Forest biomass is a basic quantity character of the forest ecosystems, and which indicates the structure and function of forest.To estimate accurately biomass, establishing biomass models is a major way.However, model of common estimated biomass is experience and its application is very limited. In this paper, we constructed a Chebyshev′s orthogonal polynomial′s property and constructed base of Continuous Functions′s Hyperspaces, and uniform Model of forest biomass was expressed as linear combination of the basis, and provided a new method to tree genus′s theoretic biomass model. This method of constructed model has complete theoretic base, so current Model of Forest Biomass was an especial example of uniform Model.Method of model′s construction was essential numerical interpolation, This interpolation firstly processed interval transform and reduced independent variable′s fields of definition. Model order was the higher, and its interpolation point was more. The estimated result was much according with practical, and also it was according with principium of stem analysis and wood growth.As disadvantages of common least squares, we applied partial least squares to calculate model of forest′s biomass. This paper indicated Moso bamboo biomass′s research was not same as other tree genus′s. On the base of Moso bamboo′s character, its biomass′s was divided into 2 phases. Finally, this paper applied uniform Model of Forest Biomass to research Moso bamboo′s biomass of the first phase, and result indicated fitted effect was better: SPRESS, h=314390，R2=0.9987, minimum variance was 0.2518，The result shows that the improved method can improve more 10 percent estimative precision than that of old method.