A method and system for coordinated decoupling control of deaerator water level and condensate pressure

Abstract

The present application relates to the technical field of adaptive control, and discloses a kind of deaerator water level and condensate pressure collaborative decoupling control method and system, comprising: according to the acquisition time sequence characteristics of deaerator, the synchronous data set of water level signal and condensate pressure signal is constructed;The control unit of deaerator is divided into working condition, and different operating condition points are obtained, and the coupling coefficient matrix of each operating condition point is analyzed based on synchronous data set;According to the historical operation data and coupling coefficient matrix of each operating condition point, the collaborative decoupling factor of each operating condition point is determined, and the adaptive decoupling control model of deaerator is constructed, and the collaborative control quantity of deaerator water level and condensate pressure is analyzed using adaptive decoupling control model;Collaborative control quantity is converted into the collaborative control instruction of deaerator, and the collaborative decoupling of deaerator water level and condensate pressure is controlled according to collaborative control instruction.The present application can improve the accuracy of water level and condensate pressure collaborative decoupling control.

Description

Technical Field

[0001] This invention relates to the field of adaptive control technology, and in particular to a method and system for the coordinated decoupling control of deaerator water level and condensate pressure. Background Technology

[0002] In energy production sectors such as thermal power and nuclear power, deaerators are core equipment for ensuring the quality of steam and water and the safe operation of the unit. The stable control of its water level and condensate pressure directly affects the reliability of the boiler feedwater system and is related to the continuity and safety of the entire energy production process.

[0003] In existing technologies, deaerator water level control and condensate pressure control mostly employ independent single-loop PID control schemes. This independent control method does not consider the dynamic coupling relationship between water level and condensate pressure. When the deaerator is operating under variable conditions such as load increase or decrease, water level changes will affect condensate pressure through thermodynamic processes and fluid dynamic characteristics. Conversely, condensate pressure fluctuations will also be transmitted to the water level control loop. This coupling interference leads to problems such as regulation lag and large overshoot in the control system, resulting in insufficient control accuracy. At the same time, the independently controlled parameters cannot be adaptively adjusted according to changes in operating conditions, causing the system's dynamic response speed to be slow and making it difficult to meet the stable control requirements under complex operating conditions. Summary of the Invention

[0004] The main objective of this invention is to provide a method for the coordinated decoupling control of deaerator water level and condensate pressure.

[0005] Another objective of this invention is to propose a collaborative decoupling control system for deaerator water level and condensate pressure.

[0006] To achieve the above objectives, a first aspect of the present invention proposes a method for the coordinated decoupling control of deaerator water level and condensate pressure, comprising:

[0007] Collect water level signals and condensate pressure signals from the deaerator, and construct a synchronous dataset of the water level signals and condensate pressure signals based on the acquisition timing characteristics of the deaerator; The operating conditions of the deaerator control unit are divided to obtain different operating condition points. The coupling coefficient matrix of each operating condition point is analyzed based on the synchronous dataset. The collaborative decoupling factor for each operating condition point is determined based on the historical operating data of each operating condition point and the coupling coefficient matrix. An adaptive decoupling control model for the deaerator is constructed based on the aforementioned collaborative decoupling factor. The collaborative control variables of the deaerator's water level and condensate pressure are then analyzed using the adaptive decoupling control model. The coordinated control quantity is converted into a coordinated control command for the deaerator, and the coordinated decoupling of the deaerator water level and condensate pressure is controlled according to the coordinated control command.

[0008] Optionally, the step of constructing a synchronous dataset of the water level signal and the condensate pressure signal based on the acquisition timing characteristics of the deaerator includes: Extract the timestamp sequences of the water level signal and the condensate pressure signal, and construct a time relationship table based on the timestamp sequences; The time series of the water level signal and the condensate pressure signal are nonlinearly aligned based on the time relationship table. The water level signal and the condensate pressure signal are resampled based on the aligned time series at a preset unified time reference point to generate a synchronous data sequence with consistent time resolution. Abnormal data points in the synchronized data sequence are identified based on a preset sliding window, and the abnormal data points are repaired by linear interpolation to obtain a synchronized dataset.

[0009] Optionally, the control unit of the deaerator is divided into operating conditions to obtain different operating point conditions, including: Extract multi-dimensional operating parameters of the deaerator from the historical operation database of the deaerator; Density analysis of the multidimensional operating parameters is performed based on a preset density clustering algorithm to obtain dense regions with similar operating characteristics during the operation of the deaerator; The operating status of the deaerator is divided according to the dense area to obtain the operating condition type; Construct a feature parameter space for each type of operating condition, and build an operating condition classifier based on preset operating condition boundary conditions and the feature parameter space; The operating condition classifier is used to classify the operating conditions of the deaerator control unit to obtain different operating condition points.

[0010] Optionally, the step of analyzing the coupling coefficient matrix of each operating point based on the synchronized dataset includes: Extract water level-pressure data pairs from each operating condition point in the synchronous dataset, and generate a training sample set for the operating condition based on the water level-pressure data pairs. A dynamic coupling model between water level and pressure is constructed based on the training sample set of the aforementioned working conditions using a pre-defined forgetting factor recursive algorithm. The coupling parameters in the dynamic coupling model are updated using the parameter update mechanism of the forgetting factor recursive algorithm to obtain real-time updated parameters; The coupling coefficient matrix for each operating condition point is determined based on the real-time updated parameters, wherein the coupling coefficient matrix characterizes the dynamic interaction strength between water level and pressure.

[0011] Optionally, determining the collaborative decoupling factor for each operating condition point based on historical operating data and the coupling coefficient matrix includes: Calculate the control performance indicators under each operating condition based on historical operating data for each operating condition point; The degree of coupling between the water level and pressure control loops is analyzed based on the coupling coefficient matrix and the control performance indicators. Based on the degree of coupling, the control characteristics and control target priorities for each operating point are determined; The collaborative decoupling factor for each operating point is determined by the control characteristics and the priority of the control objectives.

[0012] Optionally, constructing an adaptive decoupling control model for the deaerator based on the cooperative decoupling factor includes: The weight allocation ratio of the feedforward decoupling channel and the feedback correction channel is determined based on the aforementioned collaborative decoupling factor; A feedforward decoupling mathematical model for the deaerator is constructed based on the coupling coefficient matrix and the weight allocation ratio of the feedforward decoupling channel. An adaptive PID feedback controller is constructed based on the weight allocation ratio of the feedback correction channel; The feedforward decoupling mathematical model is combined in parallel with the adaptive PID feedback controller to generate a two-degree-of-freedom control structure. Stability analysis was performed on the two-degree-of-freedom control structure to obtain an adaptive decoupling control model for the deaerator.

[0013] Optionally, the step of using the adaptive decoupling control model to analyze the coordinated control quantities of the deaerator's water level and condensate pressure includes: The real-time collected water level signal and condensate pressure signal are input into the adaptive decoupling control model; In the feedforward decoupling channel of the adaptive decoupling control model, the feedforward compensation amount of the deaerator water level and condensate pressure is calculated based on the current coupling coefficient matrix and the cooperative decoupling factor. In the feedback correction channel of the adaptive decoupling control model, the feedback control quantities of the deaerator water level and condensate pressure are calculated by a preset adaptive PID algorithm. By fusing the feedforward compensation and the feedback control, a coordinated control quantity for the deaerator water level and condensate pressure is obtained.

[0014] Optionally, converting the coordinated control quantity into a coordinated control command for the deaerator includes: Based on the aforementioned coordinated control quantity, the regulating valve opening control component and the condensate pump frequency control component are separated. The control components of the regulating valve opening and the condensate pump frequency are respectively controlled by the preset actuator characteristic curves and converted into corresponding actuator commands. The actuator commands are constrained, and the constrained actuator commands are simultaneously sent to the actuators corresponding to the deaerator using a preset timing synchronization mechanism. The actuator command is verified by the execution status feedback of the actuator. When the quality of the actuator meets the preset command conditions, the actuator command is determined as the collaborative control command of the deaerator.

[0015] Optionally, the coordinated decoupling of the deaerator water level and condensate pressure according to the coordinated control command includes: Based on the execution status of the aforementioned collaborative control commands, real-time operational data on deaerator water level and condensate pressure are collected. Calculate the dynamic deviation index between the actual operating data and the preset target value, wherein the dynamic deviation index includes instantaneous deviation and cumulative deviation components; The level of coordinated control commands for deaerator water level and condensate pressure is analyzed based on the dynamic deviation index. The weight allocation of the cooperative decoupling factor is adjusted based on the cooperative control command level, and the structural parameters of the adaptive decoupling control model are optimized using the adjusted cooperative decoupling factor. The optimized adaptive decoupling control model regenerates the cooperative control quantity, and the deaerator water level and condensate pressure are controlled based on the regenerated cooperative control quantity to achieve cooperative decoupling.

[0016] To achieve the above objectives, a second aspect of the present invention provides a coordinated decoupling control system for deaerator water level and condensate pressure, comprising: A synchronous dataset construction module is used to collect water level signals and condensate pressure signals from the deaerator, and to construct a synchronous dataset of the water level signals and condensate pressure signals based on the acquisition timing characteristics of the deaerator. The coupling coefficient matrix analysis module is used to divide the operating conditions of the deaerator control unit to obtain different operating conditions, and analyze the coupling coefficient matrix of each operating condition based on the synchronous dataset. The collaborative decoupling factor determination module is used to determine the collaborative decoupling factor for each operating condition point based on the historical operating data of each operating condition point and the coupling coefficient matrix. The collaborative control quantity analysis module is used to construct an adaptive decoupling control model for the deaerator based on the collaborative decoupling factor, and to analyze the collaborative control quantity between the water level and condensate pressure of the deaerator using the adaptive decoupling control model. The collaborative decoupling control module is used to convert the collaborative control quantity into collaborative control instructions for the deaerator, and to control the collaborative decoupling of the deaerator water level and condensate pressure according to the collaborative control instructions.

[0017] The embodiments of the present invention have the following beneficial effects: This invention provides a high-quality data foundation for subsequent coupling relationship analysis by collecting water level and condensate pressure signals from the deaerator and constructing a synchronous dataset. By dividing the control unit into operating conditions and analyzing the coupling coefficient matrix at each operating point, the dynamic coupling strength of water level and pressure under different operating conditions is accurately quantified, overcoming the shortcomings of existing technologies that lack operating condition differentiation and have ambiguous coupling relationships. Based on historical operating data and the cooperative decoupling factor determined by the coupling coefficient matrix, the degree of decoupling and control priority under each operating condition are clarified. The constructed adaptive decoupling control model integrates feedforward decoupling and feedback correction, realizing active compensation for coupling interference and dynamic correction of control errors, improving the system's anti-interference capability and control accuracy. By converting cooperative control quantities into control commands and performing closed-loop optimization, the safe execution of control commands and continuous improvement of control quality are ensured, enabling the deaerator to achieve stable control of water level and condensate pressure under different operating conditions. This reduces overshoot, shortens settling time, and decreases steady-state error, significantly improving the deaerator's operational stability and control effect. Therefore, the decoupling control method and system for deaerator water level and condensate pressure proposed in this invention can solve the problem of low accuracy when performing decoupling control of water level and condensate pressure. Attached Figure Description

[0018] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 A flowchart illustrating a method for coordinated decoupling control of deaerator water level and condensate pressure, provided in an embodiment of the present invention; Figure 2 This is a structural diagram of a decoupling control system for deaerator water level and condensate pressure provided in an embodiment of the present invention. Detailed Implementation

[0019] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0021] This application provides a method for the coordinated decoupling control of deaerator water level and condensate pressure. The executing entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for coordinated decoupling control of deaerator water level and condensate pressure can be executed by software or hardware installed on a terminal device or a server device. The software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0022] The following describes, with reference to the accompanying drawings, a method and system for the coordinated decoupling control of deaerator water level and condensate pressure according to an embodiment of the present invention.

[0023] Example 1 This embodiment provides a method for the coordinated decoupling control of deaerator water level and condensate pressure, such as... Figure 1 As shown, the method includes the following steps: S1. Collect the water level signal and condensate pressure signal of the deaerator, and construct a synchronous dataset of the water level signal and the condensate pressure signal based on the acquisition timing characteristics of the deaerator.

[0024] In this embodiment of the invention, the deaerator is a key device in the energy production system used to remove dissolved gases from boiler feedwater; the water level signal is a physical quantity signal reflecting the height of the water body inside the deaerator, and the condensate pressure signal refers to the physical quantity signal of pressure generated by the condensate entering the deaerator in the pipeline.

[0025] In detail, based on the structural characteristics and process requirements of the deaerator, the water level signal acquisition location was determined to be the middle section of the side wall of the deaerator water tank, and the pressure signal acquisition location was the condensate pump outlet pipe. Differential pressure level transmitters and absolute pressure sensors were installed at the acquisition locations, respectively. The measurement range of the differential pressure level transmitters covered the normal operating water level fluctuation range of the deaerator, and the range of the absolute pressure sensors included the normal operating pressure range of the condensate system. Data acquisition was synchronously triggered by a distributed acquisition system using the differential pressure level transmitters and absolute pressure sensors, with the acquisition trigger signals provided by the same clock source to ensure consistent acquisition start times. The acquired raw signals were conditioned, including temperature compensation correction for the water level signal and pipeline resistance loss compensation for the pressure signal, generating pre-processed water level and condensate pressure signals.

[0026] Furthermore, in order to accurately reflect the dynamic correlation between water level and pressure, it is necessary to construct a high-quality synchronous dataset, which provides the necessary data foundation for subsequent working condition division and coupling analysis.

[0027] In this embodiment of the invention, the acquisition timing characteristics refer to the inherent time attributes such as the time interval and triggering method of signal acquisition during the operation of the deaerator. The synchronized dataset refers to a dataset that adjusts water level signals and condensate pressure signals from different acquisition timings to have a consistent time base, the same resolution, and a clear data correspondence.

[0028] In this embodiment of the invention, constructing a synchronous dataset of the water level signal and the condensate pressure signal based on the timing characteristics of the deaerator's data acquisition includes: Extract the timestamp sequences of the water level signal and the condensate pressure signal, and construct a time relationship table based on the timestamp sequences; The time series of the water level signal and the condensate pressure signal are nonlinearly aligned based on the time relationship table. The water level signal and the condensate pressure signal are resampled based on the aligned time series at a preset unified time reference point to generate a synchronous data sequence with consistent time resolution. Abnormal data points in the synchronized data sequence are identified based on a preset sliding window, and the abnormal data points are repaired by linear interpolation to obtain a synchronized dataset.

[0029] In detail, the timestamp sequence refers to a sequence composed of acquisition time identifiers corresponding to each acquired signal data. A high-precision time acquisition module records the acquisition time of each signal, with a sampling frequency set to 10Hz to ensure millisecond-level accuracy of the time identifiers, thus obtaining the timestamp sequence. By traversing the timestamp sequences of the water level signal and condensate pressure signal, a mapping relationship between their time points is established, clarifying the condensate pressure data acquisition period corresponding to each water level data point, resulting in a time relationship table. This time relationship table is a structured data table recording the correspondence between the acquisition times of the water level signal and the condensate pressure signal. Since the acquisition of the water level signal and condensate pressure signal may be affected by sensor response speed, transmission delay, etc., resulting in time deviations, non-linear alignment based on the time relationship table is required. Using the timestamp of the water level signal as a reference, a dynamic time warping algorithm is used to calculate the optimal matching path between the condensate pressure signal time series and the water level signal time series, adjusting the time coordinates of the condensate pressure signal. For example, when the water level signal is collected at time t1, the corresponding condensate pressure signal is actually collected at time t1+20ms. The algorithm corrects the time coordinate of the condensate pressure signal to time t1, achieving a precise time match between the two. This solves the coupling analysis error problem caused by asynchronous signal acquisition and improves the accuracy of data correlation analysis.

[0030] Specifically, a unified time reference point refers to a pre-defined fixed time interval node, and time resolution refers to the number of signal acquisitions per unit time. Resampling employs linear interpolation, with the unified time reference point time interval set to 100ms, i.e., a time resolution of 10Hz. Using the unified time reference point as the standard, data interpolation calculations are performed on the aligned water level signal and condensate pressure signal. For missing data between reference points, linear extrapolation is performed based on the values ​​of two adjacent valid data points. For example, if the aligned water level signal is 1500mm at 100ms and 1502mm at 300ms, then the water level data at the 200ms reference point is calculated as 1501mm through linear interpolation. Similarly, the resampling of the condensate pressure signal is completed.

[0031] Furthermore, the sliding window length was set to 5 data points, and the sliding step size was 1 data point. By calculating the mean and standard deviation of the data within each window, a data point was identified as an outlier if its difference from the mean exceeded three times the standard deviation. For example, if the mean water level data within a window was 1500 mm and the standard deviation was 2 mm, a data point with a value of 1493 mm would be considered an outlier, exceeding three times the standard deviation. In this case, linear interpolation was performed using the preceding (1501 mm) and following (1502 mm) data points to calculate a corrected value of 1501.5 mm. This eliminated outliers caused by sensor interference and transmission noise during signal acquisition, ensuring the accuracy of the synchronized dataset.

[0032] Furthermore, since the original signals have problems such as inconsistent acquisition timing, different resolutions, and the presence of abnormal data, directly using them for coupling analysis will lead to biased results. Therefore, it is necessary to construct a synchronous dataset to gradually optimize the data quality, providing a high-quality data foundation for subsequent analysis of the coupling relationship between water level and condensate pressure, and solving the problem of inaccurate control strategy formulation caused by poor data quality.

[0033] S2. Divide the operating conditions of the deaerator control unit to obtain different operating condition points, and analyze the coupling coefficient matrix of each operating condition point based on the synchronous dataset.

[0034] In this embodiment of the invention, the control unit refers to a collection of hardware circuits and control programs used to control the operating state of the deaerator. The operating point is the stable operating state point of the deaerator under a specific combination of operating parameters.

[0035] In this embodiment of the invention, the step of dividing the operating conditions of the deaerator control unit to obtain different operating condition points includes: Extract multi-dimensional operating parameters of the deaerator from the historical operating database of the deaerator; Density analysis of the multidimensional operating parameters is performed based on a preset density clustering algorithm to obtain dense regions with similar operating characteristics during the operation of the deaerator; The operating status of the deaerator is divided according to the dense area to obtain the operating condition type; Construct a feature parameter space for each type of operating condition, and build an operating condition classifier based on preset operating condition boundary conditions and the feature parameter space; The operating condition classifier is used to classify the operating conditions of the deaerator control unit to obtain different operating condition points.

[0036] In detail, the deaerator historical operation database stores various operating parameter data from the deaerator's long-term operation. Operating data from the past 12 months is retrieved through a database query interface, with data intervals of 1 minute to ensure data coverage of different operating scenarios. These multi-dimensional operating parameters include load rate, inlet water temperature, steam pressure, and unit operating mode. For example, the extracted load rate data ranges from 30% to 100%, inlet water temperature from 25℃ to 80℃, steam pressure from 0.3MPa to 0.8MPa, and unit operating modes include normal operation and maintenance. Next, the density clustering algorithm uses the DBSCAN algorithm. This algorithm identifies dense regions in the dataset by setting two parameters: neighborhood radius and minimum number of points. For example, a neighborhood radius of 5 (based on the distance after parameter standardization) and a minimum number of points of 50 are set to perform cluster analysis on the standardized multi-dimensional operating parameters. For example, data points are densely distributed in areas with a load rate of 80%-100%, an inlet water temperature of 60℃-80℃, and a steam pressure of 0.6MPa-0.8MPa, forming a dense region; another dense region is formed in areas with a load rate of 30%-50%, an inlet water temperature of 25℃-40℃, and a steam pressure of 0.3MPa-0.5MPa. By analyzing data density, regions with similar operating characteristics can be identified, providing a data-driven basis for classifying operating conditions.

[0037] Specifically, when classifying operating conditions based on dense areas, the operating principle of the deaerator is considered. Areas with dense data and stable operating parameters are classified as steady-state conditions; areas with continuously rising load rates and synchronous changes in other parameters are classified as increasing load conditions; areas with continuously decreasing load rates are classified as decreasing load conditions; and areas with parameter fluctuations exceeding normal ranges, sparse data points, and distance from other dense areas are classified as abnormal conditions. For example, a dense area with a stable load rate of around 90% and inlet water temperature and steam pressure fluctuations not exceeding ±5% is considered a steady-state condition; a dense area with a load rate gradually increasing from 50% to 90% is considered an increasing load condition. This clarifies the classification criteria for different operating conditions, giving the operating condition types a clear physical meaning and resolving the problem of ambiguous operating condition definitions. When constructing the feature parameter space, the value range and statistical characteristics of multidimensional operating parameters under each operating condition type are extracted. For example, the load rate range for steady-state conditions is 70%-100%, and the inlet water temperature range is 55℃-80℃. The boundary conditions for operating conditions are the critical values ​​of each parameter. For example, when the load rate change rate is greater than 0.5% / min, the system enters either a load increase or decrease condition. The operating condition classifier uses a support vector machine model, taking the parameters in the feature parameter space as input and the operating condition type as output. The model is trained using operating condition data obtained from historical clustering, enabling it to output the corresponding operating condition type based on real-time multi-dimensional operating parameters. This constructs a generalized operating condition classifier, providing efficient algorithmic support for real-time operating condition judgment and solving the problem of slow response in existing technologies.

[0038] Furthermore, when classifying operating conditions using an operating condition classifier, multi-dimensional operating parameters of the deaerator are collected in real time and input into the trained operating condition classifier to obtain the real-time operating condition type. This, combined with specific parameter values, determines the operating condition point. For example, if the real-time collected parameters are a load rate of 90%, an inlet water temperature of 70℃, and a steam pressure of 0.7MPa, the classifier classifies it as a steady-state operating condition. The operating state corresponding to this parameter combination is a steady-state operating condition point, thus achieving accurate identification of the operating condition point.

[0039] In this embodiment of the invention, the coupling coefficient matrix is ​​matrix data that characterizes the intensity of the dynamic interaction between water level and condensate pressure.

[0040] In this embodiment of the invention, the step of analyzing the coupling coefficient matrix of each operating condition point based on the synchronization dataset includes: Extract water level-pressure data pairs from each operating condition point in the synchronous dataset, and generate a training sample set for the operating condition based on the water level-pressure data pairs. A dynamic coupling model between water level and pressure is constructed based on the training sample set of the aforementioned working conditions using a pre-defined forgetting factor recursive algorithm. The coupling parameters in the dynamic coupling model are updated using the parameter update mechanism of the forgetting factor recursive algorithm to obtain real-time updated parameters; The coupling coefficient matrix for each operating condition point is determined based on the real-time updated parameters, wherein the coupling coefficient matrix characterizes the dynamic interaction strength between water level and pressure.

[0041] In detail, a water level-pressure data pair refers to a pair of data consisting of water level data and condensate pressure data corresponding to the same time reference point in the synchronous dataset. When extracting water level-pressure data pairs, based on the divided operating conditions, all data under the corresponding operating conditions are selected from the synchronous dataset and arranged into data pairs in chronological order, i.e., the operating condition training sample set. For example, in the synchronous data corresponding to a certain steady-state operating condition point, the water level at time t1 is 1500 mm and the pressure is 0.5 MPa, and the water level at time t2 is 1501 mm and the pressure is 0.502 MPa. These data pairs together form the training sample set for that operating condition point, thus providing dedicated training data for the construction of the coupled model for each operating condition point, ensuring that the model can reflect the coupling characteristics under that operating condition.

[0042] Specifically, the forgetting factor recursive algorithm uses a recursive least squares algorithm with a forgetting factor, and the system model of the dynamic coupling model is expressed as follows:

[0043] in For the system output vector, For the regression matrix, The parameter vector to be identified, This represents modeling error; The output vector includes the deaerator water level. and condensate pressure , This represents the time-varying coupling coefficient vector. When constructing the model, water level-pressure data pairs from the training sample set are substituted into the system model to initialize the model parameters. The parameters are then optimized through iterative algorithmic optimization to minimize the error between the model output and the actual data, thereby constructing a mathematical model reflecting the dynamic coupling relationship between water level and pressure. The parameter update law of the recursive least squares method includes:

[0044]

[0045]

[0046] Among them, the forgetting factor This is used to mitigate the impact of old data on parameter updates; the initial value of the covariance matrix. , It is the identity matrix. The estimated value of the parameter vector to be identified. Here is the gain matrix. The covariance matrix of condensate pressure is given. During parameter updates, the parameter estimates from the previous time step are used. The prediction error is calculated using the input and output data at the current moment, and then processed through the gain matrix. Adjust the parameter estimates and finally update the covariance matrix. .For example, The prediction error at time is minus transpose multiplied by Through the gain matrix right Make corrections to obtain Parameter estimates at time 1 This enables real-time updates of coupling parameters, allowing the model to adapt to changes in operating conditions.

[0047] Furthermore, based on real-time updated parameters The four coupling coefficients Construct the coupling coefficient matrix. Characterizes the intensity of the water level's influence on itself. Characterizing the strength of the coupling effect of pressure on water level, Characterize the strength of the coupling effect of water level on pressure. It characterizes the intensity of the pressure's influence on itself. For example, the real-time updated parameters at a certain steady-state operating point are... The corresponding coupling coefficient matrix is ​​[[0.8,0.3],[0.2,0.9]]. Simultaneously, a singularity test is performed on the coupling coefficient matrix. When the absolute value of the matrix determinant is less than a preset threshold (e.g., 0.01), it is determined to be close to singular, triggering a regularization mechanism. By adding a small perturbation, the matrix becomes invertible, quantifying the coupling strength between water level and pressure under different operating conditions. This provides a precise basis for subsequent decoupling control and solves the problem of poor decoupling effect caused by fuzzy coupling relationships.

[0048] Furthermore, precise working condition division and coupling coefficient matrix can accurately quantify the coupling relationship between water level and pressure under different working conditions. This provides a key input for determining the cooperative decoupling factor, thus leading to the need for calculating the cooperative decoupling factor.

[0049] S3. Determine the collaborative decoupling factor for each operating condition point based on the historical operating data of each operating condition point and the coupling coefficient matrix.

[0050] In this embodiment of the invention, historical operating data refers to the past operating parameters and control performance data of the deaerator at various operating conditions. The collaborative decoupling factor refers to a parameter used to adjust the degree of decoupling between the water level and pressure control loops and to coordinate control priorities.

[0051] In this embodiment of the invention, determining the collaborative decoupling factor for each operating condition point based on historical operating data and the coupling coefficient matrix includes: Calculate the control performance indicators under each operating condition based on historical operating data for each operating condition point; The degree of coupling between the water level and pressure control loops is analyzed based on the coupling coefficient matrix and the control performance indicators. Based on the degree of coupling, the control characteristics and control target priorities for each operating point are determined; The collaborative decoupling factor for each operating point is determined by the control characteristics and the priority of the control objectives.

[0052] In detail, the setpoint and actual value variation curves of water level and pressure at each operating point are extracted from historical operating data. Overshoot is calculated as the percentage difference between the actual maximum value and the setpoint, and the settling time is the time it takes for the actual value to stabilize within ±5% of the setpoint. Steady-state error is the difference between the actual value and the setpoint after stable operation. For example, under a certain load increase condition, the water level setpoint is 1500mm, the actual maximum value is 1530mm, and the overshoot is 2%. The actual value stabilizes within the range of 1495mm-1505mm in 10 seconds, with a settling time of 10 seconds. After stabilization, the actual value is 1501mm, and the steady-state error is 0.07%. When analyzing the degree of coupling, the magnitude of the coupling coefficients in the coupling coefficient matrix and the quality of the control performance indicators are considered. Coupling coefficients... and The larger the absolute value, the stronger the mutual coupling interference; at the same time, the greater the overshoot and the longer the settling time, the more significant the impact of coupling interference on the control effect. For example, the coupling coefficient of a certain load reduction condition... , Furthermore, the overshoot reaches 3% and the settling time reaches 15 seconds, indicating a strong coupling degree; a certain steady-state condition , The overshoot was 0.5% and the settling time was 5 seconds, which was determined to be a weak coupling.

[0053] Specifically, when determining control characteristics and control target priorities based on the degree of coupling, the control characteristics of strongly coupled operating conditions are characterized by large mutual interference between loops and complex dynamic responses. In these cases, the control target priorities are set to ensure that water level control and pressure control are equally important. Conversely, the control characteristics of weakly coupled operating conditions are characterized by small disturbances and stable responses. In these cases, the control target priority can be adjusted based on the operating conditions, with an emphasis on water level control (e.g., steady-state operating conditions). For example, under abnormal operating conditions, the degree of coupling is extremely strong, with control characteristics of drastic parameter fluctuations. In these cases, the control target priority is set to prioritize ensuring the stability of both water level and pressure. Under load-increasing operating conditions, the control characteristics are characterized by gradual parameter changes. In these cases, the control target priority is set to prioritize pressure control slightly over water level control to ensure stable steam supply. This clarifies the control focus for different operating conditions, making decoupling control more targeted. Based on the coupling degree analysis results, a calculation function for the cooperative decoupling factor is designed. This function comprehensively considers the system's dynamic characteristics and the priority of the control objectives. Control performance indicators include overshoot, settling time, and steady-state error. For example, under strong coupling conditions, the cooperative decoupling factor ranges from 0.8 to 1.0, enhancing the decoupling effect; under weak coupling conditions, it ranges from 0.3 to 0.7, achieving moderate decoupling while preserving the system's natural response characteristics. Combining the trace of the coupling coefficient matrix (the sum of the elements on the main diagonal of the coupling coefficient matrix) and the priority weights of the control objectives, such as when there is strong coupling and equal priorities, the cooperative decoupling factor = 0.9 × (trace of the coupling coefficient matrix / 2) + 0.1, a specific value is calculated. Thus, through the quantified cooperative decoupling factor, core adjustment parameters are provided for the decoupled control model.

[0054] Furthermore, the collaborative decoupling factor can provide appropriate decoupling strength adjustment for different operating conditions, providing key parameters for constructing an adaptive decoupling control model, thus leading to the need for control model construction.

[0055] S4. Construct an adaptive decoupling control model for the deaerator based on the aforementioned collaborative decoupling factor, and use the adaptive decoupling control model to analyze the collaborative control quantities of the deaerator's water level and condensate pressure.

[0056] In this embodiment of the invention, the adaptive decoupling control model refers to a control model that can adaptively adjust the decoupling strategy and control parameters according to changes in operating conditions and the degree of coupling.

[0057] In this embodiment of the invention, constructing an adaptive decoupling control model for the deaerator based on the cooperative decoupling factor includes: The weight allocation ratio of the feedforward decoupling channel and the feedback correction channel is determined based on the aforementioned collaborative decoupling factor; A feedforward decoupling mathematical model for the deaerator is constructed based on the coupling coefficient matrix and the weight allocation ratio of the feedforward decoupling channel. An adaptive PID feedback controller is constructed based on the weight allocation ratio of the feedback correction channel; The feedforward decoupling mathematical model is combined in parallel with the adaptive PID feedback controller to generate a two-degree-of-freedom control structure. Stability analysis was performed on the two-degree-of-freedom control structure to obtain an adaptive decoupling control model for the deaerator.

[0058] In detail, the feedforward decoupling channel refers to the control channel that compensates for coupling interference in advance, while the feedback correction channel refers to the channel that adjusts the control quantity according to the control error. The weight allocation ratio is determined based on the magnitude of the cooperative decoupling factor; the larger the cooperative decoupling factor, the higher the weight ratio of the feedforward decoupling channel; conversely, the lower the cooperative decoupling factor, the higher the weight ratio of the feedback correction channel. The weight allocation ratio refers to the proportion of the feedforward decoupling channel and the feedback correction channel in the total control quantity. For example, under strong coupling conditions, the cooperative decoupling factor is 0.9, the weight allocation ratio of the feedforward decoupling channel is 0.7, and the weight allocation ratio of the feedback correction channel is 0.3; under weak coupling conditions, the cooperative decoupling factor is 0.4, the weight allocation ratio of the feedforward decoupling channel is 0.3, and the weight allocation ratio of the feedback correction channel is 0.7, realizing the dynamic adjustment of the control channel weights, enabling the control model to optimize the control strategy according to the degree of coupling. Next, the feedforward decoupling mathematical model is a mathematical expression describing the input-output relationship of the feedforward decoupling channel. The feedforward decoupling mathematical model adopts a complete decoupling scheme based on the inverse of the coupling coefficient matrix. The output of the feedforward decoupling module is:

[0059] in This is the real-time coupled disturbance vector estimated by the coupled observer. For water level coupling interference, To mitigate pressure coupling interference, the coupling coefficient matrix is ​​inverted. Combined with the weight allocation ratio of the feedforward decoupling channel, the calculation method for the feedforward compensation is determined. For example, if the coupling coefficient matrix is ​​[[0.8,0.3],[0.2,0.9]], its inverse matrix is ​​calculated to a specific value. Combined with a weight ratio of 0.7, the specific expression of the feedforward decoupling mathematical model is obtained. This feedforward compensation can counteract coupling interference, reducing the impact of coupling on the control effect from the source.

[0060] Specifically, an adaptive PID feedback controller refers to a controller whose PID parameters can adaptively adjust according to the operating state. Adaptive PID feedback controllers employ fuzzy PID controllers, whose PID parameters (proportional coefficient, integral coefficient, and derivative coefficient) are tuned online based on the operating condition and the cooperative decoupling factor. When constructing the controller, the feedback correction module uses a modular PI controller, whose control quantity... According to error and error change rate Adaptive adjustment of PID parameters, final control quantity For example, when the error is large, the error rate of change is large, and the collaborative decoupling factor is 0.8, the fuzzy rule outputs a larger proportional coefficient and a smaller integral coefficient to speed up the system response; when the error is small and the error rate of change is small, it outputs a smaller proportional coefficient and a larger integral coefficient to improve control accuracy.

[0061] Furthermore, the feedforward decoupling mathematical model is combined in parallel with the adaptive PID feedback controller to form a two-degree-of-freedom control structure. A two-degree-of-freedom control structure simultaneously considers tracking performance and anti-interference performance. In this structure, the feedforward decoupling channel is responsible for canceling known coupling interference, while the feedback correction channel is responsible for correcting control errors and responding to unknown interference; the two work together. For example, the feedforward compensation amount calculated by the feedforward decoupling channel... The feedback control quantity calculated with the feedback correction channel Add them together to get the total control quantity. Simultaneously, stability analysis is performed using Lyapunov stability theory, and the Lyapunov function is defined as:

[0062] in For parameter estimation error, and It is a positive definite matrix. When... When this occurs, a stability monitoring mechanism is triggered, and the parameter estimates are corrected using a projection algorithm to ensure stability. Bounded, maintaining the stability of the closed-loop system.

[0063] In this embodiment of the invention, the collaborative control quantity refers to the final control quantity used to control the operation of the deaerator after integrating the feedforward compensation quantity and the feedback control quantity.

[0064] In this embodiment of the invention, the analysis of the coordinated control quantities of the deaerator's water level and condensate pressure using the adaptive decoupling control model includes: The real-time collected water level signal and condensate pressure signal are input into the adaptive decoupling control model; In the feedforward decoupling channel of the adaptive decoupling control model, the feedforward compensation amount of the deaerator water level and condensate pressure is calculated based on the current coupling coefficient matrix and the cooperative decoupling factor. In the feedback correction channel of the adaptive decoupling control model, the feedback control quantities of the deaerator water level and condensate pressure are calculated by a preset adaptive PID algorithm. By fusing the feedforward compensation and the feedback control, a coordinated control quantity for the deaerator water level and condensate pressure is obtained.

[0065] In detail, the real-time water level and condensate pressure signals are acquired by sensors, with a sampling frequency of 10Hz consistent with S1 to ensure data real-time performance and consistency. After the signals are input into the adaptive decoupled control model, the model first preprocesses the signals to remove noise interference, providing accurate input for subsequent calculations. This achieves the integration of real-time data with the control model, ensuring that the control model can perform calculations based on the current operating state and solving the problem of untimely control caused by data input lag.

[0066] Specifically, when calculating the feedforward compensation, the compensation value of the coupling interference is calculated using the feedforward decoupling mathematical model based on the coupling coefficient matrix and the cooperative decoupling factor under the current operating conditions. For example, the current coupling coefficient matrix is ​​[[0.8,0.3],[0.2,0.9]], the cooperative decoupling factor is 0.9, and the real-time coupling interference vector... The feedforward compensation amount is calculated by substituting the value [0.02MPa, 5mm] into the feedforward decoupling formula. This is a specific value used to counteract the coupling interference. The feedforward compensation amount refers to the control quantity calculated through the feedforward decoupling channel to counteract the coupling interference, thus compensating for the known coupling interference in advance and reducing the impact of the interference on the control effect. When calculating the feedback control quantity, an adaptive PID algorithm is used, based on the error between the real-time signal and the setpoint. and error change rate The PID parameters are dynamically adjusted to calculate the feedback control quantity, which is the control quantity used to correct control errors, calculated through the feedback correction channel. For example, if the water level setpoint is 1500mm and the real-time water level is 1505mm, the error... Error change rate The adaptive PID algorithm adjusts the parameters to calculate the feedback control quantity. The adjustment amount is used to lower the water level; similarly, the feedback control amount of the pressure is calculated, and the control error is corrected through the feedback mechanism, thereby improving the control accuracy.

[0067] Furthermore, when integrating feedforward compensation and feedback control quantities, a weighted sum is performed according to a preset weight allocation ratio, and the weight coefficients are dynamically adjusted based on the system's operating status. For example, if the weight of the feedforward compensation is 0.7 and the weight of the feedback control is 0.3, the total coordinated control quantity... For example, with feedforward compensation of -0.01 MPa (pressure regulation) and -3 mm (water level regulation), and feedback control of -0.02 MPa and -2 mm, the combined control results in a coordinated control of -0.013 MPa and -2.3 mm. This combines the advantages of feedforward decoupling and feedback correction to obtain the optimal coordinated control.

[0068] Furthermore, the cooperative decoupling factor provides core adjustment parameters for the construction of the control model, while the coupling coefficient matrix provides fundamental data for the feedforward decoupling mathematical model. Without the cooperative decoupling factor, the control model cannot dynamically adjust the degree of decoupling; without the coupling coefficient matrix, feedforward decoupling cannot accurately compensate for coupling interference. Therefore, the cooperative decoupling factor and the coupling coefficient matrix construct a complete adaptive decoupling control model and calculate the cooperative control quantity, providing a direct basis for the generation of subsequent control commands.

[0069] S5. Convert the coordinated control quantity into a coordinated control command for the deaerator, and control the coordinated decoupling of the deaerator water level and condensate pressure according to the coordinated control command.

[0070] In this embodiment of the invention, the coordinated control command refers to a control signal that can be recognized and executed by the deaerator actuator.

[0071] In this embodiment of the invention, converting the coordinated control quantity into a coordinated control command for the deaerator includes: Based on the aforementioned coordinated control quantity, the regulating valve opening control component and the condensate pump frequency control component are separated. The control components of the regulating valve opening and the condensate pump frequency are respectively controlled by the preset actuator characteristic curves and converted into corresponding actuator commands. The actuator commands are constrained, and the constrained actuator commands are simultaneously sent to the actuators corresponding to the deaerator using a preset timing synchronization mechanism. The actuator command is verified by the execution status feedback of the actuator. When the quality of the actuator meets the preset command conditions, the actuator command is determined as the collaborative control command of the deaerator.

[0072] In detail, the regulating valve opening control component refers to the control quantity used to regulate the opening of the regulating valve related to the deaerator, and the condensate pump frequency control component refers to the control quantity used to regulate the operating frequency of the condensate pump. When separating the control components, based on the physical meaning and controlled object of the coordinated control quantity, they are divided into an opening control component for the regulating valve and a frequency control component for the condensate pump. For example, the portion of the coordinated control quantity used to regulate water level corresponds to the regulating valve opening control component, and the portion used to regulate condensate pressure corresponds to the condensate pump frequency control component. Assuming that the water level regulation related value in the coordinated control quantity is -2.3mm, the corresponding regulating valve opening control component is -3%; and the pressure regulation related value is -0.013MPa, the corresponding condensate pump frequency control component is -0.5Hz, thus clarifying the control tasks of different actuators and enabling the control quantity to accurately correspond to the executing equipment.

[0073] Specifically, the actuator characteristic curve is the curve showing the correspondence between the control component and the actuator action obtained through experimental testing. For example, the characteristic curve of a control valve shows a linear relationship between the change in opening and the control component, while the characteristic curve of a condensate pump shows a linear relationship between the change in frequency and the control component. During conversion, the control component of the control valve opening is substituted into its characteristic curve to obtain the corresponding control valve opening command; for example, a control component of -3% corresponds to an opening command of decreasing by 3%. Similarly, the frequency control component of the condensate pump is substituted into its characteristic curve to obtain the corresponding frequency command; for example, a control component of -0.5Hz corresponds to a frequency command of decreasing by 0.5Hz. This achieves the conversion from control quantity to actuator command, enabling the control quantity to be recognized by the actuator. Constraint processing includes rate limits and amplitude limits to ensure that the commands are within the safe operating range of the equipment. The rate of change of the control valve opening is limited to ±2% / s, and the amplitude is limited to 0%-100%; the rate of change of the condensate pump frequency is limited to ±1Hz / s, and the amplitude is limited to 30Hz-50Hz. For example, if the converted control valve opening command is to decrease by 5%, and the current opening change rate is 3% / s, exceeding the limit, it will be adjusted to 2% / s, completing the 5% opening adjustment in two steps. If the frequency command is 28Hz, which is below the amplitude lower limit, it will be adjusted to 30Hz. The timing synchronization mechanism ensures that the control valve opening command and the condensate pump frequency command are issued simultaneously through hardware clock synchronization or software timestamp alignment. This avoids control misalignment caused by time differences in command issuance, guarantees the safe operation of the actuator and the coordination of control, and solves the problems of equipment damage and control imbalance caused by unconstrained commands.

[0074] Furthermore, after the actuator executes the command, the actual operating state of the actuator is collected through position and frequency sensors, such as the actual opening degree of the control valve and the actual frequency of the condensate pump. The actual operating state is compared with the command requirements. When the deviation between the actual operation and the command is less than a preset threshold (e.g., the deviation of the control valve opening degree is less than ±0.5%, and the frequency deviation is less than ±0.1Hz), it is determined that the actuator quality meets the preset command conditions, and the command is identified as a coordinated control command. If the deviation exceeds the threshold, the control quantity is readjusted and converted into a command until the conditions are met, ensuring that the control command can be executed accurately.

[0075] In this embodiment of the invention, the coordinated decoupling of the deaerator water level and condensate pressure according to the coordinated control command includes: Based on the execution status of the aforementioned collaborative control commands, real-time operational data on deaerator water level and condensate pressure are collected. Calculate the dynamic deviation index between the actual operating data and the preset target value, wherein the dynamic deviation index includes instantaneous deviation and cumulative deviation components; The level of coordinated control commands for deaerator water level and condensate pressure is analyzed based on the dynamic deviation index. The weight allocation of the cooperative decoupling factor is adjusted based on the cooperative control command level, and the structural parameters of the adaptive decoupling control model are optimized using the adjusted cooperative decoupling factor. The optimized adaptive decoupling control model regenerates the cooperative control quantity, and the deaerator water level and condensate pressure are controlled based on the regenerated cooperative control quantity to achieve cooperative decoupling.

[0076] In detail, when collecting real-time operational data, the same sensors and sampling frequency as S1 are used to ensure data consistency and accuracy. For example, after the coordinated control command is executed, water level and pressure data are collected every 100ms for 5 seconds to obtain the actual operational data sequence. When calculating the dynamic deviation index, the instantaneous deviation is the difference between the actual operational data at a certain moment and the preset target value, and the cumulative deviation component is the integral value of the instantaneous deviation over a period of time. For example, if the preset target value for water level is 1500mm, the actual value at time t1 is 1498mm, and the instantaneous deviation is -2mm; the instantaneous deviations at times t1 to t5 are -2mm, -1mm, 0mm, 1mm, and 2mm respectively, and the cumulative deviation component is 0mm. s.

[0077] Specifically, the collaborative control command levels are divided into four grades: excellent, good, qualified, and unqualified, based on the magnitude of the dynamic deviation index. For example, an instantaneous deviation absolute value of less than 1 mm and a cumulative deviation component of less than 5 mm are considered excellent. s represents excellent; the absolute value of the instantaneous deviation is less than 3mm and the cumulative deviation component is less than 15mm. s is considered good; the absolute value of the instantaneous deviation is less than 5mm and the cumulative deviation component is less than 30mm. s represents acceptable; values ​​outside the above range represent unacceptable. For example, the calculated instantaneous deviation absolute value is 2mm, and the cumulative deviation component is 10mm. The collaborative control command level is determined to be "good," thus clarifying the evaluation criteria for control quality and making the assessment of control effect more intuitive. When adjusting the weight allocation of the collaborative decoupling factor, optimization is performed based on the collaborative control command level. If the collaborative control command level is "excellent," the current weight allocation is maintained; if it is "good," the weight allocation is finely adjusted (e.g., feedforward weight ±0.05); if it is "qualified," the weight allocation is moderately adjusted (e.g., feedforward weight ±0.1); if it is "unqualified," the weight allocation is significantly adjusted (e.g., feedforward weight ±0.2). Simultaneously, the adjusted collaborative decoupling factor is used to optimize the structural parameters of the adaptive decoupling control model, such as the initial values ​​of PID parameters and the coefficients of the feedforward decoupling model. For example, if the collaborative control command level is "qualified," the feedforward decoupling channel weight is adjusted from 0.7 to 0.8, and the proportional coefficient of the PID parameters is optimized simultaneously. This adaptive optimization of the control model allows the control strategy to be dynamically adjusted according to the actual control effect.

[0078] Furthermore, the optimized adaptive decoupling control model regenerates the cooperative control quantity, repeating processes S4 and S5 to achieve closed-loop optimized control. For example, the optimized model generates new cooperative control quantities, which are converted into control commands and issued for execution. Actual operating data is collected again for evaluation until the cooperative control command level reaches excellent, ensuring that the cooperative decoupling effect between deaerator water level and condensate pressure continues to improve.

[0079] Furthermore, the cooperative control quantity is the foundation for control command conversion. Only a clear cooperative control quantity can separate the control components of each actuator and convert them into commands. At the same time, through control quality assessment and model optimization, the cooperative decoupling factor and adaptive decoupling control model are adjusted in reverse to make the subsequently generated cooperative control quantity more accurate, ensuring that the entire control process continuously adapts to changes in operating conditions and achieves stable cooperative decoupling control.

[0080] Example 2 This invention also provides a coordinated decoupling control system for deaerator water level and condensate pressure, which can be installed in electronic equipment. For example... Figure 2 As shown, the deaerator water level and condensate pressure collaborative decoupling control system 100 may include a synchronous dataset construction module 101, a coupling coefficient matrix analysis module 102, a collaborative decoupling factor determination module 103, a collaborative control quantity analysis module 104, and a collaborative decoupling control module 105.

[0081] In this embodiment, the functions of each module are as follows: The synchronous dataset construction module 101 is used to collect the water level signal and condensate pressure signal of the deaerator, and construct a synchronous dataset of the water level signal and the condensate pressure signal according to the acquisition timing characteristics of the deaerator. The coupling coefficient matrix analysis module 102 is used to divide the operating conditions of the deaerator control unit to obtain different operating conditions, and analyze the coupling coefficient matrix of each operating condition based on the synchronous dataset. The collaborative decoupling factor determination module 103 is used to determine the collaborative decoupling factor for each operating condition point based on the historical operating data of each operating condition point and the coupling coefficient matrix. The collaborative control quantity analysis module 104 is used to construct an adaptive decoupling control model for the deaerator based on the collaborative decoupling factor, and to analyze the collaborative control quantity between the water level and condensate pressure of the deaerator using the adaptive decoupling control model. The collaborative decoupling control module 105 is used to convert the collaborative control quantity into a collaborative control command for the deaerator, and to control the collaborative decoupling of the deaerator water level and condensate pressure according to the collaborative control command.

[0082] In detail, the modules in the deaerator water level and condensate pressure collaborative decoupling control system 100 described in this embodiment of the invention employ the same methods as described above. Figure 1 The method described above uses the same technical means as the coordinated decoupling control method for deaerator water level and condensate pressure, and can produce the same technical effect, so it will not be repeated here.

[0083] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0084] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0085] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for coordinated decoupling control of deaerator water level and condensate pressure, characterized in that, The method includes: Collect water level signals and condensate pressure signals from the deaerator, and construct a synchronous dataset of the water level signals and condensate pressure signals based on the acquisition timing characteristics of the deaerator; The operating conditions of the deaerator control unit are divided to obtain different operating condition points. The coupling coefficient matrix of each operating condition point is analyzed based on the synchronous dataset. The collaborative decoupling factor for each operating condition point is determined based on the historical operating data of each operating condition point and the coupling coefficient matrix. An adaptive decoupling control model for the deaerator is constructed based on the aforementioned collaborative decoupling factor. The collaborative control variables of the deaerator's water level and condensate pressure are then analyzed using the adaptive decoupling control model. The coordinated control quantity is converted into a coordinated control command for the deaerator, and the coordinated decoupling of the deaerator water level and condensate pressure is controlled according to the coordinated control command.

2. The method as described in claim 1, characterized in that, The construction of the synchronous dataset for the water level signal and the condensate pressure signal based on the acquisition timing characteristics of the deaerator includes: Extract the timestamp sequences of the water level signal and the condensate pressure signal, and construct a time relationship table based on the timestamp sequences; The time series of the water level signal and the condensate pressure signal are nonlinearly aligned based on the time relationship table. The water level signal and the condensate pressure signal are resampled based on the aligned time series at a preset unified time reference point to generate a synchronous data sequence with consistent time resolution. Abnormal data points in the synchronized data sequence are identified based on a preset sliding window, and the abnormal data points are repaired by linear interpolation to obtain a synchronized dataset.

3. The method as described in claim 1, characterized in that, The deaerator control unit is divided into operating conditions to obtain different operating point values, including: Extract multi-dimensional operating parameters of the deaerator from the historical operating database of the deaerator; Density analysis of the multidimensional operating parameters is performed based on a preset density clustering algorithm to obtain dense regions with similar operating characteristics during the operation of the deaerator; The operating status of the deaerator is divided according to the dense area to obtain the operating condition type; Construct a feature parameter space for each type of operating condition, and build an operating condition classifier based on preset operating condition boundary conditions and the feature parameter space; The operating condition classifier is used to classify the operating conditions of the deaerator control unit to obtain different operating condition points.

4. The method as described in claim 1, characterized in that, The analysis of the coupling coefficient matrix for each operating condition point based on the synchronized dataset includes: Extract water level-pressure data pairs from each operating condition point in the synchronous dataset, and generate a training sample set for the operating condition based on the water level-pressure data pairs. A dynamic coupling model between water level and pressure is constructed based on the training sample set of the aforementioned working conditions using a pre-defined forgetting factor recursive algorithm. The coupling parameters in the dynamic coupling model are updated using the parameter update mechanism of the forgetting factor recursive algorithm to obtain real-time updated parameters; The coupling coefficient matrix for each operating condition point is determined based on the real-time updated parameters, wherein the coupling coefficient matrix characterizes the dynamic interaction strength between water level and pressure.

5. The method as described in claim 1, characterized in that, The step of determining the collaborative decoupling factor for each operating condition point based on historical operating data and the coupling coefficient matrix includes: Calculate the control performance indicators under each operating condition based on historical operating data for each operating condition point; The degree of coupling between the water level and pressure control loops is analyzed based on the coupling coefficient matrix and the control performance indicators. Based on the degree of coupling, the control characteristics and control target priorities for each operating point are determined; The collaborative decoupling factor for each operating point is determined by the control characteristics and the priority of the control objectives.

6. The method as described in claim 1, characterized in that, The step of constructing an adaptive decoupling control model for the deaerator based on the cooperative decoupling factor includes: The weight allocation ratio of the feedforward decoupling channel and the feedback correction channel is determined based on the aforementioned collaborative decoupling factor; A decoupling mathematical model for the deaerator is constructed based on the coupling coefficient matrix and the weight allocation ratio of the feedforward decoupling channel. An adaptive PID feedback controller is constructed based on the weight allocation ratio of the feedback correction channel; The feedforward decoupling mathematical model is combined in parallel with the adaptive PID feedback controller to generate a two-degree-of-freedom control structure. Stability analysis was performed on the two-degree-of-freedom control structure to obtain an adaptive decoupling control model for the deaerator.

7. The method as described in claim 1, characterized in that, The analysis of the coordinated control quantities of the deaerator water level and condensate pressure using the adaptive decoupling control model includes: The real-time collected water level signal and condensate pressure signal are input into the adaptive decoupling control model; In the feedforward decoupling channel of the adaptive decoupling control model, the feedforward compensation amount of the deaerator water level and condensate pressure is calculated based on the current coupling coefficient matrix and the cooperative decoupling factor. In the feedback correction channel of the adaptive decoupling control model, the feedback control quantities of the deaerator water level and condensate pressure are calculated by a preset adaptive PID algorithm. By fusing the feedforward compensation and the feedback control, a coordinated control quantity for the deaerator water level and condensate pressure is obtained.

8. The method as described in claim 1, characterized in that, The step of converting the coordinated control quantity into a coordinated control command for the deaerator includes: Based on the aforementioned coordinated control quantity, the regulating valve opening control component and the condensate pump frequency control component are separated. The control components of the regulating valve opening and the condensate pump frequency are respectively controlled by the preset actuator characteristic curves and converted into corresponding actuator commands. The actuator commands are constrained, and the constrained actuator commands are simultaneously sent to the actuators corresponding to the deaerator using a preset timing synchronization mechanism. The actuator command is verified by the execution status feedback of the actuator. When the quality of the actuator meets the preset command conditions, the actuator command is determined as the collaborative control command of the deaerator.

9. The method as described in claim 1, characterized in that, The coordinated decoupling of deaerator water level and condensate pressure according to the coordinated control command includes: Based on the execution status of the aforementioned collaborative control commands, real-time operational data on deaerator water level and condensate pressure are collected. Calculate the dynamic deviation index between the actual operating data and the preset target value, wherein the dynamic deviation index includes instantaneous deviation and cumulative deviation components; The level of coordinated control commands for deaerator water level and condensate pressure is analyzed based on the dynamic deviation index. The weight allocation of the cooperative decoupling factor is adjusted based on the cooperative control command level, and the structural parameters of the adaptive decoupling control model are optimized using the adjusted cooperative decoupling factor. The optimized adaptive decoupling control model regenerates the cooperative control quantity, and the deaerator water level and condensate pressure are controlled based on the regenerated cooperative control quantity to achieve cooperative decoupling.

10. A collaborative decoupling control system for deaerator water level and condensate pressure, characterized in that, The system for performing the method as described in any one of claims 1-9 includes: A synchronous dataset construction module is used to collect water level signals and condensate pressure signals from the deaerator, and to construct a synchronous dataset of the water level signals and condensate pressure signals based on the acquisition timing characteristics of the deaerator. The coupling coefficient matrix analysis module is used to divide the operating conditions of the deaerator control unit to obtain different operating conditions, and analyze the coupling coefficient matrix of each operating condition based on the synchronous dataset. The collaborative decoupling factor determination module is used to determine the collaborative decoupling factor for each operating condition point based on the historical operating data of each operating condition point and the coupling coefficient matrix. The collaborative control quantity analysis module is used to construct an adaptive decoupling control model for the deaerator based on the collaborative decoupling factor, and to analyze the collaborative control quantity between the water level and condensate pressure of the deaerator using the adaptive decoupling control model. The collaborative decoupling control module is used to convert the collaborative control quantity into collaborative control instructions for the deaerator, and to control the collaborative decoupling of the deaerator water level and condensate pressure according to the collaborative control instructions.

Images