Abstract: In the optimization technology of deep peak regulation and consumption control of coal-fired units in thermal power plants, the operation data of coal-fired units is collected frequently through the data acquisition device, and the abnormal data is processed by Z-score method. First, identify and expand the deep peak regulation and consumption space of coal-fired units, analyze the cost structure, and put forward the economic compensation mechanism. Then, the optimization model of deep peak regulation and consumption control is constructed, comprehensively considering the constraints of cost, scheduling optimization and coal combustion, hill climbing and heat storage. Finally, the open source solver CPLEX, combining the branch-cut plane method and the inner point method, was used to solve the model efficiently and realize the optimization of deep peak regulation and absorption control of coal-fired units in thermal power plants. The experimental results show that the average emission concentration of NO_x, SO₂ and dust is much lower than the legal standard value, and the maximum concentration difference is 15 mg/Nm³, 10 mg/Nm³ and 3 mg/Nm³, respectively. It is proved that the optimization technology of deep peak regulation and absorption control of coal-fired power plants designed in this paper realizes the precise control and significant reduction of pollutant emission on the basis of ensuring the safe and stable operation of the units.