A deep scheduling method and system for deep peak regulation of a coal-fired unit

Abstract

The present application relates to the technical field of power system resource optimization scheduling, and particularly relates to a deep scheduling method and system for deep peak regulation of coal-fired units, which comprises the following steps: collecting multi-dimensional operation parameters of the unit, deep peak regulation instructions and carbon price signals of the carbon emission trading market, and realizing timestamp alignment; constructing a five-dimensional safety constraint model associated with the economic cost boundary; taking the maximization of peak regulation income as the target, constructing an objective function by fusing the coal purchase cost, carbon trading cost and equipment maintenance cost, solving the economic scheduling scheme by using a dynamic programming algorithm, and executing a cross-system collaborative strategy; based on the LSTM, predicting the equipment life and digital twin simulation, and triggering the cross-system correction strategy. The present application solves the business collaboration problem of economy, environmental protection and equipment reliability in deep peak regulation, and realizes the maximization of peak regulation income of coal-fired units under the power market environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system resource optimal scheduling, in particular to a deep scheduling method and system for deep load modulation of a coal-fired unit. BACKGROUND

[0002] In the early days, coal-fired units mainly undertook base load generation tasks in the power system, and the operation condition was relatively stable. The combustion system, steam turbine distribution logic and auxiliary machine configuration of the coal-fired unit were designed and optimized around the high-efficiency stable operation under rated load. The scheduling method focused on meeting the power generation demand under rated load, and lacked adaptability to deep load modulation operation scenarios of the unit. Moreover, a multi-dimensional safety constraint collaborative control mechanism for the furnace stable combustion boundary, SCR denitration efficiency lower limit, steam turbine vibration threshold, carbon emission intensity upper limit and secondary reheat system temperature difference safety domain in the deep load modulation process was not established.

[0003] With the construction of the new power system, the installed capacity of renewable energy sources such as wind power and photovoltaic power has exceeded 45% by 2023. The intermittent and volatile characteristics of renewable energy sources pose a serious challenge to the flexibility of the power grid. At the same time, the carbon emission rights trading market has started, and the carbon price signal has become an important variable in the economic scheduling of the unit. Although the application of secondary reheat technology in coal-fired units has improved energy efficiency, it has also increased the complexity of the scheduling scenario. Traditional load modulation technology has focused on a single load regulation target and cannot meet the multiple demands of deep load modulation rated load stable operation, low-carbon carbon quota trading cost constraints and energy efficiency improvement of the secondary reheat system.

[0004] Currently, although some research and applications focus on deep load modulation scheduling of coal-fired units, there are still significant deficiencies in existing technologies. On the one hand, there is a lack of comprehensive modeling of multi-dimensional safety constraints of the unit during deep load modulation. A five-dimensional safety constraint model including the stable combustion load upper limit, SCR denitration temperature lower limit, steam turbine vibration upper limit, carbon emission intensity upper limit and secondary reheat temperature difference safety threshold cannot be established, which cannot effectively guarantee the stability and reliability of the unit under low load operation. On the other hand, for the complex scenario of deep load modulation coupled with the carbon trading market, secondary reheat system and other factors, existing scheduling methods cannot achieve collaborative control and optimization of multiple systems. Moreover, they do not integrate peak shaving auxiliary service income, coal purchase cost, carbon quota trading cost and equipment maintenance cost to build a function with the goal of maximizing peak shaving income. Furthermore, they lack the prediction of the service life of steam turbine blades, SCR catalyst and other equipment based on LSTM neural networks, and the cross-system correction strategy of digital twin simulation safety boundary breakthrough scenarios integrated with APROS and EBSILON platforms. They cannot fully utilize the flexibility regulation potential of coal-fired units in the new power system, and limit the large-scale consumption of renewable energy and the low-carbon transformation process of the power system.

[0005] The invention patent with publication number CN114444785B discloses a deep scheduling method and system for deep peak regulation of coal-fired units, which builds a carbon emission intensity and transaction cost model, takes the minimum total peak regulation cost as the objective function, and solves it combined with multiple constraint conditions, but does not consider the efficiency loss of the secondary reheating system under deep peak regulation, and lacks the linkage protection mechanism of turbine vibration and SCR temperature. SUMMARY

[0006] The purpose of the present application is to solve the problems of existing technology, such as fuzzy safety boundary in deep peak regulation of coal-fired units, energy efficiency loss of secondary reheating system, lack of multi-system cooperation, lag of equipment maintenance, and extensive economic scheduling, and proposes a deep scheduling method and system for deep peak regulation of coal-fired units.

[0007] In order to achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0008] A deep scheduling method for deep peak regulation of coal-fired units, comprising the following steps:

[0009] Step S1, collecting multi-dimensional operation parameters of the coal-fired unit through the distributed control system, and synchronously accessing the deep peak regulation instruction sent by the power grid dispatching end and the carbon price signal of the carbon emission rights trading market;

[0010] Step S2, building a five-dimensional safety constraint model related to economic cost boundary, including upper limit of stable combustion load, lower limit of SCR denitration temperature, upper limit of turbine vibration, upper limit of carbon emission intensity, and secondary reheating temperature difference safety threshold;

[0011] Step S3, building an objective function including coal purchase cost, carbon quota transaction cost, and equipment maintenance cost, based on the five-dimensional safety constraint model, setting constraint conditions in each dimension, and using dynamic programming algorithm to solve the economic scheduling scheme meeting the requirements of the power market;

[0012] Step S4, based on the economic scheduling scheme, executing cross-system economic coordination strategy, including fuel cost control, maintenance risk suppression, carbon transaction linkage, energy optimization, and thermal stress loss prevention and control;

[0013] Step S5, based on the equipment life prediction and digital twin simulation of LSTM, triggering cross-system correction strategy, maintaining the safety boundary of commercial operation, and reducing the risk of loss of peak regulation auxiliary service income caused by equipment failure.

[0014] A deep scheduling system for deep peak regulation of coal-fired units, comprising:

[0015] The data acquisition module acquires multi-dimensional operation parameters of the coal-fired unit through a distributed control system, synchronously accesses a deep peak regulation instruction sent by a power grid dispatching end and a carbon price signal of a carbon emission right trading market through an edge computing unit, and realizes timestamp alignment of full-quantity parameters.

[0016] The safety modeling module constructs a five-dimensional safety constraint model, including a stable combustion load upper limit, an SCR denitration temperature lower limit, a steam turbine vibration upper limit, a carbon emission intensity upper limit and a secondary reheat temperature difference safety threshold.

[0017] The optimization solving module constructs a function with maximization of peak regulation auxiliary service income as a target, sets constraint conditions of load, environmental protection, steam turbine, secondary reheat and carbon emission based on the five-dimensional safety constraint model, and solves an economic dispatching scheme including peak regulation output distribution, carbon quota decision and marginal benefit optimization by using a dynamic programming algorithm.

[0018] The collaborative control module executes a cross-system economic collaboration strategy based on the economic dispatching scheme, and is used for realizing fuel cost control, maintenance risk inhibition, carbon transaction linkage, energy consumption optimization and thermal stress loss prevention and control.

[0019] The safety correction module predicts equipment life based on an LSTM neural network, generates an early warning signal containing peak regulation auxiliary service income loss estimation, simulates a secondary reheat system safety boundary breakthrough scenario through a digital twin body integrated with an APROS and an EBSILON platform, triggers a cross-system correction strategy, and ensures minimization of peak regulation auxiliary service income loss.

[0020] The technical scheme provided by the present application has at least the following beneficial effects:

[0021] The present application can realize collaborative control of multi-dimensional safety indexes under deep peak regulation conditions by constructing a five-dimensional safety constraint model related to an economic cost boundary, integrating safety boundaries of stable combustion load, denitration temperature, steam turbine vibration, carbon emission intensity and secondary reheat temperature difference, avoiding operation risks under single safety constraint, and improving operation stability of the unit in a wide load range.

[0022] The present application can optimize comprehensive economic income of the unit, and realize unification of economy and safety under deep peak regulation by solving a target function integrating peak regulation auxiliary service income, coal purchase cost, carbon quota transaction cost and equipment maintenance cost through a dynamic programming algorithm, and responding to a carbon price signal and a deep peak regulation instruction.

[0023] The present application can solve the problem of control fragmentation of boilers, steam turbines and environmental protection islands in traditional dispatching by executing a cross-system economic collaboration strategy and outputting collaborative control instructions to combustion, steam turbine, environmental protection, thermal auxiliary machinery and secondary reheat systems, and realizing multi-system linkage regulation.

[0024] The application can realize early warning and cross-system correction of equipment failure, reduce catalyst replacement cost and steam turbine blade water erosion risk, and reduce peak shaving income loss caused by equipment failure by predicting the remaining life of the equipment through the LSTM neural network and generating a warning signal containing peak shaving income loss estimation, combining the digital twin simulation safety boundary breakthrough scene of the integrated APROS and EBSILON platforms.

[0025] The application can improve the accuracy of carbon emission intensity calculation under high-sulfur coal working conditions by using the sulfur-carbon emission correction coefficient, and can improve the accuracy of carbon emission calculation, reduce the carbon intensity per unit of electricity, help the unit to adapt to the requirements of the carbon trading market, and realize low-carbon operation. BRIEF DESCRIPTION OF DRAWINGS

[0026] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0027] Figure 1 The method flow chart of the deep scheduling method for deep peak regulation of the coal-fired unit provided by the embodiment of the present application is shown in the figure.

[0028] Figure 2 The system architecture diagram of the deep scheduling system for deep peak regulation of the coal-fired unit provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0029] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following describes the deep scheduling method and system for deep peak regulation of the coal-fired unit according to the present application in combination with the preferred embodiments, the specific implementation, structure, features and effects of which are described in detail as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0030] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0031] The following embodiments are for illustrative purposes only and are not intended to limit the scope of the present application.

[0032] The following describes the specific scheme of the deep scheduling method and system for deep peak regulation of the coal-fired unit provided by the present application in combination with the drawings.

[0033] Referring to Figure 1 which shows a method flow chart of a deep scheduling method for deep peak regulation of a coal-fired unit according to an embodiment of the present application, the method comprises the following steps:

[0034] Step S1, collecting multi-dimensional operation parameters of the coal-fired unit through the distributed control system, synchronously accessing the deep peak regulation instruction sent by the power grid dispatching end and the carbon price signal of the carbon emission right trading market;

[0035] Step S1 further comprises the following sub-steps:

[0036] S1-1, collecting boiler parameters, including main steam pressure, main steam temperature, furnace pressure and oxygen content;

[0037] S1-2, collecting environmental protection island parameters, including SCR inlet flue gas temperature, desulfurization tower outlet sulfur dioxide concentration and dust concentration at the dust remover outlet;

[0038] S1-3, collecting turbine parameters, including turbine vibration value, turbine vacuum degree and high-pressure cylinder exhaust temperature;

[0039] S1-4, collecting fuel and load parameters, including real-time coal consumption, instantaneous coal feeding signal of the coal feeder and output power of the generator;

[0040] S1-5, collecting secondary reheat system operation parameters, including secondary reheat steam temperature, secondary reheat steam pressure, secondary reheat desuperheating water quantity and secondary reheat system bypass valve opening degree;

[0041] S1-6, synchronously collecting the deep peak regulation instruction and the carbon price signal of the carbon emission right trading market by using the edge computing unit, establishing a communication connection with the distributed control system through the OPC UA protocol, and based on the real-time clock signal of the communication interaction, time stamp calibration is performed on the deep peak regulation instruction and the carbon price signal of the carbon emission right trading market and the multi-dimensional operation parameters collected by the distributed control system, to ensure the time dimension consistency of all parameters.

[0042] It should be noted that the distributed control system adopts a redundant configuration of Siemens PCS 7 or ABB AC 800M system.

[0043] The furnace pressure in the boiler parameters is collected by the Rosemount 3051 pressure transmitter, with an accuracy of ±0.1% FS and a sampling frequency of 100Hz, for capturing combustion stability fluctuations.

[0044] The secondary reheat steam temperature is measured by an S-type thermocouple, and a temperature transmitter is used to realize a measurement accuracy of ±1℃, meeting the demand of the secondary reheat system thermal stress calculation.

[0045] The edge computing unit adopts Advantech UNO-3083G industrial computer, built-in OPC UA client server (compliant with IEC 62541 standard), and realizes timestamp alignment through the following mechanisms:

[0046] At the hardware level, a GPS clock module is installed, with a synchronization accuracy of ±1μs;

[0047] At the software level, the IEEE 1588 precision clock protocol is used to add nanosecond-level timestamps to the deep peak shaving instructions (Modbus TCP format) and carbon price signals (JSON format), and data synchronization is realized through the Linux system real-time scheduling kernel.

[0048] The furnace pressure collection point is arranged at the four corners of the boiler burner area, the pressure taking pipe adopts Φ14×2 stainless steel pipe, and the built-in anti-blocking filter screen measures the furnace negative pressure fluctuation through the differential pressure transmitter. When the pressure fluctuation amplitude exceeds ±50Pa, the combustion stability early warning is triggered, providing data support for the 20% load stable combustion boundary.

[0049] The SCR inlet flue gas temperature collection adopts array thermocouple (6-point average temperature measurement), which is arranged 1m before the SCR reactor inlet to ensure temperature field uniformity. When the flue gas temperature is <300℃, the staged heat recovery system is automatically started.

[0050] The secondary reheat system bypass valve opening is collected by a high-precision displacement sensor, and forms a closed-loop monitoring with the desuperheating water quantity (electromagnetic flowmeter, accuracy ±0.5%), providing real-time data for the bypass valve cooperative control and solving the secondary reheat temperature difference control lag problem in the prior art.

[0051] The communication between the edge computing unit and the DCS adopts OPC UA security strategy, including:

[0052] At the transport layer, TCP / IP protocol is used, and port 4840 (standard OPC UA port) is used;

[0053] At the security layer, AES-256 encryption and X.509 certificate authentication are used to prevent carbon price signal tampering;

[0054] The data model follows the OPC UA data model mapping IEC 61850-7-420, maps the peak shaving instructions to OPC UA nodes, and realizes seamless integration with the distributed control system.

[0055] The timestamp alignment adopts a bidirectional timestamp correction algorithm, and the edge computing unit sends a synchronization frame (containing a local GPS clock, accuracy ±1μs) to the distributed control system every 100ms; the distributed control system returns a response frame with its own clock, and the edge computing unit calculates the time delay compensation value through the Kalman filtering algorithm to ensure that the end-to-end synchronization error is ≤2ms (including communication delay and protocol processing time).

[0056] Step S2, construct a five-dimensional safety constraint model associated with economic cost boundary, including stable combustion load upper limit, SCR denitration temperature lower limit, steam turbine vibration upper limit, carbon emission intensity upper limit and secondary reheat temperature difference safety threshold;

[0057] In step S2, the following sub-steps are also included:

[0058] S2-1, based on the balance between furnace pressure fluctuation and peak shaving auxiliary service income, set the stable combustion load upper limit to 20% of the rated load, ensure that the low load coal purchasing cost does not exceed the peak shaving auxiliary service income;

[0059] S2-2, based on the correlation between SCR inlet flue gas temperature and catalyst maintenance cost, set the SCR denitration temperature lower limit to 300℃, avoid the cost surge caused by insufficient catalyst activity;

[0060] S2-3, based on the mapping relationship between steam turbine vibration value and blade maintenance cost, set the steam turbine vibration upper limit to 76μm, balance the acquisition of peak shaving auxiliary service income and equipment maintenance cost;

[0061] S2-4, based on the desulfurization tower outlet sulfur dioxide concentration, coal feeder instantaneous coal feeding amount signal and generator output power, calculate the carbon emission intensity upper limit, the carbon emission intensity upper limit is related to the carbon quota transaction cost, ensure that the carbon cost is controllable, the formula is expressed as:

[0062]

[0063] Among them, represents the unit power carbon emission intensity; represents the standard coal emission factor; represents the real-time coal consumption; represents the unit power carbon emission intensity; represents the sulfur fraction-carbon emission correction coefficient;

[0064] S2-5, based on the difference between the secondary reheat steam temperature and the high pressure cylinder exhaust steam temperature, set the secondary reheat temperature difference safety threshold, the secondary reheat temperature difference safety threshold is related to the equipment maintenance cost, the formula is expressed as:

[0065]

[0066] Among them, represents the secondary reheat temperature difference safety threshold; represents the secondary reheat steam temperature; represents the high pressure cylinder exhaust steam temperature; ≤80℃ represents the safety upper limit of the secondary reheat temperature difference is 80℃, when ΔT> 80℃, trigger the secondary reheat system thermal stress over-limit warning.

[0067] It should be noted that: based on the furnace pressure fluctuation to determine the combustion stability, when the unit load is lower than 20%, the furnace temperature decreases significantly, which is easy to cause unstable combustion. Through real-time monitoring of the furnace pressure fluctuation (such as when the fluctuation amplitude exceeds ± 30Pa, it is determined that the combustion is unstable), it is verified through hot state test that the 660MW unit can maintain stable combustion at 20% load (132MW), so 20% is set as the upper limit of stable combustion load to avoid the risk of low load extinction.

[0068] The lower limit of the SCR denitration temperature is the minimum inlet flue gas temperature necessary to ensure the activity of the SCR denitration catalyst, which is set to 300℃ in the present application. The activity of the SCR catalyst is highly dependent on the inlet flue gas temperature. When the flue gas temperature is lower than 300℃, the conversion rate of the catalyst to NOx will be greatly reduced, which may lead to an increase in ammonia escape rate, and in turn cause problems such as air preheater blockage, which not only affects the denitration effect, but also causes a sharp increase in the cost of catalyst replacement.

[0069] The upper limit of the steam turbine vibration is the maximum vibration value allowed to prevent water erosion of the steam turbine blades, which is set to 76μm in the present application. The last stage blades of the low-pressure cylinder of the steam turbine are prone to water erosion under low load conditions due to steam with water, and the vibration value is a key indicator of water erosion risk. According to relevant standards, 76μm is the safety threshold for steam turbine bearing vibration. When the steam turbine vibration value exceeds this upper limit, the water erosion rate of the blades will significantly increase, increasing the cost of equipment maintenance.

[0070] The calculation formula is:

[0071]

[0072] Among them, the logarithmic relationship reflects The effect of concentration on carbon emission shows a marginal decreasing effect, and the logarithmic form is more consistent with the actual chemical reaction kinetics; Sulfur dioxide concentration at the outlet of the desulfurization tower; the coefficient 0.175 represents the slope coefficient of sulfur content; 0.98 represents the reference carbon conversion rate.

[0073] The secondary reheating temperature difference reflects the thermal stress level. When ΔT> 80℃, the thermal stress of the reheater pipe exceeds the limit, which may cause creep rupture. Therefore, 80℃ is set as the threshold value, and when the temperature difference exceeds the limit, a warning is triggered. By adjusting the amount of desuperheating water, the opening degree of the bypass valve and other measures, the thermal stress is controlled to ensure the safety of the equipment.

[0074] In step S3, a target function including coal purchase cost, carbon quota transaction cost and equipment maintenance cost is constructed with the goal of maximizing the peak-shaving auxiliary service income, constraint conditions in each dimension are set based on the five-dimensional safety constraint model, and a dynamic programming algorithm is used to solve an economic dispatching scheme that meets the requirements of the power market;

[0075] In step S3, the following sub-steps are also included:

[0076] S3-1, a target function including coal procurement cost, carbon quota transaction cost, and equipment maintenance cost is constructed with the goal of maximizing peak shaving auxiliary service revenue, and the formula is expressed as:

[0077]

[0078] wherein, represents the cumulative calculation in the time range from time 1 to time N; max represents the maximum value of the target function; represents the peak shaving compensation revenue obtained by the unit participating in the power auxiliary service market; represents the coal procurement cost corresponding to the coal consumption; represents the carbon quota transaction cost corresponding to the unit carbon emission; represents the equipment maintenance cost, including the cost of secondary reheat efficiency loss; represents the target function;

[0079] The peak shaving auxiliary service revenue is associated with the power auxiliary service market rule and is executed in stages, and the peak shaving auxiliary service revenue calculation formula is expressed as:

[0080]

[0081] wherein, represents the actual peak shaving output of the unit in the t period; represents the peak shaving compensation unit price in the t period;

[0082] S3-2, based on the five-dimensional safety constraint model, set the conditions corresponding to each dimension constraint, including:

[0083] Load constraint, based on the upper limit of stable combustion load, set the load variation range to 20%-100% rated power, ensure that the load interval of economic dispatch does not break through the combustion stability boundary;

[0084] Environmental protection constraint, related to the requirement of SCR denitration temperature lower limit on environmental protection performance, set NOx emission concentration ≤25mg / Nm³ and dust emission concentration ≤2mg / Nm³, ensure that the environmental protection indicators meet the requirements in the economic optimization process;

[0085] Turbine constraint, combined with the vibration upper limit of the turbine to control the reliability of the equipment, set the variable load rate ≥1.5%Pe / min, avoid the risk of water erosion caused by vibration overrun while responding to peak shaving auxiliary service revenue;

[0086] The secondary reheat constraint is based on a secondary reheat temperature difference safety threshold, sets the secondary reheat steam temperature change rate < 1.5 ℃ / min, and the secondary reheat system bypass valve opening adjustment range is 10%-90%, to ensure that the thermal stress of the secondary reheat system in the economic dispatch is in a safe interval;

[0087] The carbon emission constraint is associated with the low-carbon requirement of the upper limit of carbon emission intensity, sets the carbon emission intensity ≤ the upper limit of carbon emission intensity, to ensure that the economic dispatch scheme meets the cost control target of the carbon trading market;

[0088] S3-3, the dynamic programming algorithm is used to solve the economic dispatch scheme meeting the requirements of the power market, including:

[0089] Peak output distribution, based on the deep peak shaving instruction and the upper limit of stable combustion load, the optimal output interval of each period is predicted;

[0090] Carbon quota decision, combined with the carbon price signal of the carbon emission right trading market and the upper limit of carbon emission intensity, the sensitivity of the carbon quota trading cost to the objective function is analyzed, and the carbon quota buying and selling opportunity is determined;

[0091] Marginal benefit optimization, based on the economic correlation of turbine maintenance cost and secondary reheat maintenance cost, the marginal balance point of device maintenance cost and peak shaving auxiliary service income is calculated.

[0092] It should be noted that: in the dispatching process of deep peak shaving of coal-fired units, in order to realize economic operation, the present application constructs an objective function with the maximum peak shaving auxiliary service income as the target. The objective function comprehensively considers various economic factors such as coal purchasing cost, carbon quota trading cost and device maintenance cost, aiming to optimize these cost factors and improve the economic benefit of coal-fired units in the process of deep peak shaving. The objective function is composed of four parts: peak shaving auxiliary service income, coal purchasing cost, carbon quota trading cost and device maintenance cost. Among them, the peak shaving auxiliary service income refers to the economic compensation obtained by the unit participating in the power auxiliary service market, which is closely related to the peak shaving capacity of the unit and the market rules; the coal purchasing cost is directly related to the coal consumption and coal price; the carbon quota trading cost depends on the carbon emission of the unit and the carbon price of the carbon emission right trading market; the device maintenance cost covers the daily maintenance, repair and additional device wear cost caused by deep peak shaving working condition and the like.

[0093] The peak shaving auxiliary service income is related to the calculation of the power auxiliary service market rules and is executed in stages. Among them, the peak shaving compensation unit price is set according to the load level, for example, 20%-30% load interval compensation 200 yuan / MWh, 30%-50% load interval compensation 100 yuan / MWh, 50%-70% load interval compensation 30 yuan / MWh, etc.

[0094] The coal purchasing cost is calculated by the instantaneous coal supply signal of the coal feeder and the real-time main steam pressure / temperature , combined with the market coal price and the operation time A to generate the cost term , and adjusted by the coal consumption correction coefficient under low load conditions.

[0095] The carbon quota trading cost is dynamically calculated by the desulfurization tower outlet concentration, real-time carbon price and power generation P , and the carbon emission intensity is introduced to calibrate the sulfur-carbon emission correction coefficient, and the final cost term is , and the digital twin correction is triggered when the carbon emission intensity exceeds 800g / kWh.

[0096] The equipment maintenance cost is mainly caused by the irreversible heat loss caused by the temperature difference exceeding the limit. When ΔT exceeds the safety threshold (80℃), the reheater pipe thermal stress increases, the steam enthalpy decreases, and the unit heat consumption rate rises, and the calculation formula is:

[0097]

[0098] Where, ΔQ represents the reheating system efficiency loss, which is obtained by ΔT table; B represents the low heat value of standard coal; A represents the operation time; η represents the boiler efficiency, which is provided by the distributed control system in real time.

[0099] The constraint condition limits the safe and economic operation boundary of the unit, the load range is 20%-100%: ensures the stability of combustion and the efficiency of denitration, NOx≤25mg / Nm³, dust≤2mg / Nm³, meets the ultra-low emission requirements, the variable load rate≥1.5%Pe / min, meets the demand of fast peak shaving of power grid, the secondary reheating steam temperature change rate<1.5℃ / min, the opening adjustment range of the bypass valve of the secondary reheating system is 10%-90%, to prevent thermal stress exceeding the limit and valve failure, and the upper limit of carbon emission intensity is set to meet the low-carbon requirement of the carbon emission right trading market. By setting the carbon emission intensity≤the upper limit of carbon emission intensity, the carbon emission of the unit can be reasonably controlled in the economic dispatching process, and the problem of increasing carbon quota purchase cost due to excessive carbon emission can be avoided.

[0100] Dynamic programming algorithm is a classic optimization algorithm, which is suitable for solving optimization problems in multi-stage decision-making process. In the economic dispatching problem of coal-fired unit deep peak shaving, the dispatching process can be divided into multiple stages, each stage corresponds to different time period or operation state. Dynamic programming algorithm can find the optimal solution of the whole problem by decomposing the original problem into a series of interrelated sub-problems and solving these sub-problems respectively.

[0101] ​The dynamic programming algorithm predicts the optimal output interval of each period based on the deep peak regulation instruction and the upper limit of stable combustion load. Considering factors such as unit start-up and shutdown cost, the output distribution is optimized to meet the grid peak regulation demand, ensure the stability and economy of operation, and maximize the peak regulation auxiliary service revenue.

[0102] The dynamic programming algorithm combines the carbon price signal of the carbon emission trading market and the upper limit of carbon emission intensity to analyze the sensitivity of carbon quota trading cost to the objective function and determine the buying and selling opportunity. When the carbon price is low and the emission is large, buy in; when the carbon price is high and the emission is small, sell out, meet the carbon emission constraint, optimize the transaction cost, and improve the economic benefit.

[0103] The dynamic programming algorithm calculates the marginal balance point of device maintenance cost and peak regulation auxiliary service revenue based on the economic correlation of turbine and secondary reheat maintenance cost. According to factors such as operating state, adjust the maintenance plan and operation strategy to realize the optimal balance of cost and benefit, and ensure the economic operation of the unit.

[0104] In step S4, the cross-system economic coordination strategy is executed based on the economic dispatching scheme, including fuel cost control, maintenance risk suppression, carbon trading linkage, energy consumption optimization and thermal stress loss prevention;

[0105] In step S4, the following sub-steps are also included:

[0106] S4-1, according to the real-time ratio of coal purchasing cost to peak regulation auxiliary service revenue, switch coal types and control the increase of coal purchasing cost not to exceed the preset proportion of peak regulation auxiliary service revenue;

[0107] S4-2, when the turbine vibration value approaches the correlation boundary of blade maintenance cost, throttle steam distribution mode is enabled to suppress device maintenance cost;

[0108] S4-3, according to the carbon price signal interval of the carbon emission trading market, dynamically adjust the SCR denitration efficiency to control the denitration cost and the carbon quota trading cost not to exceed the preset proportion of peak regulation auxiliary service revenue;

[0109] S4-4, optimize the combination of auxiliary operation sets of thermal and auxiliary systems to ensure that the power saving benefit covers the device adjustment cost;

[0110] S4-5, when the secondary reheat temperature difference approaches the safety threshold of the related device maintenance cost, start the linkage adjustment of secondary reheat desuperheating water and secondary reheat system bypass valve to avoid economic loss of shutdown maintenance.

[0111] It should be noted that: fuel cost control refers to optimizing coal purchasing and use to reduce coal purchasing cost while ensuring the peak regulation capacity and combustion stability of the unit. Through real-time monitoring and adjustment, it is ensured that the increase of coal purchasing cost will not exceed the growth of peak regulation auxiliary service revenue, so as to maintain the economic benefit of unit operation.

[0112] Maintenance risk inhibition refers to reducing equipment maintenance cost in time by monitoring equipment state, avoiding economic loss caused by equipment failure. By balancing the peak shaving auxiliary service income and equipment maintenance cost, the economic loss caused by equipment failure is avoided, and the long-term stable operation of the unit is ensured.

[0113] Carbon trading linkage refers to dynamically adjusting denitration efficiency to control carbon emission cost while meeting the low-carbon requirements of carbon emission right trading market. By controlling the denitration cost and carbon quota trading cost not to exceed the preset proportion of peak shaving auxiliary service income, the controllability of carbon emission cost is ensured, and the economic benefit of the unit is improved.

[0114] Energy consumption optimization refers to reducing energy consumption by optimizing the combination of auxiliary machine operation sets, and improving the energy efficiency of the unit. By optimizing the combination of auxiliary machine operation sets, the electricity saving income is ensured to cover the equipment adjustment cost, and the operation cost of the unit is reduced.

[0115] Thermal stress loss prevention and control refers to avoiding equipment damage caused by thermal stress overrun by adjusting desuperheating water and bypass valve, and ensuring the safe operation of the equipment. By avoiding the economic loss caused by thermal stress overrun, the continuous and stable operation of the unit is ensured, and the economic benefit is improved.

[0116] Step S5, based on the LSTM device life prediction and digital twin simulation, trigger cross-system correction strategy, maintain the safety boundary of commercial operation, reduce the risk of peak shaving auxiliary service income loss caused by equipment failure;

[0117] In step S5, the following sub-steps are also included:

[0118] S5-1, use LSTM neural network to perform time series analysis on steam turbine vibration value, SCR catalyst activity attenuation data, boiler tube wall temperature and secondary reheat system operation parameters, predict the remaining life of steam turbine last stage blade, reheater tube wall and SCR catalyst, and generate early warning signal containing potential peak shaving auxiliary service income loss estimation when the predicted life is lower than the preset threshold;

[0119] S5-2, integrate the thermal system dynamic model and real-time operation data to build a digital twin containing the secondary reheat system. The digital twin integrates the thermal system model of APROS and EBSILON platforms, which is used to simulate the following scenarios:

[0120] Dynamic distribution of secondary reheat temperature difference under deep peak shaving condition;

[0121] Quantification of economic loss when secondary reheat system operation parameter fluctuation breaks through the safety threshold;

[0122] Peak shaving capacity decline and peak shaving auxiliary service income prediction caused by equipment degradation;

[0123] S5-3, if the digital twin simulation detects a safety boundary breach, the following corrective measures are implemented, including:

[0124] When the secondary reheat temperature difference exceeds the safety threshold, adjust the load change rate, secondary reheat desuperheating water quantity, secondary reheat system bypass valve opening degree and steam turbine steam distribution parameters to balance thermal stress and output with the goal of minimizing the loss of peak shaving auxiliary service income;

[0125] When the turbine vibration value and the secondary reheat temperature difference are compound over-limit, activate the blade anti-erosion protection, switch the auxiliary drive mode and optimize the heat source distribution to ensure that the equipment maintenance cost increase does not exceed the preset proportion of the peak shaving auxiliary service income;

[0126] When the carbon emission intensity exceeds the standard, coordinate the secondary reheat system heat exchange efficiency and the coal mill operation parameters to make the carbon quota transaction cost return to the controllable interval of the target function.

[0127] It should be noted that the SCR catalyst activity is based on the back calculation of denitration efficiency and ammonia escape rate,

[0128] The denitration efficiency is calculated by the SCR inlet smoke temperature and the dust concentration at the dust remover outlet, and the ammonia escape rate is calculated by the feedback data of the environmental protection island parameters and the ammonia injection flow control instruction.

[0129] The boiler tube wall temperature is calculated by the model, and the boiler heating surface temperature field is inversely calculated by the APROS digital twin based on the secondary reheat system operation parameters.

[0130] The LSTM neural network is a special kind of recurrent neural network that can effectively handle and predict long-term dependencies in time series data. In the present invention, the LSTM neural network is used for time series analysis of turbine vibration values, SCR catalyst activity decay data, boiler tube wall temperature and secondary reheat system operation parameters to predict the remaining life of the turbine last stage blade, reheater tube wall and SCR catalyst.

[0131] The input data of the LSTM neural network includes turbine vibration values, SCR catalyst activity decay data, boiler tube wall temperature and secondary reheat system operation parameters, etc. These data are collected by the distributed control system and input into the LSTM neural network after preprocessing.

[0132] The LSTM neural network learns the long-term dependencies in the data through time series analysis of the input data, thereby predicting the remaining life of the equipment. The training process of the network is based on historical data, and by adjusting the weights and biases of the network, the error between the predicted result and the actual remaining life is minimized.

[0133] When the predicted life is lower than the preset threshold, the LSTM neural network generates a warning signal containing an estimate of potential peak shaving auxiliary service revenue loss. The warning signal is transmitted to the control center through the internal communication module of the system, and the content of the warning signal includes the device name, the predicted remaining life, the potential peak shaving auxiliary service revenue loss estimate, etc.

[0134] The digital twin is a virtual model that can reflect and predict the state and behavior of a physical system in real time. In the present invention, the digital twin is used to simulate the dynamic distribution of the secondary reheat temperature difference under deep peak shaving conditions, quantify the economic loss when the operating parameters of the secondary reheat system exceed the safety threshold, predict the decline in peak shaving capacity caused by equipment degradation, and predict the peak shaving auxiliary service revenue.

[0135] The digital twin can simulate the change of the secondary reheat temperature difference over time under deep peak shaving conditions, helping operators understand the trend of thermal stress changes in the secondary reheat system. The digital twin can predict the economic loss when the operating parameters of the secondary reheat system exceed the safety threshold, including equipment maintenance costs, downtime losses, and peak shaving auxiliary service revenue losses. The digital twin can assess the impact of equipment degradation on peak shaving capacity and predict changes in peak shaving auxiliary service revenue, providing a basis for equipment maintenance and operational strategy adjustments.

[0136] According to the simulation results of the digital twin, a cross-system correction strategy is triggered.

[0137] Please refer to Figure 2 , which shows a system architecture diagram of a deep scheduling system for deep peak shaving of a coal-fired unit according to an embodiment of the present invention, which includes:

[0138] The data acquisition module acquires multi-dimensional operating parameters of the coal-fired unit through the distributed control system, synchronously accesses the deep peak shaving instructions sent by the grid dispatching end and the carbon price signals of the carbon emissions trading market through the edge computing unit, and realizes timestamp alignment of full-quantity parameters;

[0139] The safety modeling module constructs a five-dimensional safety constraint model, including the upper limit of stable combustion load, the lower limit of SCR denitration temperature, the upper limit of turbine vibration, the upper limit of carbon emission intensity, and the secondary reheat temperature difference safety threshold;

[0140] The optimization solving module constructs a function with the goal of maximizing peak shaving auxiliary service revenue, sets constraint conditions for load, environmental protection, turbine, secondary reheat, and carbon emissions based on the five-dimensional safety constraint model, and solves the economic dispatching scheme using a dynamic programming algorithm, including peak shaving output allocation, carbon quota decision, and marginal benefit optimization;

[0141] The collaborative control module executes cross-system economic collaboration strategies based on the economic dispatching scheme, which is used to realize fuel cost control, maintenance risk suppression, carbon trading linkage, energy consumption optimization, and thermal stress loss prevention and control.

[0142] The safety correction module predicts the equipment life based on the LSTM neural network, generates an early warning signal containing the peak shaving auxiliary service benefit loss estimation, simulates the safety boundary breakthrough scene of the secondary reheat system through the digital twin of the integrated APROS and EBSILON platforms, triggers the cross-system correction strategy, and ensures the minimization of the peak shaving auxiliary service benefit loss.

[0143] In this way, the deep scheduling method and system for deep peak shaving of a coal-fired unit can be implemented.

[0144] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the foregoing embodiments of the present application have been described in detail, those skilled in the art should understand: the technical solutions recorded in the foregoing embodiments can still be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A deep scheduling method for deep peak shaving of a coal-fired unit, characterized in that, The method comprises: Step S1, collecting multi-dimensional operation parameters of the coal-fired unit through a distributed control system, synchronously accessing a deep peak shaving instruction sent by a power grid dispatching end and a carbon price signal of a carbon emission right trading market; Step S2, constructing a five-dimensional safety constraint model associated with an economic cost boundary, including a stable combustion load upper limit, an SCR denitration temperature lower limit, a steam turbine vibration upper limit, a carbon emission intensity upper limit and a secondary reheating temperature difference safety threshold; Step S3, taking maximum peak shaving auxiliary service income as a target, constructing a target function containing coal purchase cost, carbon quota transaction cost and equipment maintenance cost, setting constraint conditions of each dimension based on the five-dimensional safety constraint model, and solving an economic dispatching scheme meeting the requirements of the power market by using a dynamic programming algorithm; Step S4, based on the economic dispatching scheme, executing cross-system economic synergy strategies, including fuel cost control, maintenance risk suppression, carbon transaction linkage, energy consumption optimization and thermal stress loss prevention and control; Step S5, based on equipment life prediction and digital twin simulation of LSTM, triggering cross-system correction strategies, maintaining a commercial operation safety boundary and reducing the risk of peak shaving auxiliary service income loss caused by equipment failure; In step S2, the following sub-steps are further included: S2-1, based on the balance between furnace pressure fluctuation and peak shaving auxiliary service income, setting the stable combustion load upper limit as 20% of the rated load, to ensure that the low-load coal purchase cost does not exceed the peak shaving auxiliary service income; S2-2, based on the correlation between the SCR inlet flue gas temperature and the catalyst maintenance cost, setting the SCR denitration temperature lower limit as 300℃, to avoid a sharp increase in replacement cost caused by insufficient catalyst activity; S2-3, based on the mapping relationship between the steam turbine vibration value and the blade maintenance cost, setting the steam turbine vibration upper limit as 76μm, to balance the acquisition of peak shaving auxiliary service income and equipment maintenance cost; S2-4, based on the desulfurization tower outlet sulfur dioxide concentration, the instantaneous coal feeder coal supply signal and the generator output power, calculating the carbon emission intensity upper limit, which is associated with the carbon quota transaction cost, to ensure controllable carbon cost, and the formula is represented as: wherein, represents the unit power generation carbon emission intensity; represents the standard coal emission factor; represents the real-time coal consumption; represents the unit power generation power; represents the sulfur fraction-carbon emission correction coefficient; S2-5, based on the difference between the secondary reheating steam temperature and the high-pressure cylinder exhaust steam temperature, setting the secondary reheating temperature difference safety threshold, which is associated with the equipment maintenance cost, and the formula is represented as: wherein, represents a secondary reheat temperature difference safety threshold value; represents a secondary reheat steam temperature; represents a high-pressure cylinder exhaust temperature; ≤80℃ represents a safety upper limit of the secondary reheat temperature difference is 80℃, when ΔT>80℃, a secondary reheat system thermal stress overrun early warning is triggered; In step S3, the following sub-steps are further included: S3-1, taking maximum peak shaving auxiliary service income as a target, constructing a target function containing coal purchase cost, carbon quota transaction cost and equipment maintenance cost, and the formula is represented as: wherein, represents cumulative calculation in the time range from time 1 to time N; max represents maximum value of the objective function; represents peak regulation compensation income obtained by the unit participating in the power auxiliary service market; represents coal procurement cost corresponding to coal consumption; represents carbon quota transaction cost corresponding to unit carbon emission; represents equipment maintenance cost; represents the objective function; The peak shaving auxiliary service income is associated with the calculation of the power auxiliary service market rules and is executed in stages, and the peak shaving auxiliary service income calculation formula is represented as: wherein, represents the actual peak regulation output provided by the unit in the tth time period; represents the peak regulation compensation unit price in the tth time period; S3-2, based on the five-dimensional safety constraint model, setting the corresponding conditions of each dimension constraint, including: Load constraint, based on the stable combustion load upper limit, setting the load change range as 20%-100% of the rated power, to ensure that the load interval of the economic dispatching does not break through the combustion stability boundary; Environmental constraints, associated with the lower limit of SCR denitration temperature, set the NOx emission concentration ≤25 mg / Nm³ and the dust emission concentration ≤2 mg / Nm³ to ensure that the environmental indicators meet the requirements in the economic optimization process; Turbine constraints, combined with the upper limit of turbine vibration to control the reliability of the equipment, set the variable load rate ≥1.5%Pe / min to avoid the risk of water erosion caused by vibration overrun while responding to the peak shaving auxiliary service income; Secondary reheat constraints, based on the secondary reheat temperature difference safety threshold, set the secondary reheat steam temperature change rate <1.5 ℃ / min and the secondary reheat system bypass valve opening adjustment range of 10%-90% to ensure that the thermal stress of the secondary reheat system is in the safe interval in the economic dispatching; Carbon emission constraints, associated with the upper limit of carbon emission intensity, set the carbon emission intensity ≤ the upper limit of carbon emission intensity to ensure that the economic dispatching scheme meets the cost control target of the carbon trading market; S3-3, a dynamic programming algorithm is used to solve the economic dispatching scheme that meets the requirements of the power market, including: Peak shaving output allocation, based on the deep peak shaving instruction and the upper limit of the stable combustion load, the optimal output interval of each period is predicted; Carbon quota decision, combined with the carbon price signal of the carbon emission rights trading market and the upper limit of carbon emission intensity, the sensitivity of carbon quota transaction cost to the objective function is analyzed to determine the carbon quota buying and selling opportunity; Marginal benefit optimization, based on the economic correlation of turbine maintenance cost and secondary reheat maintenance cost, the marginal balance point of equipment maintenance cost and peak shaving auxiliary service income is calculated.

2. The method for deep scheduling of deep load following of a coal-fired unit according to claim 1, characterized in that: In step S1, the following sub-steps are also included: S1-1, collect boiler parameters, including main steam pressure, main steam temperature, furnace pressure and oxygen content; S1-2, collect environmental island parameters, including SCR inlet flue gas temperature, desulfurization tower outlet sulfur dioxide concentration and dust remover outlet dust concentration; S1-3, collect turbine parameters, including turbine vibration value, turbine vacuum and high-pressure cylinder exhaust temperature; S1-4, collect fuel and load parameters, including real-time coal consumption, instantaneous coal feeder signal and generator output power; S1-5, collect secondary reheat system operation parameters, including secondary reheat steam temperature, secondary reheat steam pressure, secondary reheat desuperheating water quantity and secondary reheat system bypass valve opening; S1-6, use the edge computing unit to synchronously collect the deep peak shaving instruction and the carbon price signal of the carbon emission rights trading market, establish a communication connection with the distributed control system through the OPC UA protocol, and based on the real-time clock signal of communication interaction, timestamp calibration is performed on the deep peak shaving instruction, the carbon price signal of the carbon emission rights trading market and the multi-dimensional operation parameters collected by the distributed control system, to ensure the time dimension consistency of all parameters.

3. The method for deep scheduling of deep load following of coal-fired units according to claim 1, characterized in that: In step S4, the following sub-steps are also included: S4-1, according to the real-time ratio of coal purchasing cost to peak shaving auxiliary service income, switch coal types and control the increase of coal purchasing cost not to exceed the preset proportion of peak shaving auxiliary service income; S4-2, when the turbine vibration value approaches the associated boundary of blade maintenance cost, the throttling steam distribution mode is enabled to suppress the equipment maintenance cost; S4-3, dynamically adjust the SCR denitration efficiency according to the carbon price signal interval of the carbon emission trading market, control the denitration cost and the carbon quota transaction cost not to exceed the preset proportion of the peak shaving auxiliary service income; S4-4, optimize the auxiliary operation station combination of the thermal system and the auxiliary system, and ensure that the power saving income covers the equipment adjustment cost; S4-5, when the secondary reheat temperature difference approaches the safety threshold related to the equipment maintenance cost, start the linkage adjustment of the secondary reheat temperature reduction water and the secondary reheat system bypass valve to avoid economic loss caused by shutdown maintenance.

4. The method for deep scheduling of a coal-fired unit for deep load following according to claim 1, wherein: In step S5, the following sub-steps are further included: S5-1, use the LSTM neural network to perform time series analysis on the steam turbine vibration value, SCR catalyst activity attenuation data, boiler tube wall temperature and secondary reheat system operation parameters, predict the remaining life of the steam turbine last stage blade, reheater tube wall and SCR catalyst, and generate a warning signal containing potential peak shaving auxiliary service income loss estimation when the predicted life is lower than the preset threshold; S5-2, integrate the thermal system dynamic model and real-time operation data to build a digital twin containing the secondary reheat system, the digital twin integrates the thermal system models of APROS and EBSILON platforms, and is used to simulate the following scenarios: Dynamic distribution of secondary reheat temperature difference under deep peak shaving conditions; Quantification of economic loss when the secondary reheat system operation parameter fluctuation breaks through the safety threshold; Peak shaving capacity decline caused by equipment degradation and peak shaving auxiliary service income prediction; S5-3, if the digital twin simulation detects a safety boundary breakthrough, the following correction measures are performed, including: When the secondary reheat temperature difference exceeds the safety threshold, adjust the load change rate, secondary reheat temperature reduction water quantity, secondary reheat system bypass valve opening degree and steam turbine steam distribution parameters to minimize the peak shaving auxiliary service income loss and balance thermal stress and output; When the steam turbine vibration value and the secondary reheat temperature difference exceed the limit, activate the blade anti-erosion protection, switch the auxiliary drive mode and optimize the heat source distribution to ensure that the equipment maintenance cost increase does not exceed the preset proportion of the peak shaving auxiliary service income; When the carbon emission intensity exceeds the standard, cooperatively adjust the secondary reheat system heat exchange efficiency and the coal mill operation parameters to make the carbon quota transaction cost return to the controllable interval of the target function.

5. A deep scheduling system for deep peak shaving of a coal-fired unit, for implementing the deep scheduling method for deep peak shaving of a coal-fired unit according to claim 1, characterized in that, It includes: A data acquisition module acquires multi-dimensional operation parameters of a coal-fired unit through a distributed control system, synchronously accesses deep peak shaving instructions sent by a power grid dispatching end and carbon price signals of a carbon emission trading market through an edge computing unit, and realizes timestamp alignment of full-quantity parameters; A safety modeling module builds a five-dimensional safety constraint model, including a stable combustion load upper limit, an SCR denitration temperature lower limit, a steam turbine vibration upper limit, a carbon emission intensity upper limit and a secondary reheat temperature difference safety threshold; An optimization solving module builds a function with the maximum peak shaving auxiliary service income as the target, sets constraint conditions of load, environmental protection, steam turbine, secondary reheat and carbon emission based on the five-dimensional safety constraint model, and solves an economic dispatching scheme including peak shaving output distribution, carbon quota decision and marginal benefit optimization by using a dynamic programming algorithm; The cooperative control module implements a cross-system economic cooperation strategy based on an economic dispatching scheme, and is used for realizing fuel cost management and control, maintenance risk inhibition, carbon transaction linkage, energy consumption optimization and thermal stress loss prevention and control. The safety correction module predicts the service life of the equipment based on an LSTM neural network, generates an early warning signal containing loss estimation of peak regulation auxiliary service benefits, simulates a secondary reheat system safety boundary breakthrough scene through a digital twin body integrated with the APROS and EBSILON platforms, triggers a cross-system correction strategy, and ensures that the loss of peak regulation auxiliary service benefits is minimized.

Images