Boiler combustion multi-target optimization control system under wide load operation

By collecting and processing data in real time during the boiler combustion process, predicting load changes and determining the optimal fuel supply and air ratio strategy, the problem of difficulty in achieving precise regulation by traditional control means is solved, and the stability and efficiency of boiler operation are improved.

CN118915442APending Publication Date: 2024-11-08STATE ENERGY CHANGZHOU NO 2 POWER GENERATION CO LTD +4
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202410953729.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-16
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

Traditional control methods are difficult to achieve precise regulation of the boiler's combustion state, resulting in problems such as unstable combustion, coking, and over-temperature burning, affecting the safe and stable operation of the unit.

Method used

A multi-objective optimization control system for boiler combustion under wide load operation is designed, including data acquisition and processing module, load prediction module, optimization scheduling module and control strategy module. Using mathematical optimization methods and artificial intelligence technology, data is collected and processed in real time. , predict load changes, determine the optimal fuel supply and air distribution strategy, and achieve real-time control.

Benefits of technology

It improves the stability and efficiency of boiler operation, avoids energy waste, and ensures efficient operation and combustion stability under different load conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118915442A_ABST
    Figure CN118915442A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of boilers, and discloses a boiler combustion multi-target optimization control system under wide load operation, which specifically comprises a data acquisition and processing module, a load prediction module, an energy efficiency evaluation module, an optimization scheduling module and a control strategy module, and the modules are interconnected. According to the invention, the load prediction module utilizes historical data and environmental conditions including weather data to predict the future load change trend, the prediction capability enables the system to adjust the fuel supply amount and the air ratio in advance, and the stability and efficiency of the system are improved; the optimal fuel supply quantity and air matching strategy is determined under different load conditions by adopting mathematical optimization methods such as linear programming, so that the system can keep efficient operation under different working conditions, and the beneficial effect that multi-target optimization control can be carried out during boiler combustion to enable a unit to operate more stably is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of boilers, in particular to a multi-objective optimization control system for boiler combustion under wide load operation. Background Art

[0002] The boiler combustion process is characterized by strong multivariable coupling, nonlinearity, time variation and large delay. Traditional control methods make it difficult to achieve precise regulation of the combustion state.

[0003] At present, the adjustment of parameters such as secondary air of large coal-fired boilers mostly relies on the experience of on-site operators, which can easily cause problems such as unstable combustion, coking, and overheating, affecting the safe and stable operation of the unit. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] In view of the deficiencies in the prior art, the present invention provides a multi-objective optimization control system for boiler combustion under wide load operation, which has the advantages of being able to perform multi-objective optimization control during boiler combustion to make the unit operation more stable, optimize the combustion of the boiler to make it stable, etc., thereby solving the problems of the above-mentioned technologies.

[0006] (II) Technical solution

[0007] To achieve the above object, the present invention provides the following technical solution: a multi-objective optimization control system for boiler combustion under wide load operation, wherein the control system specifically comprises: a data acquisition and processing module, a load prediction module, an energy efficiency evaluation module, an optimization scheduling module and a control strategy module, and the aforementioned modules are interconnected;

[0008] The data acquisition and processing module is used to collect various data in the boiler operation process in real time, including temperature wd, pressure yl and flow ll, and perform data processing and preprocessing;

[0009] The load prediction module is used to predict the load change of the boiler in the future period;

[0010] The optimization scheduling module is used to determine the best fuel supply and air ratio strategy under different load conditions by using mathematical optimization methods based on load forecasting;

[0011] The control strategy module is used to achieve real-time control of the boiler combustion process according to the optimal strategy given by the optimization scheduling module.

[0012] Preferably, the mathematical optimization method of the optimization scheduling module is: linear programming algorithm.

[0013] Preferably, the data collection in the data collection and processing module is specifically the following steps:

[0014] S1.1. Install temperature sensors, pressure sensors and flow sensors into the boiler system to monitor the operating status of the boiler in real time. The sensors send data to the data acquisition module via analog or digital signals.

[0015] S1.2. Use hardware devices such as data acquisition cards or PLC (programmable logic controller) to connect sensors and receive data sent by sensors. Data acquisition cards can convert analog data into digital signals and transmit them to computers or control systems through interfaces;

[0016] S1.3. The data acquisition module stores the real-time collected temperature, pressure and flow data in the database to ensure the timeliness and accuracy of the data.

[0017] Preferably, the processing in the data acquisition and processing module is specifically the following steps:

[0018] S1.4. After data collection, the data processing algorithm processes the collected raw data. The processing includes data filtering, data interpolation and data correction to eliminate the existing noise or outliers and ensure the accuracy and reliability of the data.

[0019] S1.5, denoising the flow data and smoothing the pressure data;

[0020] S1.6. The processed and pre-processed data is transmitted to the energy efficiency evaluation module, load forecasting module or other modules for further analysis and decision-making, and is projected onto the human-machine interface through the control system for real-time monitoring by the operator.

[0021] Preferably, the load forecasting module specifically includes the following steps:

[0022] S2.1. Identify and understand the seasonal, cyclical and daily changes in load by analyzing historical data, where the change is the boiler load data, which is the load curve and related environmental conditions over the past few days, weeks or months;

[0023] S2.2. Use time series analysis methods to establish load forecasting models to help identify and predict trends, seasonality, and cyclical changes in load changes;

[0024] S2.3. Use artificial intelligence technology in neural network load forecasting to improve forecast accuracy;

[0025] S2.4. Weather data is also integrated during load forecasting, using weather forecast data as input to improve the accuracy of the forecasting model;

[0026] S2.5, the load forecasting module adjusts the forecasting model to reflect the current and future load change trends based on the latest real-time load data and weather information;

[0027] S2.6. Use historical data to verify the model and make adjustments and optimizations based on feedback from actual operations.

[0028] Preferably, the expression of S2.1 is:

[0029] Y(t)=T(t)+S(t)+C(t)+I(t)

[0030] Among them, Y(t) is the observation value at time t, T(t) is the trend component, S(t) is the seasonal component, C(t) is the cyclical component, and I(t) is the irregular component.

[0031] Preferably, the expression in S2.3 is:

[0032]

[0033] in, is the load forecast value at time (t), X(t) is the input feature vector including past load data and weather data, and f {NN} is a neural network model, and θ is the set of parameters of the neural network.

[0034] Preferably, the expression of S2.6 is:

[0035]

[0036] in:

[0037] θ old is the parameter set of the current model;

[0038] θ new is the updated model parameter set;

[0039] η is the learning rate, which controls the step size of parameter update;

[0040] J(θ old ) is the loss function, which is used to measure the model prediction value The difference between the observed value Y(t) and the actual observed value Y(t).

[0041] Preferably, the optimization scheduling module uses a linear programming algorithm expression as follows:

[0042] Maximizeθ=f(F,A,L)

[0043] subject to Constraint1,Constraint2,

[0044] in:

[0045] F is the fuel supply;

[0046] A is the air ratio;

[0047] L is the load condition;

[0048] θ is the objective function of the optimization scheduling module;

[0049] Constraint1, Constraint2 are other constraints:

[0050] The optimal fuel supply F and air ratio A strategy under different load conditions is determined by maximizing the θ function.

[0051] Preferably, the control strategy module expression is:

[0052] u(t)=g(Optimal Strategy,x(t))

[0053] in:

[0054] u(t) is the control output at time (t), which represents the regulating operation of the combustion process;

[0055] g is the control function, which receives the optimal strategy and current system state x(t) given by the optimization scheduling module and generates the actual control operation u(t);

[0056] x(t) is the state variable of the boiler system at time (t), including temperature, pressure and flow data.

[0057] Compared with the prior art, the present invention provides a multi-objective optimization control system for boiler combustion under wide load operation, which has the following beneficial effects:

[0058] 1. The present invention uses historical data and environmental conditions, including weather data, to predict future load change trends through a load prediction module. The prediction capability enables the system to adjust the fuel supply and air ratio in advance to cope with the optimal scheduling under different load conditions, thereby improving the stability and efficiency of the system. The optimization scheduling module uses mathematical optimization methods such as linear programming based on the load prediction results to determine the optimal fuel supply and air ratio strategies under different load conditions, so that the system can maintain efficient operation under different working conditions while avoiding excessive or insufficient energy consumption, thereby improving overall energy efficiency and stability. The control strategy module dynamically adjusts the control operation according to the optimal strategy provided by the optimization scheduling module and the real-time system status including temperature, pressure and flow, thereby achieving the beneficial effect of being able to perform multi-objective optimization control during boiler combustion to make the unit run more stably.

[0059] 2. The present invention uses historical data and environmental conditions through the load forecasting module to predict future load change trends, including seasonal, periodic and daily changes, so that the system can adjust the fuel supply and air ratio in advance to optimize the combustion process and achieve stable operation under different load conditions. The optimization scheduling module uses mathematical optimization methods such as linear programming to determine the optimal fuel supply and air ratio strategy based on the load forecasting results, ensuring that the system can perform combustion adjustment in the most effective way under various load conditions, avoiding energy waste and operational instability, and achieving the beneficial effect of optimizing the combustion of the boiler and making it stable. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a schematic diagram of the system of the present invention; DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] See also Figure 1 , a multi-objective optimization control system for boiler combustion under wide load operation, wherein the control system specifically comprises: a data acquisition and processing module, a load prediction module, an energy efficiency evaluation module, an optimization scheduling module and a control strategy module, and the aforementioned modules are interconnected;

[0063] The data acquisition and processing module is used to collect various data during the operation of the boiler in real time, including temperature wd, pressure yl and flow ll, and perform data processing and preprocessing;

[0064] The load prediction module is used to predict the load change of the boiler in the future period;

[0065] The optimization scheduling module is used to determine the best fuel supply and air ratio strategy under different load conditions by using mathematical optimization methods based on load forecasting;

[0066] The control strategy module is used to achieve real-time control of the boiler combustion process according to the optimal strategy given by the optimization scheduling module.

[0067] Preferably, the mathematical optimization method of the optimization scheduling module is: linear programming algorithm.

[0068] When the boiler operating conditions change, the linear programming model adapts to the new operating environment by modifying the constraints and objective functions, and the adjustment process is relatively simple.

[0069] Preferably, the data collection in the data collection and processing module is specifically the following steps:

[0070] S1.1. Install temperature sensors, pressure sensors and flow sensors into the boiler system to monitor the operating status of the boiler in real time. The sensors send data to the data acquisition module via analog or digital signals.

[0071] S1.2. Use hardware devices such as data acquisition cards or PLC (programmable logic controller) to connect sensors and receive data sent by sensors. Data acquisition cards can convert analog data into digital signals and transmit them to computers or control systems through interfaces;

[0072] S1.3. The data acquisition module stores the real-time collected temperature, pressure and flow data in the database to ensure the timeliness and accuracy of the data.

[0073] Use hardware devices such as data acquisition cards or programmable logic controllers (PLCs) to connect sensors and receive the data they send. Data acquisition cards can convert analog data into digital signals and transmit them to computers or control systems through interfaces. Data acquisition modules store real-time collected temperature, pressure, and flow data in a database to ensure the timeliness and accuracy of the data. These data are the basis for subsequent data processing and control decisions.

[0074] Preferably, the processing in the data acquisition and processing module is specifically the following steps:

[0075] S1.4. After data collection, the data processing algorithm processes the collected raw data. The processing includes data filtering, data interpolation and data correction to eliminate the existing noise or outliers and ensure the accuracy and reliability of the data.

[0076] S1.5, denoising the flow data and smoothing the pressure data;

[0077] S1.6. The processed and pre-processed data is transmitted to the energy efficiency evaluation module, load forecasting module or other modules for further analysis and decision-making, and is projected onto the human-machine interface through the control system for real-time monitoring by the operator.

[0078] The processing part ensures the integrity and accuracy of the data and provides a reliable basis for subsequent optimization scheduling and control strategies. Through real-time monitoring and precise data processing, the system can effectively optimize the boiler combustion process, thereby improving energy utilization efficiency and ensuring stable operation.

[0079] Preferably, the load forecasting module specifically includes the following steps:

[0080] S2.1. Identify and understand the seasonal, cyclical and daily changes in load by analyzing historical data, where the change is the boiler load data, which is the load curve and related environmental conditions over the past few days, weeks or months;

[0081] S2.2. Use time series analysis methods to establish load forecasting models to help identify and predict trends, seasonality, and cyclical changes in load changes;

[0082] S2.3. Use artificial intelligence technology in neural network load forecasting to improve forecast accuracy;

[0083] S2.4. Weather data is also integrated during load forecasting, using weather forecast data as input to improve the accuracy of the forecasting model;

[0084] S2.5, the load forecasting module adjusts the forecasting model to reflect the current and future load change trends based on the latest real-time load data and weather information;

[0085] S2.6. Use historical data to verify the model and make adjustments and optimizations based on feedback from actual operations.

[0086] Through the above steps, the load forecasting module can effectively analyze and predict the load changes of the boiler system, thereby supporting optimized scheduling and resource allocation decisions and improving system operation efficiency and stability.

[0087] Preferably, the expression of S2.1 is:

[0088] Y(t)=T(t)+S(t)+C(t)+I(t)

[0089] Among them, Y(t) is the observation value at time t, T(t) is the trend component, S(t) is the seasonal component, C(t) is the cyclical component, and I(t) is the irregular component.

[0090] This formula is used to understand and predict the contribution of various influencing factors in load data, providing a basis for the subsequent establishment of an accurate load forecasting model.

[0091] Preferably, the expression in S2.3 is:

[0092]

[0093] in, is the load forecast value at time (t), X(t) is the input feature vector including past load data and weather data, and f {NN} is a neural network model, and θ is the set of parameters of the neural network.

[0094] This formula summarizes how the neural network model uses historical load data and weather data to predict the current load value. The neural network learns the complex relationships between the input data and makes load forecasts based on these relationships to improve the prediction accuracy and model accuracy.

[0095] Preferably, the expression of S2.6 is:

[0096]

[0097] in:

[0098] θ old is the parameter set of the current model;

[0099] θ new is the updated model parameter set;

[0100] η is the learning rate, which controls the step size of parameter update;

[0101] J(θ old ) is the loss function, which is used to measure the model prediction value The difference between the observed value Y(t) and the actual observed value Y(t).

[0102] Through this updating process, the model can continuously adjust the parameter θ according to the gradient of the loss function. old , so that the predicted value Gradually approaching the actual observed value Y(t), the iterative process trains the model more precisely to improve the accuracy and reliability of load forecasting.

[0103] Preferably, the optimization scheduling module uses a linear programming algorithm expression as follows:

[0104] Maximizeθ=f(F,A,L)

[0105] subject to Constraint1, Constraint2,

[0106] in:

[0107] F is the fuel supply;

[0108] A is the air ratio;

[0109] L is the load condition;

[0110] θ is the objective function of the optimization scheduling module;

[0111] Constraint1, Constraint2 are other constraints:

[0112] The optimal fuel supply F and air ratio A strategy under different load conditions is determined by maximizing the θ function.

[0113] Preferably, the control strategy module expression is:

[0114] u(t)=g(Optimal Strategy,x(t))

[0115] in:

[0116] u(t) is the control output at time (t), which represents the regulating operation of the combustion process;

[0117] g is the control function, which receives the optimal strategy and current system state x(t) given by the optimization scheduling module and generates the actual control operation u(t);

[0118] x(t) is the state variable of the boiler system at time (t), including temperature, pressure and flow data.

[0119] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. The multi-objective optimization control system for boiler combustion under wide load operation is characterized by: The control system specifically includes: a data acquisition and processing module, a load prediction module, an energy efficiency evaluation module, an optimization scheduling module and a control strategy module, and the aforementioned modules are interconnected; The data acquisition and processing module is used to collect various data in the boiler operation process in real time, including temperature wd, pressure yl and flow ll, and perform data processing and preprocessing; The load prediction module is used to predict the load change of the boiler in the future period; The optimization scheduling module is used to determine the best fuel supply and air ratio strategy under different load conditions by using mathematical optimization methods based on load forecasting; The control strategy module is used to achieve real-time control of the boiler combustion process according to the optimal strategy given by the optimization scheduling module.

2. The multi-objective optimization control system for boiler combustion under wide load operation according to claim 1 is characterized in that: The mathematical optimization method of the optimization scheduling module is: linear programming algorithm.

3. The multi-objective optimization control system for boiler combustion under wide load operation according to claim 1 is characterized in that: The data collection in the data collection and processing module is specifically the following steps: S1.

1. Install temperature sensors, pressure sensors and flow sensors into the boiler system to monitor the operating status of the boiler in real time. The sensors send data to the data acquisition module via analog or digital signals. S1.

2. Use hardware devices such as data acquisition cards or PLC (programmable logic controller) to connect sensors and receive data sent by sensors. Data acquisition cards can convert analog data into digital signals and transmit them to computers or control systems through interfaces; S1.

3. The data acquisition module stores the real-time collected temperature, pressure and flow data in the database to ensure the timeliness and accuracy of the data.

4. The multi-objective optimization control system for boiler combustion under wide load operation according to claim 3 is characterized in that: The processing in the data acquisition and processing module is specifically the following steps: S1.

4. After data collection, the data processing algorithm processes the collected raw data. The processing includes data filtering, data interpolation and data correction to eliminate the existing noise or outliers and ensure the accuracy and reliability of the data. S1.5, denoising the flow data and smoothing the pressure data; S1.

6. The processed and pre-processed data is transmitted to the energy efficiency evaluation module, load forecasting module or other modules for further analysis and decision-making, and is projected onto the human-machine interface through the control system for real-time monitoring by the operator.

5. The multi-objective optimization control system for boiler combustion under wide load operation according to claim 1 is characterized in that: The load forecasting module specifically includes the following steps: S2.

1. Identify and understand the seasonal, cyclical and daily changes in load by analyzing historical data, where the change is the boiler load data, which is the load curve and related environmental conditions over the past few days, weeks or months; S2.

2. Use time series analysis methods to establish load forecasting models to help identify and predict trends, seasonality, and cyclical changes in load changes; S2.

3. Use artificial intelligence technology in neural network load forecasting to improve forecast accuracy; S2.

4. Weather data is also integrated during load forecasting, using weather forecast data as input to improve the accuracy of the forecasting model; S2.5, the load forecasting module adjusts the forecasting model to reflect the current and future load change trends based on the latest real-time load data and weather information; S2.

6. Use historical data to verify the model and make adjustments and optimizations based on feedback from actual operations.

6. The multi-objective optimization control system for boiler combustion under wide load operation according to claim 1 is characterized in that: The S2.1 expression is: Y(t)=T(t)+S(t)+C(t)+I(t) Among them, Y(t) is the observation value at time t, T(t) is the trend component, S(t) is the seasonal component, C(t) is the cyclical component, and I(t) is the irregular component.

7. The multi-objective optimization control system for boiler combustion under wide load operation according to claim 6 is characterized in that: The expression in S2.3 is: in, is the load forecast value at time (t), X(t) is the input feature vector including past load data and weather data, and f {NN} is a neural network model, and θ is the set of parameters of the neural network.

8. The multi-objective optimization control system for boiler combustion under wide load operation according to claim 1 is characterized in that: The S2.6 expression is: in: θ old is the parameter set of the current model; θ new is the updated model parameter set; η is the learning rate, which controls the step size of parameter update; J(θ old ) is the loss function, which is used to measure the model prediction value The difference between the observed value Y(t) and the actual observed value Y(t).

9. The multi-objective optimization control system for boiler combustion under wide load operation according to claim 2 is characterized in that: The optimization scheduling module uses the linear programming algorithm expression: Maximizeθ=f(F,A,L) subject to Constraint1,Constraint2, in: F is the fuel supply; A is the air ratio; L is the load condition; θ is the objective function of the optimization scheduling module; Constraint1, Constraint2 are other constraints: The optimal fuel supply F and air ratio A strategy under different load conditions is determined by maximizing the θ function.

10. The multi-objective optimization control system for boiler combustion under wide load operation according to claim 1, characterized in that: The control strategy module expression is: u(t)=g(Optimal Strategy,x(t)) in: u(t) is the control output at time (t), which represents the regulating operation of the combustion process; g is the control function, which receives the optimal strategy and current system state x(t) given by the optimization scheduling module and generates the actual control operation u(t); x(t) is the state variable of the boiler system at time (t), including temperature, pressure and flow data.

Citation Information

Cited By

  • Hot water boiler intelligent control system and method based on thermal load prediction

    CN119468508A

  • Auxiliary control system and method for oxygen content of boiler for deep peak regulation of unit

    CN120371034A

  • Thermal power plant solid waste demand matching system and method based on big data driving

    CN121481533A