Gas turbine multi-parameter cooperative control and dynamic safety adjustment system and method

Abstract

The invention relates to the technical field of gas turbine control, in particular to a gas turbine multi-parameter cooperative control and dynamic safety adjusting system and method. According to the system, PID outputs of high-pressure compressor rotating speed tracking, power turbine rotating speed tracking, active power tracking and parameter limit control loops are generated in parallel through a multi-parameter cooperative control module, and a minimum fuel demand signal is selected as a control reference based on competitive logic; through a dynamic safety adjusting module, a dynamic threshold curve linked with the inlet air temperature is preset, action allowance is superposed, and active intervention before parameter overrun is achieved; and through the fault tolerance module, the operation is maintained by adopting historical data fitting and an online compensation strategy when the sensor has a fault. According to the method, the problems of single parameter control response delay, insufficient limit protection, poor fault tolerance and the like are solved, and the control precision, response speed and operation reliability of the gas turbine under variable load, extreme environment and sensor abnormal working conditions are improved.

Description

Technical Field

[0001] This invention relates to the field of gas turbine control technology, specifically to a multi-parameter collaborative control and dynamic safety adjustment system and method for gas turbines, applicable to the automated control and protection of gas turbines during start-up, changing operating conditions and shutdown processes, especially for optimization of complex operating conditions in power generation, ship propulsion and industrial drive scenarios. Background Technology

[0002] Gas turbines, as highly efficient power plants, are widely used in power generation, aviation, and industrial fields. Their operational stability depends on precise fuel flow regulation, speed control, and temperature protection. In existing technologies, gas turbine regulation systems often employ single-parameter feedback control, such as PID regulation only for speed or temperature. This approach suffers from the following problems under complex operating conditions (such as variable loads and extreme ambient temperatures):

[0003] 1. Insufficient multi-parameter coupling response: Parameters such as speed, temperature, and pressure affect each other, making it difficult for a single control loop to achieve global optimization;

[0004] 2. Limit protection lag: Traditional systems lag in adjustment when parameters approach fixed safety thresholds, which can easily trigger emergency shutdowns and affect operational continuity;

[0005] 3. Poor fault tolerance of sensors: The lack of a dynamic compensation mechanism for critical sensor failures increases the risk of control failure.

[0006] Therefore, there is an urgent need for an automatic control system for gas turbines that integrates multi-parameter collaborative control, dynamic limit adjustment, and fault tolerance mechanisms. Summary of the Invention

[0007] This invention aims to solve the problems of operational instability, low efficiency, and frequent shutdowns in existing gas turbine control systems caused by single-parameter feedback, lag in limit protection, and insufficient sensor fault tolerance. By integrating multi-parameter collaborative control, dynamic threshold preset, and intelligent fault compensation mechanisms, it achieves precise adjustment and proactive safety protection of the gas turbine throughout the entire process of startup, changing operating conditions, and shutdown, improving its adaptability and reliability in complex environments (such as high and low temperatures, variable loads, and sensor malfunctions).

[0008] The technical solution to achieve the above objectives is:

[0009] 1. Overall system structure (combined with...) Figure 1 )

[0010] This system consists of four core modules, which form a closed-loop control system through real-time data interaction and logical decision-making:

[0011] (1). Multi-parameter collaborative control module: Six independent PID control loops, including N1 limit, N2 limit, T4 limit, N2 tracking, N3 tracking and active power tracking, run in parallel. Each loop determines whether the generated fuel demand signal enters the subsequent competition logic based on the automatic adjustment mode of the control system (sleep mode / ready to start mode / start mode / N2 tracking control mode / N3 tracking control mode / active power tracking control mode / stop mode / emergency stop mode / fault mode). The minimum value is selected as the final control benchmark through the competition logic to avoid the limitations of single parameter control.

[0012] (2). Dynamic safety adjustment module: Built-in segmented threshold curve library (maximum operating condition limit curve of low-pressure compressor speed, maximum operating condition limit curve of high-pressure compressor speed, maximum operating condition limit curve of low-pressure turbine after-gas temperature) linked with the inlet air temperature, calculates the safety boundary in real time and presets the action margin (±20rpm or ±5℃) to achieve proactive intervention. When the parameter approaches the threshold, the fuel demand output by the limit loop in the multi-parameter collaborative control module automatically outputs the minimum value through competitive logic, triggering the switching of the PID control loop to achieve active protection.

[0013] (3). Fuel demand calculation module: Based on the maximum / minimum fuel flow curve of high pressure compressor speed (N2) superimposed with the ambient temperature correction coefficient, dynamic fuel demand signal is dynamically generated, and precise flow distribution and response are achieved through valve control.

[0014] (4) Fault-tolerant module: When the power sensor fails, the online compensation value calculated at the previous moment is superimposed with the measured flow rate. Based on the power-flow relationship fitted by the gas turbine operation history database, the current fitted power value is calculated, and the system switches to the backup regulation loop. When all channels of a critical sensor (such as the speed sensor) fail, an emergency shutdown is triggered and the post-shutdown lubrication logic is started to prevent rotor thermal deformation.

[0015] 2. Control methods and procedures (in conjunction with...) Figure 2-3 )

[0016] Step 1: Parameter Acquisition and Preprocessing

[0017] Real-time acquisition of speed (N1, N2, N3), temperature (T4), pressure, and ambient temperature signals;

[0018] The automatic control system's logic algorithm T4 uses the arithmetic average of the low-pressure turbine afterburner gas temperature and compensates for the thermocouple time constant. This signal is generated from the actual signals of 16 thermocouples (excluding unreliable calculation measurements). The thermocouple time constant is compensated based on the airflow velocity used to purge the thermocouples, T4 = T 4mean +K T dT⁴ / dt. Where T4mean K represents the average temperature measured by 16 thermocouples. T This is the correction factor for the time constant of the thermocouple after the low-pressure turbine (related to the high-pressure compressor speed N2).

[0019] Step 2: Multi-loop PID competitive control

[0020] (1) As Figure 2 As shown, the multi-parameter collaborative control module includes three limit control loops and three tracking control loops. Before calculating the PID output, setpoint limiting processing is required. The setpoint is compared with the maximum and minimum limit values, and then with the upper and lower output limits. The upper and lower output limits are calculated as follows: upper output limit = current output value + speed * running cycle; lower output limit = current output value - speed * running cycle. The limited setpoint is then used for PID calculation to obtain the loop PID output.

[0021] (2) Each PID loop uses an incremental PID algorithm to independently calculate the output:

[0022] Δu(k)=K P [e(k)-e(k-1)]+K I e(k)+K D [e(k)-2e(k-1)+e(k-2)] (1)

[0023] Where e(k), e(k-1), and e(k-2) represent the deviations between the setpoint and the feedback value at the current time, one calculation cycle ago, and two calculation cycles ago, respectively, and K P K I and K D These are the PID control parameters. To maintain optimal control performance across the entire operating range, especially in systems with nonlinear dynamic characteristics or multi-stage processes, this invention employs a piecewise PID control strategy. This strategy can adaptively switch parameters based on real-time error thresholds or process stages, thereby achieving rapid response while ensuring steady-state accuracy.

[0024]

[0025] In the formula K Pj,S ,K Ij,S ,K Dj,S (j = 1, 2, 3, 4, 5, 6) represent the small error threshold e for each control loop. S PID parameters, K Pj,M ,K Ij,M ,K Dj,M (j = 1, 2, 3, 4, 5, 6) represent the control loops corresponding to the large error threshold e. M PID parameters.

[0026] (3) Determine whether each loop is enabled based on the automatic adjustment mode of the control system, and whether the generated fuel demand signal enters the subsequent competition logic. Compare the PID output values ​​of the control loops that are currently enabled, take the minimum value among all loops as the fuel demand output value, and output it to the fuel demand calculation module for the next step of processing.

[0027] (4) For dynamic limit adjustment, when the parameter approaches the threshold, the fuel demand output by the limit loop in the multi-parameter collaborative control module automatically outputs the minimum value through the competition logic, triggering the switching of the PID control loop to achieve active protection.

[0028] (5) The PID output follows the actual output value. If the current loop is not in the enabled state, or the set value of the current loop is within the dead zone of the feedback value, the PID output value of the loop follows the fuel demand output value, including N1 limit loop, N2 limit loop, N2 control loop, T4 limit loop, N3 control loop, and DROOP active control loop.

[0029] (6) Assign the fuel demand output value to the PID output value of all control loops for calculation in the next cycle;

[0030] (7) Based on the source of the current fuel demand output value, form the current control loop flag. For example, if the current fuel demand output value comes from the T4 limit loop, then the T4 limit flag will be formed.

[0031] Step 3: Fuel Demand Calculation

[0032] The minimum and maximum fuel gas flow rates are limited by the high-pressure compressor speed and apply to all operating conditions above idle. Limiting the minimum and maximum fuel gas flow rates prevents gas turbine surge and potential combustion chamber shutdown during variable operating conditions. The fuel gas flow rate limit is determined by the high-pressure compressor speed and corrected for based on the gas turbine inlet air temperature. The curves showing the relationship between the maximum and minimum fuel gas flow rates and the high-pressure compressor speed, as well as the curves showing the relationship between the maximum and minimum fuel gas flow rate correction factors and the gas turbine inlet air temperature, are preset in the automatic control system.

[0033] When the gas turbine inlet air temperature is low, the maximum flow rate of the gas turbine is determined based on the guarantee conditions of the 1.2Ne operating condition, without correction based on the gas turbine inlet air temperature.

[0034] When the control mode is one of hibernation, ready to start, shutdown, or emergency shutdown, the fuel demand output value is 0.

[0035] Step 4: Fault Tolerance and System Recovery

[0036] When the power sensor is working normally, the control program calculates the fuel flow compensation ΔF in real time through the PID controller in each scan cycle to ensure that the power value fitted based on the fuel flow-output power fitting curve (given from the gas turbine operating history database) is consistent with the actual value measured by the sensor. Figure 3 As shown.

[0037] When all power sensors fail, the flow compensation amount ΔF calculated in the previous cycle remains unchanged. The measured flow rate is then superimposed on this, and the fitted value of the power is calculated based on the fitted curve of fuel flow rate and output power. The gas turbine regulation algorithm replaces the measured fault value and switches the current regulation loop to the standby regulation loop. In this way, the control system does not have to immediately execute the fault shutdown procedure, but continues to control under the standby regulation loop, thereby ensuring the stable operation of the unit when the power sensors fail. Operators can then intervene according to the actual situation.

[0038] The present invention has the following beneficial effects:

[0039] 1. Multi-parameter collaborative optimization: Improve control accuracy and response speed under complex operating conditions through a multi-loop PID competition mechanism;

[0040] 2. Proactive safety protection: The dynamic safety adjustment module intervenes in advance, reducing the frequency of emergency shutdowns and extending equipment life;

[0041] 3. Environmental adaptability: Temperature-based correction coefficients enhance system stability in extreme environments;

[0042] 4. High fault tolerance: It can still maintain basic control functions when the sensor fails, thus improving the reliability of the system. Attached Figure Description

[0043] Figure 1 Overall architecture diagram of gas turbine multi-parameter collaborative control and dynamic safety regulation system;

[0044] Figure 2 : Flowchart of multi-loop PID control logic;

[0045] Figure 3 : Logic diagram for power fitting calculation after power sensor failure;

[0046] Figure 4 :DROOP active control loop calculation model. Detailed Implementation

[0047] The present invention will be further described below with reference to the accompanying drawings and specific embodiments:

[0048] The core of this invention lies in constructing an intelligent control system for the entire lifecycle of a gas turbine through the deep integration of multi-parameter collaborative control and dynamic safety boundaries. The implementation process is described in detail below with typical operating conditions and system architecture. The dynamic parameter interaction logic, multi-modal switching mechanism, and fault emergency strategy involved are all based on the deep coupling of the physical characteristics of the gas turbine and the control model.

[0049] Example 1: Cold Start Full-Process Optimization Control

[0050] Step 1: System Initialization and Feature Parameter Loading

[0051] (1) System initialization: Read the single-unit characteristic parameters in the gas turbine operation history database, including the three limit adjustment curves of N1, N2 and T4, the maximum and minimum flow curves and correction coefficient curves, the thermocouple time constant correction curve, etc.

[0052] (2). Load the six preset PID parameter clusters (three for tracking loop and three for limiting loop) through the HMI interface. The tracking control loop uses fast-response PID parameters, and the limiting control loop uses strong integral suppression parameters.

[0053] Step 2: Dynamic fuel flow distribution in start-up mode

[0054] In start-up mode, the opening degree of the fuel regulating valve and the target flow rate are controlled in an open-loop manner according to the preset rate gradient, while monitoring the T4 temperature. If T4 exceeds 540℃, the opening degree of the fuel valve is immediately limited until the temperature drops. When N2 reaches 5800rpm, the high-pressure compressor speed tracking closed-loop regulation loop takes effect, and multi-loop monitoring is started at the same time. The PID competition logic intervenes, and the upper and lower limits of fuel flow and correction coefficients are dynamically calculated to ensure safe and stable entry into idle operation.

[0055] Example 2: Dynamic Adjustment of Variable Load under Extreme Environments

[0056] Scenario description: The unit is operating under high operating conditions, and the ambient temperature drops sharply from 15℃ to 0℃.

[0057] Step 1: Reconstruction of Environmental Coupling Parameters

[0058] The dynamic safety adjustment module performs threshold curve migration: it queries the preset limit adjustment curve library for the corresponding high-pressure compressor speed limit N2 at 15℃. Lim The base speed threshold is 9500 rpm, but based on changes in ambient temperature, it is corrected to 9420 rpm. The actual control boundary after adding the motion margin is 9400 rpm.

[0059] Step 2: Multi-loop contention response

[0060] Before cooling, the operating condition was 1.0Ne, and the high-pressure compressor speed was 9430 rpm. The active power tracking loop had the minimum flow rate in the competition logic. The calculation model for the DROOP active power control loop is as follows: Figure 4 As shown, it is necessary to adjust the DROOP coefficient K according to the settings on the host computer. d The power turbine speed N3 and the active power setpoint P set by the host computer. SP Calculate the setpoint P of the DROOP active control loop. DSP Then, amplitude limiting and PID output calculation are performed:

[0061]

[0062] In the formula N 3SP The set value of the power turbine speed after grid connection, P norm This refers to the rated power of the gas turbine generator set.

[0063] After cooling, the active power tracking circuit and the N2 limiting circuit select the minimum fuel output value through competitive logic. This is because the high-pressure compressor limits the N2 speed. Lim When the threshold changes, the fuel demand F output by the N2 limit control loop... N2Lim When N2 approaches the upper limit threshold, the competition logic selects F to minimize it. N2Lim As the final reference value, it switches to the limit regulation loop to reduce fuel flow and prevent overspeed.

[0064] Step 3: Synergistic Suppression of Surge Risk

[0065] During the above adjustment process, all fuel demand calculation modules must meet the dynamic constraint flow range:

[0066] Minimum flow rate F min =f min (N2)×f T (T1)

[0067] Maximum flow rate F max =f max (N2)×f T (T1):

[0068] In the formula, f min and f max f represents the interpolation function for the curves showing the relationship between minimum and maximum flow rates and the high-pressure compressor speed N2, respectively. T This is a compensation coefficient curve related to ambient temperature.

[0069] Example 3: Fault-tolerant operation under sensor failure

[0070] Scenario description: All channels of the power sensor fail.

[0071] Step 1: Fault Diagnosis and Mode Switching

[0072] Power sensor calibration: If the power sensor measurement data exceeds the range, or the data jumps by more than 10% for three consecutive cycles, it is determined to be a hard fault.

[0073] Enable backup logic: power control switches to the flow-power fitting model.

[0074] Step 2: Flow-Power Fitting Compensation

[0075] When all power sensors fail, the flow compensation ΔF(k-1) calculated in the previous cycle remains unchanged:

[0076]

[0077] In the formula P m (k-1) represents the measured power of the power sensor in the previous calculation cycle, P fit (k-1) represents the power value fitted from the flow-power curve of the gas turbine operating history database in the previous calculation cycle (after first-order filtering), K P K I and K D These are the PID control parameters.

[0078] Based on this, the measured flow rate is superimposed, and the fitted value of the power is calculated according to the fitted relationship curve between fuel flow rate and output power (this curve is given from the gas turbine operation history database):

[0079] P fit (k)=f F-P (ΔF(k-1)+F m (k)) (5)

[0080] In the formula F m (k) represents the current measured flow rate, f F-P (·) represents the fitted curve function of fuel flow and output power based on the gas turbine operating history database. The gas turbine regulation algorithm replaces the measured fault value with the calculated power fitted value and switches the current regulation loop to the standby regulation loop. This allows the control system to continue control under the standby regulation loop without immediately executing the fault shutdown procedure, ensuring stable operation of the unit even when the power sensor fails. Operators can then intervene as needed.

Claims

1. A multi-parameter coordinated control and dynamic safety regulation system for a gas turbine, characterized in that, include: ● A multi-parameter collaborative control module is used to generate PID output signals for parallel control loops of high-pressure compressor speed (N2) tracking control, power turbine speed (N3) tracking control, active power tracking control, and low-pressure compressor speed (N1) limit control, high-pressure compressor speed (N2) limit control, and low-pressure turbine afterburner gas temperature (T4) limit control loops. This is achieved through mode selection and competition logic. The minimum fuel demand signal with the highest priority is selected as the control benchmark. ● The dynamic safety adjustment module is pre-set with dynamic speed threshold curves and temperature threshold curves related to the inlet air temperature, and adds a margin of safety to form a safety boundary. When the parameters approach the threshold, it automatically switches to the limit adjustment loop to limit the fuel valve opening until the parameters return to the safe range. ● The fuel demand calculation module generates a dynamic fuel demand signal based on the PID output signal, the maximum / minimum fuel flow limit curve of the high-pressure compressor speed (N2), and the ambient temperature correction factor. Flow distribution is achieved through valve control; ● Fault-tolerant module, used to perform the following operations in the event of a power sensor failure: Calculate the power fitting value based on the historical fuel flow rate and output power fitting curve and the flow rate correction value from the previous moment. Trigger an emergency shutdown when all speed or temperature measurement channels fail.

2. The system according to claim 1, characterized in that, The threshold curve generation method of the dynamic safety adjustment module includes: a. Set a baseline threshold based on the rated operating point in the gas turbine operating history database; b. Adding motion margins (±20 rpm or ±5℃) to form a safety boundary; c. Based on the inlet air temperature, perform equidistant translation to generate multi-segment segmented constraint curves.

3. The system according to claim 1, characterized in that, The maximum / minimum fuel flow limit curves in the fuel demand calculation module are dynamically adjusted in the following ways: ●Fix the upper limit of fuel flow in low-temperature environments to avoid the risk of surge caused by over-correction; ● The safe range of fuel valve opening is adjusted in real time by combining the high-pressure compressor speed (N2) and ambient temperature correction coefficient.

4. The system according to claim 1, characterized in that, The fault-tolerant module performs the following operations when the power sensor fails: The flow compensation calculated in the previous cycle is used as the baseline value, and the current measured flow is superimposed; based on the fitted relationship curve between fuel flow and output power (generated by the gas turbine operation history database), the power fitting value is calculated and the fault data is replaced.

5. The system according to claim 1, characterized in that, The multi-parameter collaborative control module introduces the DROOP active power control loop, which adjusts the mapping relationship between the target speed and fuel demand through the droop coefficient, thereby achieving dynamic load distribution and power balance for multi-machine parallel operation.

6. A computer-readable storage medium storing a computer program, characterized in that, When the program is executed by the processor, it implements the control logic of the system according to any one of claims 1-5.

Images