CN-121978926-A - Dynamic determination method and system for sliding pressure curve of collaborative driving thermal power generating unit

CN121978926ACN 121978926 ACN121978926 ACN 121978926ACN-121978926-A

Abstract

The invention provides a method and a system for dynamically determining a sliding pressure curve of a cooperatively driven thermal power unit, and belongs to the technical field of energy conservation and consumption reduction of coal-fired units. Compared with the traditional single power-pressure sliding pressure curve, the method creatively takes the steam extraction flow as a key input variable, so that the model can accurately reflect the influence of complex working conditions such as heat supply, industrial steam extraction and the like on the optimal value of the main steam pressure, and the optimal main steam pressure under the current actual working condition can be accurately calculated by constructing a multi-element nonlinear sliding pressure curve model based on the steam extraction flow and the power generation. The system controls the unit according to the method, so that the main steam pressure tracks the optimal value in real time, and the throttling loss and the heat consumption rate of the unit in the variable working condition, especially the heating and steam extraction working condition, can be effectively reduced, thereby remarkably improving the overall operation efficiency and realizing continuous energy conservation and consumption reduction.

Inventors

LI JIAO
JING TAO
HAN LI
ZOU YANG
YANG KAIXUAN
Yang Zhenshuai
WAN CHAO
LI GAOCHAO
JIA MINGXIAO
WANG MINGYONG

Assignees

西安热工研究院有限公司
西安西热节能技术有限公司

Dates

Publication Date: 20260505
Application Date: 20260115

Claims (10)

1. A dynamic determination method for a sliding pressure curve of a collaborative driven thermal power generating unit is characterized by comprising the following steps: collecting operation parameters of the thermal power generating unit in real time and preprocessing; inputting the collected steam extraction flow and the power generation into a pre-constructed sliding pressure curve model, and calculating to obtain the optimal main steam pressure under the current working condition; Generating a control instruction according to the optimal main steam pressure, and controlling the unit to run so that the actual main steam pressure dynamically tracks the optimal main steam pressure; The dynamic operation parameters of the thermal power generating unit comprise the extraction flow of the extraction port of the steam turbine, and the actual power generation of the unit.
2. The method for dynamically determining the sliding pressure curve of the collaborative driving thermal power generating unit according to claim 1, wherein the collection object of the extraction flow comprises a high-pressure cylinder extraction port or a medium-pressure cylinder extraction port, and the collection frequency of the operation parameters is not lower than 1Hz.
3. The method for dynamically determining the sliding pressure curve of the collaborative driving thermal power generating unit according to claim 1 is characterized in that the method for preprocessing the real-time collected operation parameters is that the real-time collected operation parameters are filtered and abnormal values are removed, a sliding average filtering method is adopted to eliminate instantaneous interference, abnormal values of data are removed, and normalization processing is carried out on main steam temperature and boiler heat load.
4. The method for dynamically determining the sliding pressure curve of the thermal power generating unit driven cooperatively according to claim 1, wherein the method for constructing the sliding pressure curve model comprises the following steps of: Acquiring a historical training data set of a thermal power unit, wherein the historical training data set comprises working condition data under the combination of various power generation powers and various steam extraction flows and optimal main steam pressure reference values corresponding to the combinations; taking the steam extraction flow and the power generation power as input variables, and taking the optimal main steam pressure as output variables, constructing a sliding curve model, wherein the sliding curve model adopts a multi-element nonlinear regression model; And setting a safe operation interval of main steam pressure, taking the power generation power and the steam extraction flow as input variables, taking the optimal main steam pressure reference value as a target output variable, and training the constructed multiple nonlinear regression model based on the historical training data set.
5. The method for dynamically determining the sliding pressure curve of the collaborative driving thermal power generating unit according to claim 4, wherein the constructed multiple nonlinear regression model is as follows: Wherein, the For the parameter vector to be determined, For the steam extraction flow, P is the power generation power, Is the optimal main steam pressure.
6. The method for dynamically determining the sliding pressure curve of the collaborative driving thermal power generating unit according to claim 5, wherein the constructed multiple nonlinear regression model is trained by adopting a gradient descent method based on a historical training data set, and the method is specifically as follows: undetermined parameter vector in multi-element nonlinear regression model by mean square error loss function Initializing; Calculating the mean square error loss function with respect to Is a gradient of (2); Updating parameters of the multiple nonlinear regression model based on the calculated gradient; and when the drop of the loss function is smaller than the set threshold value, obtaining a trained model.
7. The method of claim 6, wherein in the step of updating parameters of the multiple nonlinear regression model, after each round of parameter updating, a projection step is introduced to check whether a predicted steam pressure value outputted by the model in the whole expected input range exceeds a safe interval under the current parameter, if the predicted value exceeding the boundary is found, a correction is applied to the parameter updating process, and a corrected loss function is applied The method comprises the following steps: Wherein, the Is a penalty coefficient which is a function of the penalty coefficient, Is a penalty term.
8. A cooperatively driven thermal power generating unit slip pressure curve dynamic determination system, comprising: the data acquisition and preprocessing module is used for acquiring the operation parameters of the thermal power unit in real time and preprocessing the operation parameters; The calculation module is used for inputting the collected steam extraction flow and the power generation into a pre-constructed sliding pressure curve model, and calculating to obtain the optimal main steam pressure under the current working condition; And the control module is used for generating a control instruction according to the optimal main steam pressure and controlling the unit to operate so as to dynamically track the actual main steam pressure.
9. An electronic device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the method for dynamically determining a sliding curve of a co-driven thermal power generating unit according to any one of claims 1 to 7.
10. A storage medium having stored thereon a computer program, which when executed by a processor, implements the steps of the method for dynamically determining a sliding curve of a co-driven thermal power generating unit according to any one of claims 1 to 7.

Description

Dynamic determination method and system for sliding pressure curve of collaborative driving thermal power generating unit Technical Field The invention belongs to the technical field of energy conservation and consumption reduction of coal-fired units, and particularly relates to a method and a system for dynamically determining a sliding pressure curve of a collaborative driven thermal power unit. Background The sliding pressure operation is one of key operation modes for realizing energy saving and consumption reduction of the thermal power generating unit, and is characterized in that the main steam pressure (namely the sliding pressure) is dynamically adjusted, so that the thermal efficiency of the thermal power generating unit is always kept high under different load conditions. Currently, the determination method of the sliding pressure curve mainly depends on a single core parameter of the generated power, a fixed generated power-main steam pressure corresponding relation curve is usually made through an off-line test (such as a heat consumption test), and the curve is preset in a unit control system to guide pressure adjustment in the actual running process. However, the operating characteristics of units equipped with a steam extraction system, such as cogeneration units which take on the heating task in winter, or units which supply technical steam to industrial users, are more complex. The energy balance relation between the boiler and the steam turbine can be obviously changed by the steam extraction behavior, and the extracted steam does not enter the steam turbine to do work in the subsequent stage, especially the steam inlet amount of the low-pressure cylinder is reduced, and the readjustment of power distribution among the high-pressure cylinder, the medium-pressure cylinder and the low-pressure cylinder is caused. In the dynamic process, if a fixed sliding pressure curve based on the generated power only is still adopted, the problem that the main steam pressure is seriously not matched with the actual operation condition can occur. The mismatch can bring multiple adverse effects, namely, the overall thermal efficiency of the unit is reduced due to light weight, the exhaust gas temperature of the boiler is increased, the efficiency in each stage of the steam turbine is reduced, and thus the coal consumption is increased, and the potential safety hazard of equipment can be caused due to heavy weight, such as abnormal stress and vibration increase of the last stage blade of the steam turbine caused by mismatching of steam flow and pressure, or the over pressure of the boiler caused by high pressure setting, and the equipment can be aged and damaged even after long-term operation. Therefore, under the condition of steam extraction, the traditional sliding pressure optimizing method is necessary to be improved, and the multivariable participating sliding pressure curves such as steam extraction flow, heat supply load and the like are introduced to correct in real time so as to more comprehensively reflect the actual energy distribution state of the unit, thereby realizing safe, efficient and flexible operation targets. Disclosure of Invention The invention aims to overcome the defects and provide a method and a system for dynamically determining a sliding pressure curve of a thermal power unit driven cooperatively. In order to achieve the above purpose, the invention adopts the following technical scheme: In a first aspect, the invention provides a method for dynamically determining a sliding pressure curve of a thermal power generating unit driven cooperatively, which comprises the following steps: collecting operation parameters of the thermal power generating unit in real time and preprocessing; inputting the collected steam extraction flow and the power generation into a pre-constructed sliding pressure curve model, and calculating to obtain the optimal main steam pressure under the current working condition; Generating a control instruction according to the optimal main steam pressure, and controlling the unit to run so that the actual main steam pressure dynamically tracks the optimal main steam pressure; The dynamic operation parameters of the thermal power generating unit comprise the extraction flow of the extraction port of the steam turbine, the actual power generation power of the unit, the actual pressure of main steam, the temperature of the main steam and the thermal load of the boiler. The collection object of the steam extraction flow comprises a high-pressure cylinder steam extraction port or a medium-pressure cylinder steam extraction port, and the collection frequency of the operation parameters is not lower than 1Hz. The method for preprocessing the operation parameters acquired in real time comprises the steps of filtering the operation parameters acquired in real time and removing abnormal values, namely, adopting a moving average filtering method to eliminate instantaneo