Combustion control method of garbage incinerator, garbage incineration system and related equipment

Abstract

The embodiments of the present application provide a combustion control method, a waste incineration system and related equipment for a waste incinerator. The Industrial Brain open platform periodically obtains real-time operating data collected in real time during the combustion process of the waste incinerator; inputs the real-time operating data into the operating parameter prediction model to predict the predicted values ​​of the key operating parameters of the waste incinerator in the future time period; inputs the real-time operating data and the predicted values ​​of the key operating parameters into the reinforcement learning model to decide the target setting values ​​of the key operating parameters of the waste incinerator in the future time period; sends the target setting values ​​of the key operating parameters to the combustion control system; the combustion control system performs combustion control on the waste incinerator in the future time period according to the target setting values ​​of the key operating parameters. Thus, the Industrial Brain open platform can achieve more flexible and effective combustion control by using the operating parameter prediction model and the reinforcement learning model, thereby improving the resource utilization, environmental protection performance and economic benefits of the waste incinerator.

Description

Technical Field

[0001] The present application relates to the technical field of waste incineration, and in particular to a combustion control method of a waste incinerator, a waste incineration system and related equipment. Background Art

[0002] A waste incinerator is a facility specially designed to treat solid waste (usually municipal solid waste) through high-temperature combustion. Its main function is to convert waste into ash and gas through the combustion process, thereby significantly reducing the volume of waste. In many cases, the heat energy generated during the combustion process can be recovered to generate electricity or heat, thereby realizing resource reuse.

[0003] In practice, the combustion process in a waste incinerator is complex and influenced by numerous factors, including but not limited to the calorific value of the waste, pollutant emission indicators, furnace temperature, flue gas oxygen content, or negative pressure. These factors work together to determine the resource utilization, environmental performance, and economic efficiency of the incineration process. In particular, key operating parameters such as furnace temperature, flue gas oxygen content, and negative pressure have a direct impact on the resource utilization, environmental performance, and economic efficiency of the waste incinerator. For example, an appropriate furnace temperature helps improve combustion efficiency and reduce the formation of harmful substances; a suitable flue gas oxygen content is essential for complete combustion, reducing the emission of unburned materials and carbon monoxide; and maintaining an appropriate negative pressure prevents flue gas leakage, protecting the environment and the safety of operators.

[0004] However, in practice, the setpoints for these key operating parameters often rely on manual adjustments by operators based on their personal experience. This approach makes it difficult to ensure that the incinerator is always operating optimally, as manual adjustments may not accurately respond to complex operating conditions, reducing the incinerator's resource utilization, environmental performance, and economic benefits. Summary of the Invention

[0005] The embodiments of the present application provide a combustion control method for a waste incinerator, a waste incineration system, and related equipment to improve the resource utilization, environmental protection performance, and economic benefits of the waste incinerator.

[0006] An embodiment of the present application provides a combustion control method for a waste incinerator, which is applied to an industrial brain open platform. The method includes: periodically obtaining real-time operating data collected in real time during the combustion process of the waste incinerator; inputting the real-time operating data into an operating parameter prediction model to predict the predicted values ​​of key operating parameters of the waste incinerator in future time periods; inputting the real-time operating data and the predicted values ​​of key operating parameters into a reinforcement learning model to determine the target setting values ​​of the key operating parameters of the waste incinerator in future time periods; and sending the target setting values ​​of the key operating parameters to a combustion control system, so that the combustion control system controls the combustion of the waste incinerator in the future time period according to the target setting values ​​of the key operating parameters.

[0007] An embodiment of the present application also provides a waste incineration system, including: an industrial brain open platform, a combustion control system and a waste incinerator; the industrial brain open platform is used to periodically obtain real-time operating data of the waste incinerator collected in real time during the combustion process of the waste incinerator; the real-time operating data is input into an operating parameter prediction model to predict the predicted values ​​of key operating parameters of the waste incinerator in future time periods; the real-time operating data and the predicted values ​​of key operating parameters are input into a reinforcement learning model to decide the target setting values ​​of the key operating parameters of the waste incinerator in future time periods; the target setting values ​​of the key operating parameters are sent to the combustion control system; the combustion control system is used to control the combustion of the waste incinerator in future time periods according to the target setting values ​​of the key operating parameters.

[0008] An embodiment of the present application also provides an electronic device, comprising: a memory and a processor; the memory is used to store a computer program; the processor is coupled to the memory, and is used to execute the computer program to execute the steps in the combustion control method of a waste incinerator.

[0009] An embodiment of the present application also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor is enabled to implement the steps in the combustion control method of a waste incinerator.

[0010] An embodiment of the present application also provides a computer program product, including a computer program / instruction. When the computer program / instruction is executed by a processor, the processor is enabled to implement the steps in the combustion control method of a waste incinerator.

[0011] The technical solution provided by the embodiment of the present application is that the industrial brain open platform periodically obtains real-time operating data collected in real time during the combustion process of the waste incinerator; inputs the real-time operating data into the operating parameter prediction model to predict the predicted values ​​of the key operating parameters of the waste incinerator in the future time period; inputs the real-time operating data and the predicted values ​​of the key operating parameters into the reinforcement learning model to decide the target setting values ​​of the key operating parameters of the waste incinerator in the future time period; sends the target setting values ​​of the key operating parameters to the combustion control system; the combustion control system controls the combustion of the waste incinerator in the future time period according to the target setting values ​​of the key operating parameters. Therefore, the industrial brain open platform can accurately and efficiently use the operating parameter prediction model and the reinforcement learning model to set the target setting values ​​of the key operating parameters of the waste incinerator in the future time period, can flexibly adapt to different working conditions, achieve more flexible and effective combustion control, and improve the resource utilization, environmental protection performance and economic benefits of the waste incinerator. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0013] Figure 1 A schematic structural diagram of an exemplary waste incineration system provided in an embodiment of the present application;

[0014] Figure 2 A signaling interaction diagram of a combustion control method for a waste incinerator provided in an embodiment of the present application;

[0015] Figure 3 is an exemplary application scenario diagram;

[0016] Figure 4 A schematic structural diagram of a combustion control device for a waste incinerator provided in an embodiment of the present application;

[0017] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] To make the purpose, technical solutions, and advantages of this application more clear, the technical solutions of this application will be clearly and completely described below in conjunction with the specific embodiments of this application and the corresponding drawings. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0019] In the embodiments of the present application, "at least one" refers to one or more, and "more" refers to two or more. "And / or" describes the access relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent three situations: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. In the text description of the present application, the character " / " generally indicates that the previous and next associated objects are in an "or" relationship. In addition, in the embodiments of the present application, "first", "second", "third", etc. are only used to distinguish the contents of different objects and have no other special meanings.

[0020] It should be noted that when the embodiments of this application involve user information, the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the embodiments of this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation portals for users to choose to authorize or refuse. In addition, the various models involved in this application (including but not limited to language models or large models) are in compliance with relevant laws and standards.

[0021] The following specific embodiments describe in detail the technical solutions of this application and how the technical solutions of this application solve the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The following detailed description of the technical solutions provided by each embodiment of this application is given in conjunction with the accompanying drawings.

[0022] Figure 1 This is a schematic diagram of an exemplary waste incineration system provided in an embodiment of the present application. Figure 1 The waste incineration system may include: an industrial brain open platform 10, a combustion control system 20 and a waste incinerator 30.

[0023] The Industrial Brain Open Platform is an intelligent system that integrates multiple technologies, including big data processing, cloud computing, the Internet of Things (IoT), and artificial intelligence (AI). It aims to unlock the value of data by integrating and analyzing data from various sources, including enterprise systems, factory equipment, sensor data, and personnel management data, leveraging advanced technologies such as voice interaction, image / video recognition, machine learning, and AI algorithms. By unlocking the value of massive amounts of data, the Industrial Brain Open Platform provides enterprises with powerful data analysis capabilities and intelligent decision-making support, thereby accelerating the digital transformation and intelligent upgrade of the industrial sector.

[0024] The combustion control system 20 is a control system that ensures the efficient, safe and environmentally friendly operation of the incineration process of the waste incinerator. It optimizes combustion efficiency and reduces harmful emissions by precisely controlling various parameters in the incineration process, such as temperature, oxygen content, combustion rate, etc.

[0025] A waste incinerator 30 is a facility used to process solid waste, converting it into ash, exhaust gas, and heat through high-temperature combustion. This treatment method not only reduces the volume of waste but also utilizes the heat generated during the incineration process to generate electricity or heat, thereby achieving resource recovery.

[0026] In this embodiment, the historical operating data collected during the combustion process of the waste incinerator 30 is used to train the operating parameter prediction model and the reinforcement learning model. In this way, the industrial brain open platform 10 can accurately use the operating parameter prediction model and the reinforcement learning model to set the target setting values ​​of the key operating parameters of the waste incinerator in the future time period.

[0027] In order to better understand the technical solutions provided by the embodiments of the present application, Figure 2 The signaling interaction diagram shown is introduced. Figure 2 This is a signaling interaction diagram of a combustion control method for a waste incinerator provided in an embodiment of the present application. Figure 2 , the method may include the following steps:

[0028] 201. The Industrial Brain open platform periodically obtains real-time operating data collected during the combustion process of the waste incinerator.

[0029] Specifically, see Figure 1 As shown in Figure 1, the Industrial Brain Open Platform periodically acquires real-time operating data collected during the incinerator's combustion process. In practical applications, the Industrial Brain Open Platform can periodically acquire real-time operating data collected during the incinerator's combustion process at set intervals (e.g., 1 minute, 5 minutes, 1 hour, etc.).

[0030] In this embodiment, real-time operation data refers to the operation data collected in real time during the combustion process of the waste incinerator. The real-time operation data can be various operation data, and there is no limitation on this. Furthermore, in order to accurately determine the target setting values ​​of the key operation parameters of the waste incinerator in the future time period, the real-time operation data includes at least one of the following: the number of pushes of the pusher in the recent time period, the grate action cycle, the number of grate actions, the working frequency of the primary fan, the working frequency of the secondary fan, the working frequency of the induced draft fan, the primary air volume, the secondary air volume, the flue gas flow rate, the flue gas temperature, the cooling water flow rate, the wind chamber pressure, the steam flow rate, the steam temperature, the furnace temperature, the furnace negative pressure, the flue gas oxygen content, the environmental protection index data and the key operation data of the garbage storage system; the key operation data of the garbage storage system include: the coordinate position, weight and garbage fermentation time of the garbage crane.

[0031] In this embodiment, the most recent time period may be the time period closest to the current time, such as the past minute, the past hour, etc., and can be flexibly set as needed. The real-time operating data within the most recent time period has better real-time performance and can reflect the current operating condition of the waste incinerator.

[0032] The pusher's push count refers to the number of times the pusher pushes waste from the feed port into the incinerator for combustion during the incineration process. This parameter is an important indicator of the rate at which waste enters the incinerator and is crucial for controlling the stability and efficiency of the incineration process.

[0033] Grate action cycle: refers to the time required for a complete process of grate mechanical movement, which can include multiple stages such as feeding, drying, combustion, burnout and ash discharge.

[0034] Grate Operation Times: This refers to the number of times the grate is operated within a specific time period. This parameter is very important for controlling the combustion process because it directly affects the fuel supply rate, combustion efficiency, and ash removal efficiency.

[0035] The primary fan's main function is to provide the necessary oxygen for the combustion process, ensuring that the fuel can be fully burned. Typically, the primary fan's operating frequency may vary within a large range, such as from 30 Hz to 50 Hz or even wider.

[0036] The primary function of a secondary fan in a waste incinerator is to provide excess air, ensuring sufficient oxygen in the combustion zone to completely burn the combustible gases and achieve a more thorough combustion process. This helps improve combustion efficiency and reduce harmful emissions such as carbon monoxide and unburned hydrocarbons. Typically, the operating frequency range of a secondary fan can be set within a wide range, such as from 30 Hz to 50 Hz or higher.

[0037] The main function of the induced draft fan is to maintain an appropriate negative pressure in the waste incinerator to prevent smoke leakage and ensure gas flow within the combustion chamber. Typically, the operating frequency of the induced draft fan can vary within a wide range, such as from 20 Hz to 50 Hz or higher.

[0038] Primary air volume refers to the amount of air supplied by the primary fan into the incinerator's combustion chamber. It primarily supports the initial combustion of waste, ensuring sufficient combustion of the fuel (i.e., waste). The supply of primary air is crucial for maintaining stable combustion temperatures, improving combustion efficiency, and reducing harmful emissions.

[0039] Secondary air volume refers to the additional air supplied to the combustion chamber during the waste incineration process to promote complete combustion. Unlike primary air, which primarily supports the initial combustion of waste, secondary air's primary function is to provide sufficient oxygen to ensure the complete combustion of combustible gases and fine particles in the high-temperature zone, thereby improving combustion efficiency and reducing harmful emissions such as carbon monoxide (CO), unburned hydrocarbons (HC), and fine particulate matter.

[0040] Flue gas flow refers to the velocity and total volume of flue gas generated during the waste incineration process as it passes through the incinerator, waste heat boiler, flue gas purification system, and other equipment. Monitoring and controlling flue gas flow is crucial to ensuring stable operation of the incineration system, improving combustion efficiency, and meeting environmental emission standards.

[0041] Flue gas temperature refers to the temperature of the exhaust gas (i.e., flue gas) produced by the combustion of fuels (such as garbage, coal, and natural gas) during the combustion process. Flue gas temperature is a key operating parameter that reflects the efficiency and state of the combustion process and has a significant impact on subsequent heat recovery and pollutant control.

[0042] Desuperheating water flow refers to the amount of water injected to regulate and control the temperature of flue gas or steam. The purpose is to reduce the temperature of the medium (such as superheated steam or high-temperature flue gas) by direct or indirect means to protect downstream equipment from excessive temperatures and optimize the overall efficiency of the system.

[0043] Chamber pressure refers to the pressure within the chamber (the cavity used to distribute primary or secondary air) of a waste incinerator. This pressure is crucial for ensuring even and effective distribution of air to the combustion area. It directly impacts combustion efficiency, complete fuel combustion, and pollutant generation.

[0044] Steam flow refers to the amount of steam generated by the waste incinerator. Steam flow is usually expressed in the form of mass flow (such as kg / hour or tons / hour) or volume flow (such as cubic meters / hour).

[0045] Steam temperature refers to the temperature of the steam generated by the waste incinerator. A higher steam temperature usually means a higher thermal energy content, which can improve power generation efficiency.

[0046] Furnace temperature refers to the temperature inside the combustion chamber (i.e. furnace) in a waste incinerator. It is the temperature reached by the air and combustion products (flue gas) in the furnace due to the heat released during the combustion of fuel. Furnace temperature is one of the important parameters for measuring the efficiency and safety of the combustion process.

[0047] Furnace negative pressure refers to a slightly negative pressure state inside the furnace of a waste incinerator relative to the external atmospheric pressure. Its main purpose is to ensure that the flue gas generated during the combustion process can flow along a predetermined path and prevent the flue gas from leaking into the surrounding environment, causing pollution and safety hazards.

[0048] Flue gas oxygen content refers to the oxygen content in the flue gas during the combustion process. It is one of the key factors that characterize whether the fuel is fully burned.

[0049] Environmental protection index data refers to the specific values ​​of various pollutants emitted into the atmosphere during processes such as waste incineration, including but not limited to the emission concentrations of particulate matter, sulfur dioxide, nitrogen oxides, dioxins and other pollutants.

[0050] The coordinate position of the garbage crane can refer to the spatial position of the garbage crane, which is usually recorded in the form of three-dimensional coordinates (X, Y, Z); the weight of the garbage crane can reflect the actual weight of the garbage grabbed by the garbage crane each time; the fermentation time can reflect the time from the time the garbage enters the garbage storage to the time it is put into the incinerator, usually in hours or days.

[0051] 202. The Industrial Brain open platform inputs real-time operating data into the operating parameter prediction model to predict the predicted values ​​of key operating parameters of the waste incinerator in the future time period.

[0052] Specifically, after the Industrial Brain Open Platform obtains real-time operation data, it inputs the real-time operation data into the operation parameter prediction model. Figure 1 As shown in Figure ②, the operating parameter prediction model predicts the key operating parameters of the waste incinerator in the future time period based on real-time operating data. The length of the future time period is not limited, for example, the next 5 minutes, the next 10 minutes, the next 30 minutes, etc.

[0053] In this embodiment, key operating parameters refer to important indicators that have a direct impact on the combustion efficiency, resource utilization, environmental emissions, and economic benefits of the waste incinerator. Preferably, key operating parameters include, but are not limited to, furnace temperature, furnace negative pressure, flue gas oxygen content, steam flow rate, or steam temperature.

[0054] For example, suppose the current time is 10:34 on April 18, 2025, and key operating parameters need to be predicted for the next 10 minutes (10:34 to 10:44). First, relevant real-time operating data is collected. Second, data preprocessing is performed on the real-time operating data, including but not limited to outlier removal, missing value filling, and data cleaning to ensure data quality. Third, the preprocessed real-time operating data is input into the operating parameter prediction model for prediction, resulting in predicted values ​​for key operating parameters for the next 10 minutes. For example, if the current furnace temperature is 975°C, the operating parameter prediction model predicts that the furnace temperature will fluctuate between 970°C and 980°C over the next 10 minutes. If the current flue gas oxygen content is 6%, the operating parameter prediction model predicts that the flue gas oxygen content will remain between 5.8% and 6.2% over the next 5 minutes to accommodate changes in combustion load. The current steam flow rate is 20 tons / hour, and the operating parameter prediction model predicts that the steam flow rate will increase slightly to about 20.5 tons / hour in the next 10 minutes, which may be due to the improvement of combustion efficiency.

[0055] 203. The Industrial Brain open platform inputs real-time operating data and predicted values ​​of key operating parameters into the reinforcement learning model to determine the target setting values ​​of the key operating parameters of the waste incinerator in the future time period.

[0056] Reinforcement learning (RL) is a learning method based on a trial-and-error mechanism, which learns optimal strategies through interaction with the environment. By outputting target values ​​for key operating parameters of a waste incinerator through a reinforcement learning model, the intelligence level and operational efficiency of waste combustion systems can be significantly improved, providing technical support for the green and low-carbon development of the waste incineration sector. Reinforcement learning models can dynamically adjust strategies based on environmental changes, adapting to different operating conditions and achieving more flexible and effective combustion control. The reinforcement learning model determines the optimal action based on the current state (i.e., real-time operating data and predicted values ​​of key operating parameters). The optimal action is the target value for the incinerator's key operating parameters for the future time period.

[0057] In this embodiment, the Industrial Brain Open Platform inputs the real-time operating data and the predicted values ​​of the key operating parameters of the waste incinerator in the future time period predicted by the operating parameter prediction model into the reinforcement learning model. Figure 1As shown in Figure ③, the reinforcement learning model outputs target setpoints for the incinerator's key operating parameters for a future time period. Target setpoints can be understood as ideal values ​​that the incinerator's key operating parameters should achieve within a certain time period. For example, the target setpoint for furnace temperature is 970°C; the target setpoint for flue gas oxygen content is 6%; the target setpoint for steam flow is 21 tons / hour; and the target setpoint for furnace negative pressure is -30 Pa.

[0058] 204. The industrial brain open platform sends the target setting values ​​of key operating parameters to the combustion control system.

[0059] 205. The combustion control system controls the combustion of the waste incinerator in the future time period according to the target set values ​​of key operating parameters.

[0060] See also Figure 1 As shown in Figures ③ and ④, the Industrial Brain open platform sends the target setpoints for key operating parameters to the combustion control system. The combustion control system then controls the waste incinerator's combustion in future time periods based on these target setpoints, optimizing the combustion process, improving combustion efficiency, enhancing resource utilization, and reducing pollutant emissions. For example, to achieve a furnace temperature of 970°C, the combustion control system may need to increase the fuel supply rate. This can be achieved by adjusting the feeder speed to ensure more fuel enters the furnace to provide the required heat. For another example, to maintain a flue gas oxygen content of 6%, the combustion control system fine-tunes the primary and secondary fans, controlling the ratio and flow rate of the primary and secondary air. To achieve a steam flow rate of 21 tons / hour, the combustion control system adjusts the boiler water level and feedwater flow to ensure sufficient water is converted into steam. Adjusting the combustion intensity based on actual demand affects steam production. For another example, to maintain an ideal furnace negative pressure of -30 Pa, the combustion control system adjusts the frequency of the induced draft fan. If the monitored actual negative pressure is higher than -30 Pa, the operating frequency of the induced draft fan is reduced; otherwise, the operating frequency is increased.

[0061] In actual applications, the combustion control system will continuously monitor the actual values ​​of key operating parameters and compare them with the target set values. Based on the comparison results, the waste incinerator system will be optimized in real time to ensure that the waste incinerator system always operates in the best state.

[0062] The technical solution provided by the embodiment of the present application is that the industrial brain open platform periodically obtains real-time operating data collected in real time during the combustion process of the waste incinerator; inputs the real-time operating data into the operating parameter prediction model to predict the predicted values ​​of the key operating parameters of the waste incinerator in the future time period; inputs the real-time operating data and the predicted values ​​of the key operating parameters into the reinforcement learning model to decide the target setting values ​​of the key operating parameters of the waste incinerator in the future time period; sends the target setting values ​​of the key operating parameters to the combustion control system; the combustion control system controls the combustion of the waste incinerator in the future time period according to the target setting values ​​of the key operating parameters. Therefore, the industrial brain open platform can accurately and efficiently use the operating parameter prediction model and the reinforcement learning model to set the target setting values ​​of the key operating parameters of the waste incinerator in the future time period, can flexibly adapt to different working conditions, achieve more flexible and effective combustion control, and improve the resource utilization, environmental protection performance and economic benefits of the waste incinerator.

[0063] In practical applications, there are no restrictions on the model structure of the operating parameter prediction model. Examples include, but are not limited to, ARIMA (AutoRegressive Integrated Moving Average), SARIMA (Seasonal AutoRegressive Integrated Moving Average), and LSTM (Long Short-Term Memory).

[0064] In practical applications, there is no restriction on the training method of the operating parameter prediction model, including but not limited to: supervised learning, unsupervised learning, and semi-supervised learning.

[0065] Furthermore, in order to improve the prediction accuracy of the operating parameter prediction model, the training method of the operating parameter prediction model includes: obtaining historical operating data of multiple different first time periods collected during the combustion process of the waste incinerator; constructing a first training data set and a first test data set based on the historical operating data of multiple different first time periods; with the goal of predicting the predicted values ​​of the key operating parameters of the waste incinerator in the second time period, using the first training data set to train the model to obtain an initial operating parameter prediction model, the second time period being later than the first time period; using the first test data set to adjust the parameters of the operating parameter prediction model to obtain an operating parameter prediction model.

[0066] Specifically, the first time period is a past time period, and the second time period is later than the first time period. Relatively speaking, the second time period is a future time period corresponding to the first time period. Historical operating data can be understood as operating data collected during the combustion process of the waste incinerator in the past time period. During the preparation of the training data set, historical operating data from multiple different first time periods can be selected to cover different operating conditions, thereby increasing the generalization ability of the model. Data preprocessing can also be performed on the collected historical operating data, such as removing outliers, filling missing values, standardizing or normalizing, etc., to improve data quality.

[0067] In this embodiment, the collected historical operation data is divided into a training data set and a test data set. For ease of understanding and distinction, the training data set for training the operation parameter prediction model is referred to as the first training data set, and the test data set for testing the operation parameter prediction model is referred to as the first test data set. In practical applications, there is no restriction on the way of dividing the training data set and the test data set. In practical applications, the division can be made according to the division ratio, for example, 70%-80% of the historical operation data is used for training, and 20%-30% of the historical operation data is used for testing. Alternatively, it can be divided in chronological order. The operation data of the waste incinerator is usually time series data, which has time dependency (that is, the data at the current moment is related to the data at the previous moment). For example, the historical operation data is sorted by time. The historical operation data of the earlier time period is used as the first training data set, and the historical operation data of the later time period is used as the first test data set. For example, assuming there is data from the past year (January 1, 2024 to December 31, 2024), the historical operating data from the first nine months (January 1, 2024 to September 30, 2024) can be used as the first training data set, and the historical operating data from the last three months (October 1, 2024 to December 31, 2024) can be used as the first test data set.

[0068] In this embodiment, model training is performed in conjunction with the first training dataset and the first test dataset to ensure that the model performs well on unseen data while avoiding overfitting or underfitting. First, the model is trained using the first training dataset to obtain an initial operating parameter prediction model. Next, the operating parameter prediction model is parameter-adjusted using the first test dataset to obtain a final operating parameter prediction model. In practical applications, the first test dataset is primarily used to fine-tune model parameters to improve the performance of the operating parameter prediction model on unseen data. For example, the first test dataset is used to evaluate model performance, such as calculating evaluation metrics such as mean squared error (MSE), mean absolute error (MAE), and coefficient of determination (R²). Based on the evaluation metrics, hyperparameters such as the learning rate, batch size, and number of layers are fine-tuned to optimize the operating parameter prediction model.

[0069] In some optional embodiments, operating parameter prediction models under different operating modes can be trained. Different operating modes include: high-yield mode (maximizing steam volume), load priority mode (prioritizing steam stability control), furnace temperature priority mode (prioritizing furnace temperature stability control), environmental index control priority mode, and some special system operation operations such as steam soot blowing. Different operating modes will affect the change characteristics of operating parameters. The main goal of the high-yield mode is to maximize steam production. The main goal of the load priority mode is to maintain the stability and reliability of steam supply. The main goal of the furnace temperature priority mode is to maintain the stability of the temperature in the furnace. The main goal of the environmental index control priority mode is to further reduce pollutant emissions while meeting environmental regulations and standards, such as meeting more stringent internal control standards.

[0070] In practical applications, a first training data set and a first test data set associated with each operating mode indicator can be collected; the first training data set and the first test data set associated with each operating mode can be used to train the model to obtain an operating parameter prediction model applicable to each operating mode; the operating parameter prediction models applicable to each operating mode can be combined to obtain a final operating parameter prediction model.

[0071] In some optional embodiments, the real-time operating data is input into the operating parameter prediction model to predict the predicted values ​​of the key operating parameters of the waste incinerator in the future time period as follows: the real-time operating data is input into the operating parameter prediction model, so that the operating parameter prediction model predicts the initial predicted values ​​of the key operating parameters of the waste incinerator in the future time period under different operating modes; and the predicted values ​​of the key operating parameters of the waste incinerator in the future time period are obtained according to the initial predicted values ​​of the key operating parameters of the waste incinerator in the future time period under different operating modes.

[0072] In practical applications, the initial predicted values ​​of the key operating parameters of the waste incinerator in the future time period under different operating modes can be averaged or weighted summed to obtain the predicted values ​​of the key operating parameters of the waste incinerator in the future time period.

[0073] Furthermore, in order to accurately predict the predicted values ​​of key operating parameters, the predicted values ​​of the key operating parameters of the waste incinerator in the future time period are obtained according to the initial predicted values ​​of the key operating parameters of the waste incinerator in the future time period under different operating modes. The implementation method is as follows: weighted summation is performed on the initial predicted values ​​of the key operating parameters of the waste incinerator in the future time period under different operating modes to obtain the intermediate predicted values ​​of the key operating parameters of the waste incinerator in the future time period; with the goal of meeting the process requirements of the combustion process, the intermediate predicted values ​​of the key operating parameters of the waste incinerator in the future time period are corrected to obtain the predicted values ​​of the key operating parameters of the waste incinerator in the future time period.

[0074] In practical applications, weighting based on multiple objectives (multiple operating modes) is supported. Relevant personnel can set the weight information corresponding to each operating mode according to actual needs. By assigning weights to these different operating modes, the importance of each operating mode can be flexibly adjusted according to actual needs.

[0075] In practice, the process requirements of the combustion process can constrain the numerical ranges of various operating parameters, such as the temperature range of the furnace; the corresponding pressure ranges for plenum pressure and furnace negative pressure; and the corresponding intervals for various pollutant emissions. Correcting the prediction results to meet the process requirements can ensure the safe, environmentally friendly, and efficient operation of the waste incineration process.

[0076] In this embodiment, there is no restriction on the training method of the reinforcement learning model. For example, the reinforcement learning model can be trained offline based on historical operating data. For another example, after the reinforcement learning model is trained offline based on historical operating data, it can also be updated in real time through online learning.

[0077] Furthermore, in order to improve the model performance of the reinforcement learning model, the training method of the reinforcement learning model includes: obtaining historical operating data of multiple different third time periods collected during the combustion process of the waste incinerator; for any third time period, inputting the historical operating data of the third time period into the operating parameter prediction model to predict the predicted value of the key operating parameters of the waste incinerator in the fourth time period, and the fourth time period is later than the third time period; constructing a second training data set and a second test data set based on the historical operating data of multiple different third time periods and the predicted values ​​of multiple key operating parameters in the fourth time period; using the second training data set as the target setting value of the key operating parameters of the waste incinerator in the fourth time period as the goal, model training is performed to obtain an initial reinforcement learning model; and using the second test data set to adjust the parameters of the initial reinforcement learning model to obtain a reinforcement learning model.

[0078] Specifically, the third time period is a past time period, and the fourth time period is later than the third time period. Relatively speaking, the fourth time period is a future time period corresponding to the third time period. In the preparation stage of the training data set, historical operating data of multiple different third time periods can be selected to cover different working conditions, thereby increasing the generalization ability of the model. Data preprocessing can also be performed on the collected historical operating data, for example, removing outliers, filling missing values, standardization or normalization, etc., to improve data quality. The collected historical operating data is divided into a training data set and a test data set. For ease of understanding and distinction, the training data set for training the reinforcement learning model is referred to as the second training data set, and the test data set for testing the reinforcement learning model is referred to as the second test data set.

[0079] In this embodiment, combining the second training data set and the second test data set for model training can ensure that the model can perform well on unseen data while avoiding the problems of overfitting or underfitting. First, the model is trained using the second training data set to obtain an initial reinforcement learning model; the initial reinforcement learning model is parameter-adjusted using the second test data set to obtain a reinforcement learning model. In practical applications, the second test data set is mainly used to fine-tune the model parameters so that the reinforcement learning model performs better on unseen data. For example, the second test data set is used to evaluate model performance, such as determining evaluation indicators such as the value of a weighted reward function over a period of time. The weighted reward function value is obtained by weighted summing the reward values ​​of the reward function over a period of time; hyperparameters such as the learning rate, batch size, and number of layers are fine-tuned based on the evaluation indicators to optimize the reinforcement learning model.

[0080] Furthermore, in order to improve the model performance of the reinforcement learning model and enable the reinforcement learning model to better dynamically optimize the key operating parameters of the waste incinerator, the second training data set includes multiple current state data, and the current state data includes: historical operating data of the third time period and predicted values ​​of the key operating parameters in the fourth time period; accordingly, the second training data set is used for model training to obtain the initial reinforcement learning model in the following manner: input the current state data into the initial reinforcement learning model to obtain the set values ​​of the key operating parameters of the waste incinerator in the fourth time period as determined by the reinforcement learning model; send the set values ​​of the key operating parameters in the fourth time period to the combustion control system, so that the combustion control system controls the combustion of the waste incinerator in the fourth time period according to the set values ​​of the key operating parameters; after performing combustion control based on the current state data, determine the reward value using the reward function, and adjust the model parameters of the initial reinforcement learning model according to the reward value; repeat the above steps until an initial reinforcement learning model that meets the requirements is obtained.

[0081] In practical applications, a suitable reward function is designed and used to evaluate the effectiveness of action decisions based on the current state, encouraging beneficial behaviors and punishing undesirable ones. Preferably, the reward function can be designed based on the following factors: ① Improving combustion efficiency: increase the reward value; ② Reducing pollutant emissions: increase the reward value; ③ Exceeding safety limits (such as excessive temperature or abnormal pressure): decrease the reward value; ③ Increasing the amount of electricity online per ton of garbage: increase the reward value; ④ Reducing the amount of electricity online per ton of garbage: decrease the reward value; ⑤ Improving the values ​​of other key indicators: increase the reward value; ⑥ Reducing the values ​​of other key indicators: decrease the reward value. The amount of electricity online per ton of garbage can be calculated based on the amount of steam used for power generation, the fan power, and the amount of garbage fed into the incinerator. The amount of electricity online per ton of garbage reflects the amount of electricity generated by incinerating each ton of garbage fed into the incinerator. For example, A = B × C × D / 3600; E = AF * T; N = E / M; where A represents total power generation; B represents steam generation; C represents enthalpy difference; D represents turbine efficiency; E represents on-grid power; F represents fan power; T represents fan operating time; N represents on-grid power per ton of waste; and M represents waste feed volume. Of course, the method for determining on-grid power per ton of waste can be flexibly configured as needed. Other key indicators include, but are not limited to: steam generation per unit of waste (tons / ton), which measures waste calorific value conversion efficiency; fan power consumption per unit of waste (kilowatts / ton), which reflects the energy efficiency of air supply; and electrical energy output per unit of steam production (kilowatt-hours / ton), which assesses power generation efficiency.

[0082] In reinforcement learning, adjusting model parameters based on reward values ​​is a core step in model training and optimization. This process updates the model's policy or value function, enabling the model to gradually learn to take optimal actions in different states, thereby maximizing long-term cumulative rewards. In practical applications, appropriate optimization algorithms (such as gradient descent and policy gradient) can be used to adjust model parameters based on reward values.

[0083] By adjusting the parameters of a reinforcement learning model based on reward values, the model can gradually learn to make better decisions in complex industrial environments. In a waste incinerator scenario, this adjustment process can help optimize the target setpoints of key operating parameters, thereby achieving higher combustion efficiency, lower pollutant emissions, and better economic benefits.

[0084] It is worth noting that in actual applications, the initial reinforcement learning model obtained by model training using only the second training data set can also be used as the final reinforcement learning model for predicting the target setting values ​​of key operating parameters, and there is no restriction on this.

[0085] In order to better understand the technical solution of this application, Figure 3 See the application scenario diagram shown in the figure. Figure 3 , the entire combustion control process can include the following steps:

[0086] S1. Preprocess the collected industrial data. Industrial data refers to historical data collected at industrial sites, which may include historical operating data for waste incinerators. Data preprocessing includes, but is not limited to, removing outliers, filling in missing values, and performing standardization or normalization.

[0087] S2. Model training for the operating parameter prediction model. Leverage historical operating data to build a data-driven AI (artificial intelligence) model (also known as the operating parameter prediction model).

[0088] S3. Model training of reinforcement learning model. The reinforcement learning model can be trained offline using historical operation data.

[0089] S4. Model release. After the model is released, the Industrial Brain Open Platform can use the model for reasoning.

[0090] S5. The Industrial Brain Open Platform performs online optimization. First, the Industrial Brain Open Platform performs online prediction of operating parameters. Specifically, the Industrial Brain Open Platform inputs real-time operating data into the operating parameter prediction model to predict the predicted values ​​of the key operating parameters of the waste incinerator in the future time period. Next, the Industrial Brain Open Platform inputs the real-time operating data and the predicted values ​​of the key operating parameters into the reinforcement learning model to determine the target set values ​​of the key operating parameters of the waste incinerator in the future time period. Finally, the Industrial Brain Open Platform sends the target set values ​​of the key operating parameters to the combustion control system. In this way, the combustion control system controls the combustion of the waste incinerator in the future time period based on the target set values ​​of the key operating parameters.

[0091] The waste incineration process is complex due to its physical and chemical processes, external disturbances, and the interaction between internal system variables, making it difficult to achieve efficient real-time optimization using traditional mechanism-based models. However, the technical solution provided in the embodiments of this application combines an operating parameter prediction model with a reinforcement learning model to achieve real-time intelligent decision-making of key operating parameters during the operation of the waste incinerator based on real-time operating data. This effectively reduces the labor intensity of operators, enables staff streamlining, and enables more flexible adaptation to different operating conditions, improving overall operational efficiency, increasing energy generation per ton of waste, and overall economic benefits.

[0092] Figure 4 This is a schematic diagram of the structure of a combustion control device for a waste incinerator provided in an embodiment of the present application. The device can be composed of hardware and / or software and can generally be integrated into the Industrial Brain open platform. Figure 4 , the apparatus may include:

[0093] An acquisition module 41 is used to periodically acquire real-time operating data collected in real time during the combustion process of the waste incinerator;

[0094] An operating parameter prediction module 42 is used to input real-time operating data into an operating parameter prediction model to predict the predicted values ​​of key operating parameters of the waste incinerator in a future time period;

[0095] a reinforcement learning module 43 for inputting real-time operating data and predicted values ​​of key operating parameters into a reinforcement learning model to determine target set values ​​of the key operating parameters of the waste incinerator in a future time period;

[0096] The sending module 44 is used to send the target setting values ​​of the key operating parameters to the combustion control system, so that the combustion control system can control the combustion of the waste incinerator in the future time period according to the target setting values ​​of the key operating parameters.

[0097] Optionally, the real-time operation data includes at least one of the following: the number of pushes of the pusher in the recent time period, the grate action cycle, the number of grate actions, the operating frequency of the primary fan, the operating frequency of the secondary fan, the operating frequency of the induced draft fan, the primary air volume, the secondary air volume, the flue gas flow, the flue gas temperature, the cooling water flow, the wind chamber pressure, the steam flow, the steam temperature, the furnace temperature, the furnace negative pressure, the flue gas oxygen content, the environmental protection index data and the key operation data of the garbage storage system; the key operation data of the garbage storage system include: the coordinate position, weight, fermentation time, etc. of the garbage crane.

[0098] Optionally, the key operating parameters include at least one of the following: furnace temperature, furnace negative pressure, flue gas oxygen content, steam flow rate or steam temperature.

[0099] Optionally, the training method of the operating parameter prediction model includes: obtaining historical operating data of multiple different first time periods collected during the combustion process of the waste incinerator; constructing a first training data set and a first test data set based on the historical operating data of multiple different first time periods; with the goal of predicting the predicted values ​​of key operating parameters of the waste incinerator in the second time period, using the first training data set to train the model to obtain an initial operating parameter prediction model, the second time period being later than the first time period; using the first test data set to adjust the parameters of the operating parameter prediction model to obtain an operating parameter prediction model.

[0100] Optionally, the training method of the reinforcement learning model includes: obtaining historical operating data of multiple different third time periods collected during the combustion process of the waste incinerator; for any third time period, inputting the historical operating data of the third time period into the operating parameter prediction model to predict the predicted values ​​of the key operating parameters of the waste incinerator in the fourth time period, where the fourth time period is later than the third time period; constructing a second training data set and a second test data set based on the historical operating data of multiple different third time periods and the predicted values ​​of multiple key operating parameters in the fourth time period; using the second training data set to train the model with the target setting values ​​of the key operating parameters of the waste incinerator in the fourth time period as the goal to obtain an initial reinforcement learning model; using the second test data set to adjust the parameters of the initial reinforcement learning model to obtain a reinforcement learning model.

[0101] Optionally, the second training data set includes multiple current state data, and the current state data includes: historical operating data of the third time period and predicted values ​​of key operating parameters in the fourth time period; accordingly, the second training data set is used to perform model training to obtain an initial reinforcement learning model, including: inputting the current state data into the initial reinforcement learning model to obtain the set values ​​of the key operating parameters of the waste incinerator in the fourth time period decided by the reinforcement learning model; sending the set values ​​of the key operating parameters in the fourth time period to the combustion control system, so that the combustion control system controls the combustion of the waste incinerator in the fourth time period according to the set values ​​of the key operating parameters; after performing combustion control based on the current state data, using the reward function to determine the reward value, and adjusting the model parameters of the initial reinforcement learning model according to the reward value; repeating the above steps until an initial reinforcement learning model that meets the requirements is obtained.

[0102] Optionally, the relevant factors of the reward function include one or more of the following: combustion efficiency, pollutant emissions, safety limits, and electricity generated per ton of garbage online; wherein, electricity generated per ton of garbage online is converted based on the amount of steam used for power generation, fan power, and garbage feed amount.

[0103] The detailed implementation and beneficial effects of each module in the device of this embodiment have been described in detail in the aforementioned embodiments and will not be elaborated here.

[0104] It should be noted that the execution entity of each step of the method provided in the above embodiment can be the same device, or the method can be executed by different devices.

[0105] An embodiment of the present application also provides a combustion control method for a waste incinerator, comprising: periodically acquiring real-time operating data collected in real time during the combustion process of the waste incinerator; inputting the real-time operating data into an operating parameter prediction model to predict predicted values ​​of key operating parameters of the waste incinerator in a future time period; inputting the real-time operating data and the predicted values ​​of the key operating parameters into a reinforcement learning model to determine target setting values ​​of the key operating parameters of the waste incinerator in the future time period; and performing combustion control on the waste incinerator in the future time period according to the target setting values ​​of the key operating parameters.

[0106] The detailed implementation and beneficial effects of each step in the method of this embodiment can be found in the aforementioned embodiments and will not be elaborated here.

[0107] There is no restriction on the execution entity of the combustion control method of the waste incinerator provided in the embodiment of the present application, such as a combustion control system, an industrial brain open platform or a waste incinerator system with data processing capabilities, etc.

[0108] In addition, in some of the processes described in the above embodiments and the accompanying drawings, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or may be executed in parallel. The serial numbers of the operations, such as 201, 202, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0109] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 5 As shown, the electronic device includes: a memory 51 and a processor 52;

[0110] The memory 51 is used to store computer programs and can be configured to store various other data to support operations on the computing platform. Examples of such data include instructions for any application or method operating on the computing platform, data structures, contact data, phone book data, messages, images, videos, etc.

[0111] The processor 52 is coupled to the memory 51 and is used to execute the computer program in the memory 51 to execute the steps in the combustion control method of the waste incinerator.

[0112] Optional, such as Figure 5 As shown, the electronic device also includes: a communication component 53, a display 54, a power component 55, an audio component 56 and other components. Figure 5 Only some components are shown schematically, which does not mean that the electronic device only includes Figure 5 In addition, Figure 5 The components in the dotted box are optional components, not mandatory components, and the specific components may depend on the product form of the electronic device. The electronic device of this embodiment can be implemented as a terminal device such as a desktop computer, a laptop computer, a smart phone or an IOT (Internet of Things) device, or a server device such as a conventional server, a cloud server or a server array. If the electronic device of this embodiment is implemented as a terminal device such as a desktop computer, a laptop computer, a smart phone, etc., it may include Figure 5 If the electronic device of this embodiment is implemented as a conventional server, cloud server or server array and other server-side devices, it may not include Figure 5 Components within the dotted box.

[0113] The above-mentioned memory can be implemented by any type of volatile or non-volatile memory device or a combination thereof, such as static random-access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0114] The communication component is configured to facilitate wired or wireless communication between the device containing the communication component and other devices. The device containing the communication component can access a wireless network based on a communication standard, such as 2G (2nd Generation), 3G (3rd Generation), 4G (4th Generation) / LTE (Long Term Evolution), 5G (5th Generation), or other mobile communication networks, or a combination thereof. In an exemplary embodiment, the communication component receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel.

[0115] The display includes a screen, which may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, it may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, slides, and gestures on the touch panel. The touch sensors can detect not only the boundaries of a touch or slide action, but also the duration and pressure associated with the touch or slide operation.

[0116] The power supply assembly provides power to various components of the device in which the power supply assembly is located. The power supply assembly may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device in which the power supply assembly is located.

[0117] The above-mentioned audio component can be configured to output and / or input audio signals. For example, the audio component includes a microphone (MIC). When the device where the audio component is located is in an operating mode, such as call mode, recording mode, and voice recognition mode, the microphone is configured to receive external audio signals. The received audio signal can be further stored in the memory or sent via the communication component. In some embodiments, the audio component also includes a speaker for outputting audio signals.

[0118] Accordingly, embodiments of the present application further provide a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the processor is enabled to implement the steps of the above-described method embodiments. The computer-readable storage medium may be volatile, non-volatile, or a combination thereof, and may be removable or non-removable. Examples of computer-readable storage media include, but are not limited to, phase-change random access memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, disk storage or other magnetic storage devices, or any other non-transmission media.

[0119] Accordingly, the present application embodiment also provides a computer program product, which includes a computer program or instructions. When the computer program or instructions are executed by a processor, the processor is enabled to implement the steps in the above-mentioned method embodiment. It should be understood that each process or a combination of multiple processes in the above-mentioned method flow can be implemented by a computer program or instruction. In addition, these computer programs or instructions can be applied to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device, so that the processor of the general-purpose computer, the special-purpose computer, the embedded processor or other programmable data processing device can be implemented as a device for implementing the corresponding functions in the above-mentioned method embodiment.

[0120] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0121] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A combustion control method for a waste incinerator, characterized in that: Applied to the Industrial Brain open platform, the method includes: Periodically obtain real-time operating data collected during the combustion process of the waste incinerator; The real-time operating data is input into an operating parameter prediction model, so that the operating parameter prediction model predicts initial predicted values ​​of key operating parameters of the waste incinerator in different operating modes in future time periods, and the operating parameter prediction model is obtained by combining operating parameter prediction models applicable to different operating modes, and the different operating modes include a high-yield mode with the goal of maximizing steam volume, a load priority mode with priority control of steam stability, a furnace temperature priority mode with priority control of furnace temperature stability, and an environmental protection index control priority mode; the initial predicted values ​​of key operating parameters of the waste incinerator in future time periods under different operating modes are weighted and summed to obtain intermediate predicted values ​​of key operating parameters of the waste incinerator in future time periods, wherein weight information of different operating modes is allocated according to actual needs; with the goal of meeting the process requirements of the combustion process, the intermediate predicted values ​​of key operating parameters of the waste incinerator in future time periods are corrected to obtain predicted values ​​of key operating parameters of the waste incinerator in future time periods; the process requirements of the combustion process constrain the temperature range of the furnace temperature, the pressure range of the wind chamber wind pressure, the pressure range of the furnace negative pressure, and the intervals corresponding to the emission values ​​of various pollutants; Inputting the real-time operating data and the predicted values ​​of the key operating parameters into a reinforcement learning model to determine target set values ​​of the key operating parameters of the waste incinerator in a future time period; The target setting value of the key operating parameter is sent to a combustion control system, so that the combustion control system performs combustion control on the waste incinerator in the future time period according to the target setting value of the key operating parameter.

2. The method according to claim 1, characterized in that The real-time operation data includes at least one of the following: the number of pushers in the most recent time period, the grate action cycle, the number of grate actions, the operating frequency of the primary fan, the operating frequency of the secondary fan, the operating frequency of the induced draft fan, the primary air volume, the secondary air volume, the flue gas flow, the flue gas temperature, the attemperating water flow, the wind chamber pressure, the steam flow, the steam temperature, the furnace temperature, the furnace negative pressure, the flue gas oxygen content, environmental protection index data, and key operation data of the garbage storage system; the key operation data of the garbage storage system includes at least one of the following: the coordinate position, weight, and fermentation time of the garbage crane; The key operating parameters include at least one of the following: furnace temperature, furnace negative pressure, flue gas oxygen content, steam flow rate, and steam temperature.

3. The method according to claim 1, characterized in that The training method of the operating parameter prediction model includes: Acquire historical operation data of a plurality of different first time periods collected during the combustion process of the waste incinerator; Constructing a first training data set and a first test data set based on historical operation data of multiple different first time periods; The first training data set is used to perform model training with the goal of predicting the predicted values ​​of key operating parameters of the waste incinerator in a second time period, where the second time period is later than the first time period, to obtain an initial operating parameter prediction model; The operating parameter prediction model is adjusted using the first test data set to obtain the operating parameter prediction model.

4. The method according to claim 1, wherein The training method of the reinforcement learning model includes: Acquiring historical operation data of a plurality of different third time periods collected during the combustion process of the waste incinerator; For any third time period, inputting the historical operating data of the third time period into an operating parameter prediction model to predict the predicted value of the key operating parameter of the waste incinerator in a fourth time period, wherein the fourth time period is later than the third time period; constructing a second training data set and a second test data set based on a plurality of historical operating data in different third time periods and a plurality of predicted values ​​of key operating parameters in a fourth time period; Taking the target setting values ​​of the key operating parameters of the waste incinerator in the fourth time period as the goal, model training is performed using the second training data set to obtain an initial reinforcement learning model; The initial reinforcement learning model is subjected to parameter adjustment processing using the second test data set to obtain the reinforcement learning model.

5. The method according to claim 4, characterized in that The second training data set includes a plurality of current state data, wherein the current state data includes: historical operating data of the third time period and predicted values ​​of key operating parameters in a fourth time period; Accordingly, the second training data set is used to perform model training to obtain an initial reinforcement learning model, including: Inputting the current state data into the initial reinforcement learning model to obtain set values ​​of key operating parameters of the waste incinerator in the fourth time period determined by the reinforcement learning model; sending the set values ​​of the key operating parameters in the fourth time period to the combustion control system, so that the combustion control system performs combustion control on the waste incinerator in the fourth time period according to the set values ​​of the key operating parameters; After performing combustion control based on the current state data, a reward value is determined using a reward function, and model parameters of the initial reinforcement learning model are adjusted according to the reward value; the above steps are repeated until an initial reinforcement learning model that meets the requirements is obtained.

6. The method according to claim 5, characterized in that The relevant factors of the reward function include one or more of the following: combustion efficiency, pollutant emissions, safety limits, and online electricity per ton of garbage; wherein, online electricity per ton of garbage is converted based on the amount of steam used for power generation, fan power, and garbage input amount.

7. A waste incineration system, characterized in that: include: Industrial Brain open platform, combustion control system and waste incinerator; The industrial brain open platform is used to periodically obtain real-time operating data of the waste incinerator collected in real time during the combustion process of the waste incinerator; The real-time operating data is input into an operating parameter prediction model, so that the operating parameter prediction model predicts the initial predicted values ​​of the key operating parameters of the waste incinerator in different operating modes in the future time period. The operating parameter prediction model is obtained by combining operating parameter prediction models suitable for different operating modes. The different operating modes include a high-yield mode with the goal of maximizing steam volume, a load priority mode with priority control of steam stability, a furnace temperature priority mode with priority control of furnace temperature stability, and an environmental protection index control priority mode. The initial predicted values ​​of the key operating parameters of the waste incinerator in the future time period under different operating modes are weighted and summed to obtain intermediate predicted values ​​of the key operating parameters of the waste incinerator in the future time period. In the embodiment, weight information of different operation modes is allocated according to actual needs; with the goal of meeting the process requirements of the combustion process, the intermediate predicted values ​​of the key operation parameters of the waste incinerator in the future time period are corrected to obtain the predicted values ​​of the key operation parameters of the waste incinerator in the future time period; the process requirements of the combustion process constrain the temperature range of the furnace temperature, the pressure range of the wind chamber wind pressure, the pressure range of the furnace negative pressure, and the intervals corresponding to the emission values ​​of various pollutants; the real-time operation data and the predicted values ​​of the key operation parameters are input into the reinforcement learning model to determine the target setting values ​​of the key operation parameters of the waste incinerator in the future time period; the target setting values ​​of the key operation parameters are sent to the combustion control system; The combustion control system is used to control the combustion of the waste incinerator in the future time period according to the target setting value of the key operating parameter.

8. An electronic device, characterized in that: include: memory and processor; The memory is used to store computer programs; The processor is coupled to the memory and configured to execute the computer program to perform the steps of the method according to claims 1-6.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the processor is enabled to implement the steps of the method according to claims 1 to 6.

10. A computer program product, characterized in that The computer program / instructions include computer programs / instructions which, when executed by a processor, enable the processor to implement the steps of the method according to claims 1 to 6.

Images