The invention discloses a combined heat and power generation unit SCR inlet
smoke temperature online calculation method based on a BP neural network. The method comprises the following steps of carrying out designing, carrying out multi-
coal and multi-working-condition boiler thermal tests, obtaining
test data of
coal quality, load,
heat supply steam extraction capacity and SCR inlet
smoke temperature under different test working conditions, training the BP neural network through utilization of
test data, and establishing an SCR inlet
smoke temperature initial calculation model; issuing the smoke temperature initial calculation model in a form of an online website through utilization of a network
programming technology, thereby realizing online determination of the SCR inlet smoke temperature under different operation conditions; and after the SCR inlet smoke temperature deviates, carrying out real-time online correction on the smoke temperature initial calculation model through combination of practical operation data of a thermal power generation unit. According to the method, related staff can calculate and predict the SCR inlet smoke temperature of a combined heat and power generation unit under different
coal quality and different load rates in real time, the peak-
load regulation limit of the set can be accurately determined, and when the boiler heat exchange condition is changed, the self-correction of a model can be realized.