A multi-data fusion control method for a temperature, pressure, flow and humidity integrated monitoring instrument

Through the multi-data fusion control method of the temperature and pressure flow and humidity integrated monitor, the problem of insufficient flue gas monitoring parameters dispersion control and data utilization is solved, and the accuracy of flue gas emission monitoring and the reliability of result are improved.

CN118707888BActive Publication Date: 2024-11-08NANTONG RENYUAN ENERGY SAVING & ENVIRONMENTAL PROTECTION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411205297.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-11-08
Estimated Expiration
2044-08-30

AI Technical Summary

Technical Problem

The existing flue gas emission monitoring results in low monitoring accuracy and distortion of results due to insufficient parameter dispersion control and data utilization.

Method used

A temperature-pressure, flow-wet and wet integrated monitor is used for multi-parameter integrated monitoring, and through multi-source data fusion prediction, optimization control instructions are generated to improve monitoring accuracy and result reliability.

Benefits of technology

It improves the accuracy and reliability of flue gas emission monitoring, and achieves comprehensive reflection and dynamic control of flue gas status.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118707888B_ABST
    Figure CN118707888B_ABST
Patent Text Reader

Abstract

The present invention discloses a multi-data fusion control method for a temperature-pressure-flow-humidity integrated monitor, and relates to the field of intelligent control of monitors. The method comprises: deploying a temperature-pressure-flow-humidity integrated monitor in a target smoke exhaust duct; collecting data according to preset settings; performing prediction processing after pre-processing the collected results, obtaining data corresponding to the prediction window, performing verification and adjustment, and obtaining sub-item prediction values; inputting the sub-item prediction values ​​into the control space of the monitor, generating target control instructions, and executing parameter control. The present invention solves the technical problems in the prior art of low flue gas emission monitoring accuracy and distorted monitoring results due to decentralized control of flue gas monitoring parameters and insufficient data utilization. Through integrated monitoring of temperature-pressure-flow-humidity multi-parameters and multi-source data fusion prediction, and optimizing control based on the prediction results, the technical effect of improving the accuracy of flue gas emission monitoring and the reliability of monitoring results is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of intelligent control of monitoring instruments, and specifically to a multi-data fusion control method for a temperature-pressure-flow-humidity integrated monitoring instrument. Background Art

[0002] Flue gas emission monitoring is used to control the emission status of flue gas in real time and provide a basis for flue gas treatment. Existing flue gas monitoring usually uses thermometers, pressure gauges, anemometers, hygrometers and other instruments to measure the temperature, pressure, flow rate, humidity and other parameters of the flue gas, and then transmits the measured values ​​of each parameter to the monitoring center. However, in the decentralized multi-parameter monitoring and control mode, each monitoring instrument works independently, lacks correlation analysis between parameters, and cannot fully reflect the true state of the flue gas; at the same time, the monitoring data collected by each instrument has not been fully utilized, and it is difficult to dynamically control the monitoring instrument according to changes in the flue gas state. Therefore, the existing flue gas emission monitoring has low flue gas emission monitoring accuracy and distorted monitoring results due to the decentralized control of monitoring parameters and insufficient data utilization. Summary of the invention

[0003] Based on this, it is necessary to provide a multi-data fusion control method for a temperature-pressure-flow-humidity integrated monitoring instrument that can achieve the goal of improving the accuracy of flue gas emission monitoring and the reliability of monitoring results through integrated monitoring of temperature, pressure, flow and humidity multi-parameters and multi-source data fusion prediction, and optimizing control based on the prediction results in response to the above-mentioned technical problems.

[0004] The present application provides a multi-data fusion control method for a temperature-pressure-flow-humidity integrated monitor, the method comprising: determining a target smoke exhaust duct, deploying a temperature-pressure-flow-humidity integrated monitor in the target smoke exhaust duct, and connecting the temperature-pressure-flow-humidity integrated monitor to a flue gas emission monitoring system; collecting temperature, pressure, flow rate and humidity data of flue gas in the target smoke exhaust duct through the temperature-pressure-flow-humidity integrated monitor according to a preset, and transmitting the collected results to the flue gas emission monitoring system in real time; in the flue gas emission monitoring system, after pre-processing the collected results, performing prediction processing on the pre-processed collected results to obtain temperature, pressure, flow rate and humidity data of a prediction window; verifying and adjusting the temperature, pressure, flow rate and humidity data of the prediction window to obtain sub-item prediction values ​​of temperature, pressure, flow rate and humidity; inputting the sub-item prediction values ​​of temperature, pressure, flow rate and humidity into the monitor control space, obtaining control parameters of the target monitor, and generating target control instructions based on the control parameters of the target monitor, and sending them to the temperature-pressure-flow-humidity integrated monitor to perform parameter control.

[0005] The multi-data fusion control method of the above-mentioned temperature, pressure, flow and humidity integrated monitoring instrument solves the technical problems in the prior art caused by decentralized control of flue gas monitoring parameters and insufficient data utilization, resulting in low flue gas emission monitoring accuracy and distorted monitoring results. Through integrated monitoring of temperature, pressure, flow and humidity multi-parameters and multi-source data fusion prediction, and optimized control based on the prediction results, the technical effect of improving the accuracy of flue gas emission monitoring and the reliability of monitoring results is achieved.

[0006] 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

[0007] Figure 1 It is a flow chart of a multi-data fusion control method of a temperature-pressure-flow-humidity integrated monitoring instrument in one embodiment;

[0008] Figure 2 The present invention is a flow chart of construction of a monitor control space for a multi-data fusion control method of a temperature-pressure-flow-humidity integrated monitor in one embodiment. DETAILED DESCRIPTION

[0009] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0010] like Figure 1 As shown, the present application provides a multi-data fusion control method for a temperature-pressure-flow-humidity integrated monitoring instrument, the method comprising:

[0011] Determine a target smoke exhaust duct, deploy a temperature-pressure-flow-humidity integrated monitor in the target smoke exhaust duct, and connect the temperature-pressure-flow-humidity integrated monitor to a smoke emission monitoring system;

[0012] In this application, the temperature, pressure, flow and humidity integrated monitor is a measuring instrument including temperature, flue pressure, flow rate and humidity; wherein the flow rate is designed based on the differential pressure measurement principle of the traditional pitot tube; humidity is measured using the principle of humidity-sensitive capacitance to achieve temperature control and heating of the humidity probe, prevent condensation and protect the sensor. Online monitoring of gas flow rate, dynamic pressure, flue pressure, temperature and humidity of various boilers, furnace flues and mine exhaust ducts can be performed. Multi-data fusion control involves the process of fusing information from multiple data sources and making control decisions based on these fused data. In the control system, data fusion technology can improve the accuracy and reliability of decision-making, while enhancing the robustness and adaptability of the system. This application provides a multi-data fusion control method for a temperature, pressure, flow and humidity integrated monitor, which fuses data such as temperature, pressure, flow rate and humidity from different sensors or data sources to improve the accuracy and reliability of measurement and optimize the control strategy.

[0013] According to actual needs, the smoke exhaust duct that needs to be monitored for smoke emissions is selected as the target, and the target smoke exhaust duct is random; the temperature, pressure, flow and humidity integrated monitor is deployed in the target smoke exhaust duct, and the temperature, pressure, flow and humidity integrated monitor is a device used to monitor parameters such as temperature, pressure, flow rate and humidity in real time; use appropriate communication cables or wireless transmission methods to connect the temperature, pressure, flow and humidity integrated monitor with the smoke emission monitoring system. The smoke emission monitoring system is a system specially used to monitor smoke pollutants generated in activities such as industrial production and energy utilization. It mainly includes sensors, controllers and data processing software, which can monitor the composition and data of smoke in real time and accurately. Through the above method, the target smoke exhaust duct can be successfully determined, the temperature, pressure, flow and humidity integrated monitor can be deployed, and it can be connected to the smoke emission monitoring system. This will provide reliable data support for subsequent smoke emission monitoring and help to achieve accurate control and effective management of smoke emissions.

[0014] According to the preset, the temperature, pressure, flow rate and humidity data of the smoke in the target smoke exhaust duct are collected by the temperature, pressure, flow and humidity integrated monitor, and the collected results are transmitted to the smoke emission monitoring system in real time;

[0015] Preset refers to pre-set parameters, that is, the temperature, pressure, flow and humidity integrated monitor can collect data according to the advance requirements and frequency. The temperature, pressure, flow and humidity integrated monitor will start working according to the preset, and use its built-in sensors or measuring equipment to monitor the flue gas in the target smoke exhaust duct in real time. The sensor can accurately sense and measure the key parameters of the flue gas, such as temperature, pressure, flow rate and humidity. After the data collection is completed, the collected results are transmitted to the flue gas emission monitoring system in real time. Through the above method, the timeliness and accuracy of the data are ensured. Through real-time transmission, the flue gas emission monitoring system can obtain a variety of data from the temperature, pressure, flow and humidity integrated monitor at any time, providing comprehensive and accurate data support for subsequent flue gas emission analysis, control strategy formulation, etc. This not only helps to improve the accuracy and reliability of flue gas emission monitoring, but also helps to optimize the control strategy and improve the performance of the overall system.

[0016] In the flue gas emission monitoring system, after preprocessing the collected results, prediction processing is performed on the preprocessed collected results to obtain temperature, pressure, flow rate and humidity data of the prediction window;

[0017] In the flue gas emission monitoring system, the acquisition results are preprocessed to eliminate noise, outliers or inconsistencies in the data, and to convert the data into a form suitable for subsequent analysis, including removing abnormal or invalid data caused by sensor failure, transmission error or other reasons, converting the data into a unified format and unit for subsequent processing and analysis; reducing random fluctuations and noise in the data by filtering or averaging; and estimating and supplementing missing data points by interpolation and other methods. Prediction processing is to predict the flue gas emission data in the future period of time based on preprocessing, using historical data and algorithm models. It helps to discover potential problems in advance, optimize control strategies, and prepare countermeasures in advance. Obtain the temperature, pressure, flow rate and humidity data of the prediction window, and the prediction window refers to the time window for prediction set in actual needs, such as a few minutes, hours or longer time period. The preprocessed temperature, pressure, flow rate and humidity data are input into the joint predictor to obtain the temperature, pressure, flow rate and humidity data of the prediction window, and generate the temperature, pressure, flow rate and humidity data in the prediction window through prediction processing. It provides important reference for subsequent smoke emission analysis, anomaly detection, control strategy formulation, etc. Through the above method, by preprocessing the collected results and performing predictive processing, the smoke emission monitoring system can obtain more accurate and reliable prediction data, providing strong support for the company's decision-making and control.

[0018] Pre-built joint predictors;

[0019] Receive the preprocessed acquisition results and extract the preprocessed temperature, pressure, flow rate and humidity data;

[0020] The preprocessed temperature, pressure, flow rate and humidity data are input into a joint predictor to obtain the temperature, pressure, flow rate and humidity data of a prediction window.

[0021] Building a joint predictor includes collecting historical flue gas emission data, including parameters such as temperature, pressure, flow rate and humidity, and performing necessary cleaning and sorting. According to the characteristics of the data and the prediction target, feature extraction and selection are performed, and a suitable prediction model, such as a deep learning model, is selected. According to the characteristics and needs of the data, a joint predictor is constructed, which can simultaneously process multiple parameters such as temperature, pressure, flow rate and humidity. The joint predictor is trained using historical data, and the performance of the joint predictor is improved by adjusting model parameters and using optimization algorithms. During the operation of the flue gas emission monitoring system, the real-time collected flue gas emission data is received and preprocessed to ensure the quality of the preprocessed data is reliable and provide effective input for subsequent predictions. Key parameters such as temperature, pressure, flow rate and humidity are extracted from the preprocessed data as inputs to the joint predictor to generate prediction results. The extracted preprocessed temperature, pressure, flow rate and humidity data are input into the pre-built joint predictor, and the joint predictor will perform prediction calculations based on the input data and the trained model parameters. The joint predictor will output the temperature, pressure, flow rate and humidity data within the prediction window, providing strong support for the decision-making of the flue gas emission monitoring system. Through the above method, by pre-building the joint predictor and using the pre-processed temperature, pressure, flow rate and humidity data for prediction, accurate prediction and monitoring of flue gas emissions can be achieved.

[0022] The interactive smoke emission monitoring system obtains the historical temperature, historical pressure, historical flow rate and historical humidity data of the target smoke exhaust duct, and generates a temperature sample set, a pressure sample set, a flow rate sample set and a humidity sample set, wherein each sample has a sample timestamp;

[0023] Setting an input window and an output window, wherein the output window has the same size as the prediction window;

[0024] Based on the temperature sample set, the pressure sample set, the flow rate sample set and the humidity sample set, a temperature predictor is constructed, and an input window and an output window are combined, wherein the temperature predictor takes temperature, pressure, flow rate and humidity data as input and takes temperature data as output;

[0025] Based on the temperature sample set, the pressure sample set, the flow rate sample set and the humidity sample set, a pressure predictor is constructed, and an input window and an output window are combined, wherein the pressure predictor takes the temperature, pressure, flow rate and humidity data as input and takes the pressure data as output;

[0026] Based on the temperature sample set, the pressure sample set, the flow rate sample set and the humidity sample set, a flow rate predictor is constructed, and an input window and an output window are combined, wherein the flow rate predictor takes the temperature, pressure, flow rate and humidity data as input and takes the flow rate data as output;

[0027] Based on the temperature sample set, the pressure sample set, the flow rate sample set and the humidity sample set, a humidity predictor is constructed, and an input window and an output window are combined, wherein the humidity predictor takes the temperature, pressure, flow rate and humidity data as input and takes the humidity data as output;

[0028] The temperature predictor, the pressure predictor, the flow rate predictor and the humidity predictor are combined to obtain a joint predictor.

[0029] Interact with the flue gas emission monitoring system to obtain historical data of the target exhaust duct. The data includes historical temperature, historical pressure, historical flow rate and historical humidity. Each sample is timestamped, which is helpful for the subsequent analysis of the time series characteristics of the data; from the acquired historical data, extract samples of temperature, pressure, flow rate and humidity respectively, and generate corresponding sample sets, including temperature sample set, pressure sample set, flow rate sample set and humidity sample set. Each sample set contains data points with timestamps, which provides a rich data foundation for the subsequent construction of prediction models. In order to capture the dynamic changes and time dependence of the data, set the input window and output window. The input window is used to determine the range of historical data considered by the model when predicting, while the output window is the same size as the prediction window and is used to generate prediction results for the corresponding time period. Based on the temperature sample set, pressure sample set, flow rate sample set and humidity sample set, a temperature predictor, a pressure predictor, a flow rate predictor and a humidity predictor are constructed respectively. Specifically, in the temperature predictor, based on the temperature sample set, pressure sample set, flow rate sample set and humidity sample set, a temperature predictor is constructed. Combining the input window and the output window, the temperature predictor takes temperature, pressure, flow rate and humidity data as input and temperature data as output. First, according to the characteristics of the data and the requirements of the problem, a suitable prediction model is selected, and the selected model framework is used to construct a predictor with temperature, pressure, flow rate and humidity data as input and temperature data as output. Ensure that the model can handle multivariate inputs and output the temperature prediction value of the corresponding time window. The temperature sample set is divided into a training set, a validation set and a test set. The temperature predictor is trained using the training set. The model parameters are adjusted by an optimization algorithm to minimize the prediction error. The performance of the model is evaluated using the validation set. The model is tuned as needed, such as improving the model accuracy and changing the model parameters. The final performance of the model is evaluated using the test set to ensure that it can also show good prediction capabilities on unseen data. The accuracy rate is set. When the test set inputs the temperature predictor to meet the accuracy rate, the trained temperature predictor is obtained. Based on the above method, a temperature predictor, a pressure predictor, a flow rate predictor and a humidity predictor are constructed respectively, and the constructed temperature predictor, pressure predictor, flow rate predictor and humidity predictor are obtained. They are merged to form a joint predictor. This joint predictor can simultaneously predict four parameters: temperature, pressure, flow rate and humidity, which improves the efficiency and accuracy of the prediction. Through the above method, a joint predictor based on historical data is constructed for real-time prediction of the temperature, pressure, flow rate and humidity in the target exhaust pipe. This provides a foundation for the optimization of the subsequent flue gas emission monitoring and control system.

[0030] Verify and adjust the temperature, pressure, flow rate and humidity data of the prediction window to obtain the sub-item prediction values ​​of temperature, pressure, flow rate and humidity;

[0031] Verify the temperature, pressure, flow rate and humidity data within the prediction window. This can be achieved by comparing with the statistical characteristics of historical data, such as checking whether the predicted value falls within a reasonable range. If the predicted value is obviously abnormal or unreasonable, further adjustments are required. If there are abnormal values ​​in the prediction window, such as extremely high or low values, these values ​​may be caused by model prediction errors or data noise. For these abnormal values, different processing strategies can be used, such as smoothing using sliding average, median filtering, or outlier replacement based on statistical distribution. According to the trends and patterns of historical data, trend analysis and adjustments are performed on the data within the prediction window. For example, if historical data shows that the temperature or humidity has a clear upward or downward trend in a specific time period, the predicted value can be adjusted accordingly to better reflect this trend. After the above verification and adjustment steps, more accurate sub-item prediction values ​​of temperature, pressure, flow rate and humidity can be obtained. Sub-item prediction values ​​refer to the individual prediction results of specific parameters, such as temperature, pressure, flow rate and humidity, within a specific time period or prediction window. In the flue gas emission monitoring system, each parameter is an independent variable, and their respective changes and trends have an important impact on the operation and performance of the system. Therefore, it is necessary to make a separate prediction for each parameter to obtain its possible future value or state. These separate prediction results are the sub-item prediction values. Through the above method, the temperature, pressure, flow rate and humidity data within the prediction window can be effectively calibrated and adjusted, and more accurate sub-item prediction values ​​can be obtained, providing reliable data support for flue gas emission monitoring and control systems.

[0032] Align the temperature, pressure, flow rate and humidity data of the prediction window in time to form a prediction data matrix;

[0033] Calculating historical prediction accuracy of a temperature predictor, a pressure predictor, a flow rate predictor, and a humidity predictor, and determining the confidence of each predictor according to the historical prediction accuracy;

[0034] The predictor result with the highest confidence is selected as the benchmark, and the other predictor results in the prediction data matrix are verified and adjusted according to the constraints of temperature, pressure, flow rate and humidity to obtain the adjusted prediction data matrix;

[0035] According to the adjusted prediction data matrix, the item-by-item prediction values ​​of temperature, pressure, flow rate and humidity are obtained.

[0036] The temperature, pressure, flow rate and humidity data of the prediction window are aligned in time to form a prediction data matrix. The temperature, pressure, flow rate and humidity data within the prediction window are collected to ensure that all data points are aligned at the same time interval or time point. If the data points of some parameters are missing or out of sync, interpolation or resampling is required. The time-aligned data are arranged in columns to form a prediction data matrix. Each row of the matrix represents a time point and each column represents a parameter; review the historical data, use the same predictor to predict the past data, and compare the predicted results with the actual values. Calculate the prediction error to measure the performance of the predictor. Based on the historical prediction accuracy, assign a confidence value to each predictor. Confidence can be determined based on the inverse of the prediction error, ranking, or other suitable measures. Generally, the smaller the prediction error, the higher the confidence. The predictor result with the highest confidence is selected as the benchmark. This predictor performs best on the historical data, so its prediction results are more reliable; determine the constraints between them based on the physical properties and system requirements of temperature, pressure, flow rate and humidity. For example, there is a positive or negative correlation between the parameters. Based on the results of the benchmark predictor, the results of other predictors in the prediction data matrix are verified and adjusted. If the results of other predictors deviate significantly from the benchmark results and this deviation violates the constraints, these results need to be corrected. The verified and adjusted prediction data matrix contains more reliable and consistent prediction values. Through the above methods, the consistency, reliability and accuracy of the prediction data are ensured, and valuable sub-item prediction values ​​are provided for the flue gas emission monitoring system.

[0037] Interactive smoke emission monitoring system, which obtains historical temperature, historical pressure, historical flow rate and historical humidity data of the target smoke exhaust duct, and combines the prediction window to obtain multiple historical data matrices;

[0038] Calculate the matrix similarity between the prediction data matrix and multiple historical data matrices, and obtain the matrix data volume whose matrix similarity is greater than the preset similarity;

[0039] When the matrix data volume is greater than the preset data volume, the sub-item prediction values ​​of temperature, pressure, flow rate and humidity are obtained according to the prediction data matrix.

[0040] The interactive flue gas emission monitoring system sends a data request to the flue gas emission monitoring system, specifies the target exhaust pipe, and requests to obtain historical temperature, historical pressure, historical flow rate and historical humidity data. The historical data should cover a period of time, including multiple different time windows or time periods; for each historical time period, ensure that the temperature, pressure, flow rate and humidity data are aligned at the same time interval or time point, and for each historical time period, arrange the aligned data in columns to form multiple historical data matrices. The historical data matrix represents a data set within a specific time period. Select a suitable similarity measurement method to evaluate the similarity between the prediction data matrix and the historical data matrix. For the historical data matrix, calculate its similarity with the prediction data matrix. It involves comparing the values ​​of corresponding elements in the matrix and deriving a similarity score according to the selected similarity measurement method. A similarity threshold is preset to determine which historical data matrices are similar enough to the prediction data matrix, and the number of historical data matrices with a similarity greater than the preset similarity is counted, that is, the matrix data volume. If the matrix data volume is greater than the preset data volume, that is, there are enough historical data matrices similar to the prediction data matrix, then the prediction data is considered reliable. In this case, the sub-item forecast values ​​of temperature, pressure, flow rate and humidity can be directly extracted from the forecast data matrix. The sub-item forecast values ​​reflect the future changes of each parameter within the forecast window. Through the above method, the similarity between historical data and forecast data is used to verify the reliability of the forecast results, and accurate sub-item forecast values ​​are obtained accordingly.

[0041] The sub-item predicted values ​​of temperature, pressure, flow rate and humidity are input into the monitor control space to obtain the target monitor control parameters, and the target control instructions are generated based on the target monitor control parameters and sent to the temperature, pressure, flow and humidity integrated monitor to execute parameter control.

[0042] The sub-item prediction values ​​of temperature, pressure, flow rate and humidity are input into the monitor control space, which generally refers to a virtual or physical environment for receiving, processing and executing monitor control instructions. In this space, various parameters of the monitor can be set and adjusted to achieve specific monitoring and control goals; in the monitor control space, a set of control logic is predefined, and the control logic calculates the target monitor control parameters according to the input sub-item prediction values. According to the predefined control logic and the input sub-item prediction values, the target monitor control parameters are calculated, including set points, adjustment ranges, response times, etc. According to the communication protocol and control requirements of the temperature, pressure, flow and humidity integrated monitor, the format and content of the target control instructions are determined, and specific control instructions are generated based on the target monitor control parameters. The control instructions will guide the monitor on how to adjust its operating state to achieve the ideal state indicated by the prediction value. The generated target control instructions are sent to the monitor, and after receiving the instructions, the monitor performs the corresponding parameter control operation. At the same time, the execution of the instructions and the real-time status of the monitor can be monitored through the feedback mechanism. Through the above method and steps, the sub-item prediction values ​​can be effectively converted into specific control instructions, and the precise control of the temperature, pressure, flow and humidity integrated monitor can be realized. This helps to improve the operating efficiency and accuracy of the flue gas emission monitoring system and provide strong support for environmental protection and energy utilization.

[0043] Interactive smoke emission monitoring system to obtain historical temperature, pressure, flow rate and humidity data of the target smoke exhaust duct;

[0044] Determine a plurality of historical monitoring instrument working conditions according to the historical temperature, historical pressure, historical flow rate and historical humidity data, and obtain monitoring instrument control parameters of the plurality of historical monitoring instrument working conditions, wherein each historical monitoring instrument working condition corresponds to a set of temperature, pressure, flow rate and humidity data;

[0045] The plurality of historical monitoring instrument working conditions are used as input dimensions of the monitoring instrument control space, and the corresponding monitoring instrument control parameters are used as output dimensions of the monitoring instrument control space to form a monitoring instrument control space;

[0046] The sub-item prediction values ​​of temperature, pressure, flow rate and humidity are matched with the historical data in the monitor control space to obtain the historical data point closest to the sub-item prediction value, and the monitor control parameter corresponding to the historical data point is used as the target monitor control parameter.

[0047] By interacting with the flue gas emission monitoring system, the historical temperature, historical pressure, historical flow rate and historical humidity data of the target smoke exhaust duct are obtained. The data should cover a sufficiently long period of time to include a variety of different operating conditions. Based on the historical temperature, pressure, flow rate and humidity data obtained, multiple historical monitor conditions can be defined. The historical detector condition is a combination of a specific set of temperature, pressure, flow rate and humidity data. For each historical monitor condition, it is necessary to obtain the corresponding monitor control parameters, including the monitor's set point, adjustment range, response time, etc., to guide the operation of the monitor under specific conditions. Multiple historical monitor conditions are used as the input dimension of the monitor control space, that is, each condition represents a point in the space. The corresponding monitor control parameters are used as the output dimension. Through this method, a multidimensional monitor control space can be constructed, in which each point represents a specific working condition and its corresponding control parameter. The sub-item prediction values ​​of temperature, pressure, flow rate and humidity are input into the monitor control space, and the historical data points closest to these prediction values ​​are found. This is achieved by calculating the Euclidean distance between the prediction value and the historical data point. The historical data point closest to the sub-item prediction value can be used as the monitor control parameter corresponding to the point, and it is used as the target monitor control parameter. Through the above method, the control parameter setting of the monitor is optimized based on historical data and sub-item prediction values, and the performance and accuracy of the flue gas emission monitoring system are improved.

[0048] like Figure 2 As shown, the monitor control parameters of multiple historical monitor working conditions are evaluated to obtain the control effect evaluation value of each monitor control parameter;

[0049] According to the control effect evaluation value, a monitor control parameter having a control effect evaluation value greater than a preset evaluation threshold is selected from monitor control parameters of multiple historical monitor working conditions as a preferred monitor control parameter;

[0050] A monitoring instrument control space is constructed with multiple historical monitoring instrument operating conditions as input dimensions and the corresponding optimal monitoring instrument control parameters as output dimensions.

[0051] Determine the criteria for evaluating the control effect of the monitor control parameters, including the stability of the emission concentration, the system response speed, the energy consumption efficiency, etc. Collect the actual operation data related to each historical monitor working condition, including the real-time monitoring value of the emission, the system response time, the energy consumption data, etc. According to the evaluation criteria, calculate the control effect evaluation value corresponding to each monitor control parameter. This is achieved by comparing the difference between the actual operation data and the expected target value and calculating the error rate. Set a preset evaluation threshold of the control effect evaluation value to screen out the monitor control parameters with good control effect. According to the comparison between the control effect evaluation value and the preset evaluation threshold, select the parameters with evaluation values ​​greater than the threshold from the monitor control parameters of multiple historical monitor working conditions as the preferred monitor control parameters. Take multiple historical monitor working conditions as input dimensions, take the corresponding preferred monitor control parameters as output dimensions, and construct the monitor control space based on the input dimensions and output dimensions. The monitor control space will reflect the mapping relationship between the preferred monitor control parameters and the working conditions under the historical monitor working conditions. Through the above method, the control space of the monitor constructed in this way will be based on the evaluation results of the actual operating data, which can more accurately reflect the control requirements of the monitor under different working conditions and provide a more reliable basis for subsequent parameter matching and control instruction generation.

[0052] According to the actual output data of the temperature, pressure, flow and humidity integrated monitoring instrument, combined with the sub-item prediction values ​​of temperature, pressure, flow rate and humidity, the prediction accuracy of the prediction window is calculated;

[0053] A prediction accuracy accumulator is set. When the prediction accuracy is lower than a preset threshold, the prediction window is marked as a prediction inaccuracy window, and the number of prediction inaccuracy windows is recorded in the prediction accuracy accumulator. When the prediction accuracy is higher than or equal to the preset threshold, the current prediction window is marked as a prediction accurate window.

[0054] Determine whether the count value of the prediction accuracy accumulator reaches a preset number of times threshold. When the count value reaches the preset number of times threshold, update the joint predictor according to the actual output data corresponding to the prediction inaccuracy window, and clear the prediction accuracy accumulator at the same time;

[0055] When the count value does not reach the preset number threshold, the joint predictor is kept unchanged and the prediction process of the next prediction window is continued.

[0056] The actual output data of the temperature, pressure, flow and humidity integrated monitoring instrument in the current prediction window is obtained, including the actual measured values ​​of temperature, pressure, flow rate and humidity. The actual output data is compared with the corresponding sub-item prediction value; the prediction accuracy is calculated according to the comparison result. It is realized by calculating the error rate, root mean square error, mean absolute error or other suitable accuracy indicators between the predicted value and the actual value, which are set by the technicians in this field; a prediction accuracy accumulator is set, and a counter for accumulating the number of prediction inaccuracy windows is initialized, which is set by the staff, and the calculated prediction accuracy is compared with the preset threshold. If the prediction accuracy is lower than the preset threshold, the current prediction window is marked as a prediction inaccuracy window, and the number of prediction inaccuracy windows is recorded in the prediction accuracy accumulator. If the prediction accuracy is higher than or equal to the preset threshold, the current prediction window is marked as a prediction accurate window. Check whether the count value of the prediction accuracy accumulator reaches the preset number threshold, which is set by the staff. If the count value reaches the preset number threshold, it means that there are enough prediction inaccuracies and the joint predictor needs to be updated. According to the actual output data corresponding to the prediction inaccuracy window, the joint predictor is trained or adjusted to improve its prediction accuracy. After updating the joint predictor, the prediction accuracy accumulator is cleared to restart the counting. If the count value does not reach the preset number threshold, it means that the prediction accuracy is still within an acceptable range. At this time, the joint predictor is kept unchanged and the prediction processing of the next prediction window is continued. The above preset prediction accuracy threshold and number threshold need to be reasonably set according to the actual application scenario and system requirements to ensure that the predictor is not updated too frequently and that prediction inaccuracies can be discovered and corrected in time. Through the above method, the prediction performance of the temperature, pressure, flow and humidity integrated monitor can be effectively monitored and optimized, and the accuracy and reliability of the flue gas emission monitoring system can be improved.

[0057] In summary, the beneficial effects of this method include:

[0058] The present invention solves the technical problems in the prior art of low flue gas emission monitoring accuracy and distorted monitoring results due to decentralized control of flue gas monitoring parameters and insufficient data utilization. Through integrated monitoring of temperature, pressure, flow and humidity multiple parameters and multi-source data fusion prediction, and optimized control based on the prediction results, the technical effect of improving the accuracy of flue gas emission monitoring and the reliability of monitoring results is achieved.

[0059] The specific embodiments of the multi-data fusion control method of a temperature, pressure, flow and humidity integrated monitoring instrument can be found above and will not be described here. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0060] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0061] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application.

Claims

1. A multi-data fusion control method for a temperature, pressure, flow and humidity integrated monitoring instrument, characterized in that: The method comprises: Determine a target smoke exhaust duct, deploy a temperature-pressure-flow-humidity integrated monitor in the target smoke exhaust duct, and connect the temperature-pressure-flow-humidity integrated monitor to a smoke emission monitoring system; According to the preset, the temperature, pressure, flow rate and humidity data of the smoke in the target smoke exhaust duct are collected by the temperature, pressure, flow and humidity integrated monitor, and the collected results are transmitted to the smoke emission monitoring system in real time; In the flue gas emission monitoring system, after preprocessing the collected results, prediction processing is performed on the preprocessed collected results to obtain temperature, pressure, flow rate and humidity data of the prediction window; Verify and adjust the temperature, pressure, flow rate and humidity data of the prediction window to obtain the sub-item prediction values ​​of temperature, pressure, flow rate and humidity; The sub-item predicted values ​​of temperature, pressure, flow rate and humidity are input into the monitor control space to obtain the target monitor control parameters, and the target control instructions are generated based on the target monitor control parameters and sent to the temperature, pressure, flow and humidity integrated monitor to execute parameter control.

2. The method according to claim 1, characterized in that Perform prediction processing on the preprocessed collection results to obtain the temperature, pressure, flow rate and humidity data of the prediction window, including: Pre-built joint predictors; Receive the preprocessed acquisition results and extract the preprocessed temperature, pressure, flow rate and humidity data; The preprocessed temperature, pressure, flow rate and humidity data are input into a joint predictor to obtain the temperature, pressure, flow rate and humidity data of a prediction window.

3. The method according to claim 2, characterized in that Pre-built joint predictors, including: The interactive smoke emission monitoring system obtains the historical temperature, historical pressure, historical flow rate and historical humidity data of the target smoke exhaust duct, and generates a temperature sample set, a pressure sample set, a flow rate sample set and a humidity sample set, wherein each sample has a sample timestamp; Setting an input window and an output window, wherein the output window has the same size as the prediction window; Based on the temperature sample set, the pressure sample set, the flow rate sample set and the humidity sample set, a temperature predictor is constructed, and an input window and an output window are combined, wherein the temperature predictor takes temperature, pressure, flow rate and humidity data as input and takes temperature data as output; Based on the temperature sample set, the pressure sample set, the flow rate sample set and the humidity sample set, a pressure predictor is constructed, and an input window and an output window are combined, wherein the pressure predictor takes the temperature, pressure, flow rate and humidity data as input and takes the pressure data as output; Based on the temperature sample set, the pressure sample set, the flow rate sample set and the humidity sample set, a flow rate predictor is constructed, and an input window and an output window are combined, wherein the flow rate predictor takes the temperature, pressure, flow rate and humidity data as input and takes the flow rate data as output; Based on the temperature sample set, the pressure sample set, the flow rate sample set and the humidity sample set, a humidity predictor is constructed, and an input window and an output window are combined, wherein the humidity predictor takes the temperature, pressure, flow rate and humidity data as input and takes the humidity data as output; The temperature predictor, the pressure predictor, the flow rate predictor and the humidity predictor are combined to obtain a joint predictor.

4. The method according to claim 1, characterized in that: Verify and adjust the temperature, pressure, flow rate and humidity data in the prediction window to obtain the sub-item prediction values ​​of temperature, pressure, flow rate and humidity, including: Align the temperature, pressure, flow rate and humidity data of the prediction window in time to form a prediction data matrix; Calculating historical prediction accuracy of a temperature predictor, a pressure predictor, a flow rate predictor, and a humidity predictor, and determining the confidence of each predictor according to the historical prediction accuracy; The predictor result with the highest confidence is selected as the benchmark, and the other predictor results in the prediction data matrix are verified and adjusted according to the constraints of temperature, pressure, flow rate and humidity to obtain the adjusted prediction data matrix; According to the adjusted prediction data matrix, the item-by-item prediction values ​​of temperature, pressure, flow rate and humidity are obtained.

5. The method according to claim 4, characterized in that Get itemized forecasts for temperature, pressure, flow rate, and humidity, including: Interactive smoke emission monitoring system, which obtains historical temperature, historical pressure, historical flow rate and historical humidity data of the target smoke exhaust duct, and combines the prediction window to obtain multiple historical data matrices; Calculate the matrix similarity between the prediction data matrix and multiple historical data matrices, and obtain the matrix data volume whose matrix similarity is greater than the preset similarity; When the matrix data volume is greater than the preset data volume, the sub-item prediction values ​​of temperature, pressure, flow rate and humidity are obtained according to the prediction data matrix.

6. The method according to claim 1, characterized in that Input the predicted values ​​of temperature, pressure, flow rate and humidity into the monitor control space to obtain the target monitor control parameters, including: Interactive smoke emission monitoring system to obtain historical temperature, pressure, flow rate and humidity data of the target smoke exhaust duct; Determine a plurality of historical monitoring instrument working conditions according to the historical temperature, historical pressure, historical flow rate and historical humidity data, and obtain monitoring instrument control parameters of the plurality of historical monitoring instrument working conditions, wherein each historical monitoring instrument working condition corresponds to a set of temperature, pressure, flow rate and humidity data; The plurality of historical monitoring instrument working conditions are used as input dimensions of the monitoring instrument control space, and the corresponding monitoring instrument control parameters are used as output dimensions of the monitoring instrument control space to form a monitoring instrument control space; The sub-item prediction values ​​of temperature, pressure, flow rate and humidity are matched with the historical data in the monitor control space to obtain the historical data point closest to the sub-item prediction value, and the monitor control parameter corresponding to the historical data point is used as the target monitor control parameter.

7. The method according to claim 6, characterized in that Build a monitor control space, including: Evaluate the monitor control parameters of multiple historical monitor working conditions to obtain the control effect evaluation value of each monitor control parameter; According to the control effect evaluation value, a monitor control parameter having a control effect evaluation value greater than a preset evaluation threshold is selected from monitor control parameters of multiple historical monitor working conditions as a preferred monitor control parameter; A monitoring instrument control space is constructed with multiple historical monitoring instrument operating conditions as input dimensions and the corresponding optimal monitoring instrument control parameters as output dimensions.

8. The method according to claim 1, characterized in that The method further comprises: According to the actual output data of the temperature, pressure, flow and humidity integrated monitoring instrument, combined with the sub-item prediction values ​​of temperature, pressure, flow rate and humidity, the prediction accuracy of the prediction window is calculated; A prediction accuracy accumulator is set. When the prediction accuracy is lower than a preset threshold, the prediction window is marked as a prediction inaccuracy window, and the number of prediction inaccuracy windows is recorded in the prediction accuracy accumulator. When the prediction accuracy is higher than or equal to the preset threshold, the current prediction window is marked as a prediction accurate window. Determine whether the count value of the prediction accuracy accumulator reaches a preset number of times threshold. When the count value reaches the preset number of times threshold, update the joint predictor according to the actual output data corresponding to the prediction inaccuracy window, and clear the prediction accuracy accumulator at the same time; When the count value does not reach the preset number threshold, the joint predictor is kept unchanged and the prediction process of the next prediction window is continued.