Soft measurement method and system for flue gas flow of coal-fired boiler, medium and program product

Abstract

The invention provides a soft measurement method and system for flue gas flow of a coal-fired boiler, a medium and a program product, and relates to the technical field of electric digital data processing. Firstly, historical equipment operation data of a power plant distributed control system and a continuous emission monitoring system in a set time period are collected, and the historical equipment operation data cover boiler load, air volume and other information. And meanwhile, historical flue gas flow measured data at flue measuring points with different heights in the same set period are obtained. Secondly, determining a flue gas flow prediction model through machine learning by taking historical flue gas flow actual measurement data as dependent variables and historical equipment operation data as covariables; after current equipment operation data are obtained, the current equipment operation data are input into the model to obtain flue gas flow prediction data of different heights, and finally the prediction data are sent to a management end to be displayed, by implementing the method, the limitation of traditional single-point measurement can be broken through, and the flue gas flow condition in a flue can be comprehensively reflected.

Description

Technical Field

[0001] The present application relates to the technical field of electronic digital data processing, and in particular to a soft measurement method, system, medium and program product for flue gas flow of a coal-fired boiler. Background Art

[0002] With the rapid development of my country's economy, energy demand continues to grow. Coal-fired boilers, as important energy conversion equipment, play an important role in industrial production, residential heating and other fields. However, with the enhancement of environmental awareness and the implementation of energy-saving and emission reduction policies, higher requirements are put forward for the operation efficiency and emission control of coal-fired boilers. During the operation of the boiler, the flue gas flow rate is a key operating parameter, which not only affects the combustion efficiency and operation stability of the boiler, but also directly related to the emission level of flue gas pollutants. In HJ∕T75-2017 "Technical Specifications for Continuous Monitoring of Flue Gas (SO2, NOx, Particulate Matter) Emissions from Fixed Pollution Sources", the flow rate accuracy and calculation method are required. When the flow rate is >10m / s, the relative error does not exceed ±10%, and when the flow rate is ≤10m / s, the relative error does not exceed ±12%. Therefore, accurate measurement and control of flue gas flow is crucial for the overall performance optimization of the boiler.

[0003] Flue gas flow measurement is a key parameter to understand boiler combustion conditions, optimize the combustion process, and improve boiler thermal efficiency. The traditional flue gas flow measurement methods are mainly the following: 1. Pitot tube method: The flue gas flow rate is calculated by measuring the difference between the dynamic pressure and static pressure of the flue gas. This method is simple and easy to implement, but has low accuracy and is easily affected by the uneven distribution of the flow field in the pipeline.

[0004] 2. Matrix flow meter: Based on the differential pressure measurement principle, the flue surface flow velocity is measured through a full-section multi-point layout to obtain more representative data.

[0005] However, with the increasing demand for refined management in power plants, traditional hardware measurement devices have exposed many problems. On the one hand, it can only measure fixed positions in the flue. Due to the complex airflow distribution in the flue, there are differences in flue gas flow at different heights and in different areas. Single-point measurement data cannot fully reflect the flue gas flow in the entire flue, which makes the combustion optimization adjustment based on this limited and difficult to achieve precise control. On the other hand, hardware measurement devices are greatly affected by environmental factors, such as high temperature, dust, and corrosive gases in the flue, which can easily lead to a decline in sensor performance and an increase in measurement errors. The long-term operating stability is poor, and frequent maintenance and calibration are required, which increases the operating costs of the power plant. Therefore, it is difficult for existing technologies to accurately measure the flue gas flow in the flue at a low cost. Summary of the invention

[0006] The present application provides a soft measurement method, system, medium and program product for the flue gas flow of a coal-fired boiler, which are used to achieve accurate and stable monitoring of the flue gas flow in different height areas in the flue of a coal-fired boiler.

[0007] In the first aspect, the present application provides a soft measurement method for the flue gas flow of a coal-fired boiler, which is applied to a soft measurement system, and the method includes: collecting historical equipment operation data of a set time period of a power plant distributed control system and a continuous emission monitoring system, and the historical equipment operation data includes boiler load, total boiler air volume, flue gas temperature, type of coal and quality of coal; obtaining historical flue gas flow measurement data measured at flue measurement points within different height ranges using high-precision equipment within the set time period, and the historical flue gas flow measurement data includes at least flue gas flow rate, temperature and pressure; using the historical flue gas flow measurement data as a dependent variable, and using the historical equipment operation data as a covariate, and combining the dependent variable and the covariate for machine learning to determine a flue gas flow prediction model; after obtaining the current equipment operation data, inputting the current equipment operation data into the flue gas flow prediction model to obtain flue gas flow prediction data at different heights; and sending the flue gas flow prediction data to a management end for display.

[0008] By adopting the above technical solution, we first collect the historical equipment operation data of the power plant's distributed control system and continuous emission monitoring system, as well as the historical flue gas flow measurement data at different flue measurement points. These multi-source data provide rich information for subsequent machine learning. Machine learning is performed to determine the prediction model using the historical flue gas flow measurement data as the dependent variable and the historical equipment operation data as the covariate, so that the model can make predictions based on multiple factors. The current equipment operation data is input into the model to obtain the predicted data and displayed on the management side, realizing the soft measurement of the flue gas flow of coal-fired boilers, providing operators with comprehensive and intuitive flue gas flow information, which helps to timely understand the boiler operation status and improve the convenience and accuracy of boiler operation management.

[0009] In combination with some embodiments of the first aspect, in some embodiments, after the step of inputting the current equipment operation data into the flue gas flow prediction model to obtain the flue gas flow prediction data at different heights, it also includes: obtaining the current flue gas flow measured data through high-precision equipment detection; comparing the flue gas flow prediction data with the current flue gas flow measured data to determine a prediction evaluation result; and adjusting the flue gas flow prediction model according to the prediction evaluation result.

[0010] By adopting the above technical solution, after obtaining the flue gas flow prediction data, high-precision equipment is used to obtain the current flue gas flow measured data. Comparing the predicted data with the measured data can directly reflect the accuracy of the prediction model. The model is adjusted according to the prediction evaluation results obtained by comparison, forming a closed-loop feedback mechanism. The prediction evaluation results are used as the basis for adjustment to continuously optimize the model, thereby improving the accuracy of flue gas flow prediction.

[0011] In combination with some embodiments of the first aspect, in some embodiments, after taking the historical flue gas flow measured data as the dependent variable and the historical equipment operation data as the covariate, and combining the dependent variable and the covariate for machine learning to determine the flue gas flow prediction model, it also includes: obtaining environmental characteristic data of different seasons; classifying the boiler operating condition information of each season according to the environmental characteristic data, and establishing a seasonal operating condition feature library, which includes boiler operating condition characteristic parameter sets for spring, summer, autumn and winter; based on the seasonal operating condition feature library, performing feature extraction and seasonal classification on the historical operation data to obtain a seasonal historical equipment operation data set; using the seasonal historical operation data set as training data, adjusting the flue gas flow prediction model to obtain a seasonal new flue gas flow prediction model; when obtaining the current boiler operation data, judging the current season according to the current date, and determining the flue gas flow prediction data through a flue gas flow prediction model that matches the current season.

[0012] By adopting the above technical solution, the environmental characteristic data of different seasons are obtained and classified to establish a seasonal operating condition characteristic library. Taking into account the impact of seasonal factors on boiler operating conditions, the historical operating data is processed based on the library to obtain a seasonal historical equipment operation data set, which is used to train the model to obtain a new seasonal model. The corresponding seasonal model prediction is matched according to the current date, making full use of the characteristics of boiler operation in different seasons. The operating conditions of the boiler will also change due to different environmental temperatures, humidity, etc. in different seasons. This seasonal adjustment model method can make the prediction more in line with the actual situation, improve the accuracy of flue gas flow prediction in different seasons, and provide strong guarantees for the precise operation and regulation of boilers in different seasons.

[0013] In combination with some embodiments of the first aspect, in some embodiments, when obtaining the current boiler operation data, the current season is judged according to the current date, and after the step of determining the flue gas flow prediction data through a flue gas flow prediction model that matches the current season, it also includes: determining the normal fluctuation range of flue gas flow in different seasons according to the seasonal historical equipment operation data set; establishing an adaptive warning threshold based on the flue gas flow prediction data, and the adaptive warning threshold is dynamically adjusted with the season and boiler operating conditions; when the flue gas flow prediction data exceeds the adaptive warning threshold, a warning signal is triggered; and according to the frequency and degree of the warning signal, the new flue gas flow prediction model is optimized and updated.

[0014] By adopting the above technical solution, the seasonal historical equipment operation data set is used to determine the normal fluctuation range of flue gas flow in different seasons, and then an adaptive early warning threshold is established. When the flue gas flow forecast data exceeds the threshold and triggers the early warning signal, the abnormality can be discovered in time. According to the frequency and degree of the early warning signal, the model is optimized and updated. On the one hand, it can quickly detect abnormal boiler operation, remind the staff to deal with it in time, and ensure the safe and stable operation of the boiler; on the other hand, the relevant data of the early warning signal is used to optimize the model, which can make the model better adapt to complex and changeable operating conditions, improve the reliability and adaptability of the model, make subsequent predictions more accurate, and further ensure the long-term stable operation of the boiler.

[0015] In combination with some embodiments of the first aspect, in some embodiments, after the step of inputting the current equipment operation data into the flue gas flow prediction model to obtain the flue gas flow prediction data at different heights, it also includes: monitoring the changing trend of the flue gas flow prediction data; when it is monitored that the smoke gas flow prediction data continues to rise or fall, judging whether there is an abnormal situation; if there is an abnormal situation, determining the height position information of the abnormal situation when it occurs in the flue, and sending the height position information to the client.

[0016] By adopting the above technical solution, the trend of flue gas flow prediction data changes can be monitored, and abnormal dynamics of continuous rise or fall can be discovered in time. Once an abnormality is detected, it is determined whether there is an abnormal situation, and the height position information of the abnormality in the flue is determined, and then the information is sent to the client. This enables staff to quickly locate areas in the flue where problems may exist, such as local blockages, equipment failures, etc., providing an accurate basis for timely and targeted maintenance measures, avoiding the expansion of problems, reducing the impact on the normal operation of the boiler, reducing maintenance costs, and improving the reliability and safety of boiler operation.

[0017] In combination with some embodiments of the first aspect, in some embodiments, after the step of inputting the current equipment operation data into the flue gas flow prediction model to obtain the flue gas flow prediction data at different heights, it also includes: collecting the working status data of the dust removal equipment in the flue gas system; and correcting the flue gas flow prediction results based on the working status data.

[0018] By adopting the above technical solution, the working status data of the dust removal equipment in the flue gas system is collected and acquired. The working status of the dust removal equipment will affect the properties and flow of the flue gas. It is combined with the flue gas flow prediction results for correction, taking into account the impact of the dust removal equipment on the flue gas flow. Because the operating status of the dust removal equipment is different, the treatment effect on the flue gas is different, which in turn affects the flue gas flow. Through correction, the prediction results are more in line with the actual flue gas flow, which improves the accuracy of the prediction, provides more accurate data support for boiler combustion adjustment, environmental monitoring, etc., and ensures the economy and environmental protection of boiler operation.

[0019] In combination with some embodiments of the first aspect, in some embodiments, after the step of sending the flue gas flow prediction data to the management end for display, it also includes: when the accuracy of the prediction result is lower than a preset threshold, outputting a model prediction alarm; collecting the equipment operation data when the alarm is generated and the corresponding flue gas flow measured data; using the equipment operation data when the alarm is generated and the corresponding flue gas flow measured data to retrain the flue gas flow prediction model.

[0020] By adopting the above technical solution, when the accuracy of the prediction result is lower than the preset threshold, the model prediction alarm is output to promptly remind relevant personnel that there is a problem with the model. The equipment operation data at the time of the alarm and the corresponding flue gas flow measured data are collected, and the model is retrained using these data. This process can promptly discover and solve the problem of inaccurate model prediction, avoid wrong decisions caused by inaccurate prediction data, and retrain the model to adapt the model to the new operating conditions, restore and improve the prediction accuracy, ensure the reliability of flue gas flow prediction, and ensure the accuracy and effectiveness of boiler operation management.

[0021] In a second aspect, the present application provides a soft measurement system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the soft measurement system to perform the method described in the first aspect and any possible implementation of the first aspect.

[0022] In a third aspect, the present application provides a computer-readable storage medium, comprising instructions, which, when executed on a soft measurement system, enable the soft measurement system to execute the method described in the first aspect and any possible implementation manner of the first aspect.

[0023] In a fourth aspect, the present application provides a computer program product. When the computer program product is run on a soft measurement system, the soft measurement system executes the method described in the first aspect and any possible implementation manner of the first aspect.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. The method collects historical data from multiple sources (including historical equipment operation data of the power plant's distributed control system and continuous emission monitoring system, and historical flue gas flow measurement data at flue measuring points at different heights), and uses these data as the basis to determine the flue gas flow prediction model through machine learning, and then inputs the current equipment operation data into the model to obtain the predicted data and display it on the management side. This effectively solves the technical problems in the prior art that traditional hardware measurement devices measure data incompletely, are greatly affected by the environment, and are difficult to accurately measure the flue gas flow. This achieves the technical effect of comprehensively and intuitively obtaining flue gas flow information and improving the convenience and accuracy of boiler operation management.

[0025] 2. The method adopts the method of obtaining environmental characteristic data of different seasons, establishing a seasonal operating condition characteristic library based on the data, using the library to process historical operating data to obtain a seasonal historical equipment operation data set, using the data set to train a new seasonal model, and matching the corresponding seasonal model according to the current date for prediction. Therefore, the technical problem of inaccurate flue gas flow prediction caused by not considering the influence of seasonal factors on boiler operating conditions in the existing technology is effectively solved, thereby achieving the technical effect of improving the accuracy of flue gas flow prediction in different seasons and providing strong guarantee for the precise operation and control of boilers.

[0026] 3. Due to the adoption of the technical means of monitoring the flue gas flow prediction data change trend, judging and determining the height position information of the abnormality in the flue when an abnormality occurs, and then sending the information to the client, it effectively solves the technical problem that it is difficult to quickly locate the abnormal area of ​​the flue in the existing technology, resulting in untimely problem handling, and thus achieves the technical effect of quickly locating the problem area of ​​the flue, facilitating timely maintenance, and ensuring the reliable and safe operation of the boiler. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flow chart of a soft measurement method of flue gas flow of a coal-fired boiler in an embodiment of the present application; Figure 2 is another flow chart of the soft measurement method of the flue gas flow rate of a coal-fired boiler in an embodiment of the present application; Figure 3 It is a schematic diagram of a physical device structure of a soft measurement system in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification and appended claims of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to and includes any or all possible combinations of one or more listed items.

[0029] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.

[0030] For ease of understanding, the following is a description of the process of the method provided by this implementation. Figure 1 , which is a flow chart of a soft measurement method for flue gas flow of a coal-fired boiler in an embodiment of the present application.

[0031] S101, collecting historical equipment operation data of a set time period of the power plant distributed control system and continuous emission monitoring system, the historical equipment operation data including boiler load, boiler total air volume, flue gas temperature, coal type and coal quality; The soft measurement system first establishes a stable communication connection with the power plant's distributed control system (DCS) and continuous emissions monitoring system (CEMS). Considering that the system architecture and data interface of different power plants may be different, the soft measurement system adopts a standardized data interface protocol to ensure compatibility with various mainstream DCS and CEMS systems and achieve seamless data connection.

[0032] During the data collection process, multi-threaded parallel collection technology can be used. The system assigns an independent thread to each data collection task. For example, for different types of data such as boiler load, boiler total air volume, flue gas temperature, coal type and coal quality, the corresponding collection threads are started respectively. This can significantly improve the efficiency of data collection, reduce the collection time, and ensure that the acquired data has a high timeliness.

[0033] Specifically, for boiler load data, the soft measurement system obtains it from the DCS. The DCS collects boiler steam flow, pressure and other parameters in real time, and calculates the boiler load through the built-in calculation module based on thermodynamic principles and specific formulas. The soft measurement system reads the data from the DCS regularly, and uses the timestamp to record the exact time of data collection. When obtaining the total air volume data of the boiler, the soft measurement system reads the air volume sensor data installed on the air duct from the DCS. These sensors, such as matrix flowmeters and pitot tube flowmeters, use the differential pressure measurement principle. Taking the differential pressure air volume sensor as an example, it measures the pressure difference in the air duct, combines the geometric parameters of the air duct, and calculates the air volume based on the Bernoulli equation. The flue gas temperature data is obtained from the temperature sensor in the CEMS. These sensors are mostly thermocouples or thermal resistors, and use the thermoelectric effect or the characteristics of resistance changing with temperature to measure temperature. The soft measurement system collects the signal of the temperature sensor in real time, and uses cold end compensation technology to eliminate the influence of the cold end temperature change of the thermocouple on the measurement results. Since the temperature distribution in the flue may be uneven, the system will integrate the data of multiple temperature sensors and use a weighted average algorithm to calculate a more representative flue gas temperature. In terms of obtaining coal type and coal quality data, the soft measurement system first obtains the identification information of the coal type from the DCS. The coal type here can be different coal qualities, and the common ones are bituminous coal, anthracite, lignite, etc. For coal quality data, if the power plant is equipped with online coal quality monitoring equipment, the system will connect to it to obtain coal quality parameters such as calorific value, sulfur content, and ash content in real time. If there is no online monitoring equipment, the system will estimate the coal quality parameters based on the coal procurement records and regular laboratory test reports, combined with the coal consumption. For example, based on the coal procurement batch and the corresponding laboratory test report, as well as the consumption of the batch of coal in the boiler, the average calorific value of the currently used coal is estimated by the weighted average method. At the same time, the system will establish a coal quality database to record the quality data and usage of different batches of coal for subsequent analysis and model training.

[0034] S102, obtaining historical flue gas flow measurement data measured at flue measurement points within different height ranges using high-precision equipment within the set time period, the historical flue gas flow measurement data at least including flue gas flow velocity, temperature and pressure; The soft measurement system collects the historical measured data of flue gas flow at different measuring points in the flue within a set time period from the continuous emission monitoring system through a standardized interface. In order to obtain accurate and reliable measured data, these measuring points are set at different heights of the flue, such as hoisting at the upper, middle and lower cross sections. On each cross section, the floating measuring device will set several measuring points in the flue longitudinally and transversely to comprehensively monitor the flue gas flow on the cross section. These floating measuring devices use high-precision measurement principles such as thermal and ultrasonic measurement principles to directly obtain parameters such as flue gas flow rate, temperature and pressure. Considering the influence of the complex flow form of the flow field in the flue, at least 4 measuring points are arranged on each floating measuring device for cross-checking to improve the reliability of the measurement. The soft measurement system establishes a stable connection with these floating measuring devices and regularly collects the flue gas flow rate, temperature and pressure data of each measuring point. In order to avoid data omission, the soft measurement system adopts a heartbeat packet mechanism. If the data of a measuring point is not received for a certain period of time, a query request will be actively sent to the measuring point. At the same time, the soft measurement system has a data verification function, which can detect and eliminate obviously distorted measurement data. These multi-point flue gas flow data collected in real time will be bound to the corresponding timestamp and saved for subsequent model training.

[0035] S103, taking the historical measured smoke flow data as a dependent variable, taking the historical equipment operation data as a covariate, and combining the dependent variable and the covariate to perform machine learning to determine a smoke flow prediction model; The soft measurement system uses the collected historical equipment operation data and the historical flue gas flow multi-point measured data at the flue measuring points within the corresponding time period to train the flue gas flow prediction model through a machine learning algorithm.

[0036] Specifically, before model training, the soft measurement system will pre-process the collected data. First, the data quality is checked, and the data with missing or abnormal problems is marked or eliminated. Then the data is normalized, and the historical equipment operation data including boiler load, boiler total air volume, flue gas temperature, coal type and coal quality are standardized. The data is scaled to conform to the standard normal distribution. The formula is:

[0037] In the formula, mean is the mean value and σ is the standard deviation.

[0038] In addition, the soft measurement system will also perform feature engineering, construct and select model input features, and combine professional knowledge of the boiler combustion process to use feature extraction methods to construct more representative feature variables, such as wind-coal ratio, unit steam production coal consumption and other comprehensive features. At the same time, high-dimensional features are selected through methods such as linear correlation analysis, redundant features are eliminated, and dimensionality reduction is performed to improve the efficiency of model training and prediction.

[0039] Then analyze the key factors that affect the flue gas flow of coal-fired boilers, where the key factors include at least boiler load, boiler total air volume, and flue gas temperature. Specifically, the Pearson correlation coefficient method and the Kendall rank correlation coefficient method can be used to perform correlation analysis on the key factor data, where the Pearson correlation coefficient method formula is as follows: ; In the formula, and represent the observed values ​​of two variables, and and Respectively represent their means. By calculating the correlation coefficient in the sample data of the data set, we can get a value between -1 and 1, thereby judging the strength of the linear relationship between the two variables. The closer it is to 1, the higher the positive correlation.

[0040] The Kendall correlation coefficient formula is as follows: ; In the formula, is the number of predicted values ​​of flue gas flow of coal-fired boiler, and is the corresponding rank in the first dataset, and is the corresponding rank in the second dataset, is a symbolic function, when ,but The function value is 1 when ,but The function value is -1, otherwise The function value is 0.

[0041] The purpose of calculating the Pearson correlation coefficient method and the Kendall rank correlation coefficient method is to find out the variables that are positively correlated with the flue gas of coal-fired boilers from the linear and nonlinear relationships for subsequent algorithm modeling. After the Pearson correlation coefficient method and the Kendall rank correlation coefficient method were calculated, the variables with a correlation greater than 0.4, that is, medium correlation, were obtained, which can include boiler load, boiler total air volume, flue gas temperature, and coal type and quality.

[0042] After the characteristic data is prepared, the soft measurement system uses the multi-point measured data of historical flue gas flow as the dependent variable and the processed historical equipment operation data as the covariate. Here, the BP neural network can be used to build a regression model. The back propagation neural network model uses the prediction method of multiple linear regression, and its formula is ; In the formula, is the flue gas flow output value of the coal-fired boiler predicted by the back propagation neural network model, is the activation function, The input value of the back propagation neural network model is the key data of the flue gas flow of the coal-fired boiler, namely the boiler load, the total air volume of the boiler, the flue gas temperature, the type of coal and the quality of the coal. is the hidden node connection weight, is the implicit node threshold, is the hidden node threshold, is the number of hidden nodes; The back propagation neural network model architecture is described again. The hidden layer is 2 layers, and the activation function of the output layer uses Tanh, and its formula is: , In the formula, is the input value. During the model training process, when the input value is large, the output of the hyperbolic tangent function is close to 1, and when the input value is small, the output is close to -1.

[0043] After the model is determined, the soft measurement system will divide the pre-processed historical data into a training set and a test set. Generally speaking, the training set accounts for 70%-80% of the total data, and the test set accounts for 20%-30%. The system will use the training set to train the model, and by continuously adjusting the model parameters, minimize the error between the model's prediction results and the historical flue gas flow measured data.

[0044] In terms of model quality assessment, 30% of the data set can be selected to construct a validation set, and the variable values ​​contained in the validation set are used as inputs. The back propagation neural network model is used for calculation, and the predicted flue gas flow output value of the coal-fired boiler is obtained. The flue gas flow data measured by high-precision equipment is compared and evaluated, and the average relative error is less than 1.90%. Through this method, the flue gas flow can be measured with high quality without relying on instruments, and the flue gas emissions of coal-fired boilers can be effectively monitored and managed, contributing to environmental protection and resource conservation.

[0045] S104, after obtaining the current equipment operation data, input the current equipment operation data into the smoke flow prediction model to obtain smoke flow prediction data at different heights; After the flue gas flow prediction model is trained, the soft measurement system can be used for online prediction. The soft measurement system continuously collects the current operation data of the boiler from the DCS and CEMS through standard interfaces, including boiler load, air volume, coal consumption and other information. After receiving new data, the soft measurement system will first perform necessary verification, such as smoothing and filtering several consecutive sets of data, or performing differential verification with the last data to eliminate the impact of collection anomalies.

[0046] Then the soft measurement system will extract the characteristic variables from these current operating data, normalize and homogenize them, and the processing method is consistent with the preprocessing process during model training. The current operating characteristic data after processing is input into the flue gas flow prediction model. The model calls the corresponding algorithm module to predict the flue gas flow values ​​in the flue corresponding to different height areas based on the input data. Considering the large internal space of the flue, the soft measurement system can divide the flue into multiple subdivided areas in the height direction and predict the flow rate for each thin layer space separately. For example, the entire flue height can be divided into three equal areas: upper, middle and lower, and each area is about 5 meters high. After model prediction, the soft measurement system can obtain parameters such as flue gas flow velocity, temperature and pressure in each thin layer area, and form three-dimensional distribution information of flue gas flow within the flue height range. This overcomes the limitations of single-point measurement of sensors and realizes full-flow channel measurement of the soft measurement system.

[0047] In some embodiments, after obtaining the predicted data of flue gas flow, the soft measurement system can also obtain the measured data of flue gas flow at the current moment through high-precision equipment. Here, high-precision equipment refers to sensors installed in the flue that can directly measure the parameters of flue gas flow, such as thermal air flow meters, which can measure the real-time flow rate, temperature and pressure of flue gas. The soft measurement system will obtain these measured data and compare them with the flue gas flow data predicted by the model to determine the deviation between the two. If the measured data and the predicted data are highly consistent, it means that the prediction effect is good; if there is a large deviation, it means that the predicted data is inaccurate. The evaluation result of this prediction, that is, the accuracy of the prediction, can be determined by comparison. The soft measurement system will adjust the flue gas flow prediction model based on this result, such as increasing the training data set, adjusting the model algorithm parameters, etc., to improve the accuracy of subsequent predictions. In this way, by using the closed-loop feedback of measured data and predicted data, the model can be continuously optimized to make its prediction results more accurate and reliable.

[0048] In some embodiments, the soft measurement system can also monitor the changing trend of the flue gas flow prediction data. The monitoring mentioned here refers to the continuous tracking of the dynamic changes of the flue gas flow prediction value. If the predicted data is monitored to show an abnormal situation of continuous rise or fall, the soft measurement system will determine whether this represents an abnormal problem in the actual flue. If it is determined that there is an abnormality, the soft measurement system can further analyze the changing pattern of the predicted data to determine at which height position of the flue the abnormality occurs. For example, if only the predicted flow in the uppermost area increases significantly, it can be determined that the abnormality occurs in the upper area. Finally, the soft measurement system will send the abnormal height position information to the client, such as the mobile terminal of the monitoring center or maintenance personnel, which can help locate the abnormal position in time and take targeted maintenance measures. In this way, by monitoring the changes in the predicted data and determining the abnormal position information, the ability to identify and respond to flue abnormalities can be improved to ensure the safe and stable operation of the flue system.

[0049] S105. Send the flue gas flow prediction data to the management end for display.

[0050] The obtained flue gas flow prediction data is packaged in the form of data frames and pushed to the monitoring main station through the network interface. The monitoring main station has a three-dimensional data visualization module, which can clearly present the flue gas flow in different height areas in a color-coded manner, generating a three-dimensional effect of airflow movement in the flue. Monitoring personnel can intuitively understand the airflow distribution in the flue, which helps to find problem areas. In addition, the monitoring main station also supports customized multi-point data extraction and comparison. Through soft measurement and comprehensive analysis of a small amount of actual measurement point data, the flue gas flow state in the flue can be comprehensively evaluated, providing a basis for subsequent model optimization and boiler adjustment.

[0051] In the embodiment of the present application, due to the use of historical equipment operation data of the power plant's distributed control system and continuous emission monitoring system, as well as historical flue gas flow measurement data at different flue measurement points at different heights, and based on these data, a flue gas flow prediction model is determined through machine learning, and then the current equipment operation data is input into the model to obtain prediction data and displayed on the management side. Therefore, it is possible to integrate multi-source data for accurate prediction without being limited by single-point measurement and complex environmental interference, and effectively solves the technical problems in the prior art of traditional hardware measurement devices that measurement data is incomplete, greatly affected by the environment, and difficult to accurately measure the flue gas flow, thereby achieving comprehensive and intuitive acquisition of flue gas flow information.

[0052] In some embodiments, the soft measurement system can also collect the working status data of the dust removal equipment in the flue gas system. This is because the working status of the dust removal equipment will have a certain impact on the flue gas flow rate. For example, if the filter bag of the dust removal equipment is clogged, it will increase the resistance of the flue gas to pass through, thereby reducing the flue gas flow rate. To this end, the soft measurement system needs to establish a communication connection with the dust removal equipment to obtain its working parameters, such as the inlet flue gas volume, outlet flue gas volume, pressure drop, filter bag resistance and other data, which reflect the working status of the dust removal equipment. After obtaining the working status data of the dust removal equipment, the soft measurement system combines it with the flue gas flow prediction results and corrects it through relevant models. This correction takes into account the impact of the actual operating conditions of the dust removal equipment on the results, and can make the flue gas flow prediction value more accurately reflect the actual situation.

[0053] In addition, after the flue gas flow prediction model has been running for a period of time, the soft measurement system also needs to verify whether its prediction accuracy meets the requirements. One way to verify is to compare the predicted data with the actual flue gas flow data obtained through sampling measurement. If the prediction accuracy is lower than the preset threshold, it indicates that there is a problem with the model and it needs to be optimized. At this time, the soft measurement system will output a model prediction alarm, indicating that an inaccurate prediction has occurred. At the same time, the soft measurement system will collect the boiler operation data when the alarm is generated, as well as the actual measured data of the flue gas flow at the corresponding moment. Then, the model is retrained using these data. By adding these new data samples, the adaptability of the model to the actual situation can be improved, thereby improving the subsequent prediction accuracy. In this way, through closed-loop inspection, alarm, and retraining of the model, the soft measurement system can continuously optimize the effect of flue gas flow prediction and ensure its accuracy and reliability.

[0054] After combining the above content, the following is a more detailed description of the process of the method provided by this implementation. Figure 2 , is another flow chart of the soft measurement method of the flue gas flow rate of a coal-fired boiler in an embodiment of the present application.

[0055] S201, obtaining environmental characteristic data of different seasons; The soft measurement system first establishes a stable data connection with the meteorological data provider to obtain environmental characteristic data in different seasons. To ensure the accuracy and timeliness of the data, the system uses multi-source data fusion technology. In addition to obtaining conventional meteorological information such as temperature, humidity, and air pressure from a professional meteorological data platform, it also uses environmental monitoring equipment installed around the power plant to collect localized environmental data such as wind speed, wind direction, and light intensity. These localized monitoring devices communicate with the soft measurement system through a wireless sensor network, and upload data to the system in real time in a low-power, high-reliability transmission method.

[0056] In terms of data collection frequency, the system dynamically adjusts according to seasonal changes and actual needs. During the seasonal transition period, the data collection frequency can be set higher due to frequent environmental changes, while during the relatively stable season, the collection frequency can be adjusted slightly lower. This ensures that sufficiently detailed environmental change information is obtained while avoiding excessive data collection that leads to waste of system resources.

[0057] S202, classifying boiler operating condition information of each season according to the environmental characteristic data, and establishing a seasonal operating condition characteristic library, wherein the seasonal operating condition characteristic library includes boiler operating condition characteristic parameter sets for spring, summer, autumn and winter; After the soft measurement system collects the environmental characteristic data of different seasons, it further classifies and models the boiler's operating conditions based on these data. Specifically, the operating condition classification module in the soft measurement system reads the environmental characteristic data and extracts the key parameters that affect the boiler's operating conditions, such as ambient temperature, humidity, atmospheric pressure, etc. Then, combined with the expert experience knowledge base, the boiler operating state corresponding to different parameter value ranges is determined. After statistical analysis, the soft measurement system obtains the characteristic parameter ranges of the boiler's operating conditions in different seasons, such as load size, air volume level, combustion mode, etc. These parameter ranges form the operating condition feature set corresponding to the season, and are constructed into a seasonal operating condition feature library and stored in the system database. The feature library contains the operating condition feature parameter sets of the four seasons of spring, summer, autumn, and winter. Each season set includes about 5 feature parameters. For example, the summer operating condition feature set can include the parameters "low load, small air volume, and the main coal is bituminous coal". The establishment of this feature library enables the soft measurement system to obtain the empirical knowledge of boiler operating conditions in different seasons.

[0058] S203, based on the seasonal operating condition feature library, feature extraction and seasonal classification are performed on the historical operation data to obtain a seasonal historical equipment operation data set; On the basis of establishing the seasonal operating condition feature library, the soft measurement system further processes the large amount of historical boiler operation data that has been collected to identify the season to which it belongs and classify it. Specifically, the system first extracts environmental characteristic parameters such as temperature and humidity from the historical operating data using relevant feature extraction algorithms. These parameters are then matched with the seasonal operating condition feature library to determine the season category to which they are most likely to belong. For example, if the extracted parameters include "temperature 20°C, humidity 40%", it can be determined that the boiler operating condition corresponding to the operating data is closest to the characteristics of spring and autumn. Finally, based on the matching results, the software system automatically classifies the historical operating data into seasonal categories, forming corresponding spring data sets, summer data sets, etc., to obtain the final seasonal historical equipment operation data set.

[0059] S204, using the seasonal historical operation data set as training data, adjusting the flue gas flow prediction model to obtain a seasonal new flue gas flow prediction model; On the basis of the classified data sets, the soft measurement system uses these data sets to train and adjust the corresponding seasonal flue gas flow prediction models accordingly. Specifically, the system uses the spring data set as training data, and uses machine learning algorithms to adjust the original general model so that its seasonal parameters adapt to the spring operating conditions, thereby obtaining a spring-specific flue gas flow prediction model. Similarly, the summer, autumn, and winter data sets are used to obtain the corresponding seasonal models. Compared with the general model, these seasonal models are specifically tuned using the actual operating data of the corresponding season, which can more accurately reflect the operating characteristics of each season, improve seasonal adaptability, and facilitate modeling and prediction of flue gas flow changes in different seasons.

[0060] After the model training is completed, the soft measurement system establishes an index to record the models corresponding to different seasons for subsequent automatic selection. After the four-season model is determined, if new data is collected in the subsequent operation, the system can determine the season by matching with the seasonal operating condition feature library, and then automatically select the corresponding model for flue gas flow prediction, without manual judgment, to achieve flexible calling of seasonal models.

[0061] S205. When obtaining the current boiler operation data, the current season is determined according to the current date, and the flue gas flow prediction data is determined by a flue gas flow prediction model that matches the current season; After the seasonal flue gas flow prediction model is determined, the soft measurement system enters the online prediction stage. When the system obtains the real-time operation data of the boiler through DCS and CEMS, it will first use the current time information to determine the season. Specifically, the system has a built-in season judgment module associated with the calendar. According to the current month and date, the current season can be directly determined, such as March 15th is spring. Then the system will select the model corresponding to the current season from the alternative four-season models. For example, the spring model will be selected on March 15th as the online model for flue gas flow prediction. Finally, the soft measurement system inputs the real-time operation characteristic data of the boiler into the online model, calls the prediction function of the model, and outputs the flue gas flow prediction results for different height areas. The entire prediction model selection and use process is fully automated without manual participation, realizing the dynamic switching of seasonal models.

[0062] S206. Determine the normal fluctuation range of flue gas flow in different seasons based on seasonal historical equipment operation data sets; The soft measurement system further analyzes and processes a large number of seasonal historical equipment operation data sets that have been obtained to determine the normal fluctuation range of flue gas flow in different seasons. Specifically, the software system uses statistical methods to calculate the mean and standard deviation of each characteristic parameter in the seasonal data set for each season. Taking the winter data set as an example, the mean and standard deviation of the winter ambient temperature, the mean and standard deviation of the boiler load, etc. can be obtained. Then, the software system combines expert experience and knowledge, takes measurement error as a consideration, and determines the range of normal fluctuations, such as the normal fluctuation of winter load within the range of plus or minus 10% of the mean. Finally, the software system summarizes and constructs a knowledge base of the normal fluctuation range of each characteristic parameter in different seasons, and records the normal minimum and maximum values ​​of each characteristic parameter in spring, summer, autumn and winter for subsequent judgment of abnormalities.

[0063] S207, establishing an adaptive warning threshold based on the flue gas flow prediction data, and the adaptive warning threshold is dynamically adjusted according to the season and boiler operating conditions; On the basis of obtaining the normal fluctuation range of flue gas flow in different seasons, the soft measurement system can establish an adaptive early warning threshold to realize the prediction of abnormal warning. Specifically, the software system combines the flue gas flow value predicted by the current model to determine the corresponding seasonal category and characteristic parameter interval. Then, the system calculates the upper and lower limits of the prediction point according to the normal fluctuation range of the parameters of the season, that is, it is considered abnormal if it exceeds the upper and lower limits. This method of dynamically calculating the threshold based on each prediction result realizes an adaptive early warning mechanism. Compared with fixed static thresholds, adaptive thresholds can better determine abnormalities, avoid the problem of overcorrection of fixed thresholds under different parameters, and improve the accuracy of early warning.

[0064] S208, when the flue gas flow prediction data exceeds the adaptive warning threshold, triggering a warning signal; On the basis of establishing an adaptive warning threshold in the soft measurement system, when the real-time predicted flue gas flow data exceeds the currently calculated warning threshold, the system will trigger an abnormal warning signal. Specifically, the software system continuously monitors the flue gas flow rate, temperature, pressure and other parameters predicted at different heights at each moment. These parameters and the corresponding adaptive warning thresholds will be updated in real time. Once a parameter is detected to have exceeded the threshold, the system will start the warning module and generate a warning signal through obvious visual and auditory means, such as flashing warning lights and alarms. At the same time, the system will record the predicted flue gas flow data that triggered the warning, the corresponding height area, the time when the warning was generated, and other information. The generation of the warning signal can quickly remind the operator to pay attention to the abnormal operation of the boiler, and can combine the recorded information to determine the location of the abnormality, which is convenient for subsequent processing.

[0065] S209. Optimize and update the new flue gas flow prediction model according to the frequency and degree of the early warning signal.

[0066] After the warning signal is generated, the soft measurement system will statistically analyze the frequency and degree of the warning as feedback for the optimization of the smoke flow prediction model. Specifically, the system will count the number of warnings triggered in each height area. If the warning is frequent in a certain area, it means that the prediction deviation of the model in this area is large and needs to be optimized. At the same time, the system evaluates the degree of the warning signal, that is, the magnitude of the predicted data exceeding the threshold. The greater the excess, the greater the prediction error. The system will use the frequency and degree as the goal of model optimization, and automatically select a new data sample set for model retraining, such as selecting more historical data samples in the area, and using regularization methods to enhance the robustness of the model and reduce the frequency of warnings. After repeated iterative training, until the frequency and degree of warnings are reduced, it is considered that the model performance has improved and the update is completed. This closed-loop update can continuously improve the prediction accuracy of the model in error-prone areas.

[0067] In the embodiment of the present application, by acquiring environmental characteristic data of different seasons to establish a seasonal operating condition characteristic library, seasonal classification and processing are performed on historical operating data and a seasonal model is trained to achieve accurate prediction of flue gas flow in different seasons and abnormal monitoring and early warning. This not only effectively solves the problems of inaccurate flue gas flow prediction and inability to timely detect abnormal boiler operation caused by the failure of the prior art to consider seasonal factors, but also further improves the intelligence level of boiler operation management, reduces operating risks, and ensures long-term stable, efficient, and environmentally friendly operation of the boiler.

[0068] The soft measurement system in the embodiment of the present invention is described below from the perspective of hardware processing. Figure 3 , is a schematic diagram of a physical device structure of a soft measurement system in an embodiment of the present application.

[0069] It should be noted that Figure 3 The structure of the soft measurement system shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0070] like Figure 3 As shown, the soft measurement system includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage part 308 to the random access memory (RAM) 303, such as executing the method described in the above embodiment. In RAM 303, various programs and data required for system operation are also stored. CPU 301, ROM 302 and RAM 303 are connected to each other through bus 304. Input / output (I / O) interface 305 is also connected to bus 304.

[0071] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a liquid crystal display (LCD) and an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read therefrom is installed into the storage section 308 as needed.

[0072] In particular, according to an embodiment of the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication part 309, and / or installed from a removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are performed.

[0073] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, apparatus, or device.

[0074] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram may represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the box may also occur in an order different from that marked in the accompanying drawings.

[0075] Specifically, the soft measurement system of this embodiment includes a processor and a memory. The memory stores a computer program. When the computer program is executed by the processor, the soft measurement method for the flue gas flow of a coal-fired boiler provided in the above embodiment is implemented.

[0076] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the soft measurement system described in the above embodiment; or may exist independently without being assembled into the soft measurement system. The above storage medium carries one or more computer programs, and when the above one or more computer programs are executed by a processor of the soft measurement system, the soft measurement system implements the soft measurement method for the flue gas flow of a coal-fired boiler provided in the above embodiment.

[0077] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0078] As used in the above embodiments, the term "when..." may be interpreted to mean "if..." or "after..." or "in response to determining..." or "in response to detecting...", depending on the context. Similarly, the phrases "upon determining..." or "if (the stated condition or event) is detected" may be interpreted to mean "if determining..." or "in response to determining..." or "upon detecting (the stated condition or event)" or "in response to detecting (the stated condition or event)", depending on the context.

[0079] Those skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by computer programs to instruct related hardware, and the programs can be stored in computer-readable storage media. When the programs are executed, they can include the processes of the above-mentioned method embodiments. The aforementioned storage media include: ROM or random access memory RAM, magnetic disk or optical disk and other media that can store program codes.

Claims

1. A soft measurement method for flue gas flow of a coal-fired boiler, applied to a soft measurement system, characterized in that: The method comprises: Collect historical equipment operation data of a set time period of a power plant distributed control system and a continuous emission monitoring system, wherein the historical equipment operation data includes boiler load, boiler total air volume, flue gas temperature, coal type and coal quality; Obtaining historical flue gas flow measured data measured at flue measuring points within different height ranges using high-precision equipment within the set time period, wherein the historical flue gas flow measured data at least includes flue gas flow velocity, temperature and pressure; Taking the historical flue gas flow measured data as the dependent variable, taking the historical equipment operation data as the covariate, and combining the dependent variable and the covariate to perform machine learning to determine the flue gas flow prediction model; After obtaining the current equipment operation data, the current equipment operation data is input into the smoke flow prediction model to obtain the smoke flow prediction data at different heights; The flue gas flow prediction data is sent to the management end for display.

2. The method according to claim 1, characterized in that After the step of inputting the current equipment operation data into the smoke flow prediction model to obtain the smoke flow prediction data at different heights, the method further includes: Obtain current flue gas flow measured data through high-precision equipment detection; Determine the prediction evaluation result by comparing the smoke flow prediction data with the current smoke flow measured data; The flue gas flow prediction model is adjusted according to the prediction evaluation result.

3. The method according to claim 1, characterized in that After the step of using the historical smoke flow measured data as the dependent variable, using the historical equipment operation data as the covariate, and combining the dependent variable and the covariate to perform machine learning to determine the smoke flow prediction model, the method further includes: Obtain environmental characteristic data in different seasons; Classifying boiler operating condition information of each season according to the environmental characteristic data, and establishing a seasonal operating condition characteristic library, wherein the seasonal operating condition characteristic library includes boiler operating condition characteristic parameter sets in spring, summer, autumn and winter; Based on the seasonal operating condition feature library, feature extraction and seasonal classification are performed on the historical equipment operation data to obtain a seasonal historical equipment operation data set; Using the seasonal historical operation data set as training data, adjusting the flue gas flow prediction model to obtain a seasonal new flue gas flow prediction model; When acquiring the current boiler operation data, the current season is determined according to the current date, and the flue gas flow prediction data is determined by a flue gas flow prediction model that matches the current season.

4. The method according to claim 3, characterized in that When obtaining the current boiler operation data, after the step of determining the current season according to the current date and determining the flue gas flow prediction data by a flue gas flow prediction model matching the current season, the method further includes: Determine the normal fluctuation range of flue gas flow in different seasons based on the seasonal historical equipment operation data set; Establishing an adaptive warning threshold based on the flue gas flow prediction data, wherein the adaptive warning threshold is dynamically adjusted according to the season and boiler operating conditions; When the flue gas flow prediction data exceeds the adaptive warning threshold, triggering a warning signal; The new flue gas flow prediction model is optimized and updated according to the frequency and degree of the early warning signal.

5. The method according to claim 1, characterized in that After the step of inputting the current equipment operation data into the smoke flow prediction model to obtain the smoke flow prediction data at different heights, the method further includes: Monitoring the change trend of the flue gas flow prediction data; When the flue gas flow forecast data is monitored to continue to rise or fall, determine whether there is an abnormal situation; If there is an abnormal situation, the height position information when the abnormal situation occurs in the flue is determined, and the height position information is sent to the client.

6. The method according to claim 1, characterized in that After the step of inputting the current equipment operation data into the smoke flow prediction model to obtain the smoke flow prediction data at different heights, the method further includes: Collect and gather working status data of dust removal equipment in the flue gas system; The smoke flow prediction result is corrected in combination with the working status data.

7. The method according to claim 1, characterized in that After the step of sending the smoke flow prediction data to the management terminal for display, the method further includes: When the accuracy of the prediction result is lower than the preset threshold, the model prediction alarm is output; Collect equipment operation data and corresponding flue gas flow measured data when the alarm is generated; The smoke flow prediction model is retrained using the equipment operation data at the time of the alarm and the corresponding smoke flow measured data.

8. A soft measurement system, characterized in that: The soft measurement system comprises: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code comprises computer instructions, and the one or more processors call the computer instructions so that the soft measurement system executes the method as described in any one of claims 1 to 7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instructions are executed on a soft measurement system, the soft measurement system is caused to execute the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product runs on a soft measurement system, the soft measurement system is enabled to perform the method according to any one of claims 1 to 7.

Images