Method and system for controlling steam temperature of waste incineration boiler based on hierarchical control strategy

Abstract

The invention discloses a waste incineration boiler steam temperature control method and system based on a hierarchical control strategy, and the method comprises the steps: collecting data through a high-precision sensor, and constructing a fault detection model for monitoring the system fault and abnormal conditions in real time; when the fault detection model detects a fault, an improved PID control algorithm considering mutual influence of characteristics is used to quickly respond to and adjust the instantaneous change of the steam temperature at the outlet of the waste heat boiler of the waste incineration power plant; constructing a medium-term prediction model by using the adjusted outlet steam temperature of the waste heat boiler of the waste incineration power plant, and predicting the relationship between the main steam flow and the outlet steam temperature of the boiler in real time; a long-term prediction model is constructed through historical data, and the large inertia effect and the long-term change trend of the waste heat boiler of the waste incineration power plant are predicted; and collecting and analyzing suggestions of a user on system operation, and dynamically adjusting a control strategy to improve system performance and user satisfaction. And load fluctuation and environment change can be adapted more quickly, and stable operation of the system is ensured.

Description

Technical Field

[0001] The invention relates to the technical field of boiler control technology, and in particular to a steam temperature control method and system for a garbage incineration boiler using a hierarchical control strategy. Background Art

[0002] As an important way of waste treatment, waste incineration power plants can not only effectively reduce the amount of waste, but also use the heat energy generated by waste combustion for power generation and heating. Therefore, how to improve the operating efficiency and stability of waste incineration power plants has become a research hotspot in related fields. As one of the core equipment of waste incineration power plants, the control of the outlet steam temperature of waste heat boilers is crucial. With the development of technology, traditional boiler temperature control methods have gradually evolved, covering from simple PID control to complex adaptive control and model predictive control. However, these methods still have many shortcomings in practical applications, especially in dealing with complex load changes and environmental interference. Traditional control strategies often find it difficult to achieve precise temperature control. In addition, with the improvement of sensor technology and data processing capabilities, the use of high-precision sensors for real-time data acquisition, combined with advanced algorithms for data analysis and fault detection, has gradually become an important means to improve system performance.

[0003] The existing technology has the following major deficiencies in the temperature control of waste incineration boilers. First, although the traditional PID control algorithm is widely used in industrial control, its single feedback mechanism is difficult to cope with complex multi-variable coupling and nonlinear dynamic characteristics, resulting in insufficient response speed and stability of the system when the load fluctuates and the environment changes. Secondly, most of the existing fault detection methods are based on single sensor data, lacking the ability to fuse and comprehensively analyze multi-sensor data, making it difficult to achieve early detection and prevention of system failures. In addition, traditional prediction models often ignore the large inertia effect and long-term change trend of the system, and cannot accurately predict and adjust the operating status of the boiler, resulting in increased energy consumption and increased equipment wear. Finally, the role of user feedback in traditional system optimization has been ignored, and there is a lack of dynamic adjustment mechanism based on user feedback, making it difficult to respond to user needs in a timely manner and optimize system performance. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problems solved by the present invention are: the traditional PID control algorithm has the problems of slow response speed and insufficient stability when dealing with complex load fluctuations and environmental changes, and the traditional prediction model of multi-sensor data fusion and fault detection often ignores the large inertia effect and long-term change trend of the system, and cannot accurately predict and adjust the operating status of the boiler.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a steam temperature control method for a garbage incineration boiler with a hierarchical control strategy, comprising:

[0007] Use high-precision sensors to collect data and build fault detection models to monitor system faults and abnormal conditions in real time;

[0008] When the fault detection model detects a fault, an improved PID control algorithm that considers the mutual influence of characteristics is used to quickly respond to and adjust the instantaneous change of the outlet steam temperature of the waste heat boiler of the waste incineration power plant;

[0009] A medium-term prediction model is constructed using the adjusted outlet steam temperature of the waste heat boiler of the waste incineration power plant to predict the relationship between the main steam flow and the outlet steam temperature of the boiler in real time;

[0010] Build a long-term prediction model through historical data to predict the large inertia effect and long-term change trend of waste heat boilers in waste incineration power plants;

[0011] Collect and analyze user suggestions on system operation, and dynamically adjust control strategies to improve system performance and user satisfaction.

[0012] As a preferred solution of the steam temperature control method of a garbage incineration boiler of the hierarchical control strategy described in the present invention, the data includes steam temperature data, steam pressure data, steam flow data, combustion condition data in the boiler, environmental parameter data, fault detection data, user feedback data and historical data;

[0013] High-precision sensors include RTD temperature sensors, pressure transmitters, piezoresistive pressure sensors, vortex flowmeters, electromagnetic flowmeters, flame detectors, flue gas analyzers, thermopile sensors, environmental sensors, vibration sensors, leakage detectors and user feedback collection devices.

[0014] As a preferred solution of the steam temperature control method of a waste incineration boiler of the hierarchical control strategy described in the present invention, the construction of a fault detection model to monitor system faults and abnormal conditions in real time includes combining data collected by an RTD temperature sensor, a pressure transmitter, a piezoresistive pressure sensor and a vibration sensor, using a naive leaf Bayes classifier to construct a real-time fault detection model to predict in real time whether there are faults and abnormalities;

[0015] Feature extraction is performed on each sensor data, and the integration function and normalization function are used to generate a comprehensive feature vector. The formula is expressed as:

[0016]

[0017] The eigenvectors are combined into a comprehensive eigenmatrix, and the formula is expressed as;

[0018]

[0019] Assuming that the features of the comprehensive feature matrix X are independent, a multivariate normal distribution is used to represent each category C k And introduce the real-time fault detection model of normalized function;

[0020] The Naive Bayes classifier assumes that the features are independent of each other:

[0021]

[0022] Assume that features are independent:

[0023]

[0024] Where, T(t) represents the temperature data collected by the RTD temperature sensor, P(t) represents the pressure data collected by the pressure transmitter, R(t) represents the pressure data collected by the piezoresistive pressure sensor, V(t) represents the vibration data collected by the vibration sensor, and μ T represents the average value of temperature data, represents the variance of temperature data, μ P represents the average value of pressure data, represents the variance of pressure data, μ R represents the average value of the piezoresistive pressure data, represents the variance of the piezoresistive pressure data, μ V represents the average value of vibration data, represents the variance of vibration data, X represents the comprehensive feature matrix, P(C k |X) indicates that category C is given feature X. k The posterior probability, P(X|C k ) indicates that in category C k The likelihood of feature X, P(C k ) indicates category C k The prior probability, μ k Indicates category C k The mean vector of k Indicates category C k The covariance matrix of , d represents the feature dimension, K represents the total number of categories, P(X| C k =Faul) represents the likelihood of observing the feature vector X when the system is in a faulty state, P(C k =Faul) represents the prior probability that the system is in a fault state;

[0025] A threshold θ is calculated based on the median and standard deviation of historical data. When the posterior probability P(Fault|X) of the fault category exceeds the threshold θ, a fault alarm is triggered. The formula is expressed as:

[0026] θ=α·median(P(C k =Fault|X history ))+β·σ(P(C k =Fault|X history ))

[0027] Among them, P(C k =Fault|X history ) represents the posterior probability of the fault category under the given feature X, θ represents the fault detection threshold, α and β represent the threshold adjustment weight parameters, median(P(C k =Fault|X history )) is the median of historical failure data, σ(P(C k =Fault|X history )) represents the standard deviation of historical fault data, X represents the comprehensive feature matrix, ∑ Fault represents the covariance matrix of the fault category, μ Fault represents the mean vector of fault categories, K represents the total number of categories;

[0028] The threshold of θ is between 0 and 1, P(C k =Faul|X)>θ when a fault is detected.

[0029] As a preferred solution of the steam temperature control method of a waste incineration boiler of the hierarchical control strategy described in the present invention, when the fault detection model detects a fault, an improved PID control algorithm that considers the mutual influence of features is used to quickly respond to and adjust the instantaneous change of the outlet steam temperature of the waste heat boiler of the waste incineration power plant, including using a real-time fault detection model to perform fault detection on the collected features, and if a fault is detected, triggering the control strategy;

[0030] When the fault detection model detects a fault, an improved PID control algorithm that considers the mutual influence of characteristics is used to adjust the boiler outlet steam temperature;

[0031] The improved PID control algorithm is used to obtain a complete function for adjusting the boiler outlet steam temperature;

[0032] When P(C k =Faul|X)>θ, the improved PID control algorithm considering the mutual influence of characteristics is triggered:

[0033]

[0034] Where u(t) represents the control signal of the system at time t, K prepresents the proportional gain coefficient, X represents the comprehensive characteristic matrix, e(t) represents the difference between the set value and the actual measured value at time t, K i Indicates the integral gain coefficient, K d represents the differential gain coefficient;

[0035] The value threshold of u(t) is between 0 and 1. When 0≤u(t)<0.5, the heater is turned off;

[0036] When 0.5≤u(t)<1, the heater operates at 50% power;

[0037] When u(t)=1, the heater operates at maximum power.

[0038] As a preferred solution of the steam temperature control method of the waste incineration boiler of the hierarchical control strategy described in the present invention, the medium-term prediction model is constructed by using the adjusted outlet steam temperature of the waste heat boiler of the waste incineration power plant, and the real-time prediction of the relationship between the main steam flow rate and the boiler outlet steam temperature includes constructing a medium-term prediction model by combining the data of the main steam flow rate and the boiler outlet steam temperature through a multivariate nonlinear regression model, and the formula is expressed as follows:

[0039]

[0040] Wherein, Q(t) represents the main steam flow at time t, T(t) represents the boiler outlet steam temperature at time t, and α, β, γ, δ, ∈, η represent model parameters.

[0041] As a preferred solution of the steam temperature control method of the waste incineration boiler of the hierarchical control strategy described in the present invention, a long-term prediction model is constructed through historical data to predict the large inertia effect and long-term change trend of the waste heat boiler of the waste incineration power plant, including designing a long-term prediction model combining time series and cumulative effect in combination with the boiler outlet steam temperature and historical temperature data, and the formula is expressed as follows:

[0042]

[0043] in, represents the predicted boiler outlet steam temperature at time t, α represents a constant term, φ k represents the coefficient of the autoregressive model, T(tk) represents the temperature data at the kth moment in the past, θ k represents the coefficients of the moving average model, ∈ t -k represents the error term at the kth moment in the past, β represents the coefficient of the integral term, δ represents the coefficient of the nonlinear cumulative term, and η is the constant term;

[0044] The value threshold is greater than 0.

[0045] As a preferred solution of the steam temperature control method of a waste incineration boiler with a hierarchical control strategy described in the present invention, wherein: collecting and analyzing user suggestions on system operation, dynamically adjusting the control strategy to improve system performance and user satisfaction includes collecting user feedback in real time and performing classification and statistical analysis, and triggering a multi-level user feedback system when the system detects that multiple user feedbacks are concentrated on the same issue and the feedback score of the issue is lower than 4.0 / 5;

[0046] The multi-level user feedback system includes dividing the priority assessment of user feedback into user satisfaction score, problem severity, feedback frequency and environmental urgency, assigning different weights to each, and then comprehensively calculating the priority score of the feedback;

[0047] The user satisfaction score is expressed as:

[0048]

[0049] Among them, R i represents the satisfaction score of the i-th user, ranging from 1 to 5, N represents the total number of users who provide feedback, and S u represents the user satisfaction score;

[0050] The severity of the problem is expressed as:

[0051]

[0052] Among them, X j represents the score of the jth influencing factor, ranging from 1 to 10, w sj Indicates the weight of the factor, M indicates the total number of influencing factors, I s Indicates the severity of the problem;

[0053] Feedback frequency reflects the number of times the same issue is reported by different users. The more times, the higher the priority. Feedback frequency is expressed as:

[0054] F f =log(1+F)

[0055] Where F represents the total number of feedbacks, F f Indicates the feedback frequency;

[0056] The degree of environmental urgency is expressed as:

[0057]

[0058] Among them, Y k represents the urgency score of the kth environmental factor, ranging from 1 to 10, K represents the total number of environmental factors, E e Indicates the degree of environmental urgency;

[0059] The final priority scoring formula is expressed as:

[0060] Priority Score = w1·S u +w2.I s +w3.F f +w4.E e

[0061] Among them, Priority Score represents the comprehensive priority score, w1 represents the weight of the user satisfaction score, S u represents the user satisfaction score, N represents the total number of users who provide feedback, and R i represents the satisfaction score of the i-th user, w2 represents the weight of the problem severity, and I s Indicates the severity of the problem, M indicates the total number of influencing factors, X j represents the score of the jth influencing factor, w sj represents the weight of this factor, w3 represents the weight of the feedback frequency, F f represents the feedback frequency, F represents the total number of feedbacks, w4 represents the weight of the environmental urgency, E e represents the degree of environmental urgency, K represents the total number of environmental factors, Y k represents the urgency score of the kth environmental factor;

[0062] The Priority Score value threshold is 1 to 10;

[0063] When Priority Score>8, it is a high priority feedback. The system will respond within 24 hours and start to solve the high priority feedback problem.

[0064] When 5<Priority Score≤8, it is medium priority feedback. The system will respond and start solving medium priority feedback issues within 3 days.

[0065] When Priority Score ≤ 5, it is low priority feedback. The system will review and start to resolve low priority feedback issues within one week.

[0066] The steam temperature control system of the garbage incineration boiler with hierarchical control strategy includes:

[0067] The safety monitoring layer uses high-precision sensors to collect data and build fault detection models to monitor system failures and abnormal conditions in real time;

[0068] The fast response layer, when the fault detection model detects a fault, uses an improved PID control algorithm that takes into account the mutual influence of characteristics to quickly respond to and adjust the instantaneous change of the outlet steam temperature of the waste heat boiler of the waste incineration power plant;

[0069] The mid-term response layer uses the adjusted outlet steam temperature of the waste heat boiler of the waste incineration power plant to build a mid-term prediction model to predict the relationship between the main steam flow and the outlet steam temperature of the boiler in real time;

[0070] The long-term response layer builds a long-term prediction model through historical data to predict the large inertia effect and long-term change trend of waste heat boilers in waste incineration power plants;

[0071] The user feedback layer collects and analyzes users’ suggestions on system operation and dynamically adjusts control strategies to improve system performance and user satisfaction.

[0072] A computer device comprises: a memory and a processor; the memory stores a computer program, wherein the processor implements the steps of any one of the methods of the present invention when executing the computer program.

[0073] A computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of any one of the methods of the present invention.

[0074] Beneficial effects of the present invention: The steam temperature control method of the garbage incineration boiler with a hierarchical control strategy provided by the present invention improves the accuracy of temperature control by using an improved PID control algorithm, and also significantly improves the response speed of the system, and can adapt to load fluctuations and environmental changes more quickly to ensure the stable operation of the system. It improves the accuracy and timeliness of fault detection, reduces energy consumption, and increases the life of the equipment. By optimizing the control strategy and improving the temperature control accuracy, the energy consumption and emissions of the boiler system are reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0076] Figure 1 An overall flow chart of a method for controlling steam temperature of a garbage incineration boiler using a hierarchical control strategy provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0077] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0078] Example 1

[0079] Reference Figure 1 , which is an embodiment of the present invention, provides a steam temperature control method for a garbage incineration boiler with a hierarchical control strategy, comprising:

[0080] S1: Use high-precision sensors to collect data and build fault detection models to monitor system faults and abnormal conditions in real time.

[0081] The data include steam temperature data, steam pressure data, steam flow data, boiler combustion data, environmental parameter data, fault detection data, user feedback data and historical data;

[0082] High-precision sensors include RTD temperature sensors, pressure transmitters, piezoresistive pressure sensors, vortex flowmeters, electromagnetic flowmeters, flame detectors, flue gas analyzers, thermopile sensors, environmental sensors, vibration sensors, leakage detectors and user feedback collection devices.

[0083] Building a fault detection model to monitor system faults and abnormalities in real time includes combining the data collected by the RTD temperature sensor, pressure transmitter, piezoresistive pressure sensor and vibration sensor, using the Naive Leaf Bayes classifier to build a real-time fault detection model to predict in real time whether there are faults and abnormalities;

[0084] Feature extraction is performed on each sensor data, and the integration function and normalization function are used to generate a comprehensive feature vector. The formula is expressed as:

[0085]

[0086] The eigenvectors are combined into a comprehensive eigenmatrix, and the formula is expressed as;

[0087]

[0088] Assuming that the features of the comprehensive feature matrix X are independent, a multivariate normal distribution is used to represent each category C k And introduce the real-time fault detection model of normalized function;

[0089] The Naive Bayes classifier assumes that the features are independent of each other:

[0090]

[0091] Assume that features are independent:

[0092]

[0093] Where, T(t) represents the temperature data collected by the RTD temperature sensor, P(t) represents the pressure data collected by the pressure transmitter, R(t) represents the pressure data collected by the piezoresistive pressure sensor, V(t) represents the vibration data collected by the vibration sensor, and μ T represents the average value of temperature data, represents the variance of temperature data, μ P represents the average value of pressure data, represents the variance of pressure data, μ R represents the average value of the piezoresistive pressure data, represents the variance of the piezoresistive pressure data, μ V represents the average value of vibration data, represents the variance of vibration data, X represents the comprehensive feature matrix, P(C k |X) indicates that category C is given feature X. k The posterior probability, P(X|C k ) indicates that in category C k The likelihood of feature X, P(C k ) indicates category C k The prior probability, μ k Indicates category C k The mean vector of k Indicates category C k The covariance matrix of , d represents the feature dimension, K represents the total number of categories, P(X| C k =Faul) represents the likelihood of observing the feature vector X when the system is in a faulty state, P(C k =Faul) represents the prior probability that the system is in a fault state;

[0094] A threshold θ is calculated based on the median and standard deviation of historical data. When the posterior probability P(Fault|X) of the fault category exceeds the threshold θ, a fault alarm is triggered. The formula is expressed as:

[0095] θ=α·median(P(C k =Fault|X history ))+β·σ(P(C k =Fault|x history ))

[0096] Among them, P(C k =Fault|X history) represents the posterior probability of the fault category under the given feature X, θ represents the fault detection threshold, α and β represent the threshold adjustment weight parameters, median(P(C k =Fault|X history )) is the median of historical failure data, σ(P(C k =Fault|X history )) represents the standard deviation of historical fault data, X represents the comprehensive feature matrix, ∑ Fault represents the covariance matrix of the fault category, μ Fault represents the mean vector of fault categories, K represents the total number of categories;

[0097] The threshold of θ is between 0 and 1, P(C k =Faul|X)>θ when a fault is detected.

[0098] By collecting and processing a variety of sensor data in real time, we can fully understand the operating status of the boiler and improve the accuracy and response speed of fault detection. This ensures the comprehensiveness and accuracy of the data and provides a reliable data basis for subsequent fault detection and temperature control. The application of high-precision sensors improves the accuracy of data collection, and data fusion technology improves the effectiveness of the fault detection model, which can detect potential problems in time before a fault occurs, thereby reducing system downtime and maintenance costs. By extracting and comprehensively analyzing the features of sensor data, a comprehensive feature vector can be generated, which facilitates the processing and analysis of complex data and further improves the accuracy of fault detection.

[0099] S2: When the fault detection model detects a fault, an improved PID control algorithm that considers the mutual influence of characteristics is used to quickly respond to and adjust the instantaneous change of the outlet steam temperature of the waste heat boiler of the waste incineration power plant.

[0100] Using the real-time fault detection model, the collected features are used for fault detection. If a fault is detected, the control strategy is triggered.

[0101] When the fault detection model detects a fault, an improved PID control algorithm that considers the mutual influence of characteristics is used to adjust the boiler outlet steam temperature;

[0102] The improved PID control algorithm is used to obtain a complete function for adjusting the boiler outlet steam temperature;

[0103] When P(C k =Faul|X)>θ, the improved PID control algorithm considering the mutual influence of characteristics is triggered:

[0104]

[0105] Where u(t) represents the control signal of the system at time t, Kp represents the proportional gain coefficient, X represents the comprehensive characteristic matrix, e(t) represents the difference between the set value and the actual measured value at time t, K i Indicates the integral gain coefficient, K d represents the differential gain coefficient;

[0106] The value threshold of u(t) is between 0 and 1. When 0≤u(t)<0.5, the heater is turned off;

[0107] When 0.5≤u(t)<1, the heater operates at 50% power;

[0108] When u(t)=1, the heater operates at maximum power.

[0109] When a fault occurs, rapid response is the key to ensure stable operation of the system. The improved PID control algorithm improves control accuracy and response speed by comprehensively considering the mutual influence of various characteristics, thereby effectively maintaining the stability of steam temperature. Traditional PID control algorithms have problems such as slow response speed and insufficient control accuracy when dealing with complex system dynamic changes. The improved PID control algorithm can better cope with instantaneous temperature changes, achieve rapid adjustment and response, avoid the impact of temperature fluctuations on system operation, and improve system stability and safety.

[0110] S3: Use the adjusted outlet steam temperature of the waste heat boiler of the waste incineration power plant to build a medium-term prediction model to predict the relationship between the main steam flow and the outlet steam temperature of the boiler in real time.

[0111] A medium-term prediction model is constructed by combining the data of main steam flow and boiler outlet steam temperature with a multivariate nonlinear regression model. The formula is expressed as follows:

[0112]

[0113] Wherein, Q(t) represents the main steam flow at time t, T(t) represents the boiler outlet steam temperature at time t, and α, β, γ, δ, ∈, η represent model parameters.

[0114] The medium-term prediction model can accurately predict and adjust the relationship between the main steam flow and the boiler outlet steam temperature to ensure the stable operation of the system in the medium term. This is especially important for systems with large load fluctuations, which helps to optimize resource allocation and improve system efficiency. The construction of the medium-term prediction model is based on the analysis of the adjusted temperature data, which can predict the steam flow and temperature change trends in the future in advance, thereby providing a reliable basis for system regulation. Through real-time prediction and adjustment, temperature fluctuations can be effectively reduced and the operating efficiency and stability of the system can be improved.

[0115] S4: Build a long-term prediction model through historical data to predict the large inertia effect and long-term change trend of waste heat boilers in waste incineration power plants.

[0116] Combining the boiler outlet steam temperature and historical temperature data, a long-term prediction model combining time series and cumulative effects is designed. The formula is expressed as:

[0117]

[0118] in, represents the predicted boiler outlet steam temperature at time t, α represents a constant term, φ k represents the coefficient of the autoregressive model, T(tk) represents the temperature data at the kth moment in the past, θ k represents the coefficients of the moving average model, ∈ t -k represents the error term at the kth moment in the past, β represents the coefficient of the integral term, δ represents the coefficient of the nonlinear cumulative term, and η is the constant term;

[0119] The value threshold is greater than 0.

[0120] The long-term prediction model helps predict the long-term change trend and large inertia effect of the system, and provides a forward-looking judgment on the future operating status. It is of great significance for formulating long-term operation strategies and optimizing system performance. The long-term prediction model is based on historical data and can identify the long-term change laws and trends of the system through time series analysis methods. By predicting long-term changes, more scientific maintenance plans and operation strategies can be formulated to avoid performance degradation and failures caused by inertia effects and long-term changes in the system, and improve the reliability and operation efficiency of the system.

[0121] S5: Collect and analyze user suggestions on system operation, and dynamically adjust control strategies to improve system performance and user satisfaction.

[0122] Collect user feedback in real time and perform classification and statistical analysis. When the system detects that multiple user feedbacks are concentrated on the same issue and the feedback score of the issue is lower than 4.0 / 5, the multi-level user feedback system is triggered;

[0123] The multi-level user feedback system includes dividing the priority assessment of user feedback into user satisfaction score, problem severity, feedback frequency and environmental urgency, assigning different weights to each, and then comprehensively calculating the priority score of the feedback;

[0124] The user satisfaction score is expressed as:

[0125]

[0126] Among them, R irepresents the satisfaction score of the i-th user, ranging from 1 to 5, N represents the total number of users who provide feedback, and S u represents the user satisfaction score;

[0127] The severity of the problem is expressed as:

[0128]

[0129] Among them, X j represents the score of the jth influencing factor, ranging from 1 to 10, w sj Indicates the weight of the factor, M indicates the total number of influencing factors, I s Indicates the severity of the problem;

[0130] Feedback frequency reflects the number of times the same issue is reported by different users. The more times, the higher the priority. Feedback frequency is expressed as:

[0131] F f =log(1+F)

[0132] Among them, F represents the total number of feedbacks, F f Indicates the feedback frequency;

[0133] The degree of environmental urgency is expressed as:

[0134]

[0135] Among them, Y k represents the urgency score of the kth environmental factor, ranging from 1 to 10, K represents the total number of environmental factors, E e Indicates the degree of environmental urgency;

[0136] The final priority scoring formula is expressed as:

[0137] Priority Score = w1·S u +w2.I s +w3.F f +w4.E e

[0138] Among them, Priority Score represents the comprehensive priority score, w1 represents the weight of the user satisfaction score, S u represents the user satisfaction score, N represents the total number of users who provide feedback, and R i represents the satisfaction score of the i-th user, w2 represents the weight of the problem severity, and I s Indicates the severity of the problem, M indicates the total number of influencing factors, X j represents the score of the jth influencing factor, w sj represents the weight of this factor, w3 represents the weight of the feedback frequency, Ff represents the feedback frequency, F represents the total number of feedbacks, w4 represents the weight of the environmental urgency, E e represents the degree of environmental urgency, K represents the total number of environmental factors, Y k represents the urgency score of the kth environmental factor;

[0139] The value threshold of Priority Score is 1 to 10;

[0140] When Priority Score>8, it is a high priority feedback. The system will respond within 24 hours and start to solve the high priority feedback problem.

[0141] When 5<Priority Score≤8, it is medium priority feedback. The system will respond and start solving medium priority feedback issues within 3 days.

[0142] When Priority Score ≤ 5, it is low priority feedback. The system will review and start to resolve low priority feedback issues within one week.

[0143] User feedback is an important basis for system optimization. By dynamically adjusting the control strategy, not only can the system performance be improved, but also user satisfaction can be improved. This ensures the adaptability and flexibility of the system in actual operation, and can be continuously optimized according to user needs. The collection and analysis of user feedback data can help identify problems and deficiencies in system operation, and by calculating feedback priorities, resources can be reasonably allocated to prioritize solving important and urgent problems. Dynamically adjusting the control strategy enables the system to be continuously optimized according to user needs and actual operation conditions, improving system operation efficiency and user satisfaction.

[0144] On the other hand, this embodiment also provides a waste incineration boiler steam temperature control system with a hierarchical control strategy, which includes:

[0145] In the fast response layer, when the fault detection model detects a fault, an improved PID control algorithm that considers the mutual influence of characteristics is used to quickly respond to and adjust the instantaneous change of the outlet steam temperature of the waste heat boiler of the waste incineration power plant.

[0146] The medium-term response layer uses the adjusted outlet steam temperature of the waste heat boiler of the waste incineration power plant to build a medium-term prediction model to predict the relationship between the main steam flow and the outlet steam temperature of the boiler in real time.

[0147] The long-term response layer builds a long-term prediction model through historical data to predict the large inertia effect and long-term change trend of the waste heat boiler of the waste incineration power plant.

[0148] The user feedback layer collects and analyzes users’ suggestions on system operation and dynamically adjusts control strategies to improve system performance and user satisfaction.

[0149] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0150] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0151] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0152] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0153] Example 2

[0154] Experiments were conducted in a waste incineration power plant. During the operation of the system, high-precision sensors were used to collect real-time data. The data collection frequency was set to once per second to ensure that the slight changes in the system could be captured. The data collection time was one month to ensure the representativeness and integrity of the data. The collected data was input into the fault detection model through the data acquisition system to monitor system faults and abnormalities in real time. The fault detection model is based on multi-sensor data fusion technology and can detect abnormalities in multi-dimensional data such as temperature, pressure, vibration and flow. Once an abnormality is detected, the fault detection model immediately records the time and type of fault. When the fault detection model detects a fault, the improved PID control algorithm is immediately started. The parameters of the PID controller are optimized based on historical data to ensure that it can achieve the best control effect under various operating conditions. The medium-term prediction model is constructed using the adjusted waste heat boiler outlet steam temperature data. The medium-term prediction model uses a machine learning algorithm to predict the relationship between the main steam flow and the boiler outlet steam temperature in real time. The long-term prediction model is constructed based on historical data to predict the large inertia effect and long-term change trend of the waste heat boiler system. The long-term prediction model uses time series analysis technology to accurately predict the long-term operating status of the system. Collect feedback from operators and maintenance personnel, analyze feedback data, and dynamically adjust control strategies. User feedback data is collected through questionnaires and system logs. The experimental data is shown in Table 1.

[0155] Table 1 Experimental data table

[0156]

[0157] The adjusted boiler outlet steam temperature is stable between 350°C and 360°C, with a small fluctuation range, indicating that the improved PID control algorithm of the present invention effectively maintains the stability of the steam temperature. Compared with the traditional control method, the temperature fluctuation is smaller and the control accuracy is higher. The pressure change data is between 5.1MPa and 5.3MPa, showing good stability. This shows that the system can maintain the stability of the pressure while adjusting the steam temperature, avoiding pressure fluctuations caused by temperature changes. The vibration amplitude data are all lower than 0.3mm / s, indicating that the vibration amplitude of the system is very small during operation and the equipment runs smoothly. This is attributed to the effectiveness of the fault detection model, which can detect and handle faults in time and reduce equipment vibration caused by faults. The flow change data is between 12.4t / h and 12.6t / h, indicating that the medium-term prediction model can accurately predict and adjust the relationship between the main steam flow and the boiler outlet steam temperature, ensuring the stable operation of the system. The fault response time data is between 1.8 seconds and 2.2 seconds, showing that the system can respond quickly and adjust the boiler outlet steam temperature. This reflects the advantages of the present invention in fast fault detection and response.

[0158] The user satisfaction scores are all above 4.7 points, indicating that users are very satisfied with the system operation effect and fault handling capabilities. This reflects the effectiveness of the dynamic adjustment mechanism based on user feedback in improving system performance and user satisfaction.

[0159] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A steam temperature control method for a garbage incineration boiler based on a hierarchical control strategy, characterized in that: include: Use high-precision sensors to collect data and build fault detection models to monitor system faults and abnormal conditions in real time; When the fault detection model detects a fault, an improved PID control algorithm that considers the mutual influence of characteristics is used to quickly respond to and adjust the instantaneous change of the outlet steam temperature of the waste heat boiler of the waste incineration power plant; A medium-term prediction model is constructed using the adjusted outlet steam temperature of the waste heat boiler of the waste incineration power plant to predict the relationship between the main steam flow and the outlet steam temperature of the boiler in real time; Build a long-term prediction model through historical data to predict the large inertia effect and long-term change trend of waste heat boilers in waste incineration power plants; Collect and analyze user suggestions on system operation, and dynamically adjust control strategies to improve system performance and user satisfaction.

2. The method for controlling steam temperature of a garbage incineration boiler with a hierarchical control strategy as claimed in claim 1, characterized in that: The data include steam temperature data, steam pressure data, steam flow data, boiler combustion data, environmental parameter data, fault detection data, user feedback data and historical data; High-precision sensors include RTD temperature sensors, pressure transmitters, piezoresistive pressure sensors, vortex flowmeters, electromagnetic flowmeters, flame detectors, flue gas analyzers, thermopile sensors, environmental sensors, vibration sensors, leakage detectors and user feedback collection devices.

3. The method for controlling steam temperature of a garbage incineration boiler using a hierarchical control strategy as claimed in claim 2, characterized in that: Building a fault detection model to monitor system faults and abnormalities in real time includes combining the data collected by the RTD temperature sensor, pressure transmitter, piezoresistive pressure sensor and vibration sensor, using the Naive Leaf Bayes classifier to build a real-time fault detection model to predict in real time whether there are faults and abnormalities; Feature extraction is performed on each sensor data, and the integration function and normalization function are used to generate a comprehensive feature vector. The formula is expressed as: The eigenvectors are combined into a comprehensive eigenmatrix, and the formula is expressed as; Assuming that the features of the comprehensive feature matrix X are independent, a multivariate normal distribution is used to represent each category C k And introduce the real-time fault detection model of normalized function; The Naive Bayes classifier assumes that the features are independent of each other: Assume that features are independent: Where, T(t) represents the temperature data collected by the RTD temperature sensor, P(t) represents the pressure data collected by the pressure transmitter, R(t) represents the pressure data collected by the piezoresistive pressure sensor, V(t) represents the vibration data collected by the vibration sensor, and μ T represents the average value of temperature data, represents the variance of temperature data, μ P represents the average value of pressure data, represents the variance of pressure data, μ R represents the average value of the piezoresistive pressure data, represents the variance of the piezoresistive pressure data, μ V represents the average value of vibration data, represents the variance of vibration data, X represents the comprehensive feature matrix, P(C k |X) indicates that category C is given feature X. k The posterior probability, P(X|C k ) indicates that in category C k The likelihood of feature X, P(C k ) indicates category C k The prior probability, μ k Indicates category C k The mean vector of k Indicates category C k The covariance matrix of , d represents the feature dimension, K represents the total number of categories, P(X|C k =Faul) represents the likelihood of observing the feature vector X when the system is in a faulty state, P(C k =Faul) represents the prior probability that the system is in a fault state; A threshold θ is calculated based on the median and standard deviation of historical data. When the posterior probability P(Fault|X) of the fault category exceeds the threshold θ, a fault alarm is triggered. The formula is expressed as: θXα·median(P(C). k NFault|X history ))+β·σ(P(C). k NFault|X history )) Among them, P(C k =Fault|X history ) represents the posterior probability of the fault category under the given feature X, θ represents the fault detection threshold, α and β represent the threshold adjustment weight parameters, median(P(C k =Fault|X history )) is the median of historical failure data, σ(P(C k =Fault|X history )) represents the standard deviation of historical fault data, X represents the comprehensive feature matrix, Σ Fault represents the covariance matrix of the fault category, μ Fault represents the mean vector of fault categories, K represents the total number of categories; The threshold of θ is between 0 and 1, P(C k =Faul|X)>θ when a fault is detected.

4. The method for controlling steam temperature of a garbage incineration boiler with a hierarchical control strategy as claimed in claim 3 is characterized in that: When the fault detection model detects a fault, an improved PID control algorithm that considers the mutual influence of features is used to quickly respond to and adjust the instantaneous change of the outlet steam temperature of the waste heat boiler of the waste incineration power plant. This includes using the real-time fault detection model to perform fault detection on the collected features, and if a fault is detected, triggering the control strategy; When the fault detection model detects a fault, an improved PID control algorithm that considers the mutual influence of characteristics is used to adjust the boiler outlet steam temperature; The improved PID control algorithm is used to obtain a complete function for adjusting the boiler outlet steam temperature; When P(C k =Faul|X)>θ, the improved PID control algorithm considering the mutual influence of characteristics is triggered: Where u(t) represents the control signal of the system at time t, K p represents the proportional gain coefficient, X represents the comprehensive characteristic matrix, e(t) represents the difference between the set value and the actual measured value at time t, K i Indicates the integral gain coefficient, K d represents the differential gain coefficient; The value threshold of u(t) is between 0 and 1. When 0≤u(t)<0.5, the heater is turned off; When 0.5≤u(t)<1, the heater operates at 50% power; When u(t)=1, the heater operates at maximum power.

5. The method for controlling steam temperature of a garbage incineration boiler with a hierarchical control strategy as claimed in claim 4, characterized in that: A medium-term prediction model is constructed using the adjusted steam temperature at the outlet of the waste incineration power plant's waste heat boiler to predict the relationship between the main steam flow rate and the steam temperature at the boiler outlet in real time. Specifically, a medium-term prediction model is constructed by combining the data of the main steam flow rate and the steam temperature at the boiler outlet through a multivariable nonlinear regression model, and the formula is expressed as: Where, Q(t) represents the main steam flow rate at time t, T(t) represents the steam temperature at the boiler outlet at time t, and α, β, γ, δ, ∈, η represent model parameters.

6. The method for controlling steam temperature of a garbage incineration boiler using a hierarchical control strategy as claimed in claim 5, characterized in that: A long-term prediction model is constructed through historical data to predict the large inertia effect and long-term change trend of the waste heat boiler of the waste incineration power plant. Specifically, a long-term prediction model combining time series and cumulative effect is designed by combining the steam temperature at the boiler outlet and historical temperature data, and the formula is expressed as: in, represents the predicted boiler outlet steam temperature at time t, α represents a constant term, φ k represents the coefficient of the autoregressive model, T(tk) represents the temperature data at the kth moment in the past, θ k represents the coefficients of the moving average model, ∈ t -k represents the error term at the kth moment in the past, β represents the coefficient of the integral term, δ represents the coefficient of the nonlinear cumulative term, and η is the constant term; The value threshold is greater than 0.

7. The method for controlling steam temperature of a garbage incineration boiler with a hierarchical control strategy as claimed in claim 6, characterized in that: Collect and analyze the suggestions of users on the system operation, and dynamically adjust the control strategy to improve the system performance and user satisfaction. Specifically, collect user feedback in real time, classify and statistically analyze it. When the system detects that multiple user feedbacks focus on the same problem and the feedback score of this problem is lower than 4.0 / 5, trigger a multi-level user feedback system; The multi-level user feedback system includes evaluating the priority of user feedback in terms of user satisfaction score, problem severity, feedback frequency, and environmental urgency, assigning different weights respectively, and then comprehensively calculating the priority score of the feedback; The user satisfaction score is expressed as: Among them, R i represents the satisfaction score of the i-th user, ranging from 1 to 5, N represents the total number of users who provide feedback, and S u represents the user satisfaction score; The problem severity is expressed as: Among them, X j represents the score of the jth influencing factor, ranging from 1 to 10, w sj Indicates the weight of the factor, M indicates the total number of influencing factors, I s Indicates the severity of the problem; The feedback frequency reflects the number of times the same problem is feedback by different users. The more times, the higher the priority. The feedback frequency is expressed as: F f =log(1+F) Among them, F represents the total number of feedbacks, F f Indicates the feedback frequency; The environmental urgency is expressed as: Among them, Y k represents the urgency score of the kth environmental factor, ranging from 1 to 10, K represents the total number of environmental factors, E e Indicates the degree of environmental urgency; The formula for the final priority score is expressed as: Priority Score=w1·S u +w2.I s +w3.F f +w4.E e Among them, Priority Score represents the comprehensive priority score, w1 represents the weight of the user satisfaction score, S u represents the user satisfaction score, N represents the total number of users who provide feedback, and R i represents the satisfaction score of the i-th user, w2 represents the weight of the problem severity, and I s Indicates the severity of the problem, M indicates the total number of influencing factors, X j represents the score of the jth influencing factor, w sj represents the weight of this factor, w3 represents the weight of the feedback frequency, F f represents the feedback frequency, F represents the total number of feedbacks, w4 represents the weight of the environmental urgency, E e represents the degree of environmental urgency, K represents the total number of environmental factors, Y k represents the urgency score of the kth environmental factor; The value range of the Priority Score is from 1 to 10; When Priority Score > 8, it is a high-priority feedback, and the system responds within 24 hours and starts to solve the high-priority feedback problem; When 5 < Priority Score ≤ 8, it is a medium-priority feedback, and the system responds within 3 days and starts to solve the medium-priority feedback problem; When Priority Score ≤ 5, it is a low-priority feedback, and the system reviews and starts to solve the low-priority feedback problem within a week.

8. A waste incineration boiler steam temperature control system adopting a hierarchical control strategy as described in any one of claims 1-7, characterized in that: The safety monitoring layer collects data using high-precision sensors and constructs a fault detection model to monitor system faults and abnormal conditions in real time; The fast response layer, when the fault detection model detects a fault, uses an improved PID control algorithm considering the mutual influence of characteristics to quickly respond to and adjust the instantaneous change of the steam temperature at the outlet of the waste heat boiler of the waste incineration power plant; The medium-term response layer constructs a medium-term prediction model using the adjusted steam temperature at the outlet of the waste heat boiler of the waste incineration power plant to predict the relationship between the main steam flow rate and the steam temperature at the boiler outlet in real time; The long-term response layer constructs a long-term prediction model through historical data to predict the large inertia effect and long-term change trend of the waste heat boiler of the waste incineration power plant; The user feedback layer collects and analyzes users’ suggestions on system operation and dynamically adjusts control strategies to improve system performance and user satisfaction.

9. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that when the processor executes the computer program, the steps of the method for controlling the steam temperature of a waste incineration boiler using a hierarchical control strategy are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program, when executed by a processor, implements the steps of a method for controlling steam temperature of a garbage incineration boiler using a hierarchical control strategy.

Images