～ 0 .4 m;从温度剖面等值线图判断 ,中试规模的城市污泥堆肥 ,其合理通风量小于 0 .79m3 / min· m3 。

The studies of spatial variability had been undertaken during geology, agrology and ecology since 1980's, but the theory of spatial variability had not been used in the composting science. Composting temperature varied at different positions in the composting pile, which was the spatial variability of temperature, and it was a universal phenomenon during the composting process.The composting temperature was high at some positions and low at other positions because of the temperature spatial variability. If the temperature was too high or too low, the process of eliminating pathogen or stabilization of organic matter was impeded. So it was necessary to know the spatial pattern of temperature in the pile, and know the location and ratio of composting material that had achieved the standard of land application.The composting material that could not achieve the standard should be composted again. The quality of compost was improved when improving the composting technologies-such as adjusting the mixing ratio and the strategy of aeration-according to the temperature spatial pattern. The spatial variability of temperature was well expressed when analyzing and explaining the spatial data of temperature with the Geostatistics method.The aims of this study were to analyze the temperature spatial variability in composing pile, and optimally interpolate the unmeasured temperature data depending on the measured data.The aerated static pile was designed in a building with the one 6^0m×2^0m bay with cement floors and walls. Aeration boards were laid on the bottom of the bay. The pile contained sewage sludge (dry matter=19^3 percent) and bulking agent with the ratio of 1:1 by volume. A bulking agent layer on the aeration board dispersed air through the 1^0 m high mix layer. The pile was covered with a 20-cm insulating layer of recycled compost. There were two chambers built on the bottom of the bay and two aerating quantities, 0^79m+3/(min·m+3) and 2^03m+3/(min·m+3), were tested. The temperature sensors were inserted into the different designed positions of the pile to detect the temperatures on the two profiles with dimensions of 6^0m×1^0m. Using all available data of main nodes with the grids of 0^5m by 0^1m analyzed the temperature spatial structure. The detected temperature during the geometrical center of the pile was fed back to the computerized control system. A three-stage control algorithm controlled the damper duty cycle of the aerator. The primary composting process had been undertaken for 16 days. A computer installed with a automatic control software (Compsoft) logged temperatures of all positions and generated reports. The semivariogram analysis was carried out to study the temperature spatial variability. The semivariogram of temperature increased gradually with the increasing of distance. When the distance was longer than α (correlation range), the semivariogram was invariable. The temperature changing during the composting process was decided by the determined variability and the randomicity variability, but the changing and distribution of temperatures during the field were usually decided by the determined variability. The semivariogram function following to the 0° direction was different from that of the 90° direction, which showed that the temperature spatial variability held the characteristic of orientation. The different varying principles of temperature showed the semivariogram functions were not completely similar when the directions changed. The temperature semivariogram functions of horizontal and vertical direction were simulated with the globosity model and the linearity model, respectively. The correlation ranges of profiles along the horizontal direction were 0.90 and 1.25m respectively and that following to the vertical direction were 0.75 and 1.00m respectively. So the temperature change in the pile was not completely unattached. It was important to study the correlation field of sampling spots and then decide the dimension of experimental area and the sampling locations. The unmeasured comp