Application of Hopfield network integrating with SA operator in fitting of micro-drill´s main lips

For measurement chips of micro-drill´s main lips, a novel data processing was presented. While the line of micro-drill´s main lips is fitted, the special structure´s Hopfield neural network is designed according to normal equation. The neural network has 2 neurons, the weights of which are elements of normal equation´s symmetry matrix, the input numerical value of which are the vector consisted of equation´s right terms, and asynchronous work means is adopted. And via integrating simulated annealing algorithm, the iteration is carried at every control temperature in the light of Metropolis acceptive criterion, the probability of convergence global optical solution is 1 with iteration operator. While the temperature approximate to desired value and the network attain to stability states, the states vector can be taken as coefficients of fitted main lips line, According to which the straightness errors or chips of micro-drill´s main lips line can be solved.