Energy-saving control system and method for waste heat boiler and desulfurization fan

Abstract

The present invention relates to the technical field of waste heat utilization and tail gas treatment in industrial production lines, and specifically to an energy-saving control system and method for a waste heat boiler and a desulfurization fan. The energy-saving control system for the waste heat boiler and the desulfurization fan comprises a data acquisition module, a data processing module, a model training module and a control module. The data acquisition module acquires data, the data processing module preprocesses the acquired data, the model training module develops a prediction model and an analysis model to generate an energy-saving adjustment strategy, and the control module realizes remote monitoring and energy-saving adjustment of the waste heat boiler and the desulfurization fan based on the adjustment scheme predicted by the training module, thereby solving the problem of existing energy-saving control methods for waste heat boilers and desulfurization fans, in which the operation of the fan needs to rely on manual experience to control the corresponding equipment, so errors may occur in the control data due to operational errors, making it difficult to achieve refined and dynamic management, leading to energy waste.

Description

Technical Field

[0001] The present invention relates to the technical field of waste heat utilization and tail gas treatment in industrial production lines, and in particular to an energy-saving control system and method for a waste heat boiler and a desulfurization fan. Background Art

[0003] In industrial production, many production lines use equipment such as boilers and kilns. These equipment will generate a large amount of waste heat and exhaust gas during operation. In order to effectively utilize the waste heat and treat the exhaust gas, waste heat boilers and desulfurization devices are often installed, and induced draft fans are required to assist this process.

[0004] The prior art has an energy-saving control method for waste heat boilers and desulfurization fans. The operation of the fan needs to rely on manual experience to control the corresponding equipment. Therefore, errors in the control data may occur due to operational errors, making it difficult to achieve refined and dynamic management, resulting in energy waste. Summary of the invention

[0006] The purpose of the present invention is to provide an energy-saving control system and method for a waste heat boiler and a desulfurization fan, aiming to solve the problem of energy-saving control methods for existing waste heat boilers and desulfurization fans. The operation of the fan needs to rely on manual experience to control the corresponding equipment. Therefore, errors in the control data may occur due to operational errors, making it difficult to achieve refined and dynamic management, resulting in energy waste.

[0007] To achieve the above-mentioned purpose, in a first aspect, the present invention provides an energy-saving control system for a waste heat boiler and a desulfurization fan, comprising a data acquisition module, a data processing module, a model training module and a control module; The data processing module is connected to the data acquisition module; the model training module is connected to the data processing module; the control module is connected to the model training module; The data acquisition module is used to access the data ports of the kiln central control system and the waste heat boiler central control system based on the data acquisition interface and protocol; The data processing module is used to clean, convert and store the collected raw data to ensure the availability of the data; The model training module is used to develop prediction models and analysis models to generate optimal energy-saving adjustment strategies based on processed data and corresponding machine learning algorithms; The control module is used to remotely monitor and perform energy-saving regulation on the waste heat boiler and the desulfurization fan based on the energy-saving regulation strategy generated by the model training module.

[0008] Wherein, the data acquisition module includes a data docking submodule, a data definition submodule and a data acquisition submodule, the data definition submodule is connected to the data docking submodule; the data acquisition submodule is connected to the data definition submodule; The data docking submodule is used to connect with the data interfaces opened by the kiln central control system and the waste heat boiler central control system, and obtain the data required by the system PLC through the OPC service.

[0009] The data definition submodule is used to define the interface address, request mode, frequency and parameters based on the OPC UA software; The data acquisition submodule is used to read pressure data according to the output mode of the pressure sensor.

[0010] Wherein, the pressure sensor adopts a silicon resonant pressure sensor.

[0011] Wherein, the data processing module includes a data cleaning submodule, a data conversion submodule and a data storage submodule, the data cleaning submodule is connected to the data acquisition module; the data conversion submodule is connected to the data cleaning submodule; the data storage submodule is connected to the data conversion submodule; The data cleaning submodule is used to remove duplicate data, correct data errors, process missing data values, and then verify the data to process abnormal values; The data conversion submodule is used to convert the processed data into a corresponding data format; The data storage submodule is used to store the processed data based on MySQL as a data storage database.

[0012] Wherein, the model training module includes a model creation submodule, a model development submodule and a data evaluation submodule, the model creation submodule is connected to the data processing module; the model development submodule is connected to the model creation submodule; the data evaluation submodule is connected to the model development submodule; The model creation submodule is used to select the corresponding development model platform; The model development submodule is used to import pre-processed data in the MySQL database based on a pre-selected development platform to train and develop the model; The data evaluation submodule is used to evaluate the model based on indicators such as accuracy, recall, F1 score, ROC curve, etc. of the calculation model.

[0013] Wherein, the model training module also includes a data prediction submodule, and the data prediction submodule is connected to the data evaluation submodule; The data prediction submodule is used to use the trained model to predict the new fan frequency and each section pressure and other data when different inverter frequencies are set for two fans.

[0014] Wherein, the control module includes a real-time control submodule and a user control submodule, the real-time control submodule is connected to the model training module; the user control submodule is connected to the real-time control submodule; The real-time control submodule is used to generate a frequency adjustment suggestion for the frequency converter according to the prediction result generated by the training model, and to control the frequency of the frequency converter; The user control submodule is used to provide a remote operation interface for operators to remotely monitor each segment of real-time data and adjust the fan frequency.

[0015] In a second aspect, the present invention further provides an energy-saving control method for a waste heat boiler and a desulfurization fan, comprising: The data acquisition module is used to access the data ports of the kiln central control system and the waste heat boiler central control system based on the data acquisition interface and protocol; The collected raw data is cleaned, converted and stored through the data processing module to ensure the availability of the collected data; Through the model training module, based on the processed data and the corresponding machine learning algorithm, the prediction model and analysis model are developed to generate the optimal energy-saving adjustment strategy; The control module is used to remotely monitor and energy-savingly adjust the waste heat boiler and the desulfurization fan based on the energy-saving adjustment strategy generated by the model training module.

[0016] The present invention discloses an energy-saving control system for a waste heat boiler and a desulfurization fan, wherein the data acquisition module is used to access the data ports of the kiln central control system and the waste heat boiler central control system based on the data acquisition interface and protocol, the data processing module cleans, converts and stores the collected raw data to ensure the availability of the collected data, the model training module develops a prediction model and an analysis model based on the processed data and the corresponding machine learning algorithm to generate an optimal energy-saving adjustment strategy, and the control module is used to remotely monitor and energy-savingly adjust the waste heat boiler and the desulfurization fan based on the energy-saving adjustment strategy generated by the model training module, thereby solving the existing energy-saving control method for waste heat boilers and desulfurization fans. The operation of the fan needs to rely on manual experience to control the corresponding equipment, so errors may occur in the control data due to operational errors, making it difficult to achieve refined and dynamic management, leading to energy waste. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art are briefly introduced below.

[0019] Figure 1 It is a schematic diagram of the overall structure of an energy-saving control system for a waste heat boiler and a desulfurization fan of the present invention.

[0020] Figure 2 It is a structural schematic diagram of a data acquisition module of an energy-saving control system for a waste heat boiler and a desulfurization fan of the present invention.

[0021] Figure 3 It is a structural schematic diagram of a data processing module of an energy-saving control system for a waste heat boiler and a desulfurization fan of the present invention.

[0022] Figure 4 It is a structural schematic diagram of a model training module of an energy-saving control system for a waste heat boiler and a desulfurization fan of the present invention.

[0023] Figure 5 It is a structural schematic diagram of a control module of an energy-saving control system for a waste heat boiler and a desulfurization fan of the present invention.

[0024] Figure 6 The present invention is a flow chart of an energy-saving control method for a waste heat boiler and a desulfurization fan.

[0025] 101-data acquisition module, 102-data processing module, 103-model training module, 104-control module, 105-data docking submodule, 106-data definition submodule, 107-data acquisition submodule, 108-data cleaning submodule, 109-data conversion submodule, 110-data storage submodule, 111-model creation submodule, 112-model development submodule, 113-data evaluation submodule, 114-data prediction submodule, 115-real-time control submodule, 116-user control submodule. DETAILED DESCRIPTION

[0027] Embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, but should not be construed as limiting the present invention.

[0028] In the first aspect, the present invention provides an energy-saving control system for a waste heat boiler and a desulfurization fan, see Figure 1-Figure 5 ,in, Figure 1 It is a schematic diagram of the overall structure of an energy-saving control system for a waste heat boiler and a desulfurization fan of the present invention. Figure 2 It is a structural schematic diagram of a data acquisition module of an energy-saving control system for a waste heat boiler and a desulfurization fan of the present invention. Figure 3It is a structural schematic diagram of a data processing module of an energy-saving control system for a waste heat boiler and a desulfurization fan of the present invention. Figure 4 It is a structural schematic diagram of a model training module of an energy-saving control system for a waste heat boiler and a desulfurization fan of the present invention, Figure 5 It is a structural schematic diagram of a control module of an energy-saving control system for a waste heat boiler and a desulfurization fan of the present invention.

[0029] An energy-saving control system for a waste heat boiler and a desulfurization fan of the present invention comprises a data acquisition module 101, a data processing module 102, a model training module 103 and a control module 104, wherein the data processing module 102 is connected to the data acquisition module 101; the model training module 103 is connected to the data processing module 102; the control module 104 is connected to the model training module 103; The data acquisition module 101 is used to access the data ports of the kiln central control system and the waste heat boiler central control system based on the data acquisition interface and protocol; The data processing module 102 is used to clean, convert and store the collected raw data to ensure the availability of the data; The model training module 103 is used to develop a prediction model and an analysis model to generate an optimal energy-saving adjustment strategy based on the processed data and the corresponding machine learning algorithm; The control module 104 is used to remotely monitor and perform energy-saving regulation on the waste heat boiler and the desulfurization fan based on the energy-saving regulation strategy generated by the model training module.

[0030] In this embodiment, the data acquisition module 101 accesses the data of the kiln central control system and the waste heat boiler central control system through the data acquisition interface and protocol. The data processing module 102 cleans, converts and stores the collected raw data to ensure the availability of the collected data. The model training module 103 develops prediction models and analysis models based on the processed data and the corresponding machine learning algorithm to generate the optimal energy-saving adjustment strategy. The control module 104 realizes remote monitoring and energy-saving adjustment of the waste heat boiler and the desulfurization fan based on the optimal adjustment scheme predicted by the training module, thereby solving the existing energy-saving control method of the waste heat boiler and the desulfurization fan. The operation of the fan needs to rely on manual experience to control the corresponding equipment. Therefore, errors may occur in the control data due to operational errors, making it difficult to achieve refined and dynamic management, resulting in energy waste.

[0031] Further, the data acquisition module 101 includes a data docking submodule 105, a data definition submodule 106 and a data acquisition submodule 107, wherein the data definition submodule 106 is connected to the data docking submodule 105; the data acquisition submodule 107 is connected to the data definition submodule 106; The data docking submodule 105 is used to connect with the data interfaces opened by the kiln central control system and the waste heat boiler central control system, and obtain the data required by the system PLC through the OPC service.

[0032] The data definition submodule 106 is used to define the interface address, request mode, frequency and parameters based on the OPC UA software; The data acquisition submodule 107 is used to read pressure data according to the output mode of the pressure sensor.

[0033] In this embodiment, the data docking submodule 105 is connected to the data interface opened by the kiln central control system and the waste heat boiler central control system, and the data required by the system PLC is obtained through the OPC service. The data definition submodule 106 predefines the required data blocks in the PLC, creates the OPC service, configures the OPC and associates the data in the OPC. The interface address, request method, frequency, parameters, etc. are defined by the OPC UA software in the industrial computer, and the connection is established by wired mode to collect data. The data acquisition submodule 107 reads the pressure data by wired mode according to the output mode of the pressure sensor, digital signal / analog signal.

[0034] Further, the data processing module 102 includes a data cleaning submodule 108, a data conversion submodule 109 and a data storage submodule 110, wherein the data cleaning submodule 108 is connected to the data acquisition module 101; the data conversion submodule 109 is connected to the data cleaning submodule 108; the data storage submodule 110 is connected to the data conversion submodule 109; The data cleaning submodule 108 is used to remove duplicate data, correct data errors, process missing data values, and then verify the data to process abnormal values; The data conversion submodule 109 is used to convert the processed data into a corresponding data format; The data storage submodule 110 is used to store the processed data based on MySQL as a data storage database.

[0035] In this embodiment, the data cleaning submodule 108 is based on Python as a development language, and designs and implements data cleaning algorithms including denoising, deduplication, filling missing values, etc., designs and implements data conversion algorithms, supports conversion between multiple data formats, and designs and implements data conversion algorithms through the data conversion submodule 109, supports conversion between multiple data formats, and implements data format conversion of the processed data, and converts it into corresponding data encoding. The data storage submodule 110 is used to store the cleaned and converted data based on the MySQL database. In Python, the pandas library is used for data cleaning: the drop_duplicates() method is used to remove duplicate data, and the fillna() method is used to fill missing values. The pandas library is used to convert the data format: the to_csv() method is used to save the data as a CSV file, and the to_json() method is used to convert the data into JSON format. MySQL is selected as the database for data storage, and the cleaned and converted data is stored in the MySQL database. Pymysql is used for database operations: the database table structure is designed, and the corresponding table is created, and the cleaned and converted data is inserted into the table, and then the index is established and the storage parameters are adjusted to optimize the database. The dataset is divided into a training set (70%) and a test set (30%).

[0036] Further, the model training module 103 includes a model creation submodule 111, a model development submodule 112, a data evaluation submodule 113 and a data prediction submodule 114, wherein the model creation submodule 111 is connected to the data processing module 102; the model development submodule 112 is connected to the model creation submodule 111; the data evaluation submodule 113 is connected to the model development submodule 112, and the data prediction submodule 114 is connected to the data evaluation submodule 113; The model creation submodule 111 is used to select a corresponding development model platform; The model development submodule 112 is used to import pre-processed data in the MySQL database based on a pre-selected development platform to train and develop the model; The data evaluation submodule 113 is used to evaluate the model based on the accuracy, recall, F1 score, ROC curve and other indicators of the calculation model; The data prediction submodule 114 is used to use the trained model to predict the new fan frequency and each section pressure and other data when two fans have different inverter frequency settings.

[0037] In this embodiment, the model creation submodule 111 is based on Python as a development platform, and the cleaned and converted data in the MySQL database is imported through the model development submodule 112. Scikit-learn is used to implement linear regression, logistic regression, decision tree and other algorithms or deep learning models for model training and development. Then, the data evaluation submodule 113 uses recurrent neural network (RNN) and reinforcement learning model to train and analyze the machine learning model, and evaluates the module by calculating the accuracy, recall rate, F1 score, ROC curve and other indicators of the model. Then, the data prediction submodule 114 predicts the pressure of each section when the frequency of the two fans is set to different inverters. Directly call the linear regression, logistic regression, decision tree and other models or deep learning models in sklearn: recurrent neural network (RNN), reinforcement learning model to train the above data set (training set). After the model is trained, the test set is used for prediction, and then the accuracy, ROC, F1 and other results are calculated and evaluated and the evaluation results are printed. According to the results, a suitable model is selected, and the trained model is used to predict the new fan frequency and each section pressure data. The model is retrained regularly with new real-time operation data to ensure its accuracy.

[0038] Furthermore, the control module 104 includes a real-time control submodule 115 and a user control submodule 116, wherein the real-time control submodule 115 is connected to the model training module 103; the user control submodule 116 is connected to the real-time control submodule 115; The real-time control submodule 115 is used to generate a frequency adjustment suggestion for the frequency converter according to the prediction result generated by the training model, and control the frequency of the frequency converter; The user control submodule 116 is used to provide a remote operation interface for operators to remotely monitor each segment of real-time data and adjust the fan frequency.

[0039] In this embodiment, the real-time control submodule 115 is used to generate frequency adjustment suggestions for the inverter according to the model analysis results, select a suitable communication protocol to adjust the frequency of the inverter, and feedback the adjustment results. The selected model is exported to a deployable format (such as ONNX, TensorFlow SavedModel), and the model is integrated into the real-time control module 104 through API or directly embedded in the model. The real-time data is processed and stored as described above. The above module is used to delineate the acceptable pressure range of the kiln and waste heat boiler, and then the predicted value of the pressure under each frequency is calculated by the model, and the lowest frequency in the result is found by the search algorithm. PROFINET with the characteristics of high efficiency and real-time performance is used as a communication method for data interaction with the inverter to realize real-time control of the inverter; data interaction with the host computer is carried out through the OPC UA service to realize data collection and data control of the control model. The user control submodule 116 is used to provide an operation interface for operators to remotely monitor each segment of real-time data and adjust the fan frequency.

[0040] When using the energy-saving control system of the waste heat boiler and the desulfurization fan described in the present invention, the data docking submodule 105 is connected to the data interface opened by the kiln central control system and the waste heat boiler central control system, and the data required by the system PLC is obtained through the OPC service. The data definition submodule 106 pre-defines the required data blocks in the PLC, creates the OPC service, configures the OPC and associates the data in the OPC. In the industrial computer, the OPCUA software is used to define the interface address, request method, frequency, parameters, etc., and a connection is established through a wired method to collect data. The data acquisition submodule 107 reads the pressure data through a wired method according to the output mode of the pressure sensor, digital signal / analog signal. The data cleaning submodule 108 is based on Python as the development language, and designs and implements data cleaning algorithms including denoising, deduplication, filling missing values, etc., designs and implements data conversion algorithms, and supports conversion between multiple data formats. The data conversion submodule 109 is used to design and implement data conversion algorithms, and supports conversion between multiple data formats. The processed data is converted into data format and converted into corresponding data encoding. The data storage submodule 110 is used to store the cleaned and converted data based on the MySQL database. In Python, the pandas library is used for data cleaning: the drop_duplicates() method is used to remove duplicate data, and the fillna() method is used to fill missing values. The pandas library is used to convert the data format: the to_csv() method is used to save the data as a CSV file, and the to_json() method is used to convert the data into JSON format. Select MySQL as the database for data storage, store the cleaned and converted data in the MySQL database, use pymysql to perform database operations: design the database table structure, create the corresponding table, insert the cleaned and converted data into the table, and then create indexes and adjust storage parameters to optimize the database. Divide the data set into a training set (70%) and a test set (30%). The model creation submodule 111 is based on Python as a development platform. The cleaned and converted data in the MySQL database is imported through the model development submodule 112. Scikit-learn is used to implement linear regression, logistic regression, decision tree and other algorithms or deep learning models for model training and development. Then, the data evaluation submodule 113 uses a recurrent neural network (RNN) and a reinforcement learning model to train and analyze the machine learning model. The module is evaluated by calculating the model's accuracy, recall rate, F1 score, ROC curve and other indicators. Then, the data prediction submodule 114 predicts the pressure of each section when the frequency of the two fans is set to different inverters.Directly call the linear regression, logistic regression, decision tree and other models or deep learning models in sklearn: recurrent neural network (RNN), reinforcement learning model to train the above data set (training set). After the model is trained, use the test set for prediction, and then use accuracy, ROC, F1, etc. to calculate and evaluate the results and print the evaluation results. Select a suitable model based on the results, and use the trained model to predict the new fan frequency and each section pressure and other data. The new real-time operation data regularly retrains the model to ensure the accuracy of the model. The real-time control submodule 115 is used to generate frequency adjustment suggestions for the inverter according to the model analysis results, select a suitable communication protocol to adjust the frequency of the inverter, and feedback the adjustment results. The selected model is exported to a deployable format (such as ONNX, TensorFlow SavedModel), and the model is integrated into the real-time control module 104 through API or directly embedded in the model. The real-time data is processed and stored as above. The above module is used to delineate the acceptable pressure range of the kiln and waste heat boiler, and then the predicted value of the pressure at each frequency is calculated by the model, and the lowest frequency in the result is found using the search algorithm. PROFINET, which has the characteristics of high efficiency and real-time performance, is used as a communication method for data interaction with the frequency converter to achieve real-time control of the frequency converter; data interaction with the host computer is carried out through the OPC UA service to realize data collection and data control of the control model. The user control submodule 116 is used to provide an operation interface for operators to remotely monitor each segment of real-time data and adjust the fan frequency, so as to solve the existing energy-saving control method of waste heat boiler and desulfurization fan. The operation of the fan needs to rely on manual experience to control the corresponding equipment. Therefore, errors may occur in the control data due to operational errors, making it difficult to achieve refined and dynamic management, resulting in energy waste.

[0041] In the second aspect, the present invention also provides an energy-saving control method for a waste heat boiler and a desulfurization fan, see Figure 6 ,in, Figure 6 It is a flow chart of an energy-saving control method for a waste heat boiler and a desulfurization fan of the present invention, and the energy-saving control method for the waste heat boiler and the desulfurization fan includes: S1 is used to access the data ports of the kiln central control system and the waste heat boiler central control system through the data acquisition module 101 based on the data acquisition interface and protocol; In the embodiment of the present invention, the data docking submodule 105 is connected to the data interface opened by the kiln central control system and the waste heat boiler central control system, and the data required by the system PLC is obtained through the OPC service. The data definition submodule 106 predefines the required data blocks in the PLC, creates the OPC service, configures the OPC and associates the data in the OPC. The interface address, request mode, frequency, parameters, etc. are defined by the OPC UA software in the industrial computer, and the connection is established by wired mode to collect data. The data acquisition submodule 107 reads the pressure data by wired mode according to the output mode of the pressure sensor, digital signal / analog signal.

[0042] S2 cleans, converts and stores the collected raw data through the data processing module 102 to ensure the availability of the collected data; In the embodiment of the present invention, the data cleaning submodule 108 is based on Python as a development language, and designs and implements data cleaning algorithms including denoising, deduplication, filling missing values, etc., designs and implements data conversion algorithms, supports conversion between multiple data formats, and designs and implements data conversion algorithms through the data conversion submodule 109, supports conversion between multiple data formats, and implements data format conversion of the processed data, and converts it into corresponding data encoding. The data storage submodule 110 is used to store the cleaned and converted data based on the MySQL database. In Python, the pandas library is used for data cleaning: the drop_duplicates() method is used to remove duplicate data, and the fillna() method is used to fill missing values. The pandas library is used to convert the data format: the to_csv() method is used to save the data as a CSV file, and the to_json() method is used to convert the data into JSON format. MySQL is selected as the database for data storage, and the cleaned and converted data is stored in the MySQL database. Pymysql is used for database operations: the database table structure is designed, and the corresponding table is created, and the cleaned and converted data is inserted into the table, and then the index is established and the storage parameters are adjusted to optimize the database. Divide the data set into training set (70%) and test set (30%) S3 develops a prediction model and an analysis model to generate an optimal energy-saving adjustment strategy based on the processed data and the corresponding machine learning algorithm through the model training module 103; In the embodiment of the present invention, the model creation submodule 111 is based on Python as a development platform, and the cleaned and converted data in the MySQL database is imported through the model development submodule 112. Scikit-learn is used to implement linear regression, logistic regression, decision tree and other algorithms or deep learning models for model training and development. Then, the data evaluation submodule 113 uses a recurrent neural network (RNN) and a reinforcement learning model to train and analyze the machine learning model, and the module is evaluated by calculating the accuracy, recall rate, F1 score, ROC curve and other indicators of the model. Then, the data prediction submodule 114 predicts the pressure of each section when the frequency of the two fans is set to different inverters. Directly call the linear regression, logistic regression, decision tree and other models or deep learning models in sklearn: recurrent neural network (RNN), reinforcement learning model to train the above data set (training set). After the model is trained, the test set is used for prediction, and then the result is calculated and evaluated using accuracy, ROC, F1, etc. and the evaluation result is printed. According to the result, a suitable model is selected, and the trained model is used to predict the new fan frequency and each section pressure data. The model is retrained regularly with new real-time operation data to ensure its accuracy.

[0043] S4 is used to remotely monitor and adjust the waste heat boiler and the desulfurization fan for energy saving based on the energy saving adjustment strategy generated by the model training module through the control module 104; In an embodiment of the present invention, the real-time control submodule 115 is used to generate a frequency adjustment suggestion for the frequency converter according to the model analysis results, select a suitable communication protocol to adjust the frequency of the frequency converter, and feedback the adjustment result. The selected model is exported to a deployable format (such as ONNX, TensorFlow SavedModel), and the model is integrated into the real-time control module 104 through an API or directly embedded in the model. The real-time data is processed and stored as described above. The above module is used to delineate the acceptable pressure range of the kiln and the waste heat boiler, and then the predicted value of the pressure under each frequency is calculated by the model, and the lowest frequency in the result is found by a search algorithm. PROFINET with the characteristics of high efficiency and real-time performance is used as a communication method for data interaction with the frequency converter to realize real-time control of the frequency converter; data interaction with the host computer is carried out through the OPC UA service to realize data collection and data control of the control model. The user control submodule 116 is used to provide an operation interface for operators to remotely monitor each segment of real-time data and adjust the fan frequency.

[0044] The present invention provides an energy-saving control method for a waste heat boiler and a desulfurization fan, wherein the data acquisition module 101 is used to access the data ports of the kiln central control system and the waste heat boiler central control system based on the data acquisition interface and protocol, the data processing module 102 cleans, converts and stores the collected raw data to ensure the availability of the collected data, the model training module 103 develops a prediction model and an analysis model based on the processed data and the corresponding machine learning algorithm to generate an optimal energy-saving adjustment strategy, and the control module 104 is used to remotely monitor and energy-savingly adjust the waste heat boiler and the desulfurization fan based on the energy-saving adjustment strategy generated by the model training module, thereby solving the existing energy-saving control method for waste heat boilers and desulfurization fans. The operation of the fan needs to rely on manual experience to control the corresponding equipment, so errors may occur in the control data due to operational errors, making it difficult to achieve refined and dynamic management, leading to energy waste.

[0045] What is disclosed above is only one or more preferred embodiments of the present application, and cannot be used to limit the scope of rights of the present application. Ordinary technicians in this field can understand that all or part of the processes of implementing the above embodiments and equivalent changes made according to the claims of the present application are still within the scope covered by the present application.

Claims

1. An energy-saving control system for a waste heat boiler and a desulfurization fan, characterized in that: It includes data acquisition module, data processing module, model training module and control module; The data processing module is connected to the data acquisition module; the model training module is connected to the data processing module; the control module is connected to the model training module; The data acquisition module is used to access the data ports of the kiln central control system and the waste heat boiler central control system based on the data acquisition interface and protocol; The data processing module is used to clean, convert and store the collected raw data to ensure the availability of the data; The model training module is used to develop prediction models and analysis models to generate energy-saving adjustment strategies based on processed data and corresponding machine learning algorithms; The control module is used to remotely monitor and perform energy-saving regulation on the waste heat boiler and the desulfurization fan based on the energy-saving regulation strategy generated by the model training module.

2. The energy-saving control system for waste heat boiler and desulfurization fan according to claim 1, characterized in that: The data acquisition module includes a data docking submodule, a data definition submodule and a data acquisition submodule, wherein the data definition submodule is connected to the data docking submodule; the data acquisition submodule is connected to the data definition submodule; The data docking submodule is used to connect with the data interfaces opened by the kiln central control system and the waste heat boiler central control system, and obtain the data required by the system PLC through the OPC service; The data definition submodule is used to define the interface address, request mode, frequency and parameters based on the OPC UA software; The data acquisition submodule is used to read pressure data according to the output mode of the pressure sensor.

3. The energy-saving control system for waste heat boiler and desulfurization fan according to claim 2, characterized in that: The pressure sensor adopts a silicon resonant pressure sensor.

4. The energy-saving control system for waste heat boiler and desulfurization fan according to claim 3, characterized in that: The data processing module includes a data cleaning submodule, a data conversion submodule and a data storage submodule, wherein the data cleaning submodule is connected to the data acquisition module; the data conversion submodule is connected to the data cleaning submodule; and the data storage submodule is connected to the data conversion submodule; The data cleaning submodule is used to remove duplicate data, correct data errors, process missing data values, and then verify the data to process abnormal values; The data conversion submodule is used to convert the processed data into a corresponding data format; The data storage submodule is used to store the processed data based on MySQL as a data storage database.

5. The energy-saving control system for waste heat boiler and desulfurization fan according to claim 4, characterized in that: The model training module includes a model creation submodule, a model development submodule and a data evaluation submodule, wherein the model creation submodule is connected to the data processing module; the model development submodule is connected to the model creation submodule; and the data evaluation submodule is connected to the model development submodule; The model creation submodule is used to select the corresponding development model platform; The model development submodule is used to import pre-processed data in the MySQL database based on a pre-selected development platform to train and develop the model; The data evaluation submodule is used to evaluate the model based on indicators such as accuracy, recall, F1 score, ROC curve, etc. of the calculation model.

6. The energy-saving control system for waste heat boiler and desulfurization fan according to claim 5, characterized in that: The model training module also includes a data prediction submodule, and the data prediction submodule is connected to the data evaluation submodule; The data prediction submodule is used to use the trained model to predict the new fan frequency and each section pressure and other data when different inverter frequencies are set for two fans.

7. The energy-saving control system for waste heat boiler and desulfurization fan according to claim 6, characterized in that: The control module includes a real-time control submodule and a user control submodule, wherein the real-time control submodule is connected to the model training module; and the user control submodule is connected to the real-time control submodule; The real-time control submodule is used to generate a frequency adjustment suggestion for the frequency converter according to the prediction result generated by the training model, and to control the frequency of the frequency converter; The user control submodule is used to provide a remote operation interface for operators to remotely monitor each segment of real-time data and adjust the fan frequency.

8. An energy-saving control method for a waste heat boiler and a desulfurization fan, applied to an energy-saving control system for a waste heat boiler and a desulfurization fan according to any one of claims 1 to 7, characterized in that: include: The data acquisition module is used to access the data ports of the kiln central control system and the waste heat boiler central control system based on the data acquisition interface and protocol; The collected raw data is cleaned, converted and stored through the data processing module to ensure the availability of the collected data; Through the model training module, based on the processed data and the corresponding machine learning algorithm, the prediction model and analysis model are developed to generate the optimal energy-saving adjustment strategy; The control module is used to remotely monitor and energy-savingly adjust the waste heat boiler and the desulfurization fan based on the energy-saving adjustment strategy generated by the model training module.

Images