Intelligent optimization method and system for wide load operation of thermal power generating unit

Abstract

The present application relates to the intelligent optimization method and system of wide load operation of thermal power unit, aiming at the problems of lack of quantitative target, high safety risk and low economic benefit of monitoring adjustment under deep peak regulation condition, an innovative solution is proposed. Through real-time collection of 50 DCS measurement point data such as power generation load rate 71.1%, main steam pressure 13.98MPa, back pressure 22.9kPa, combined with real-time electricity price 0.38 yuan / kWh and comprehensive standard coal unit price 700 yuan / ton in power market, a multi-dimensional analysis model is constructed.

Description

Technical Field

[0001] This invention belongs to the field of thermal power unit operation control technology, specifically relating to intelligent optimization methods and systems for wide-load operation of thermal power units. Background Technology

[0002] In the context of the electricity spot market, thermal power generating units face the challenge of normalized deep peak shaving. With the increasing proportion of renewable energy grid connection, the load regulation range of thermal power units has expanded to 20%-100%. While the 50%-100% basic peak shaving range has design parameters to guide it, the 20%-50% deep peak shaving range has long lacked technical specifications. Monitoring personnel are forced to rely on experience for adjustments, leading to three major technical bottlenecks: First, safety risks and economic losses coexist. During wide-load operation, frequent occurrences of main steam pressure fluctuations exceeding limits (±0.5MPa) and abnormal back pressure increases (>25kPa) occur. Each 1kPa increase in back pressure alone increases coal consumption by 1.5g / kWh. Industry data from 2024 shows that the number of unplanned outages due to uncontrolled deep peak shaving conditions increased by 23% year-on-year. Second, the lack of profit allocation technology. During periods of zero or even negative electricity prices in the spot market, the traditional "profit-attributing-to-electricity" coal consumption analysis model becomes ineffective. The shift supervisor lacks real-time benefit calculation tools, making it impossible to dynamically optimize the allocation ratio of power generation, heating (5.8 million GJ / year), and steam supply. According to a report from the Northern Power Regulatory Bureau, the annual benefit loss due to improper heat and power allocation exceeds 3 million yuan. Thirdly, monitoring and adjustments lack quantifiable targets. When adjusting parameters such as main steam pressure and oxygen content, operators rely solely on historical experience. At 70% load, the boiler efficiency of a 300MW unit fluctuates between 93.0% and 94.5%, equivalent to a daily coal consumption difference of 12 tons. While the existing DCS system can display real-time data, it lacks a benefit-oriented optimization target closed-loop mechanism. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent optimization method and system for wide-load operation of thermal power units. This invention proposes an intelligent optimization method and system for wide-load operation of thermal power units, including:

[0004] The intelligent optimization method and system for wide-load operation of thermal power units includes the following steps: Real-time data acquisition: The following parameters of the DCS system are acquired through the data transmission interface:

[0005] The power generation load factor was 71.1%.

[0006] The main steam pressure was optimized from 13.98 MPa to 14.08 MPa (an increase of 0.1 MPa);

[0007] The back pressure decreased from 22.9 kPa to 21.79 kPa (a decrease of 1.11 kPa);

[0008] The furnace efficiency increased from 93.67% to 93.79% (an increase of 0.12 percentage points);

[0009] The exhaust gas temperature decreased by 4.15℃;

[0010] The ratio of wind-to-coal mixture decreased by 0.21 t / t.

[0011] Steady-state processing: The fixed window method (window duration 5 minutes) was used to process data from 50 measuring points, including power generation load, main steam temperature, reheat temperature, oxygen content, and fan power consumption rate (forced draft fan ≤ 0.85%, induced draft fan ≤ 1.2%).

[0012] Safety Analysis: Based on 15 years of unit service data, constraints were applied using an asymptotic optimization method (safety boundary coefficient 0.5).

[0013] Main steam temperature ≤540℃;

[0014] Reheat temperature ≤535℃;

[0015] Main steam pressure fluctuation ≤ ±0.2MPa;

[0016] Back pressure fluctuation ≤ ±1.5 kPa;

[0017] Indicator Optimization: Establish a combustion-efficiency composite model and output the following optimization target values ​​to the synchronous performance system:

[0018] The target back pressure is 21.8 kPa (the actual adjusted value is 21.79 kPa).

[0019] The target value for wind-fired coal was reduced by 0.21 t / t;

[0020] The target value for flue gas temperature was reduced by 4.15℃.

[0021] Furthermore, the data collection includes the real-time clearing price of the electricity market (0.38 yuan / kWh), the unified clearing price of the entire network, the unified clearing price of Hohhot East / West, and the comprehensive standard coal unit price (700 yuan / ton) and carbon cost in the management data.

[0022] Furthermore, the asymptotic optimization method limits the power consumption rate of the coal mill to ≤0.35% and the power consumption rate of the primary air fan to ≤0.28% through a safety boundary coefficient (section 3.2.3, Auxiliary Machine Energy Consumption Parameters).

[0023] Furthermore, the combustion-efficiency composite model, based on boiler design principles, optimizes the oxygen setpoint to 2.8%-3.2%, and calculates the air-coal ratio using the following formula:

[0024] Air-to-coal ratio = Primary air volume / Coal volume

[0025] Furthermore, the synchronous performance system displays the following comparative data in real time:

[0026] The target curve (14.08 MPa) and the actual curve (14.08 MPa) of the main steam pressure optimization;

[0027] Back pressure optimization target curve (21.8 kPa) and actual value curve (21.79 kPa);

[0028] The target curve for furnace efficiency optimization (93.79%) and the actual curve (93.79%).

[0029] Furthermore, it includes: a data acquisition module: real-time acquisition of data from 50 measuring points in the DCS system and 10 types of meter reading data from the electricity market;

[0030] Steady-state processing module: Uses a fixed window algorithm to output a 5-minute steady-state dataset;

[0031] Safety optimization module: Stores historical safety data of the unit and outputs main steam temperature / pressure safety boundary values;

[0032] Indicator generation module: Runs the combustion-efficiency composite model to generate optimization targets for air-coal ratio and flue gas temperature;

[0033] Performance synchronization module: Pushes target values ​​to the monitoring interface via API.

[0034] Furthermore, the index generation module outputs a back pressure optimization target value of 21.8 kPa, and the actual value after adjustment by the monitoring personnel reaches 21.79 kPa (deviation 0.01 kPa). The main steam pressure target value of 14.08 MPa is completely consistent with the actual value of 14.08 MPa.

[0035] Furthermore, the system reduces coal consumption for power supply by 1.2 g / kWh, specifically achieved as follows:

[0036] The coal consumption was reduced by 0.21 t / h through optimization of the air-to-coal ratio;

[0037] Boiler efficiency was improved by 0.12% through flue gas temperature optimization;

[0038] Annual savings of 4,800 tons of standard coal.

[0039] Furthermore, the display interface of the performance synchronization module in the Northern Company's real-time production system includes:

[0040] Left side: Optimize the target curve (red) and the actual value curve (blue);

[0041] Right side area: Real-time deviation alarm list (triggered when back pressure deviation > 0.5 kPa);

[0042] Bottom area: Data on the safe service life of the unit (15 years).

[0043] Furthermore, through technical optimizations such as reducing back pressure by 1.11 kPa, increasing main steam pressure by 0.1 MPa, and increasing furnace efficiency by 0.12%, the system achieves an improvement of 420 yuan / h in monitoring quality benefits and an annual economic benefit of 3.36 million yuan.

[0044] Advantages of this invention:

[0045] A data acquisition network was constructed based on 50 high-precision measuring points (power generation load, main steam pressure, back pressure, etc.), and a fixed window method was used for 5-minute steady-state processing. Compared with the traditional sliding window method, the data processing efficiency was improved by 10 times, ensuring that the main steam pressure control accuracy reached ±0.02MPa at a load rate of 71.1%. A safety boundary coefficient of 0.5 was introduced through a progressive optimization algorithm to strictly limit the main steam temperature to within 540℃ and control the back pressure fluctuation to the ±1.5kPa threshold, effectively avoiding the risks of over-temperature and over-pressure. An innovative wide-load analysis model was developed, establishing a coupling relationship between the air-coal ratio and the flue gas temperature based on boiler design principles. In the 300MW load range, through optimization of the oxygen setpoint (2.8%-3.2%), a core breakthrough was achieved: a reduction of 0.21t / t in the air-coal ratio and a decrease of 4.15℃ in the flue gas temperature. This model simultaneously optimized the energy consumption of auxiliary equipment, reducing the power consumption rate of the forced draft fan to 0.85% and the power consumption rate of the induced draft fan to 1.2%, which is 15% lower than the industry average. The main steam pressure was precisely increased by 0.1 MPa, enhancing the steam's work capacity. An innovative performance synchronization mechanism was designed, pushing optimized targets to the monitoring interface in real time. The system displays the deviation curve (±0.01 kPa accuracy) between the target back pressure value of 21.8 kPa and the actual value of 21.79 kPa. When the deviation exceeds 0.5 kPa, an audible and visual alarm is automatically triggered. Application at Baotou No. 1 Thermal Power Plant shows that the target boiler efficiency of 93.79% and the actual value of 93.79% are perfectly aligned across all time periods, completely eliminating human adjustment errors. A 1.11 kPa reduction in back pressure significantly improved the efficiency of the air-cooled system. Through the synergistic optimization of a 1.11 kPa reduction in back pressure, a 0.1 MPa increase in main steam pressure, and a 0.12% increase in boiler efficiency, a 1.2 g / kWh reduction in coal consumption for power generation was achieved. This translates to an annual saving of 4,800 tons of standard coal per unit, equivalent to an economic benefit of 3.36 million yuan. The emission reduction benefits are particularly outstanding: CO2 emissions reduction of 12,000 tons is equivalent to planting 65,000 trees, SO2 emissions reduction of 41 tons improves regional acid rain pollution, and NOx emissions reduction of 36 tons reduces the probability of smog formation. Furnace efficiency improved by 0.12 percentage points, and boiler thermal efficiency exceeded the design limit.

[0046] A safety boundary model was constructed using 15 years of unit operating history data. An asymptotic optimization algorithm was employed to strictly limit the main steam temperature within the design threshold of 540℃, and control back pressure fluctuations within a safe range of ±1.5 kPa. Compared to traditional experience-based adjustment methods, the incidence of over-temperature and over-pressure risks was reduced by 37%, and unplanned downtime was shortened by 52%. The introduction of a safety boundary coefficient of 0.5 ensured the safe operation of the auxiliary systems even under stringent conditions of ≤0.35% power consumption of the coal mill and ≤0.28% power consumption of the primary air fan. The combustion-efficiency composite model broke through the limitations of traditional single-variable adjustment, establishing a coupling relationship between the air-coal ratio and flue gas temperature based on boiler thermodynamic principles. Through precise control of the oxygen setpoint of 2.8%-3.2%, a core breakthrough was achieved: a 0.21 t / t reduction in the air-coal ratio and a 4.15℃ decrease in flue gas temperature. This technology improved boiler efficiency by 0.12 percentage points to 93.79%, exceeding the unit's design limit, and reduced power supply coal consumption by 1.2 g / kWh, reaching an industry-leading level. An innovative target value synchronization mechanism was developed, achieving complete overlap between the target and actual main steam pressure of 14.08 MPa on the monitoring interface, with back pressure control accuracy reaching the 0.01 kPa level. The system employs a fixed window method for data processing, improving efficiency tenfold compared to the traditional sliding window method. It completes data analysis of 50 monitoring points and outputs optimization instructions within 5 minutes, with a response speed meeting the second-level scheduling requirements of the spot market. A 1.11 kPa reduction in back pressure directly improves the efficiency of the air-cooled system, while a 0.1 MPa increase in main steam pressure enhances the steam's work capacity. The synergy of these two factors results in a monitoring quality improvement of 420 yuan / hour. Annual savings of 4800 tons of standard coal are achieved, equivalent to an economic benefit of 3.36 million yuan. Significant environmental benefits are also observed: a 12,000-ton reduction in CO2 emissions is equivalent to the annual carbon sequestration of 55 hectares of forest; a 41-ton reduction in SO2 emissions improves regional acid rain pollution; and a 36-ton reduction in NOx emissions lowers the probability of smog formation. The system is seamlessly integrated with the real-time production platform of Northern Power Company. The three-zone layout of the performance interface (target curve / alarm list / service data) conforms to the operating habits of monitoring personnel. Software copyright certification (2024SR1221315) ensures technical reliability, achieving 100% parameter matching in 300MW subcritical air-cooled units, and providing a standardized solution for wide-load operation in the thermal power industry. Detailed Implementation

[0047] How to use:

[0048] Operators log into the Northern Company's real-time production system and select the "Wide Load Optimization" module in the monitoring interface. The system automatically connects to the DCS data transmission interface and collects real-time data from 50 key measuring points, including a power generation load rate of 71.1%, main steam pressure of 13.98 MPa, back pressure of 22.9 kPa, and boiler efficiency of 93.67%. During the initialization phase, manual input of management data is required: a comprehensive standard coal price of 700 yuan / ton, a real-time clearing electricity price of 0.38 yuan / kWh, and carbon cost parameters. The system then initiates a fixed-window steady-state processing program, filtering the data every 5 minutes to eliminate transient fluctuations.

[0049] Based on a 15-year safe service database of the unit, the system automatically loads an asymptotic optimization algorithm to generate safety protection thresholds: main steam temperature upper limit 540℃, reheat temperature upper limit 535℃, main steam pressure fluctuation range ±0.2MPa, and back pressure fluctuation range ±1.5kPa. When the real-time main steam pressure reaches 14.08MPa, the system displays a red target curve coinciding with the blue actual curve in the curve area on the left side of the monitoring interface. If the back pressure exceeds 22.4kPa (the trigger point of the safety boundary coefficient 0.5), the alarm area on the right immediately flashes a red warning and sounds a buzzer.

[0050] The combustion-efficiency composite model runs automatically every 5 minutes, outputting optimization instructions based on the air-coal ratio calculation formula (primary air volume / coal consumption): adjusting the oxygen setpoint to the 2.8%-3.2% range, reducing the air-coal ratio by a target of 0.21 t / t, and reducing the flue gas temperature by a target of 4.15℃. Monitoring personnel manually adjust the operation of the blower frequency converter based on the flashing yellow instruction markers displayed on the interface. When the air-coal ratio decreases from the baseline value by 0.21 t / t, the system automatically records a reduction in coal consumption of 0.21 tons / hour and updates the cumulative coal savings in the bottom service data area.

[0051] The synchronous performance module pushes three core optimization targets to the monitoring interface in real time: back pressure target value of 21.8 kPa (displayed as green numbers), furnace efficiency target value of 93.79% (displayed as a gold progress bar), and main steam pressure target of 14.08 MPa (bar comparison chart). Operators can fine-tune the air-cooled island fan speed by observing the 0.01 kPa deviation between the actual back pressure value of 21.79 kPa and the target. When the real-time furnace efficiency reaches 93.79%, the system automatically generates a before-and-after optimization comparison chart in the curve area, indicating a 0.12% improvement.

[0052] Before each shift handover, the shift leader retrieves three key results from the system's "Efficiency Dashboard" interface: the vacuum improvement data corresponding to a 1.11 kPa reduction in back pressure, the steam work capacity calculation value for a 0.1 MPa increase in main steam pressure, and the coal consumption conversion result for a 0.12% increase in furnace efficiency. The system automatically generates hourly efficiency curves, displaying a monitoring quality efficiency growth point of 420 yuan / h. During monthly analysis, the "Annualized Forecast" function is used to view the 3.36 million yuan economic benefit analysis report, and the coal procurement plan for the following month is optimized based on the 4,800 tons of standard coal savings.

[0053] When the electricity spot market experiences zero electricity price, the system automatically switches to "cost-first mode": the safety boundary coefficient is tightened to 0.4, the back pressure fluctuation range is narrowed to ±1.0 kPa, and the wind-to-coal ratio optimization target is increased to a reduction of 0.25 t / t. Monitoring personnel, based on the orange warning prompts displayed on the interface, stabilize the load rate within the safe range of 70%-75%, ensuring minimum cost operation during periods of zero revenue. Shift handover records must highlight the flue gas temperature control value (2°C lower than the benchmark) and the compliance status of the fan power consumption rate under these special operating conditions.

[0054] Example:

[0055] Example 1: Safety and Efficiency Optimization of Unit #1 at Baotou No.1 Thermal Power Plant

[0056] Implementation Scenarios

[0057] In April 2025, Unit #1 (a 300MW subcritical air-cooled unit) of the Baotou No.1 Thermal Power Plant of Northern United Power entered the deep peak-shaving phase, with the load rate fluctuating frequently between 50% and 75%. Monitoring personnel activated the wide-load intelligent optimization system, collecting key parameters in real time through the DCS interface, including a power generation load rate of 71.1%, main steam pressure of 13.98MPa, back pressure of 22.9kPa, and boiler efficiency of 93.67%. During the system initialization phase, the real-time clearing price of electricity in the spot market (0.38 yuan / kWh) and the comprehensive standard coal price (700 yuan / ton) were entered, along with the unit's 15-year safe service database. The system uses an asymptotic optimization algorithm (safety boundary coefficient 0.5) to dynamically constrain operating parameters: when the main steam temperature reaches 538℃ (design threshold 540℃), an automatic airflow increase command is triggered, stabilizing the temperature at 535℃ by increasing the secondary airflow by 10%. The back pressure monitoring module detected that the inlet pressure of the air-cooled island had risen to 23.2 kPa (exceeding the safety threshold of 1.5 kPa) and immediately executed third-order control:

[0058] 1. Increase the air-cooled fan speed to 85%.

[0059] 2. Lower the circulating water temperature by 2℃

[0060] 3. The back pressure curve on the right side of the monitoring interface for adjusting the extraction steam volume is displayed in real time as it decreases from 22.9 kPa to 21.79 kPa. A deviation of 0.01 kPa is confirmed by a green indicator light.

[0061] Combustion optimization results: The combustion-efficiency composite model outputs oxygen setpoint instructions every 5 minutes. Under a load rate of 71.1%, the system reduced the oxygen content from 3.5% to 3.0%, and the air supply volume decreased by 8%. The real-time monitoring value of the air-coal ratio decreased by 0.21 t / t from the baseline value, corresponding to a reduction of 0.21 tons of coal consumption per hour. The flue gas temperature monitoring curve showed a decrease from 138℃ to 133.85℃ (a decrease of 4.15℃), and the furnace efficiency increased to 93.79%. The left area of ​​the performance synchronization interface displays a comparison of the furnace efficiency curves before and after optimization.

[0062] Red target curve: 93.79%

[0063] The blue actual curve: 93.79% perfect overlap of the two lines confirms the control accuracy.

[0064] Economic benefit verification: After 24 hours of continuous operation, the system's benefit dashboard generates the following data:

[0065] The back pressure decreased by 1.11 kPa, increasing the vacuum level by 0.8%.

[0066] The main steam pressure was increased to 14.08 MPa to enhance the steam's work capacity.

[0067] A total of 5.04 tons of coal were saved (equivalent to a benefit of 3,780 yuan). Monthly statistics show that coal consumption for power generation decreased by 1.2 g / kWh. The back pressure curve comparison chart confirms that CO2 emissions can be reduced by 12,000 tons annually.

[0068] Example 2: Cost Control Optimization During Deep Peak Shaving Periods

[0069] Implementation Scenarios

[0070] From 2:00 AM to 5:00 AM on January 15, 2025, zero electricity price occurred in the spot electricity market. The load factor of Unit #2 (300MW) at Baotou Power Plant dropped to 55%. The system automatically switched to "cost-priority mode," tightening the safety boundary coefficient to 0.4 and narrowing the back pressure fluctuation range to ±1.0kPa. An orange warning pop-up appeared on the monitoring interface: "Minimum cost operation strategy activated during zero electricity price period." When the back pressure rose to 22.0kPa (only 0.5kPa away from the safety threshold), the safety interlock control executed a level four response:

[0071] 1. The power consumption rate of the coal mill is limited to 0.33% (design upper limit 0.35%).

[0072] 2. The primary air fan power consumption rate was reduced to 0.26%.

[0073] 3. Main steam temperature is forcibly locked at 535℃.

[0074] 4. A 5% reduction in industrial steam supply ensures safe power generation. The main steam pressure curve shows that the actual value remains stable within the range of 13.95-14.05 MPa (fluctuation ≤0.1 MPa), fully complying with the ±0.2 MPa safety standard.

[0075] Combustion system deep optimization: Model output limit optimization instructions:

[0076] The oxygen level setpoint dropped to 2.8% (below the normal lower limit).

[0077] The ratio of coal to wind power decreased by another 0.04 t / t (cumulative decrease of 0.25 t / t).

[0078] The target value for flue gas temperature was increased by 2℃, and monitoring personnel, following the yellow flashing instruction marker, adjusted the blower opening from 65% to 60%. Real-time data showed that the flue gas temperature dropped from 136℃ to 129.8℃ (a decrease of 6.2℃ from the baseline), and the furnace efficiency improved to 93.92%.

[0079] Zero-price period benefits: System records during 3 hours of operation:

[0080] To avoid losses of 123,000 yuan due to excessive power generation.

[0081] Saved 1.89 tons of standard coal.

[0082] The carbon cost reduction of 840 yuan generates a special operating condition report at the bottom of the performance interface, noting "Back pressure control deviation of 0.08 kPa, better than design standard".

[0083] Example 3: Application of system calibration after annual overhaul

[0084] Implementation Scenarios

[0085] Following the overhaul of Unit #1 in June 2025, thermal control personnel initiated the model calibration procedure. They uploaded the historical DCS data package (including turbine flow path modification data) after the overhaul. The system automatically compared 300 sets of operating parameters and found that the back pressure control deviation was >0.3 kPa for three consecutive days. This triggered a self-learning program to refit the combustion model. The model iteration process...

[0086] 1. Data collection for new operating conditions: The design value of main steam pressure has been increased to 14.2 MPa.

[0087] 2. Corrected safety boundary: The back pressure fluctuation zone was reduced from ±1.5 kPa to ±1.2 kPa.

[0088] 3. Updated the air-to-coal ratio formula coefficient: The weighting of air volume / coal consumption in the calculation was adjusted by 12%. The first load point (75% load) test after calibration showed:

[0089] Target back pressure 21.5 kPa → Actual value 21.49 kPa

[0090] The furnace efficiency target was 93.85% → the actual value was 93.86%. The system generated a calibration report pop-up window on the monitoring interface, along with a comparison chart of the old and new models.

[0091] Further Improvement in Economic Efficiency: First Month of Operational Data After Calibration:

[0092] The average back pressure dropped to 21.65 kPa (a further decrease of 0.14 kPa compared to before calibration).

[0093] Furnace efficiency improved to 93.91% (breaking historical records).

[0094] Monthly coal savings reached 420 tons (15% higher than expected), and annual statistics showed economic benefits of 3.58 million yuan. The performance interface was also updated with the "Service Life: 16 years" data.

[0095] Example 4: Emergency Response to Frequency Fluctuations in Regional Power Grids

[0096] Implementation scenario: In August 2025, the frequency of the Hohhot-Baotou West Power Grid suddenly dropped to 49.5Hz, and the system automatically triggered the "primary frequency regulation priority mode". Under the condition of 68% load factor, the main steam pressure fluctuation limit of ±0.2MPa was broken, allowing short-term fluctuations to ±0.3MPa to improve the frequency regulation response speed.

[0097] Safety and efficiency balance control

[0098] 1. The asymptotic optimization algorithm dynamically relaxes the safety boundary coefficient to 0.6.

[0099] 2. The back pressure fluctuation threshold is temporarily expanded to ±2.0 kPa.

[0100] 3. Synchronous performance interface switching between red and yellow dual-color warning mode: When the main steam pressure jumps from 14.08MPa to 14.38MPa within 30 seconds, the system executes compensation control:

[0101] Connect the high-voltage bypass by 10%.

[0102] Water flow rate increased by 8%

[0103] After the back pressure control target value was temporarily increased to 22.5 kPa and the frequency was restored, the main steam pressure curve showed that it dropped back to 14.08 MPa within 5 minutes.

[0104] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the invention. Any person skilled in the art can make various modifications and refinements without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be defined by the claims.

Claims

1. An intelligent optimization method and system for wide-load operation of thermal power units, characterized in that: The following steps are included: Real-time data acquisition: Acquire the following parameters of the DCS system through the data transmission interface: The power generation load factor was 71.1%. The main steam pressure was optimized from 13.98 MPa to 14.08 MPa (an increase of 0.1 MPa); The back pressure decreased from 22.9 kPa to 21.79 kPa (a decrease of 1.11 kPa); The furnace efficiency increased from 93.67% to 93.79% (an increase of 0.12 percentage points); The exhaust gas temperature decreased by 4.15℃; The ratio of wind-to-coal mixture decreased by 0.21 t / t. Steady-state processing: The fixed window method (window duration 5 minutes) was used to process data from 50 measuring points, including power generation load, main steam temperature, reheat temperature, oxygen content, and fan power consumption rate (forced draft fan ≤ 0.85%, induced draft fan ≤ 1.2%). Safety Analysis: Based on 15 years of unit service data, constraints were applied using an asymptotic optimization method (safety boundary coefficient 0.5). Main steam temperature ≤540℃; Reheat temperature ≤535℃; Main steam pressure fluctuation ≤ ±0.2MPa; Back pressure fluctuation ≤ ±1.5 kPa; Indicator Optimization: Establish a combustion-efficiency composite model and output the following optimization target values ​​to the synchronous performance system: The target back pressure is 21.8 kPa (the actual adjusted value is 21.79 kPa). The target value for wind-fired coal was reduced by 0.21 t / t; The target value for flue gas temperature was reduced by 4.15℃.

2. The intelligent optimization method and system for wide-load operation of thermal power units according to claim 1, characterized in that: The data collection includes the real-time clearing price of the electricity market (0.38 yuan / kWh), the unified clearing price of the entire network, the unified clearing price of Hohhot East / West, and the comprehensive standard coal unit price (700 yuan / ton) and carbon cost in the management data.

3. The intelligent optimization method and system for wide-load operation of thermal power units according to claim 1, characterized in that: The asymptotic optimization method limits the power consumption rate of the coal mill to ≤0.35% and the power consumption rate of the primary air fan to ≤0.28% by using a safety boundary coefficient (section 3.2.3, Auxiliary machine energy consumption parameters).

4. The intelligent optimization method and system for wide-load operation of thermal power units according to claim 1, characterized in that: The combustion-efficiency composite model is based on boiler design principles, optimizes the oxygen setpoint to 2.8%-3.2%, and calculates the air-coal ratio using the following formula: Air-to-coal ratio = Primary air volume / Coal volume 5. The intelligent optimization method and system for wide-load operation of thermal power units according to claim 1, characterized in that: The synchronous performance system displays the following comparative data in real time: The target curve (14.08 MPa) and the actual curve (14.08 MPa) of the main steam pressure optimization; Back pressure optimization target curve (21.8 kPa) and actual value curve (21.79 kPa); The target curve for furnace efficiency optimization (93.79%) and the actual curve (93.79%).

6. The intelligent optimization method and system for wide-load operation of thermal power units according to claims 1-5, characterized in that: include: Data acquisition module: Real-time acquisition of data from 50 measuring points in the DCS system and 10 types of meter reading data from the electricity market; Steady-state processing module: Uses a fixed window algorithm to output a 5-minute steady-state dataset; Safety optimization module: Stores historical safety data of the unit and outputs main steam temperature / pressure safety boundary values; Indicator generation module: Runs the combustion-efficiency composite model to generate optimization targets for air-coal ratio and flue gas temperature; Performance synchronization module: Pushes target values ​​to the monitoring interface via API.

7. The intelligent optimization method and system for wide-load operation of thermal power units according to claim 6, characterized in that: The indicator generation module outputs a back pressure optimization target value of 21.8 kPa. After adjustment by the monitoring personnel, the actual value reached 21.79 kPa (deviation 0.01 kPa). The main steam pressure target value of 14.08 MPa is completely consistent with the actual value of 14.08 MPa.

8. The intelligent optimization method and system for wide-load operation of thermal power units according to claim 6, characterized in that: The system reduces coal consumption for power supply by 1.2 g / kWh, specifically achieved as follows: The coal consumption was reduced by 0.21 t / h through optimization of the air-to-coal ratio; Boiler efficiency was improved by 0.12% through flue gas temperature optimization; Annual savings of 4,800 tons of standard coal.

9. The intelligent optimization method and system for wide-load operation of thermal power units according to claim 6, characterized in that: The display interface of the performance synchronization module in the Northern Company's real-time production system includes: Left side: Optimize the target curve (red) and the actual value curve (blue); Right side area: Real-time deviation alarm list (triggered when back pressure deviation > 0.5 kPa); Bottom area: Data on the safe service life of the unit (15 years).

10. The intelligent optimization method and system for wide-load operation of thermal power units according to claim 6, characterized in that: The system achieves an improvement of 420 yuan / h in monitoring quality benefits and an annual economic benefit of 3.36 million yuan through technical optimizations that reduce back pressure by 1.11 kPa, increase main steam pressure by 0.1 MPa, and increase furnace efficiency by 0.12%.