大气污染优化控制理论的一些新进展

本文介绍了国内学者的最新研究成果，包括了大气污染优化控制理论在以下几个具体应用领域的进展：(1)工业污染源的优化布局。应用伴随方法计算目标函数的梯度以求解优化问题，给出了理论框架。进行了数值试验。(2)短期空气质量的动态控制。应用实际气象数据和污染源数据进行了优化控制数值模拟。(3)化学风险的预先评价与控制。应用伴随方法，对运动化学源进行风险评价。规划最佳运输路线；针对禁止化学武器公约中处理日本二战期间遗弃化学武器的有关问题。妥善选择化武销毁工厂的地址。以上优化控制问题，以气象预报模式和大气污染预报模式为基础，而以伴随方法和最优化算法为求解途径，数值试验表明理论的正确性和方法的高效率。新发展的理论和方法与数值模式和观测技术紧密结合，并且顺应了计算机速度和容量高速增长的趋势。从而具有广阔的应用前景。

Some New Progresses in the Theory for Optimal Control of Atmospheric Pollution

Some new works by Chinese researchers are introduced here, which include progresses in the following fields:(1) Optimal planning of industrial pollutant sources. By using the adjoint method to calculate the gradient of the object function and then to solve the optimal problem, the theoretical formulation is constructed. Numerical experiments show the validity and high efficiency of the method developed. (2) Dynamic control of air quality. Real meteorological data and pollutant source data are applied for numerical simulation and the method and algorithms are verified. (3) Pre assessment and control of chemical risk. The adjoint method is introduced for the assessment of the moving chemical sources, and then for planning optimal paths to reduce risk. Focusing on problems associated with the abandoned chemical weapons by Japanese army during World War II, methods of finding appropriate location for destroying the chemical weapons are suggested. The above optimal problems are based on numerical models to predict weather condition and pollutants in the atmosphere. The adjoint method and optimization algorithms are applied to solve the optimal problems. Theoretical formulations are verified through numerical experiments, and methods are shown to be of high efficiency. Methods developed here have broad prospect of application as they are combined with the techniques of numerical models and observation, as well as the rapid growth of computer speed and memory.