Three-layer dynamic optimization control system for deep peak regulation and primary frequency modulation of thermal power generating unit

Abstract

The invention discloses a three-layer dynamic optimization control system for deep peak regulation and primary frequency modulation of a thermal power generating unit, and belongs to the technical field of thermal power generation frequency modulation. A power grid frequency response model based on a traditional unit comprises a global coordination layer, an optimization control layer and an execution control layer. According to the signal uplink path, the execution control layer uploads unit state parameters to the optimization control layer in real time, and the optimization control layer receives signals and uploads dynamic response characteristics to the global coordination layer; the global coordination layer issues an optimization instruction, distributes total frequency modulation power and a power distribution proportion of a boiler and a steam turbine, the optimization control layer receives the optimization instruction from the global coordination layer, optimizes and calculates a unit target power value and a control boundary and issues a signal to the execution control layer, and the execution control layer receives the signal and controls the unit target power value and the control boundary. And stabilizing power grid fluctuation by utilizing turbine and boiler characteristic response. The control precision of the wide load range is improved, the combustion process is optimized, coal consumption and emission are reduced, and the energy efficiency of the unit is improved.

Description

Technical Field

[0001] This invention belongs to the field of frequency regulation technology for thermal power generation, and in particular relates to a three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation of thermal power units. Background Technology

[0002] Thermal power units play a crucial role in ensuring the safety of the power grid and regulating the system during the implementation of new power systems, thus maintaining a high proportion of renewable energy integration. However, with the continuous increase in renewable energy capacity, the uncertainty of renewable energy output poses a significant threat to the safe operation of the power grid. Therefore, the primary frequency regulation capability of thermal power units used for deep peak shaving has become a critical component.

[0003] Existing technologies generally suffer from three major bottlenecks: First, the use of a two-layer control architecture or centralized control makes it difficult to achieve dynamic power distribution between the boiler and the turbine, resulting in lag in response under low load conditions; second, reliance on fixed dead zones and preset response coefficients makes it impossible to adapt to nonlinear characteristics under deep peak shaving (such as valve flow curve distortion and increased combustion lag); and third, the safety protection mechanism is singular (only main steam pressure or flow limit), lacking a coordinated strategy with energy efficiency optimization, resulting in increased coal consumption during frequency regulation.

[0004] Therefore, there is an urgent need for a new technical solution to address this problem. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation of thermal power units, which is used to solve the problem of poor coordination between deep peak shaving and primary frequency regulation under low load conditions in the prior art.

[0006] The technical solution adopted in this invention is to provide a three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation of thermal power units, including a global coordination layer, an optimization control layer, and an execution control layer. The global coordination layer is used to dynamically allocate the total frequency regulation power demand and the boiler-turbine power allocation ratio. The optimization control layer is used to dynamically adjust the dead-zone frequency range and the response coefficient K. p The execution control layer is used for closed-loop feedback and adaptive optimization;

[0007] The system response signal's uplink path involves the execution control layer uploading unit status parameters in real time to the optimization control layer. The optimization control layer receives the signal, calculates the dynamic response characteristics, and uploads them to the global coordination layer.

[0008] The system response signal downlink path is initiated by the global coordination layer issuing global optimization commands to allocate total frequency regulation power and the power distribution ratio between the boiler and the turbine. The optimization control layer receives the optimization commands from the global coordination layer and corrects the dead-zone parameters under the frequency requirements. The signal is then transmitted to the execution control layer. Based on the commands from the global coordination layer and the constraints from the optimization control layer, the execution control layer updates the dynamic characteristic parameters of the unit, monitors the real-time power changes of the unit, and implements closed-loop feedback and adaptive optimization.

[0009] The unit status parameters include basic parameters, dynamic parameters, and execution feedback parameters. The basic parameters include the unit's main steam pressure, actual unit power, turbine valve opening, boiler fuel quantity, and air supply. The dynamic parameters are constructed based on frequency deviation, power change rate, and pressure change rate. The execution feedback parameters are constructed based on frequency regulation delay time, overshoot, and steady-state error.

[0010] The dynamic response characteristics include unit limits, unit nonlinear characteristics, and constraints. The unit limits include the maximum and minimum output capacity under the current unit load. The unit nonlinear characteristics include valve flow rates under low load. The constraints include the main steam pressure safety threshold and NO0. x Emission concentration.

[0011] The optimized control layer has a safety mechanism with dual constraints of main steam pressure limiting and power increment.

[0012] The optimized control layer dynamically adjusts the dead zone based on the unit load rate. Under low load conditions, the dead zone is widened, and under high load conditions, the dead zone is reduced.

[0013] The dynamic characteristic parameters of the unit include the dynamically adjustable quantities on the turbine side and the boiler side, the turbine opening degree, the boiler fuel quantity and air supply, and the unit power, which are obtained using an ARX linear time series model.

[0014] Through the above design scheme, the present invention can bring the following beneficial effects:

[0015] 1. Breakthrough in control performance: Through a three-layer dynamic collaborative architecture (global coordination layer + optimization control layer + execution layer), the response speed and flexibility of frequency regulation are greatly improved. The lower limit of deep peak load regulation of traditional thermal power units is extended to 30%Pe, which is more adaptable to the needs of high-proportion renewable energy grids.

[0016] 2. Synergistic optimization of energy efficiency and safety: By dynamically allocating the boiler-turbine power ratio and dynamically adjusting control parameters, the combustion process is optimized, reducing coal consumption and emissions, and improving unit energy efficiency. It also pioneers a dual protection mechanism of pressure closed-loop limiting and power increment to reduce the accident trigger rate.

[0017] 3. Universality of technical solution: By adopting the ARX linear time series model to replace complex algorithms, the computational complexity is reduced, and the adaptive dead-zone adjustment mechanism is compatible with different unit models, thus reducing the transformation cost.

[0018] 4. It is suitable for thermal power units that require frequent peak shaving, and is especially suitable for future power systems that are dominated by new energy sources. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the architecture of a three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation of thermal power units according to the present invention.

[0020] Figure 2 This is a flowchart of the dynamic dead zone adjustment logic of a three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation of thermal power units according to the present invention.

[0021] Figure 3 This is a block diagram illustrating the feedforward compensation algorithm implementation of a three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation of thermal power units according to the present invention.

[0022] Figure 4 This is a schematic diagram of the safety constraint calculation model of a three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation of thermal power units according to the present invention. Detailed Implementation

[0023] The present invention will be further described below with reference to the accompanying drawings and specific embodiments:

[0024] like Figure 1 As shown, the present invention discloses a three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation of thermal power units. Based on the grid frequency response model of traditional units, the system includes a global coordination layer, an optimization control layer, and an execution control layer. The system response signal path is divided into an uplink path and a downlink path.

[0025] For the signal uplink path, the execution control layer uploads the unit status parameters in real time, including basic parameters and dynamic parameters. The basic parameters include the unit's main steam pressure, the unit's actual power, the turbine control valve opening, the boiler fuel quantity, and the air supply.

[0026] Dynamic parameters, based on the following formula, are constructed based on frequency deviation (Δf) and power change rate. Pressure change rate Dynamic parameter coefficients:

[0027]

[0028] In the formula, f actual Indicates the actual frequency of the power grid; f nominal The rated frequency is indicated by P; the unit output power is indicated by P; and the main steam pressure is indicated by p.

[0029] The uploaded unit execution feedback parameters are used to construct a system based on the frequency modulation action delay time (T). d ), overshoot (M), steady-state error (e) ss Execution feedback parameters:

[0030]

[0031] In the formula, t response Indicates the time when the frequency modulation command is issued; t command Indicates the moment when the unit output reaches the commanded amplitude; P max P represents the actual peak output; target Indicates the target power for frequency modulation; P actual This indicates the actual power.

[0032] The execution control layer uploads the above signals to the optimization control layer. The optimization control layer receives the signals and calculates the unit limits, unit nonlinear characteristics, and constraints, including:

[0033] Optimize the maximum output capacity of the control layer under the current unit load, based on the following formula:

[0034] P max =max(P boilermax P turbinemax )-P curren1

[0035] In the formula, P boilermax P represents the maximum allowable output of the boiler under its current combustion conditions. turbinemax P represents the maximum output of the steam turbine at the current valve opening. current This indicates the current actual effort exerted.

[0036] The minimum output capacity of the control layer under the current unit load is optimized based on the following formula:

[0037] P min =min(P boilermin P turbinemin P emissionlimit )-P current

[0038] In the formula, P boilermin Indicates the minimum stable combustion load of the boiler; P turbinemin P represents the minimum output of the steam turbine at the current valve opening. emis sionlimit This indicates the minimum load required to meet emission standards.

[0039] Based on the current valve status and steam status, the optimized control layer approximates the dynamic characteristics of valve flow (Q) under low load using the following formula:

[0040]

[0041] In the formula, C v denoted by ; f(x) represents the valve flow coefficient; Δp represents the pressure difference across the valve; and ρ represents the steam density.

[0042] Based on the system design and safe operation, the optimized control layer calculates the main steam pressure safety threshold using the following formula:

[0043] p min ≤p(t)≤p max

[0044] In the formula, p min This indicates the minimum allowable main steam pressure; p max This indicates the maximum allowable main steam pressure.

[0045] The optimized control layer further optimizes the emission standards for pollutants based on the following formula, and uses NO as the basis for the national emission standards. x For example, let's calculate environmental emission limits:

[0046]

[0047] In the formula, C NOx Indicates NO x Emission concentration; m NOx Indicates NO x Discharge mass flow rate; C limit This indicates emission limits.

[0048] After the optimized control layer completes the calculation of the above response characteristics, it uploads the above dynamic response characteristics to the global coordination layer.

[0049] The system response signal downlink path is initiated by the global coordination layer issuing optimization commands. The global coordination layer allocates the total frequency regulation power demand and the respective power allocation ratios of the boiler and turbine. The global coordination layer then issues optimization command signals, which are transmitted to the optimization control layer.

[0050] The optimization control layer corrects dead-time parameters based on frequency requirements, such as... Figure 2 The diagram shown is a flowchart of the dynamic dead zone adjustment logic according to an embodiment of the present invention. The dead zone is dynamically adjusted based on the unit load rate to avoid pressure fluctuations caused by frequent actions at low loads, which would expand the dead zone. During low load periods (e.g., 20%–50%), due to poor combustion stability, the dead zone needs to be widened to avoid control oscillations. During high load periods (>80%), a smaller dead zone is used to quickly respond to grid AGC (Automatic Generation Control) commands.

[0051] The following is the dead-zone adjustment logic, where P is the unit load factor:

[0052] If 20% < P < 30%, set the dead zone to ±1.5%.

[0053] If 30% < P < 50%, set the dead zone to ±1.0%.

[0054] If 50% < P < 80%, set the dead zone to ±0.5%.

[0055] If 80% < P < 100%, set the dead zone to ±0.3%.

[0056] To further improve the response speed, set a response coefficient that is updated in real time according to the unit's frequency regulation performance and the system frequency deviation, and optimize the control layer to dynamically adjust the response coefficient K p based on the following formula:

[0057]

[0058] where K p,bass is the basic response coefficient calibrated by the unit's frequency regulation performance; Δf is the system frequency deviation; f deadband is the frequency modulation dead zone threshold; α and β are weighting coefficients.

[0059] The optimization control layer further adopts safety boundary protection, performs closed-loop amplitude limiting on the main steam pressure and power increment constraint to prevent affecting the safe operation of the unit. According to the heat storage capacity of the unit and the dynamic characteristics of the boiler, limit the single-frequency modulation power increment (ΔP cmd ), based on the following formula:

[0060] ΔP cmd = min(ΔP req , ΔP max ·(1 - e -t / τ ))

[0061] where, ΔP req is the frequency modulation demand power increment; ΔP max is the maximum instantaneous power increment allowed by the unit; τ is the heat storage time constant of the unit, that is, the inertial characteristic of the unit to buffer power changes by releasing or absorbing internal heat storage.

[0062] The optimization control layer completes the above optimization calculations, sends out signals, and transmits the signals to the execution control layer. The execution control layer receives the upper signals, and according to the instructions from the global coordination layer and the constraints pointed out by the optimization control layer, the execution control layer adjusts the dynamically adjustable amounts on the steam turbine side and the boiler side, changes the opening of the steam turbine, changes the fuel supply and air supply of the boiler, monitors the real-time power change of the unit, and performs closed-loop feedback and adaptive optimization of the actual output and the target output according to the following formula.

[0063] Calculate the actual output (Pactual ) and target output (P) target )deviation:

[0064] ΔP(t k )=|P actual (t k )-P target (t k )|.

[0065] And perform threshold comparison:

[0066] ΔP(t k >0.02P e

[0067] t k =5k (k=1,2,…).

[0068] The execution control layer compares the actual output with the target output every 5 seconds. If the deviation is greater than 0.02Pe (rated power), the signal is fed back to the optimization control layer, triggering a recalculation of the feedforward quantity. The optimization control layer updates the data in real time and transmits the optimized data back to the execution control layer.

[0069] To accelerate the updating of unit dynamic characteristic parameters and improve system response speed, the execution control layer further adopts the ARX linear time series model to update the unit dynamic characteristic parameters in real time. Taking unit power (P) as an example, it is as follows:

[0070] P(k)=a1P(k-1)+b1u(k-1)+b2u(k-2)

[0071] In the formula, P is the unit power; u is the valve opening command; k is time k; a1, b1, and b2 are dynamic update coefficients that quantify the impact of historical outputs and inputs on the current state.

[0072] The implementation of the present invention is not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and are included within the protection scope of the present invention.

Claims

1. A three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation in thermal power units, characterized in that: It includes a global coordination layer, an optimization control layer, and an execution control layer. The global coordination layer is used to dynamically allocate the total frequency regulation power demand and the boiler-turbine power allocation ratio. The optimization control layer is used to dynamically adjust the dead-zone frequency range and the response coefficient K. p The execution control layer is used for closed-loop feedback and adaptive optimization; The system response signal's uplink path involves the execution control layer uploading unit status parameters in real time to the optimization control layer. The optimization control layer receives the signal, calculates the dynamic response characteristics, and uploads them to the global coordination layer. The system response signal downlink path is initiated by the global coordination layer issuing global optimization commands to allocate total frequency regulation power and the power distribution ratio between the boiler and the turbine. The optimization control layer receives the optimization commands from the global coordination layer and corrects the dead-zone parameters under the frequency requirements. The signal is then transmitted to the execution control layer. Based on the commands from the global coordination layer and the constraints from the optimization control layer, the execution control layer updates the dynamic characteristic parameters of the unit, monitors the real-time power changes of the unit, and implements closed-loop feedback and adaptive optimization.

2. The three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation of thermal power units according to claim 1, characterized in that: The unit status parameters include basic parameters, dynamic parameters, and execution feedback parameters. The basic parameters include the unit's main steam pressure, actual unit power, turbine valve opening, boiler fuel quantity, and air supply. The dynamic parameters are constructed based on frequency deviation, power change rate, and pressure change rate. The execution feedback parameters are constructed based on frequency regulation delay time, overshoot, and steady-state error.

3. The three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation of thermal power units according to claim 1, characterized in that: The dynamic response characteristics include unit limits, unit nonlinear characteristics, and constraints. The unit limits include the maximum and minimum output capacity under the current unit load. The unit nonlinear characteristics include valve flow rates under low load. The constraints include the main steam pressure safety threshold and NO0. x Emission concentration.

4. The three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation of thermal power units according to claim 1, characterized in that: The optimized control layer has a safety mechanism with dual constraints of main steam pressure limiting and power increment.

5. A three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation of thermal power units according to claim 1, characterized in that: The optimized control layer dynamically adjusts the dead zone based on the unit load rate. Under low load conditions, the dead zone is widened, and under high load conditions, the dead zone is reduced.

6. The three-layer dynamic optimization control system for deep peak shaving and primary frequency regulation of thermal power units according to claim 1, characterized in that: The dynamic characteristic parameters of the unit include the dynamically adjustable quantities on the turbine side and the boiler side, the turbine opening degree, the boiler fuel quantity and air supply, and the unit power, which are obtained using an ARX linear time series model.

Images