Abstract: The troubles confronted with the use of the traditional coal blending methods are heavy reliance on experience of operators, delay in data feedback, operational complexity and high labor intensity. Through analysis of the current state of the coal blending system applied at Mengda Coal Preparation Plant, and study of the intelligent blending algorithm and control scheme, an intelligent coal blending system is developed. The system can execute coal blending in proportion or according to coal quality under fuzzy control, and is capable of working out automatically an optimum blending scheme according to requirements of users as well as the quality and quantity of the coal stored in each bunker. Practices shows that the system can make online adjustment of the proportions of blending coal based on coal quality data; with the use of the system at Mengda Plant, the calorific value of blended coal is increased by 9.85 J, creating an additional revenue of 734,500 yuan for the plant per year; the economic dispute aroused by the plant′s inability to have the calorific value of blended product accurately controlled at the required level can be avoided; three operators have been reduced, saving a labor cost of 300,000 yuan a year. The use of the system enables coal preparation plants to operate with an intelligent, highly efficient and systematic coal blending system; the system developed is of positive significance for upgrading of coal product and improvement of coal cleaning efficiency; and the system has a high reproducibility and can find widespread applications in other plant with similar conditions.