The invention discloses a
surface heat flow identification three-dimensional effect
correction method based on a neural network. The
correction method comprises the following steps: mounting a plurality of temperature sensors on the inner wall surface of an area around a stationary point
heat flow; utilizing temperature data of each
internal temperature measuring point; obtaining a
heat flow on acorresponding heating
surface point through a one-dimensional
heat flow identification method; and then introducing an
artificial neural network algorithm, carrying out normalization
processing on theidentified heat flow corresponding to each measuring point in the previous step to serve as an input sequence of the neural network, and carrying out training in the neural network to obtain an output anti-normalization result to serve as a heat flow identification value of a
region of interest. According to the
correction method provided by the invention, the
time complexity of three-dimensionalidentification is avoided, meanwhile, the good
noise resistance of the sequential
function method and the strong nonlinearity of the neural network are combined, a traditional model can be greatly simplified, the identification precision of the stationary point heat flow is improved, and the real-time performance of on-line identification is ensured.