Intelligent boiler combustion optimization control system and control method thereof

Abstract

The invention discloses an intelligent boiler combustion optimization control system which comprises a plurality of sensors arranged at different positions of a boiler; the data transmission module is used for transmitting the data acquired by the sensor to a data stream processing system through a wireless or wired network; the data stream processing system is used for preprocessing original data generated in the boiler operation process; the data fusion module is used for fusing the multi-dimensional data from each sensor; the combustion control decision module generates a combustion control decision based on the evaluation result of the combustion state; the real-time monitoring module is used for calculating the combustion efficiency, the pollutant emission amount and the oxygen content of the boiler in real time in the combustion control process, the technical problems of a traditional boiler in the aspects of combustion efficiency, emission control, safety, fault early warning, energy saving and the like are solved, and therefore the overall operation efficiency and the environment-friendly performance of the boiler are improved.

Description

Technical Field

[0001] The invention relates to an intelligent boiler combustion optimization control system and a control method thereof. Background Art

[0002] With the advancement of industrialization, boilers, as important energy equipment, are widely used in power generation, heating, and various industrial production processes. Traditional boiler control systems mostly rely on manual operation or simple control logic. This approach has many limitations in improving boiler operating efficiency and reducing pollutant emissions. The following are the main shortcomings of existing technologies:

[0003] Traditional boiler systems often rely on a few monitoring points, and most monitoring data is collected through manual inspections, making it difficult to timely grasp the boiler's true operating status. Due to the lack of sufficient sensor distribution and real-time data collection, it is often impossible to fully monitor the boiler's combustion status, resulting in potential problems in boiler operation not being discovered in a timely manner.

[0004] Boiler systems operate in a complex environment, involving multiple parameters such as temperature, pressure, and oxygen content, all of which are strongly correlated. Traditional systems employ relatively simple data analysis methods, often failing to effectively integrate multidimensional data from various sensors. This makes it difficult to assess the boiler's combustion status from a global perspective, hindering combustion optimization and efficiency improvements.

[0005] Traditional boiler control systems typically utilize fixed combustion control logic, which is unable to flexibly adjust to real-time boiler state changes. This is especially true when faced with nonlinear factors such as load fluctuations and changes in fuel quality. Existing systems often lag behind in their control strategies, leading to reduced boiler combustion efficiency and even high pollutant emissions.

[0006] Traditional boiler control systems often fail to optimize the combustion process in real time, resulting in low combustion efficiency and excessive fuel consumption. Incomplete combustion and improperly adjusted combustion parameters can lead to insufficient oxygen or excessive pollutant emissions. Especially in industrial environments, lax boiler emissions control can violate environmental regulations and impose a burden on the environment.

[0007] Existing boiler systems lack intelligent anomaly monitoring mechanisms, making them unable to accurately detect abnormalities that may arise during combustion, such as temperature fluctuations and excessive or insufficient oxygen. This results in the system being unable to respond promptly. The lack of effective emergency response and automatic adjustment capabilities often requires manual intervention, increasing operational risks and potentially damaging boiler equipment.

[0008] Traditional boiler operation is mostly regulated based on maximum load, neglecting to optimize combustion efficiency under varying loads. This lack of intelligent energy-saving measures results in significant unnecessary energy waste during boiler operation, while also failing to adequately reduce pollutant emissions and comply with environmental policies and regulations. Summary of the Invention

[0009] The purpose of the present invention is to provide an intelligent boiler combustion optimization control system and a control method thereof. By integrating sensors, data transmission, data stream processing, data fusion, intelligent decision-making and real-time monitoring, the technical difficulties of traditional boilers in combustion efficiency, emission control, safety, fault warning and energy saving are solved, thereby improving the overall operating efficiency and environmental performance of the boiler.

[0010] The technical solution adopted by the present invention to solve its technical problem is:

[0011] An intelligent boiler combustion optimization control system, comprising:

[0012] Several sensors are placed at different locations on the boiler to collect data generated during the operation of the boiler in real time;

[0013] A data transmission module, configured to transmit the data collected by the sensor to a data stream processing system via a wireless or wired network;

[0014] Data stream processing system, used to pre-process the raw data generated during boiler operation;

[0015] The data fusion module is used to fuse the multi-dimensional data from various sensors to generate a comprehensive evaluation of the boiler combustion status information;

[0016] The combustion control decision module generates combustion control decisions based on the evaluation results of the combustion state and adjusts the combustion parameters of the boiler;

[0017] The real-time monitoring module is used to calculate the boiler's combustion efficiency, pollutant emissions, and oxygen content in real time during the combustion control process, and trigger an alarm mechanism when an abnormality occurs to automatically adjust the combustion control parameters.

[0018] Preferably, the sensors include a temperature sensor, a pressure sensor, an oxygen concentration sensor, a flue gas emission sensor, a fuel flow sensor, and an air volume sensor. The output signals of the sensors are sent to the data stream processing system through the data transmission module for use by the data fusion module.

[0019] Another technical problem to be solved by the present invention is to provide an intelligent boiler combustion optimization control method, comprising the following steps:

[0020] By installing a number of sensors, data generated during the operation of the boiler is collected in real time; the data from the sensors is transmitted to the data stream processing system via a wireless or wired network;

[0021] Pre-process the data generated during boiler operation;

[0022] Fuse multi-dimensional data from various sensors to generate a comprehensive assessment of boiler combustion status information;

[0023] Generate combustion control decisions based on the evaluation results of the combustion state and adjust the combustion parameters of the boiler;

[0024] During the combustion control process, the boiler's combustion efficiency, pollutant emissions, and oxygen content are calculated in real time. If any abnormality occurs, the alarm mechanism is triggered and the combustion control parameters are automatically adjusted.

[0025] Preferably, the sensors include a temperature sensor, a pressure sensor, a flow sensor, an oxygen sensor and a flue gas analyzer. The temperature sensor is used to collect temperature data in the boiler, the pressure sensor is used to collect working pressure data in the boiler, the flow sensor is used to collect flow data in the boiler, the oxygen sensor is used to collect oxygen concentration data during boiler combustion, and the flue gas analyzer is used to collect pollutant concentration data in the flue gas discharged by the boiler.

[0026] As a preferred method, the method for preprocessing the data generated during the operation of the boiler is:

[0027] Assume that the sensor data is X = {x1, x2, …, xn}. The value range of each sensor is different. The normalization formula is as follows:

[0028]

[0029] Among them, xi is the original data, μi is the mean of the i-th sensor data, and σi is its standard deviation;

[0030] Use mean filling to handle missing values:

[0031]

[0032] Where N is the number of valid data, replacing missing values;

[0033] Use a sliding window to construct samples, each sample contains T time steps of data:

[0034] X t ={x t-T+1 ,x t-T+2 ,...,x t}

[0035] Each input sample contains T time steps of data.

[0036] As a preferred method, the multi-dimensional data from various sensors are integrated to generate a comprehensive evaluation of the boiler combustion state information as follows:

[0037] A neural network is used to fuse multi-dimensional data from various sensors to perform a comprehensive assessment of the boiler combustion status, as follows:

[0038] Input layer: The data of each sensor is used as input. If the number of sensors is n, the size of the input layer is n;

[0039] The input is x = [x1, x2, …, xn], and the goal of the neural network is to output a combustion state assessment based on this input data;

[0040] Hidden layer: Design multiple fully connected layers. Let the output of the k-th layer be hk, the number of neurons in this layer be mk, and the calculation formula of the k-th layer be:

[0041] h k =f(W k h k-1 +b k )

[0042] Where Wk is the weight matrix, bk is the bias term, f is the activation function, and the output of the input layer is regarded as the 0th layer, denoted as h0 = x;

[0043] Assume that the output is a scalar y, which represents the evaluation value of the combustion state, and uses the Softmax activation function:

[0044]

[0045] Where yi represents the predicted probability of the i-th category, and C is the number of categories;

[0046] The loss function is the mean square error:

[0047]

[0048] Where y^ is the predicted value of the model, y is the true value, and N is the number of samples;

[0049] Use the gradient descent method to minimize the loss function. The formula for updating the weights by gradient descent is:

[0050]

[0051] Where η is the learning rate, is the gradient of the loss function with respect to the weights;

[0052] The dataset is divided into training set and validation set, and the generalization ability of the model is evaluated through cross-validation;

[0053] The evaluation indicator used is mean square error;

[0054]

[0055] When new sensor data arrives, it is input into the trained neural network model to calculate the evaluation result y of the combustion state pred .

[0056] As a preferred method, a combustion control decision is generated based on the evaluation result of the combustion state, and the combustion parameters of the boiler are adjusted:

[0057] According to the output y pred , mapping the evaluation result of the combustion state to the combustion control signal u(t);

[0058] According to the generated control signal u(t), the fuel flow rate mf(t), air flow rate ma(t), furnace temperature Tf(t) and exhaust gas temperature Ts(t) are adjusted as follows:

[0059] m f (t) = K f u(t)

[0060] m a (t) = K a ·m f (t)

[0061] T f (t) = T f，set +K T ·(u(t)-u ref )

[0062] T s (t) = T s，set +K s ·(m f (t)-m a (t))

[0063] The control signal u(t) is a control decision signal generated based on the current operating state, target state, and external environmental data of the boiler. The adjustment coefficient Kf of the fuel flow rate mf(t) and the ratio coefficient Ka between the air flow rate and the fuel flow rate are set according to the design parameters and operating requirements of the boiler. The adjustment coefficient KT of the furnace temperature Tf(t) is set according to the combustion characteristics and operating state of the boiler. The adjustment coefficient Ks of the exhaust temperature Ts(t) is set according to the exhaust system and thermal efficiency requirements of the boiler.

[0064] According to the error between the evaluation result ypred and the set target, the corresponding control signal is adjusted by the PID controller:

[0065]

[0066] Where e(t) is the evaluation error, and Kp, Ki, and Kd are the proportional, integral, and derivative gains.

[0067] As a preferred method, during the combustion control process, the combustion efficiency, pollutant emissions, and oxygen content of the boiler are calculated in real time. If an abnormality occurs, an alarm mechanism is triggered and the combustion control parameters are automatically adjusted as follows:

[0068] The combustion efficiency of the boiler is calculated in real time, and the combustion efficiency is calculated by the following formula:

[0069]

[0070] Among them, Qoutput is the actual output heat energy, Qinput is the theoretical heat energy released by the fuel;

[0071] The pollutant emissions from the boiler are monitored in real time. The pollutants include nitrogen oxides NOx, carbon dioxide CO2 and carbon monoxide CO, and the emissions are calculated using the following formula:

[0072] NOx emissions = k NOx ·(T flame -1000)·(λ-1)

[0073]

[0074] CO emissions = k CO ·(1-η 燃烧 )

[0075] Where Tflame is the combustion temperature, λ is the excess air coefficient, mfuel is the fuel flow rate, Ccarbon is the carbon content in the fuel, and HCO2 is the amount of CO2 produced per unit carbon. The oxygen content O2 in the boiler flue gas is measured in real time and compared with the set oxygen content target to ensure the integrity of the combustion process.

[0076] When the combustion efficiency falls below a set threshold, pollutant emissions exceed the standard, or the oxygen content is abnormal, an alarm mechanism is triggered to further adjust the combustion control parameters to optimize the boiler's combustion process. The alarm trigger conditions include:

[0077] The combustion efficiency is lower than the set lower limit;

[0078] Pollutant emissions exceed legal emission standards;

[0079] The oxygen content is lower than the set minimum value;

[0080] According to the alarm results, the combustion control parameters are automatically adjusted, including but not limited to adjusting the fuel flow, air flow and excess air coefficient.

[0081] Another technical problem to be solved by the present invention is to provide an electronic device, including a memory, a processor and a computer program stored in the memory and runnable on the processor, wherein when the processor executes the program, the intelligent boiler combustion optimization control method as described above is implemented.

[0082] Another technical problem to be solved by the present invention is to provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the intelligent boiler combustion optimization control method as described above.

[0083] The beneficial effects of the present invention are:

[0084] Real-time data collection and monitoring issues:

[0085] By installing sensors at various locations throughout the boiler, various data generated during operation (such as temperature, pressure, and oxygen content) are collected in real time, addressing the issue of delayed or incomplete data collection during operation. Sensors in various locations ensure comprehensive monitoring of the boiler's internal conditions. However, the data collected by these sensors is diverse and complex, making it difficult for traditional single-data analysis methods to fully and accurately reflect the boiler's combustion status. The data fusion module integrates data from various sensors in multiple dimensions, enabling a comprehensive assessment of the boiler's combustion status and improving the accuracy and reliability of data analysis.

[0086] The combustion process requires flexible adjustments based on the real-time, changing boiler status to ensure optimal combustion efficiency and minimized pollutant emissions. The combustion control decision module generates control decisions based on a comprehensive assessment of the combustion status and adjusts the boiler's combustion parameters, addressing the challenges of traditional boiler systems with lags in regulation or their inability to adapt to transient changes. The real-time monitoring module accurately calculates the boiler's combustion efficiency, pollutant emissions, and oxygen content, addressing the issues of low energy efficiency and excessive emissions during boiler operation. By precisely controlling the combustion process, combustion efficiency can be effectively optimized and pollutant emissions reduced, complying with environmental regulations.

[0087] The system detects abnormal boiler operating conditions through real-time monitoring and automatically triggers alarms. When anomalies occur (such as decreased combustion efficiency or excessive emissions), the system automatically adjusts combustion control parameters to avoid human error and damage to boiler equipment. This addresses the lack of adaptive regulation and emergency response in traditional boiler systems. By optimizing combustion control parameters, energy savings can be achieved during boiler operation, reducing fuel consumption. This issue is often difficult to precisely control and optimize in traditional boiler control systems, so the system's intelligent combustion optimization control helps achieve energy savings and consumption reductions. Specific implementation methods

[0088] The principles and features of the present invention are described below. The examples provided are intended to illustrate the present invention only and are not intended to limit the scope of the present invention. The following paragraphs describe the present invention in more detail by way of example. The advantages and features of the present invention will become more apparent from the following description and claims.

[0089] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this invention pertains. The terms used in this specification of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items. Example

[0090] An intelligent boiler combustion optimization control system, comprising:

[0091] Several sensors are placed at different locations on the boiler to collect data generated during the operation of the boiler in real time;

[0092] A data transmission module, configured to transmit the data collected by the sensor to a data stream processing system via a wireless or wired network;

[0093] Data stream processing system, used to pre-process the raw data generated during boiler operation;

[0094] The data fusion module is used to fuse the multi-dimensional data from various sensors to generate a comprehensive evaluation of the boiler combustion status information;

[0095] The combustion control decision module generates combustion control decisions based on the evaluation results of the combustion state and adjusts the combustion parameters of the boiler;

[0096] The real-time monitoring module is used to calculate the boiler's combustion efficiency, pollutant emissions, and oxygen content in real time during the combustion control process, and trigger an alarm mechanism when an abnormality occurs to automatically adjust the combustion control parameters.

[0097] By real-time monitoring of various operating data of the boiler, including combustion efficiency, oxygen content, etc., the system can accurately adjust the combustion parameters of the boiler, thereby maximizing combustion efficiency, reducing energy waste, and lowering fuel consumption; real-time monitoring and calculation of pollutant emissions (such as NOx, CO2, etc.) can help to promptly detect excessive emissions, and optimize the combustion process by adjusting control parameters, reducing the emission of harmful gases and meeting environmental protection requirements.

[0098] The system features a built-in alarm mechanism. When abnormal conditions such as low combustion efficiency and excessive pollutant emissions occur, it triggers an alarm and automatically adjusts control parameters, preventing unstable boiler operation and potentially damaging equipment or creating safety hazards. Through an integrated real-time monitoring module and data stream processing system, boiler operators can monitor the boiler's operating status at all times. Comprehensively evaluated boiler combustion status information helps managers make more accurate decisions, enabling refined management and remote control.

[0099] Precise combustion control and parameter adjustment can effectively control energy loss during boiler operation, reduce fuel waste, thereby achieving energy conservation and consumption reduction, and improving the overall economic efficiency of boiler operation; the system's automated and intelligent design reduces the need for manual intervention, and can automatically adjust the boiler's operating parameters based on real-time data, improving the system's operational stability and automation level; through continuous monitoring and real-time adjustment of the boiler's operating conditions, it can greatly reduce equipment failures and maintenance requirements, extend the boiler's service life, and thus reduce maintenance and repair costs; the system can detect anomalies in boiler operation in real time, avoid safety accidents caused by human errors or equipment failures, and enhance the safety of boiler operation.

[0100] The sensors include a temperature sensor, a pressure sensor, an oxygen concentration sensor, a flue gas emission sensor, a fuel flow sensor, and an air volume sensor. The output signals of the sensors are sent to the data stream processing system through the data transmission module for use by the data fusion module.

[0101] By configuring multiple sensors (such as temperature, pressure, and oxygen concentration sensors), comprehensive monitoring of the boiler's key operating parameters is possible. The data provided by these sensors is transmitted in real time to the data stream processing system, enabling the system to instantly obtain boiler status information and ensure constant visibility of the boiler's operating status. This real-time monitoring capability helps promptly identify potential anomalies, reduces the risk of equipment failure, and improves system safety and stability.

[0102] The data transmission module feeds data from various sensors into the data stream processing system, which is then processed by the data fusion module. This allows for the integration and analysis of data from multiple dimensions. By fusing data from multiple sensors, the overall operating status of the boiler can be more accurately assessed, and combustion control strategies can be optimized in real time, thereby improving boiler efficiency, reducing energy consumption, and lowering emissions. This data fusion capability enables more intelligent and precise management of boiler systems.

[0103] Comprehensive sensor monitoring and data fusion modules optimize the combustion process, adjusting parameters such as air volume, fuel flow, and oxygen concentration in real time to ensure optimal boiler operation. This not only improves combustion efficiency and reduces fuel waste, but also effectively reduces pollutant emissions, meeting environmental protection requirements and minimizing environmental impact. Real-time monitoring of flue gas emissions and oxygen concentration enables better control of boiler emissions, achieving energy conservation and emission reduction goals.

[0104] An intelligent boiler combustion optimization control method includes the following steps:

[0105] By installing a number of sensors, data generated during the operation of the boiler is collected in real time; the data from the sensors is transmitted to the data stream processing system via a wireless or wired network;

[0106] Pre-process the data generated during boiler operation;

[0107] Fuse multi-dimensional data from various sensors to generate a comprehensive assessment of boiler combustion status information;

[0108] Generate combustion control decisions based on the evaluation results of the combustion state and adjust the combustion parameters of the boiler;

[0109] During the combustion control process, the boiler's combustion efficiency, pollutant emissions, and oxygen content are calculated in real time. If any abnormality occurs, the alarm mechanism is triggered and the combustion control parameters are automatically adjusted.

[0110] By monitoring the boiler's operating data in real time and evaluating the combustion status based on data fusion, the system can dynamically adjust combustion control parameters (such as temperature, oxygen content, and fuel flow) to ensure the boiler operates at optimal efficiency. This not only improves energy efficiency but also effectively reduces fuel waste, achieving energy conservation. The system calculates and monitors key parameters such as pollutant emissions and oxygen content in real time and automatically adjusts combustion control parameters based on the evaluation results, thereby optimizing the combustion process. This intelligent control can effectively reduce the emission of harmful gases (such as carbon dioxide and carbon monoxide), reduce environmental pollution, and comply with environmental protection standards. During boiler operation, if any abnormal conditions occur (such as decreased combustion efficiency or excessive pollutant emissions), the system will promptly trigger an alarm mechanism and automatically adjust the combustion control parameters. This real-time monitoring and self-adjustment capability can significantly improve boiler safety, reduce human intervention and operational errors, improve the automation level of the boiler system, and reduce maintenance costs.

[0111] The sensors include a temperature sensor, a pressure sensor, a flow sensor, an oxygen sensor and a flue gas analyzer. The temperature sensor is used to collect temperature data inside the boiler, the pressure sensor is used to collect working pressure data inside the boiler, the flow sensor is used to collect flow data inside the boiler, the oxygen sensor is used to collect oxygen concentration data during the boiler combustion process, and the flue gas analyzer is used to collect pollutant concentration data in the flue gas discharged by the boiler.

[0112] The method for preprocessing the data generated during boiler operation is as follows:

[0113] Assume that the sensor data is X = {x1, x2, …, xn}. The value range of each sensor is different. The normalization formula is as follows:

[0114]

[0115] Among them, xi is the original data, μi is the mean of the i-th sensor data, and σi is its standard deviation;

[0116] Use mean filling to handle missing values:

[0117]

[0118] Where N is the number of valid data, replacing missing values;

[0119] Use a sliding window to construct samples, each sample contains T time steps of data:

[0120] X t ={x t-T+1 ,x t-T+2 ,...,x t}

[0121] Each input sample contains T time steps of data.

[0122] Through normalization (converting sensor data to a mean of 0 and a standard deviation of 1), data from different sensors is converted to the same scale. This avoids deviations caused by differences in sensor data ranges, making model training more stable and accelerating convergence, thereby improving prediction accuracy.

[0123] Using mean filling to handle missing values ​​ensures data integrity when there is less missing data, while preserving the statistical characteristics of sensor data. By replacing missing values, the impact of data loss on subsequent analysis and model training is avoided, ensuring data continuity and consistency.

[0124] By constructing samples using a sliding window, each sample contains T time steps of data, capturing the temporal characteristics of boiler operation. This approach enables the model to identify dynamic patterns and trends in boiler operation (for example, changes in combustion efficiency), thereby improving the model's predictive capabilities for time series data and enabling better combustion control optimization and anomaly detection.

[0125] The method of fusing multi-dimensional data from various sensors to generate a comprehensive evaluation of boiler combustion status information is as follows:

[0126] A neural network is used to fuse multi-dimensional data from various sensors to perform a comprehensive assessment of the boiler combustion status, as follows:

[0127] Input layer: The data of each sensor is used as input. If the number of sensors is n, the size of the input layer is n;

[0128] The input is x = [x1, x2, …, xn], and the goal of the neural network is to output a combustion state assessment based on this input data;

[0129] Hidden layer: Design multiple fully connected layers. Let the output of the k-th layer be hk, the number of neurons in this layer be mk, and the calculation formula of the k-th layer be:

[0130] h k =f(W k h k-1 +b k )

[0131] Where Wk is the weight matrix, bk is the bias term, f is the activation function, and the output of the input layer is regarded as the 0th layer, denoted as h0 = x;

[0132] Assume that the output is a scalar y, which represents the evaluation value of the combustion state, and uses the Softmax activation function:

[0133]

[0134] Where yi represents the predicted probability of the i-th category, and C is the number of categories;

[0135] The loss function is the mean square error:

[0136]

[0137] Where y^ is the predicted value of the model, y is the true value, and N is the number of samples;

[0138] Use the gradient descent method to minimize the loss function. The formula for updating the weights by gradient descent is:

[0139]

[0140] Where η is the learning rate, is the gradient of the loss function with respect to the weights;

[0141] The dataset is divided into training set and validation set, and the generalization ability of the model is evaluated through cross-validation;

[0142] The evaluation indicator uses mean square error;

[0143]

[0144] When new sensor data arrives, it is input into the trained neural network model to calculate the evaluation result y of the combustion state pred .

[0145] Neural networks are capable of handling complex nonlinear relationships. Through their multi-layered structure and extensive training data, they can automatically extract features from sensor data and identify underlying patterns that influence boiler combustion conditions. This enables the model to more accurately assess boiler combustion conditions than traditional methods, reducing human assumptions and errors.

[0146] Since the neural network can receive new sensor data and make predictions in real time after training, the model can immediately calculate the combustion status assessment results when new data arrives. This allows the system to monitor the combustion status in real time during boiler operation, promptly identify potential problems, ensure safe operation of the boiler, and provide rapid response capabilities.

[0147] Through automatic learning and optimization using a neural network, the system can automatically process large amounts of sensor data and generate comprehensive evaluation results without human intervention. This not only reduces the workload of manual analysis but also enhances the system's intelligence, enabling it to autonomously adapt to complex operating conditions and reduce the possibility of human error.

[0148] Method for generating combustion control decisions and adjusting boiler combustion parameters based on combustion state evaluation results:

[0149] According to the output y pred , mapping the evaluation result of the combustion state to the combustion control signal u(t);

[0150] According to the generated control signal u(t), the fuel flow rate mf(t), air flow rate ma(t), furnace temperature Tf(t) and exhaust gas temperature Ts(t) are adjusted as follows:

[0151] m f (t) = K f u(t)

[0152] m a (t) = K a ·m f (t)

[0153] T f (t) = T f，set +K T ·(u(t)-u ref )

[0154] T s (t) = T s，set +K s ·(m f (t)-m a (t))

[0155] The control signal u(t) is a control decision signal generated based on the current operating state, target state, and external environmental data of the boiler. The adjustment coefficient Kf of the fuel flow rate mf(t) and the ratio coefficient Ka between the air flow rate and the fuel flow rate are set according to the design parameters and operating requirements of the boiler. The adjustment coefficient KT of the furnace temperature Tf(t) is set according to the combustion characteristics and operating state of the boiler. The adjustment coefficient Ks of the exhaust temperature Ts(t) is set according to the exhaust system and thermal efficiency requirements of the boiler.

[0156] According to the error between the evaluation result ypred and the set target, the corresponding control signal is adjusted by the PID controller:

[0157]

[0158] Where e(t) is the evaluation error, and Kp, Ki, and Kd are the proportional, integral, and derivative gains.

[0159] By mapping the combustion state evaluation result (ypred) into the combustion control signal u(t) and using a PID controller to precisely adjust the fuel flow rate, air flow rate, furnace temperature, and exhaust gas temperature, key combustion parameters can be dynamically adjusted based on the actual operating state and target state of the boiler. This precise control helps achieve higher combustion efficiency, reduce energy waste, and improve boiler performance.

[0160] By monitoring the boiler's operating status in real time and adjusting combustion parameters based on the assessment results, it is possible to effectively prevent abnormalities or failures during boiler operation. For example, timely adjustments to air or fuel flow can prevent over- or under-combustion, thereby reducing the risk of failure and extending the boiler's service life. Furthermore, PID control can smoothly regulate boiler operation, avoiding drastic fluctuations and improving system stability.

[0161] This method not only considers the boiler's internal operating state but also comprehensively considers the impact of external environmental data on the combustion process. By adjusting control factors (such as the fuel flow rate to air flow rate ratio and the furnace temperature control factor), the system can adjust the boiler's operating mode in real time according to different operating conditions and environmental conditions, ensuring optimal combustion control under different external environments and load conditions.

[0162] During the combustion control process, the boiler's combustion efficiency, pollutant emissions, and oxygen content are calculated in real time. If an abnormality occurs, an alarm mechanism is triggered and the combustion control parameters are automatically adjusted as follows:

[0163] The combustion efficiency of the boiler is calculated in real time, and the combustion efficiency is calculated by the following formula:

[0164]

[0165] Among them, Qoutput is the actual output heat energy, Qinput is the theoretical heat energy released by the fuel;

[0166] The pollutant emissions from the boiler are monitored in real time. The pollutants include nitrogen oxides NOx, carbon dioxide CO2 and carbon monoxide CO, and the emissions are calculated using the following formula:

[0167] NOx emissions = k NOx ·(T flame -1000)·(λ-1)

[0168]

[0169] CO emissions = k CO ·(1-η 燃烧 )

[0170] Where Tflame is the combustion temperature, λ is the excess air coefficient, mfuel is the fuel flow rate, Ccarbon is the carbon content in the fuel, and HCO2 is the amount of CO2 produced per unit carbon. The oxygen content O2 in the boiler flue gas is measured in real time and compared with the set oxygen content target to ensure the integrity of the combustion process.

[0171] When the combustion efficiency falls below a set threshold, pollutant emissions exceed the standard, or the oxygen content is abnormal, an alarm mechanism is triggered to further adjust the combustion control parameters to optimize the boiler's combustion process. The alarm trigger conditions include:

[0172] The combustion efficiency is lower than the set lower limit;

[0173] Pollutant emissions exceed legal emission standards;

[0174] The oxygen content is lower than the set minimum value;

[0175] According to the alarm results, the combustion control parameters are automatically adjusted, including but not limited to adjusting the fuel flow, air flow and excess air coefficient.

[0176] By real-time monitoring of combustion efficiency, pollutant emissions, and oxygen levels, the boiler can be ensured to operate efficiently and environmentally friendly. If combustion efficiency is low or pollutant emissions exceed standards, the system can promptly identify and take action to reduce energy waste and pollutant emissions, thereby improving the boiler's overall energy efficiency and ensuring compliance with environmental standards. By optimizing the combustion process, negative environmental impacts can be minimized, carbon emissions can be reduced, and compliance with environmental regulations can be achieved.

[0177] This solution provides real-time feedback and automatic adjustment capabilities. If low combustion efficiency, excessive pollutant emissions, or abnormal oxygen levels occur, the system triggers an alarm and automatically adjusts combustion control parameters (such as fuel flow, air flow, and excess air coefficient). This automatic adjustment mechanism not only improves the boiler's response speed and operating efficiency, but also reduces reliance on manual operation, enabling rapid response to changes in boiler status and ensuring stable operation.

[0178] By monitoring and adjusting key combustion process indicators such as combustion efficiency, emissions, and oxygen content, safety risks and legal liabilities arising from improper boiler operation or excessive pollution can be effectively avoided. When abnormal conditions occur, alarm mechanisms are activated promptly, prompting operators to intervene or automatically making adjustments to prevent boiler failure or excessive emissions, ensuring safe and stable operation of the equipment while also helping boiler operators avoid legal and financial penalties for excessive emissions.

[0179] This embodiment also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the intelligent boiler combustion optimization control method described above is implemented.

[0180] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the intelligent boiler combustion optimization control method as described above is implemented.

[0181] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0182] Those skilled in the art will clearly understand that for the sake of convenience and brevity in description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0183] The above embodiments of the present invention are not intended to limit the scope of protection of the present invention, and the implementation methods of the present invention are not limited thereto. All other modifications, replacements or changes made to the above structures of the present invention based on the above contents of the present invention, in accordance with common technical knowledge and customary means in this field, without departing from the above basic technical ideas of the present invention, should fall within the scope of protection of the present invention.

Claims

1. An intelligent boiler combustion optimization control system, characterized in that: include: Several sensors are placed at different locations on the boiler to collect data generated during the operation of the boiler in real time; A data transmission module, configured to transmit the data collected by the sensor to a data stream processing system via a wireless or wired network; Data stream processing system, used to pre-process the raw data generated during boiler operation; The data fusion module is used to fuse the multi-dimensional data from various sensors to generate a comprehensive evaluation of the boiler combustion status information; The combustion control decision module generates combustion control decisions based on the evaluation results of the combustion state and adjusts the combustion parameters of the boiler; The real-time monitoring module is used to calculate the boiler's combustion efficiency, pollutant emissions, and oxygen content in real time during the combustion control process, and trigger an alarm mechanism when an abnormality occurs to automatically adjust the combustion control parameters.

2. The intelligent boiler combustion optimization control method according to claim 1 is characterized in that: The sensors include a temperature sensor, a pressure sensor, an oxygen concentration sensor, a flue gas emission sensor, a fuel flow sensor, and an air volume sensor. The output signals of the sensors are sent to the data stream processing system through the data transmission module for use by the data fusion module.

3. An intelligent boiler combustion optimization control method, characterized in that: The following steps are involved: By installing a number of sensors, data generated during the operation of the boiler is collected in real time; the data from the sensors is transmitted to the data stream processing system via a wireless or wired network; Pre-process the data generated during boiler operation; Fuse multi-dimensional data from various sensors to generate a comprehensive assessment of boiler combustion status information; Generate combustion control decisions based on the evaluation results of the combustion state and adjust the combustion parameters of the boiler; During the combustion control process, the boiler's combustion efficiency, pollutant emissions, and oxygen content are calculated in real time. If any abnormality occurs, the alarm mechanism is triggered and the combustion control parameters are automatically adjusted.

4. The intelligent boiler combustion optimization control method according to claim 3 is characterized in that: The sensors include a temperature sensor, a pressure sensor, a flow sensor, an oxygen sensor and a flue gas analyzer. The temperature sensor is used to collect temperature data inside the boiler, the pressure sensor is used to collect working pressure data inside the boiler, the flow sensor is used to collect flow data inside the boiler, the oxygen sensor is used to collect oxygen concentration data during the boiler combustion process, and the flue gas analyzer is used to collect pollutant concentration data in the flue gas discharged by the boiler.

5. The intelligent boiler combustion optimization control method according to claim 3 is characterized in that: The method for preprocessing the data generated during boiler operation is as follows: Assume that the sensor data is X = {x1, x2, …, xn}. The value range of each sensor is different. The normalization formula is as follows: Among them, xi is the original data, μi is the mean of the i-th sensor data, and σi is its standard deviation; Use mean filling to handle missing values: Where N is the number of valid data, replacing missing values; Use a sliding window to construct samples, each sample contains T time steps of data: X t ={x t-T+1 ,x t-T+2 ,...,x t } Each input sample contains T time steps of data.

6. The intelligent boiler combustion optimization control method according to claim 5, characterized in that: The method of fusing multi-dimensional data from various sensors to generate a comprehensive evaluation of boiler combustion status information is as follows: A neural network is used to fuse multi-dimensional data from various sensors to perform a comprehensive assessment of the boiler combustion status, as follows: Input layer: The data of each sensor is used as input. If the number of sensors is n, the size of the input layer is n; The input is x = [x1, x2, …, xn], and the goal of the neural network is to output a combustion state assessment based on this input data; Hidden layer: Design multiple fully connected layers. Let the output of the k-th layer be hk, the number of neurons in this layer be mk, and the calculation formula of the k-th layer be: h k =f(W k h k-1 +b k ) Where Wk is the weight matrix, bk is the bias term, f is the activation function, and the output of the input layer is regarded as the 0th layer, denoted as h0 = x; Assume that the output is a scalar y, which represents the evaluation value of the combustion state, and uses the Softmax activation function: Where yi represents the predicted probability of the i-th category, and C is the number of categories; The loss function is the mean square error: Where y^ is the predicted value of the model, y is the true value, and N is the number of samples; Use the gradient descent method to minimize the loss function. The formula for updating the weights by gradient descent is: Where η is the learning rate, is the gradient of the loss function with respect to the weights; The dataset is divided into training set and validation set, and the generalization ability of the model is evaluated through cross-validation; The evaluation indicator used is mean square error; When new sensor data arrives, it is input into the trained neural network model to calculate the evaluation result y of the combustion state pred .

7. The intelligent boiler combustion optimization control method according to claim 6, characterized in that: Method for generating combustion control decisions and adjusting boiler combustion parameters based on combustion state evaluation results: According to the output y pred , mapping the evaluation result of the combustion state to the combustion control signal u(t); According to the generated control signal u(t), the fuel flow rate mf(t), air flow rate ma(t), furnace temperature Tf(t) and exhaust gas temperature Ts(t) are adjusted as follows: m f (t)=K f ·u(t) m a (t)=K a ·m f (t) T f (t)=T f，set +K T ·(u(t)-u ref ) T s (t)=T s，set +K s ·(m f (t)-m a (t)) The control signal u(t) is a control decision signal generated based on the current operating state, target state, and external environmental data of the boiler. The adjustment coefficient Kf of the fuel flow rate mf(t) and the ratio coefficient Ka between the air flow rate and the fuel flow rate are set according to the design parameters and operating requirements of the boiler. The adjustment coefficient KT of the furnace temperature Tf(t) is set according to the combustion characteristics and operating state of the boiler. The adjustment coefficient Ks of the exhaust temperature Ts(t) is set according to the exhaust system and thermal efficiency requirements of the boiler. According to the error between the evaluation result ypred and the set target, the corresponding control signal is adjusted by the PID controller: Where e(t) is the evaluation error, and Kp, Ki, and Kd are the proportional, integral, and derivative gains.

8. The intelligent boiler combustion optimization control method according to claim 7, characterized in that: During the combustion control process, the boiler's combustion efficiency, pollutant emissions, and oxygen content are calculated in real time. If an abnormality occurs, an alarm mechanism is triggered and the combustion control parameters are automatically adjusted as follows: The combustion efficiency of the boiler is calculated in real time, and the combustion efficiency is calculated by the following formula: Among them, Qoutput is the actual output heat energy, Qinput is the theoretical heat energy released by the fuel; The pollutant emissions from the boiler are monitored in real time. The pollutants include nitrogen oxides NOx, carbon dioxide CO2 and carbon monoxide CO, and the emissions are calculated using the following formula: NOx emissions = k NOx ·(T flame -1000)·(λ-1) CO emissions = k CO ·(1-η 燃烧 ) Where Tflame is the combustion temperature, λ is the excess air coefficient, mfuel is the fuel flow rate, Ccarbon is the carbon content in the fuel, and HCO2 is the amount of CO2 produced per unit carbon. The oxygen content O2 in the boiler flue gas is measured in real time and compared with the set oxygen content target to ensure the integrity of the combustion process. When the combustion efficiency falls below a set threshold, pollutant emissions exceed the standard, or the oxygen content is abnormal, an alarm mechanism is triggered to further adjust the combustion control parameters to optimize the boiler's combustion process. The alarm trigger conditions include: The combustion efficiency is lower than the set lower limit; Pollutant emissions exceed legal emission standards; The oxygen content is lower than the set minimum value; According to the alarm results, the combustion control parameters are automatically adjusted, including but not limited to adjusting the fuel flow, air flow and excess air coefficient.

9. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for optimizing combustion control of an intelligent boiler as claimed in any one of claims 3 to 8 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the intelligent boiler combustion optimization control method according to any one of claims 3 to 8 is implemented.

Images