用过渡金属元素的原子半径、价电子数、主量子数、d电子数和f电子数等为人工神经网络的输入值，用已知的二元连续固溶系相图液相线和固相线的实测数据为输出值训练人工神经网络．并用其对未知二元连续固溶系的液相线和固相线进行计算机预报．若干二元系计算结果与实测结果相当接近． The atomic radius, valence number, main quantum number, d electron number and f electron number of the transition metal elements are used as the input values ​​of the artificial neural network. The binary phase solid solution phase and the solid phase line The measured data for the output training artificial neural network. And use it to predict the liquidus and solidus of unknown binary continuous solid solution system. Some binary system calculated results and the measured results are quite close.