Energy efficiency control method and system for shaft furnace combustion

Abstract

The invention relates to the technical field of shaft furnace combustion control, in particular to a shaft furnace combustion energy efficiency control method and system.The method comprises the steps that combustion voiceprint features are extracted by obtaining combustion noise of a shaft furnace cavity, the combustion state is recognized, a graded early warning system is arranged according to the combustion state, and a combustion early warning result is obtained; according to the invention, a combustion early warning result is established, a fuel flow value is dynamically adjusted according to the combustion early warning result, and the combustion early warning result is dynamically adjusted by establishing a self-adaptive waste heat adjusting center, so that waste heat distribution is carried out, and the problems that the fine change of the combustion state is difficult to capture in time and abnormal combustion early signals are easy to miss in a traditional monitoring mode are effectively solved; the combustion noise is collected, the combustion voiceprint features are extracted, the combustion state is dynamically monitored, accurate recognition is achieved, the energy efficiency of the combustion process is remarkably improved through the grading early warning system and dynamic adjustment of the fuel flow, and energy waste and equipment faults caused by combustion abnormity are reduced.

Description

Technical Field

[0001] The invention relates to the technical field of shaft furnace combustion control, and in particular to an energy efficiency control method and system for shaft furnace combustion. Background Art

[0002] With the continuous improvement of the demand for industrial automation and energy management, the combustion efficiency and energy utilization rate of the vertical furnace, as a common high-temperature industrial equipment, have an important impact on the overall production efficiency and energy consumption. The efficiency of vertical furnace combustion directly affects the energy cost and production stability. Therefore, how to achieve accurate and efficient energy management has become a key technical challenge in modern industrial production. At present, the combustion process of the vertical furnace mainly relies on traditional physical quantities such as temperature and pressure for monitoring and adjustment. However, these monitoring methods often cannot reflect the combustion status in real time and comprehensively, and the adjustment of energy efficiency is relatively lagging.

[0003] Although existing control technologies are able to perform combustion regulation to a certain extent through feedback of physical quantities, in actual applications, these methods often ignore the acoustic characteristics of the combustion process, resulting in the inability to accurately capture subtle changes in the combustion state, making it difficult to detect abnormal combustion phenomena in a timely manner under complex operating conditions, which in turn affects combustion efficiency and energy utilization.

[0004] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present disclosure, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art known to those skilled in the art. Summary of the invention

[0005] The present invention provides an energy efficiency control method and system for vertical furnace combustion, which can effectively solve the problems in the background technology.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is:

[0007] A method for controlling energy efficiency of vertical furnace combustion, the method comprising:

[0008] Collecting combustion noise from a vertical furnace cavity, extracting combustion sound pattern features according to the combustion noise, and identifying a combustion state according to the combustion sound pattern features;

[0009] A graded warning system is set according to the combustion state, the graded warning system performs abnormal management according to the recognition result of the combustion state, obtains the combustion warning result, and dynamically adjusts the fuel flow value according to the combustion warning result;

[0010] An adaptive waste heat regulation center is established to dynamically adjust the waste heat distribution strategy according to the combustion warning results.

[0011] Furthermore, extracting combustion soundprint features according to the combustion noise includes:

[0012] Using a filter to perform denoising on the combustion noise to obtain an audio preprocessing signal;

[0013] Performing time window segmentation on the audio preprocessing signal to obtain a number of time frames, and performing amplitude-frequency conversion on the time frames to obtain a combustion noise spectrum;

[0014] Using a filter bank to weight the combustion noise spectrum to obtain combustion noise spectrum characteristics;

[0015] The combustion noise spectrum feature is logarithmically transformed, and the transformed combustion noise spectrum feature is converted into a cepstrum domain to obtain a combustion soundprint feature.

[0016] Furthermore, identifying the combustion state according to the combustion soundprint feature includes:

[0017] Collecting historical combustion noise signals to construct a training sample set, wherein the historical combustion noise signals include normal combustion signals and abnormal combustion signals;

[0018] Determine the kernel function type and the penalty factor according to the signal characteristics of the historical combustion noise signal in combination with a hyperparameter tuning algorithm to obtain a pre-recognition model, wherein the kernel function is used to determine the boundary for distinguishing the combustion state, and the penalty factor is used to process the deviation during model training;

[0019] The pre-recognition model is trained using the training sample set to obtain a combustion state recognition model, and the combustion soundprint feature is recognized according to the combustion state to obtain the combustion state.

[0020] Furthermore, a hierarchical warning system is set according to the combustion state, including:

[0021] Outputting a combustion state recognition result according to the combustion state recognition model and calculating a confidence score;

[0022] Acquire historical combustion data, and set combustion trend change criteria according to the historical combustion data;

[0023] Assigning weights to the confidence score and the combustion trend change standard respectively, and obtaining a final warning value;

[0024] A graded warning is set according to the final warning value.

[0025] Furthermore, the confidence score is calculated, including:

[0026] Acquire combustion state differentiation conditions according to the historical combustion noise signal, and set an initial decision boundary;

[0027] Setting boundary constraints according to a support vector machine algorithm, optimizing the initial decision boundary using a mathematical method according to the boundary constraints, and obtaining an execution decision boundary;

[0028] The distance between each fuel noise signal point and the execution decision boundary is calculated, and the distance is represented as a confidence score of each fuel noise signal point.

[0029] Furthermore, the kernel function type and penalty factor are determined according to the signal characteristics of the historical combustion noise signal in combination with a hyperparameter tuning algorithm, including:

[0030] extracting combustion noise data features according to the historical combustion noise signal;

[0031] According to the characteristics of the combustion noise data, a kernel function candidate set is set, a training sample set is used to perform simulation training on several kernel functions in the kernel function candidate set, and a kernel function is selected based on the training results;

[0032] Initial penalty factor candidate values ​​are set according to the historical combustion noise signal, the initial penalty factor candidate values ​​are cross-combined using a grid search method, and simulation training is performed using the training sample set to determine the penalty factor.

[0033] Furthermore, the fuel flow value is dynamically adjusted according to the combustion warning result, including:

[0034] Recording the abnormal combustion time according to the combustion warning result, and obtaining a corresponding confidence score, and calculating and obtaining a thermal efficiency loss value according to the abnormal combustion time and the confidence score;

[0035] The thermal efficiency loss value is fed back to a fuel supply system, and the fuel supply system controls a fuel flow value based on the thermal efficiency loss value.

[0036] Furthermore, dynamically adjusting the waste heat distribution strategy according to the combustion warning result includes:

[0037] A joint control channel is established based on the hierarchical early warning system, and the combustion state of the vertical furnace cavity is monitored in real time to obtain real-time monitoring results;

[0038] When the combustion state is abnormal, the overflow heat is extracted according to the real-time monitoring result, and the waste heat distribution system is collaboratively controlled to extract the overflow heat;

[0039] When the combustion state is restored, the residual heat is scheduled based on a preset allocation strategy.

[0040] An energy efficiency control system for vertical furnace combustion, the system comprising:

[0041] A combustion noise state recognition module collects the combustion noise of the vertical furnace cavity, extracts the combustion sound pattern characteristics according to the combustion noise, and recognizes the combustion state according to the combustion sound pattern characteristics;

[0042] A graded warning adjustment module is used to set a graded warning system according to the combustion state, wherein the graded warning system performs abnormal management according to the recognition result of the combustion state, obtains the combustion warning result, and dynamically adjusts the fuel flow value according to the combustion warning result;

[0043] The residual heat regulation and allocation module establishes an adaptive residual heat regulation center to dynamically adjust the residual heat allocation strategy according to the combustion warning results.

[0044] Furthermore, the combustion noise state recognition module includes:

[0045] A combustion noise denoising processing unit, which uses a filter to denoise the combustion noise to obtain an audio preprocessing signal;

[0046] A combustion-noise spectrum acquisition unit is configured to divide the audio preprocessing signal into time windows to obtain a plurality of time frames, and to perform amplitude-frequency conversion on the plurality of time frames to obtain a combustion-noise spectrum;

[0047] A spectrum weighted feature extraction unit, which uses a filter bank to weight the fuel noise spectrum to obtain fuel noise spectrum features;

[0048] The combustion soundprint acquisition unit performs logarithmic transformation on the combustion noise spectrum feature, and performs cepstrum domain conversion on the transformed combustion noise spectrum feature to acquire the combustion soundprint feature.

[0049] The technical solution of the present invention can achieve the following technical effects:

[0050] The present invention effectively solves the problem that traditional monitoring methods are difficult to capture subtle changes in the combustion state in a timely manner and are prone to miss early signals of abnormal combustion. By collecting combustion noise and extracting combustion soundprint features, the combustion state is dynamically monitored and accurate identification is achieved. Through the hierarchical early warning system and dynamic adjustment of the fuel flow, the energy efficiency of the combustion process is significantly improved, and energy waste and equipment failures caused by abnormal combustion are reduced.

[0051] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1 A schematic flow chart of an energy efficiency control method for vertical furnace combustion;

[0054] Figure 2 A schematic diagram of the process of extracting combustion voiceprint features;

[0055] Figure 3 A schematic diagram of a flow chart set up for a hierarchical early warning system;

[0056] Figure 4 This is the relationship diagram of the combustion noise status identification module. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0058] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0059] Embodiment 1;

[0060] like Figure 1 As shown, the present application provides a method for controlling energy efficiency of vertical furnace combustion, the method comprising:

[0061] S10: collecting combustion noise from the vertical furnace cavity, extracting combustion sound pattern features according to the combustion noise, and identifying the combustion state according to the combustion sound pattern features;

[0062] S20: a hierarchical warning system is set according to the combustion state, the hierarchical warning system performs abnormal management according to the recognition result of the combustion state, obtains the combustion warning result, and dynamically adjusts the fuel flow value according to the combustion warning result;

[0063] S30: Establish an adaptive waste heat regulation center to dynamically adjust the waste heat distribution strategy according to the combustion warning results.

[0064] Specifically, a multi-channel noise collection device is first installed in the vertical furnace cavity. In some embodiments, a high-sensitivity piezoelectric sensor or microphone array is used. Such devices have good anti-interference performance and can work stably in high temperature and harsh industrial environments. At the same time, the noise collection device has a wide frequency response range. In this embodiment, 10 Hz to 10 kHz is used to capture the low-frequency and high-frequency noise characteristics in the vertical furnace combustion process. Then the collected combustion noise signal is subjected to noise reduction processing. On the basis of the denoised signal, a time-frequency analysis method, such as short-time Fourier transform, wavelet transform, etc., can be used to extract the combustion soundprint features to analyze the spectral characteristics of the signal. According to the extracted combustion soundprint features, a pattern recognition algorithm (such as a decision tree, a neural network, etc.) is used to identify the combustion state to determine whether it is in a normal combustion state or an abnormal combustion state. Then, a hierarchical early warning system is constructed according to the identification result of the combustion state, and the early warning result is obtained according to the hierarchical early warning system. The fuel flow rate is dynamically adjusted to ensure that the combustion process is in the optimal state. Then, an adaptive waste heat regulation center is established according to the combustion early warning result. The center can monitor the temperature, pressure and other data of the waste heat source in real time according to the real-time combustion state, and dynamically adjust the waste heat flow direction.

[0065] The technical solution of the present invention effectively solves the problem that traditional monitoring methods are difficult to capture subtle changes in the combustion state in time and are prone to miss early signals of abnormal combustion. By collecting combustion noise and extracting combustion soundprint features, the combustion state is dynamically monitored and accurate identification is achieved. Through the hierarchical early warning system and dynamic adjustment of the fuel flow, the energy efficiency of the combustion process is significantly improved, and energy waste and equipment failures caused by abnormal combustion are reduced.

[0066] Further, if Figure 2 As shown, the combustion soundprint features are extracted based on the combustion noise, including:

[0067] S11: using a filter to perform denoising on the combustion noise to obtain an audio preprocessing signal;

[0068] S12: dividing the audio preprocessing signal into time windows to obtain a number of time frames, and performing amplitude-frequency conversion on the time frames to obtain a combustion noise spectrum;

[0069] S13: weighting the combustion noise spectrum using a filter bank to obtain combustion noise spectrum features;

[0070] S14: performing logarithmic transformation on the combustion noise spectrum features, and performing cepstrum domain transformation on the transformed combustion noise spectrum features to obtain combustion soundprint features.

[0071] As a preferred embodiment of the above, the collected combustion noise is first denoised, and the denoising scheme is as follows: first, a high-pass filter (such as a band-pass filter) is used to set a suitable range for denoising. In some embodiments, the lower limit of the filter is set to 20 Hz and the upper limit is set to 10 kHz to retain only the frequency components related to the combustion process. Then, a band-stop filter is used to remove the factory environmental noise (such as mechanical equipment vibration, fan noise, etc.) in the signal to obtain a purer combustion noise signal. Then, the denoised signal is the audio preprocessing signal, which is saved as a digital signal. The audio preprocessing signal is then divided into time windows. In at least one embodiment, the window size is set to 50 ms to 100 ms, and the audio signal is segmented according to the window size. Each time window contains a time frame. During processing, the signal of each time frame will be subjected to a separate spectrum analysis, and then each time frame is converted into an amplitude-frequency conversion. The amplitude-frequency conversion scheme is as follows: the signal data in each time frame is prepared as an N-point discrete time series. Assuming that there are N data points in a time frame, for each The signal data of each time frame is converted into a frequency domain signal by fast Fourier transform. The basic operation of fast Fourier transform is to decompose the time domain signal into the sum of different frequency components, so as to obtain the amplitude of each frequency. After the fast Fourier transform, the amplitude of each frequency component corresponds to the intensity of the frequency component in the signal. For each frequency point, the amplitude is the modulus value of the complex number. Then, the intensity distribution of the signal at different frequencies is analyzed by calculating the spectrum amplitude value, and the spectrum is output. Then, the spectrum is weighted using a filter bank. For example, the high-frequency noise characteristics of combustion can be emphasized by weighting the high-frequency part, or the low-frequency noise characteristics can be emphasized by weighting the low-frequency part. After the weighting operation is completed, the combustion noise spectrum characteristics are obtained. In order to enhance the characteristics of the low-frequency signal, the combustion noise spectrum characteristics are logarithmically transformed, and the spectrum characteristic values ​​are logarithmized. Finally, the combustion noise spectrum characteristics after logarithmic transformation are converted into the cepstrum domain. Discrete cosine transform (DCT) or discrete Fourier transform (DFT) can be used to complete this conversion to obtain the combustion soundprint characteristics, which are used as input data for subsequent combustion state recognition.

[0072] Furthermore, the combustion state is identified according to the combustion soundprint characteristics, including:

[0073] Collect historical combustion noise signals and construct a training sample set. The historical combustion noise signals include normal combustion signals and abnormal combustion signals.

[0074] According to the signal characteristics of the historical combustion noise signal combined with the hyperparameter tuning algorithm, the kernel function type and penalty factor are determined to obtain the pre-identification model. The kernel function is used to determine the boundary of distinguishing the combustion state, and the penalty factor is used to deal with the deviation during model training.

[0075] The pre-recognition model is trained using a training sample set to obtain a combustion state recognition model, and the combustion soundprint features are recognized according to the combustion state to obtain the combustion state.

[0076] In this embodiment, firstly, the historical combustion noise signals are extracted according to the operation records of the shaft furnace to ensure that they include sufficient sample data of two types: normal combustion signals and abnormal combustion signals. The collected historical combustion noise signals are labeled according to their corresponding combustion states (normal or abnormal), and organized into a training sample set. Each sample should include the corresponding combustion soundprint feature and its corresponding combustion state label. Then, a machine learning algorithm is used to build a pre-recognition model. For example, a support vector machine or a random forest is used to determine the optimal hyperparameters of the pre-recognition model through hyperparameter tuning algorithms such as grid search and random search. The key hyperparameters include: kernel function type and penalty factor. The main function of the kernel function is to map the data to a higher-dimensional space so that the data that is inseparable in the low-dimensional space is The data becomes separable in high-dimensional space. Simply put, the kernel function adds additional features so that the data can be effectively separated by a simple decision boundary (such as a hyperplane) in a higher-dimensional space. The penalty factor is used to control the complexity of the model to avoid overfitting or underfitting. A smaller penalty factor allows more training errors, while a larger penalty factor tends to reduce errors but may cause overfitting. The extracted training sample set is then used to train the pre-recognition model. Once the pre-recognition model is trained, it can be applied to real-time monitoring. In practical applications, the collected combustion noise signal is preprocessed and feature extracted, and then input into the trained pre-recognition model. The model identifies the current combustion state based on the input combustion soundprint features.

[0077] Furthermore, if Figure 3 As shown, a graded warning system is set up according to the combustion status, including:

[0078] S21: outputting a combustion state recognition result according to the combustion state recognition model and calculating a confidence score;

[0079] S22: Obtain historical combustion data, and set combustion trend change standards according to the historical combustion data;

[0080] S23: assigning weights to the confidence score and the combustion trend change standard respectively, and obtaining a final warning value;

[0081] S24: Set a graded warning based on the final warning value.

[0082] Specifically, first, a combustion state recognition result is output according to the combustion recognition model. The result is often divided into two categories: "normal combustion" and "abnormal combustion". After each recognition result, the confidence score of the result is calculated to measure the accuracy of the recognition result. In some embodiments, a support vector machine or a probability regression method is used to calculate the confidence score. Then, based on the historical combustion data, a combustion trend change standard is set to judge the fluctuations in the combustion process. For example, the standard includes: the range of combustion temperature changes (such as combustion temperature fluctuations exceeding a set threshold indicate abnormalities), fuel flow rate change trends (such as a sudden increase or decrease in fuel flow rate exceeding a certain ratio, it is considered that there may be an abnormality). Then, according to the confidence score of the combustion state recognition result, a corresponding weight can be set. A recognition result with a higher confidence score can be given a higher weight, and a recognition result with a lower confidence score should have its influence reduced. The weight setting method can be adopted as follows: Confidence score>0.8, indicating that the model is not accurate in judging the combustion state. Often have confidence, at this time, the confidence score should have a greater impact on the final warning value, therefore, give a higher weight, such as 0.7, 0.5 < confidence score ≤ 0.8, the confidence score has a moderate impact on the warning value, the weight can be set to 0.5 for reference, confidence score ≤ 0.5, the weight is set to 0.3, according to the severity and urgency of the combustion trend change standard, also set the corresponding weight, combine the confidence score and the weight of the combustion trend change standard, calculate the final warning value, set the graded warning according to the final warning value, set the reference scheme as follows: warning value> 0.8, the system judges it as a serious abnormality, triggers an emergency warning, and needs to take immediate measures, such as adjusting the fuel flow, suspending operation or conducting equipment inspections, 0.5 < warning value ≤ 0.8, the system judges it as a moderate abnormality, triggers an intermediate warning, needs to strengthen monitoring and real-time adjustment, may require manual intervention, warning value ≤ 0.5, the system considers the state to be normal or slightly abnormal, triggers a low-level warning, continues monitoring, and conducts periodic inspections.

[0083] Specifically, the confidence score is calculated, including:

[0084] Obtain combustion state distinction conditions based on historical combustion noise signals and set initial decision boundaries;

[0085] Boundary constraints are set according to the support vector machine algorithm, and mathematical methods are used to optimize the initial decision boundary according to the boundary constraints to obtain the execution decision boundary;

[0086] The distance from each fuel noise signal point to the execution decision boundary is calculated, and the distance is expressed as the confidence score of each fuel noise signal point.

[0087] As a preferred embodiment of the above, the combustion state distinction condition is defined according to the historical combustion noise signal. For example, the performance of the noise signal in different combustion states in terms of frequency and amplitude can be analyzed to extract the features that can distinguish normal combustion from abnormal combustion, and set the initial decision boundary. The initial decision boundary is a dividing line that separates the normal combustion and abnormal combustion data points in the multidimensional feature space. Then, the support vector machine algorithm is used to train the support vector machine model according to the features of the historical combustion noise signal (such as spectrum, time domain features, etc.). The goal of the support vector machine is to find an optimal decision boundary to separate the samples of normal combustion and abnormal combustion; then, the boundary constraint condition is set according to the support vector machine: First, the support vector machine algorithm will find a line or a hyperplane in the multidimensional feature space to separate the data points of normal and abnormal combustion. In order to ensure the accuracy of the boundary, the goal of the support vector machine is to maximize the interval between the boundary and the data point, and to improve the robustness of the model by introducing a soft interval (that is, allowing a small amount of misclassification); according to the boundary constraints, the optimal execution decision boundary is obtained. After the initial decision boundary is optimized, the operator needs to calculate the distance from each combustion noise signal point to the decision boundary. This distance reflects the proximity of the signal point to the decision boundary. By calculating the distance from each signal point to the execution decision boundary and normalizing it, the confidence score of each signal point is obtained.

[0088] Furthermore, the kernel function type and penalty factor are determined based on the signal characteristics of the historical combustion noise signal and the hyperparameter tuning algorithm, including:

[0089] Extracting combustion noise data features based on historical combustion noise signals;

[0090] According to the characteristics of the combustion noise data, a kernel function candidate set is set, several kernel functions in the kernel function candidate set are simulated and trained respectively using a training sample set, and a kernel function is selected based on the training results;

[0091] The initial penalty factor candidate values ​​are set according to the historical combustion noise signal, the grid search method is used to cross-combine the initial penalty factor candidate values, and the training sample sets are used to perform simulation training to determine the penalty factors.

[0092] In this embodiment, firstly, a kernel function candidate set is set by extracting combustion noise data features, such as spectral features, time domain features, etc., including linear kernel functions, radial basis kernel functions, and polynomial kernel functions, and then a kernel function is selected according to the number and complexity of the combustion noise data features: a training sample set is used to perform simulation training on each kernel function in the kernel function candidate set, and its performance on the training set is evaluated: the training process can adopt a support vector machine algorithm, and under each kernel function, the support vector machine will learn how to segment normal combustion and abnormal combustion data through the optimal hyperplane, and then the performance of each kernel function is evaluated through cross-validation, and the cross-validation method is as follows: one subset is used to test the model, and the remaining K-1 subsets are used to train the model. Cross-validation usually performs K iterations, each time using a different subset as the validation set, and finally calculating the average evaluation result of all iterations; finally, a kernel function is selected based on the evaluation result; then, in some embodiments, the method for setting the penalty factor is as follows: first, select a number of initial candidate values ​​of the penalty factor, usually ranging from 0.01 to 100. For example, the candidate set of the penalty factor can be set to {0.1, 1, 10, 100}, and then a grid search method can be used to traverse multiple candidate values, and combine them in turn, and finally obtain the optimal penalty factor combined with cross-validation. This penalty factor can effectively balance the bias and variance of the model and improve the generalization ability of the support vector machine model.

[0093] Furthermore, the fuel flow value is dynamically adjusted according to the combustion warning result, including:

[0094] According to the combustion warning results, the abnormal combustion time is recorded and the corresponding confidence score is obtained. The thermal efficiency loss value is calculated based on the abnormal combustion time and the confidence score.

[0095] The thermal efficiency loss value is fed back to the fuel supply system, and the fuel supply system controls the fuel flow value based on the thermal efficiency loss value.

[0096] Specifically, first, when the system detects abnormal combustion, it records the duration of the abnormal combustion. The abnormal combustion time refers to the time period when the combustion state is judged to be abnormal. The specific operation is: when abnormal combustion is detected, the system automatically starts timing until the combustion state returns to normal. The system records this time period as the abnormal combustion time and stores the time value for subsequent calculations. Then, the confidence score is obtained according to the combustion state recognition model, and the thermal efficiency loss value is calculated. The thermal efficiency loss refers to the energy loss caused by the abnormal combustion state during the combustion process. Abnormal combustion state usually leads to incomplete combustion or energy waste, thereby reducing the overall thermal efficiency. The calculation formula is as follows: L heat =T ab ×S conf ×η,T ab It indicates the duration of abnormal combustion, Sconf represents the confidence score of abnormal combustion, η represents the confidence score of abnormal combustion, and then the thermal efficiency loss value is fed back to the fuel supply system. The fuel supply system will dynamically adjust the fuel flow rate according to the fed-back thermal efficiency loss value. If the thermal efficiency loss value is high, it means incomplete combustion or too much fuel. The system will reduce the fuel flow rate and adjust it to an appropriate value to optimize the combustion process.

[0097] Furthermore, the waste heat distribution strategy is dynamically adjusted according to the combustion warning results, including:

[0098] Establish a joint control channel based on the hierarchical early warning system, and monitor the combustion status of the vertical furnace cavity in real time to obtain real-time monitoring results;

[0099] When the combustion state is abnormal, the overflow heat is extracted according to the real-time monitoring results, and the waste heat distribution system is coordinated to export the overflow heat;

[0100] When the combustion state is restored, the waste heat is scheduled based on the preset allocation strategy.

[0101] As a preferred embodiment of the above, first, a joint control channel is established between the vertical furnace control system and the waste heat distribution system according to the hierarchical warning system. This channel is used to transmit combustion status information in real time so as to dynamically adjust the waste heat distribution strategy according to the combustion warning results. Then, the combustion status is monitored in real time by deploying multiple sensors (such as temperature sensors, pressure sensors, gas concentration sensors, etc.) in the vertical furnace cavity. The monitoring data will be transmitted to the combustion status recognition system in real time. The system determines whether the combustion status is normal by analyzing the real-time data, and outputs the combustion status warning result according to the preset threshold. If combustion abnormality occurs, the system will automatically trigger the waste heat adjustment process. When the system recognizes that the combustion status is abnormal (such as excessive temperature fluctuations, abnormal oxygen concentration, etc.), it is necessary to trigger the waste heat adjustment mechanism according to the degree of abnormality of the combustion status: when the combustion is incomplete or abnormal, excess heat will be generated, which is called overflow heat. The waste heat distribution system will automatically adjust the waste heat flow direction according to the size and type of the overflow heat. For example, the overflow heat can be directed to the waste heat boiler, air preheater or other energy recovery device to achieve heat recovery. When the combustion state returns to normal levels, the operator should ensure that the system returns to the normal waste heat distribution strategy in time. At this time, the waste heat distribution system will re-dispatch the heat energy flow according to the preset distribution strategy to ensure the efficient operation of the system after the combustion state is restored.

[0102] Embodiment 2:

[0103] Based on the same inventive concept as the energy efficiency control method for shaft furnace combustion in the aforementioned embodiment, the present invention further provides an energy efficiency control system for shaft furnace combustion, the system comprising:

[0104] The combustion noise state recognition module collects the combustion noise of the vertical furnace cavity, extracts the combustion sound pattern characteristics according to the combustion noise, and recognizes the combustion state according to the combustion sound pattern characteristics;

[0105] The graded warning adjustment module sets up a graded warning system according to the combustion state. The graded warning system performs abnormal management according to the recognition result of the combustion state, obtains the combustion warning result, and dynamically adjusts the fuel flow value according to the combustion warning result;

[0106] The waste heat regulation and allocation module establishes an adaptive waste heat regulation center to dynamically adjust the waste heat allocation strategy according to the combustion warning results.

[0107] The above adjustment system in the present invention can effectively implement the energy efficiency control method of vertical furnace combustion, and the technical effects that can be achieved are as described in the above embodiments, which will not be repeated here.

[0108] Further, if Figure 4 As shown, the combustion noise state recognition module includes:

[0109] The combustion noise denoising processing unit uses a filter to denoise the combustion noise and obtain an audio preprocessing signal;

[0110] The combustion-noise spectrum acquisition unit divides the audio preprocessing signal into time windows to obtain a number of time frames, and performs amplitude-frequency conversion on the time frames to obtain the combustion-noise spectrum;

[0111] The spectrum weighted feature extraction unit uses a filter bank to weight the fuel noise spectrum to obtain the fuel noise spectrum features;

[0112] The combustion soundprint acquisition unit performs logarithmic transformation on the combustion noise spectrum characteristics, and performs cepstrum domain transformation on the transformed combustion noise spectrum characteristics to obtain the combustion soundprint characteristics.

[0113] Similarly, the above-mentioned optimization schemes for the system can also respectively achieve the optimization effects corresponding to the method in Example 1, which will not be repeated here.

[0114] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application as defined therein, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method for controlling energy efficiency of vertical furnace combustion, characterized in that: The method comprises: Collecting combustion noise from a vertical furnace cavity, extracting combustion sound pattern features according to the combustion noise, and identifying a combustion state according to the combustion sound pattern features; A graded warning system is set according to the combustion state, the graded warning system performs abnormal management according to the recognition result of the combustion state, obtains the combustion warning result, and dynamically adjusts the fuel flow value according to the combustion warning result; An adaptive waste heat regulation center is established to dynamically adjust the waste heat distribution strategy according to the combustion warning results.

2. The energy efficiency control method for vertical furnace combustion according to claim 1, characterized in that: Extracting combustion soundprint features according to the combustion noise includes: Using a filter to perform denoising on the combustion noise to obtain an audio preprocessing signal; Performing time window segmentation on the audio preprocessing signal to obtain a number of time frames, and performing amplitude-frequency conversion on the time frames to obtain a combustion noise spectrum; Using a filter bank to weight the combustion noise spectrum to obtain combustion noise spectrum characteristics; The combustion noise spectrum feature is logarithmically transformed, and the transformed combustion noise spectrum feature is converted into a cepstrum domain to obtain a combustion soundprint feature.

3. The energy efficiency control method for vertical furnace combustion according to claim 2, characterized in that: Identifying the combustion state according to the combustion soundprint feature includes: Collecting historical combustion noise signals to construct a training sample set, wherein the historical combustion noise signals include normal combustion signals and abnormal combustion signals; Determine the kernel function type and the penalty factor according to the signal characteristics of the historical combustion noise signal in combination with a hyperparameter tuning algorithm to obtain a pre-recognition model, wherein the kernel function is used to determine the boundary for distinguishing the combustion state, and the penalty factor is used to process the deviation during model training; The pre-recognition model is trained using the training sample set to obtain a combustion state recognition model, and the combustion soundprint feature is recognized according to the combustion state to obtain the combustion state.

4. The energy efficiency control method for vertical furnace combustion according to claim 3, characterized in that: A hierarchical warning system is set up according to the combustion state, including: Outputting a combustion state recognition result according to the combustion state recognition model and calculating a confidence score; Acquire historical combustion data, and set combustion trend change criteria according to the historical combustion data; Assigning weights to the confidence score and the combustion trend change standard respectively, and obtaining a final warning value; A graded warning is set according to the final warning value.

5. The energy efficiency control method for vertical furnace combustion according to claim 4, characterized in that: Calculate confidence scores, including: Acquire combustion state differentiation conditions according to the historical combustion noise signal, and set an initial decision boundary; Setting boundary constraints according to a support vector machine algorithm, optimizing the initial decision boundary using a mathematical method according to the boundary constraints, and obtaining an execution decision boundary; The distance between each fuel noise signal point and the execution decision boundary is calculated, and the distance is represented as a confidence score of each fuel noise signal point.

6. The energy efficiency control method for vertical furnace combustion according to claim 3, characterized in that: The kernel function type and penalty factor are determined according to the signal characteristics of the historical combustion noise signal in combination with a hyperparameter tuning algorithm, including: extracting combustion noise data features according to the historical combustion noise signal; According to the characteristics of the combustion noise data, a kernel function candidate set is set, a training sample set is used to perform simulation training on several kernel functions in the kernel function candidate set, and a kernel function is selected based on the training results; Initial penalty factor candidate values ​​are set according to the historical combustion noise signal, the initial penalty factor candidate values ​​are cross-combined using a grid search method, and simulation training is performed using the training sample set to determine the penalty factor.

7. The energy efficiency control method for vertical furnace combustion according to claim 1, characterized in that: Dynamically adjust the fuel flow value according to the combustion warning results, including: Recording the abnormal combustion time according to the combustion warning result, and obtaining a corresponding confidence score, and calculating and obtaining a thermal efficiency loss value according to the abnormal combustion time and the confidence score; The thermal efficiency loss value is fed back to a fuel supply system, and the fuel supply system controls a fuel flow value based on the thermal efficiency loss value.

8. The energy efficiency control method for vertical furnace combustion according to claim 1, characterized in that: Dynamically adjusting the waste heat distribution strategy according to the combustion warning result includes: A joint control channel is established based on the hierarchical early warning system, and the combustion state of the vertical furnace cavity is monitored in real time to obtain real-time monitoring results; When the combustion state is abnormal, the overflow heat is extracted according to the real-time monitoring result, and the waste heat distribution system is collaboratively controlled to extract the overflow heat; When the combustion state is restored, the residual heat is scheduled based on a preset allocation strategy.

9. An energy efficiency control system for vertical furnace combustion, characterized in that: The system comprises: A combustion noise state recognition module collects the combustion noise of the vertical furnace cavity, extracts the combustion sound pattern characteristics according to the combustion noise, and recognizes the combustion state according to the combustion sound pattern characteristics; A graded warning adjustment module is used to set a graded warning system according to the combustion state, wherein the graded warning system performs abnormal management according to the recognition result of the combustion state, obtains the combustion warning result, and dynamically adjusts the fuel flow value according to the combustion warning result; The residual heat regulation and allocation module establishes an adaptive residual heat regulation center to dynamically adjust the residual heat allocation strategy according to the combustion warning results.

10. The energy efficiency control system for vertical furnace combustion according to claim 9, characterized in that: The combustion noise state recognition module comprises: A combustion noise denoising processing unit, which uses a filter to denoise the combustion noise to obtain an audio preprocessing signal; A combustion-noise spectrum acquisition unit is configured to divide the audio preprocessing signal into time windows to obtain a plurality of time frames, and to perform amplitude-frequency conversion on the plurality of time frames to obtain a combustion-noise spectrum; A spectrum weighted feature extraction unit, which uses a filter bank to weight the fuel noise spectrum to obtain fuel noise spectrum features; The combustion soundprint acquisition unit performs logarithmic transformation on the combustion noise spectrum feature, and performs cepstrum domain transformation on the transformed combustion noise spectrum feature to acquire the combustion soundprint feature.

Images