A pressure balance control method for coordinated operation of a quenching boiler and a cracking furnace

Abstract

The present invention relates to the field of chemical production control technology and discloses a pressure balance control method for the coordinated operation of a quenching boiler and a cracking furnace. The method comprises: collecting data in real time through pressure sensors deployed at the inlet and outlet of the quenching boiler and temperature sensors in the reaction section of the cracking furnace, constructing a pressure state matrix, and predicting pressure distribution characteristics in combination with pressure gradient changes and flow path information. A coordinated control model is used to analyze the pressure intensity distribution, flow delay characteristics, and change trends, dynamically divide subunits, calculate the pressure intensity gradient, and generate the spatiotemporal distribution characteristics of the pressure balance. Based on this characteristic, the system automatically adjusts the valve opening, flow rate, and coolant distribution to achieve optimal resource allocation for the quenching boiler and the cracking furnace. The present invention improves pressure control accuracy and resource utilization, ensures stable equipment operation, and enhances chemical production efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of chemical production control, in particular to a pressure balance control method for the coordinated operation of a quenching boiler and a cracking furnace. Background Art

[0002] In the chemical production process, the coordinated operation of quench boilers and cracking furnaces is a critical link, and their pressure balance control directly affects the safety, stability, and efficiency of production. With the continuous development of the chemical industry, the demand for automation and intelligent production processes is becoming increasingly higher, and traditional pressure balance control methods are gradually showing some shortcomings.

[0003] Traditional control methods often rely on data from a single pressure or temperature sensor, making it difficult to fully and accurately reflect the pressure state during the coordinated operation of the quench boiler and cracking furnace. Due to the complex spatial relationships and interactions between the multiple pressure sensors deployed at the quench boiler inlet and outlet and the multiple temperature sensors in the cracking furnace reaction section, and the set distance between adjacent pressure sensors, the collection and processing of single data cannot fully account for the spatial and temporal distribution characteristics of pressure.

[0004] In actual operation, pressure balance is affected by a variety of factors, such as flow rate fluctuations, pipeline resistance, and equipment operating status. Traditional methods struggle to comprehensively analyze and predict these factors, resulting in inadequate and inaccurate pressure balance control. For example, when pressure fluctuations or anomalies occur in the system, traditional methods may be unable to quickly identify the root cause and impact, making it difficult to implement effective control measures in a timely manner. This can lead to decreased production efficiency and even safety accidents.

[0005] Furthermore, as chemical production continues to expand, the structures and operating parameters of quench boilers and cracking furnaces are becoming increasingly complex, placing increasing demands on pressure balance control. Traditional control methods lack the ability to dynamically model and optimize the system, making it difficult to adapt to the pressure balance control requirements under varying operating conditions. For example, under varying production loads, the system's pressure distribution and trends can vary significantly. Traditional methods are unable to adjust control strategies in a timely manner to these changes, impacting production stability and efficiency.

[0006] Furthermore, existing pressure balance control methods also have shortcomings in resource regulation. Traditional methods often employ fixed control strategies and fail to dynamically allocate resources based on the real-time distribution characteristics of pressure balance, resulting in inefficient resource utilization. For example, when pressure balance deviates, traditional methods may fail to accurately allocate resources to the physical units requiring regulation, thus affecting the speed and effectiveness of pressure balance recovery. Summary of the Invention

[0007] The object of the present invention is to provide a pressure balance control method for the coordinated operation of a quenching boiler and a cracking furnace to solve the problems raised in the above background technology.

[0008] To achieve the above object, the present invention provides the following technical solution: a pressure balance control method for the coordinated operation of a quenching boiler and a cracking furnace, the method comprising:

[0009] The pressure sensor and the temperature sensor are used to collect pressure data and temperature data during the coordinated operation of the quenching boiler and the cracking furnace to generate a pressure state data set;

[0010] Receive real-time data from multiple pressure sensors and multiple temperature sensors through the collaborative control server to construct a pressure state matrix;

[0011] Determining distribution characteristics of pressure balance based on the pressure state data set and real-time data from each sensor, wherein determining the distribution characteristics of pressure balance includes: processing a pressure state matrix, extracting pressure characteristics based on the pressure state data set, and predicting pressure distribution based on pressure gradient changes and flow path information, outputting spatiotemporal distribution characteristics of pressure balance through a collaborative control model, and updating the pressure state data set based on the spatiotemporal distribution characteristics;

[0012] According to the distribution characteristics, the resources of the quench boiler and the cracking furnace are dynamically regulated.

[0013] Preferably, the determination of the distribution characteristics of the pressure balance includes:

[0014] Processing the pressure state matrix to extract pressure intensity distribution, flow delay characteristics and pressure change trends;

[0015] Performing pressure intensity modeling on the pressure state matrix according to the pressure intensity distribution and flow delay characteristics, dividing the quench boiler and the cracking furnace into a plurality of subunits and marking the unit identifiers, performing correlation matching with the pressure state data set according to the pressure intensity of the subunits, and marking the unit identifiers in the pressure state data set;

[0016] Calculating a pressure intensity gradient according to the position of the pressure sensor, predicting the distribution of pressure balance according to the pressure intensity gradient and the pressure change trend, and calculating pressure prediction information for each subunit;

[0017] Constructing a collaborative control model, taking the pressure prediction information as an input parameter of the collaborative control model, performing spatial correlation modeling on the pressure prediction information through the collaborative control model, and outputting spatiotemporal distribution characteristics of pressure balance;

[0018] The pressure state data set is updated according to the spatiotemporal distribution characteristics to obtain the distribution characteristics of the pressure balance.

[0019] Preferably, the processing of the pressure state matrix includes:

[0020] The pressure state matrix is ​​normalized, the pressure hotspot area in the matrix is ​​intercepted by a sliding window, the hotspot area is noise filtered, and the pressure change trend is calculated by a differential decomposition algorithm;

[0021] Calculating the spatial correlation characteristics of the pressure state matrix, calculating the interference intensity between units, the pipeline stability coefficient and the pressure blind zone index based on the spatial correlation characteristics, constructing a feature fusion network, and calculating the flow delay characteristics through the feature fusion network;

[0022] The time domain features and frequency domain features collected by each pressure sensor are extracted, and the pressure feature vector of the sensor is calculated based on the phase difference between the time domain features and the frequency domain features. Based on the pressure feature vector, feature matching is performed on pressure sensors at different positions to calculate the pressure change trend.

[0023] Preferably, the pressure intensity modeling of the pressure state matrix includes:

[0024] According to the pressure intensity distribution, pressure intensity sampling points are extracted from each frame of data, and the sampling points are correlated and mapped with flow delay characteristics to generate a pressure intensity map. The pressure intensity maps collected by multiple sensors are spatially aligned to calculate the pressure intensity distribution of the unit;

[0025] Set the interference threshold value, locate the interference source according to the pressure intensity value of the multi-frame pressure state matrix, and calculate the interference intensity difference. If the interference intensity difference is greater than or equal to the interference threshold value, it means that there is a pressure blind spot in the unit. Perform flow model constraint compensation on the current unit, iteratively correct the pressure intensity distribution of the current unit according to the flow resistance loss model corresponding to the current unit, and calculate the intensity compensation value of the blind spot pressure based on the correction result;

[0026] The pressure intensity model is performed on the pressure state matrix according to the pressure intensity distribution, and the pressure model of the unit is marked with a time delay through the flow delay feature.

[0027] Preferably, the calculating of the pressure intensity gradient according to the position of the pressure sensor includes:

[0028] Based on multiple sets of pressure coverage data, the pressure intensity change points are extracted and mapped to a unified coordinate framework according to the deployment location of the sensor. The change points are fitted using a spatial interpolation algorithm to generate a unit pressure field model.

[0029] Performing sampling at equal intervals along the flow path of the pressure field model, calculating the pressure decay rate, interference fluctuation index, and pressure change slope of the path based on the sampling results, and calculating the pressure change parameter based on the pressure decay rate, interference fluctuation index, and pressure change slope;

[0030] Based on the deployment parameters and acquisition accuracy of the pressure sensors, the distribution characteristics of the pressure changes in each frame of data are projected onto the pressure field model. The pressure field model is partitioned along the flow direction according to the number of sensors. The changing patterns of the pressure changes within the partitions are analyzed, and the pressure distribution characteristics are calculated based on these changing patterns.

[0031] The pressure intensity gradient is calculated according to the pressure change parameter and the pressure distribution characteristics. The calculation process of the pressure intensity gradient includes: based on the coordinate position range from the first pressure sensor to the last pressure sensor, selecting spatial coordinate points in the sensor deployment direction, accumulating the product of the pressure field strength characteristic weight value and the pressure distribution characteristic weight value within the spatial resolution range, and superimposing the influence value of the sensor acquisition frequency on the pressure intensity change rate.

[0032] Preferably, the calculating of the pressure prediction information of each subunit includes:

[0033] Using the main flow path of the pressure field model as a baseline and the peak position of the pressure change in each frame of data as a reference point, the pressure offset is calculated and a pressure distribution curve is drawn according to the coordinates;

[0034] Correcting the growth rate and direction of the pressure change trend according to the pressure intensity gradient;

[0035] Starting from the most recent pressure distribution point, the distribution curve is continuously drawn according to the correction results of the growth rate and direction to generate the pressure distribution points for the next period until the distribution points cover the entire target unit and the pressure prediction information is generated.

[0036] Preferably, the constructing of the collaborative control model includes:

[0037] The input layer is used to organize the pressure prediction information into spatial distribution data and perform normalization processing;

[0038] The feature fusion layer is used to extract the unit-related features of pressure by processing spatial distribution data and construct the dependency relationship between physical units;

[0039] The resource allocation layer is used to integrate the correlation between pressure balance in spatial units and generate resource regulation strategies.

[0040] Preferably, the obtaining of distribution characteristics of pressure balance includes:

[0041] According to the spatiotemporal distribution characteristics of the pressure balance output by the collaborative control model, the identification of the subunit is matched with the spatiotemporal distribution characteristics;

[0042] The unit data in the pressure state dataset are reorganized according to the spatiotemporal characteristics to generate a unit distribution map sorted by the pressure balance intensity;

[0043] According to the reorganized unit distribution map, the optimized pressure balance distribution characteristics are output.

[0044] Preferably, the dynamic regulation of the resources of the quench boiler and the cracking furnace includes:

[0045] Mapping unit identifications to units of pressure balance distribution characteristics one by one;

[0046] According to the temporal and spatial distribution characteristics of pressure balance, the execution actions of control resources are controlled, including valve opening adjustment, flow regulation and coolant distribution operations;

[0047] Resources are dynamically allocated to corresponding physical units based on the spatial distribution of pressure balance and preset control strategies.

[0048] Preferably, the execution action of the control resource includes:

[0049] When the pressure balance in the target unit reaches a preset intensity threshold, an opening increase instruction of the adjacent valve is triggered;

[0050] Dynamically combine available fluid resources according to the flow regulation strategy to generate a coolant allocation vector;

[0051] A pipeline parameter of a target unit transmission node is adjusted based on the coolant distribution vector.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] By deploying multiple pressure sensors at the inlet and outlet of the quench boiler, and multiple temperature sensors in the cracking furnace reaction section, it is possible to collect pressure and temperature data in real time and comprehensively during system operation, generate a pressure state data set, and provide an accurate and rich data basis for subsequent pressure balance control.

[0054] The collaborative control server receives real-time data from multiple sensors and constructs a pressure state matrix. By processing this matrix, it can extract key information such as pressure intensity distribution, flow delay characteristics, and pressure change trends, thereby providing a deeper understanding of the system's pressure state. Normalization, noise filtering, and differential decomposition of the pressure state matrix effectively improve the quality and reliability of the data, ensuring the accuracy of subsequent analysis.

[0055] To determine the distribution characteristics of pressure balance, we modeled the pressure intensity within the pressure state matrix, divided the quench boiler and cracking furnace into multiple subunits, and labeled these subunits, enabling refined system management. Correlating and matching the subunit pressure intensity with the pressure state dataset allows for more precise understanding of the pressure conditions within each subunit. Furthermore, by calculating the pressure intensity gradient and predicting the pressure balance distribution, we provide a scientific basis for dynamic resource regulation.

[0056] The collaborative control model constructed includes an input layer, a feature fusion layer, and a resource allocation layer. It can model the spatial correlation of pressure prediction information and output the spatiotemporal distribution characteristics of pressure balance. This enables the system to comprehensively understand the distribution of pressure balance in both time and space, providing strong support for the precise allocation and dynamic adjustment of resources. Updating the pressure state dataset based on the spatiotemporal distribution characteristics can continuously optimize the system's control strategy and improve control accuracy and adaptability.

[0057] When dynamically regulating the resources of the quench boiler and cracking furnace, the system maps unit identifiers to units with pressure balance distribution characteristics. This allows precise control of actions such as valve opening adjustment, flow regulation, and coolant distribution based on the spatiotemporal distribution of pressure balance, achieving dynamic resource allocation. When the pressure balance in a target unit reaches a preset intensity threshold, it triggers an increase in the opening of adjacent valves, enabling timely response to pressure changes and ensuring stable system operation. The system dynamically combines available fluid resources according to the flow regulation strategy to generate a coolant distribution vector, and adjusts the pipeline parameters of the target unit's transmission nodes based on this vector, improving resource utilization and system operational efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a working principle diagram of the pressure balance control method for the coordinated operation of the quenching boiler and the cracking furnace according to the present invention;

[0059] Figure 2 Flowchart for pressure balance distribution characterization;

[0060] Figure 3 Flowchart for stress state matrix processing;

[0061] Figure 4 Flowchart for pressure intensity gradient calculation. DETAILED DESCRIPTION

[0062] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0063] See also Figures 1-4 The present invention relates to a pressure balance control method for the coordinated operation of a quench boiler and a cracking furnace. The method includes multiple pressure sensors deployed at the inlet and outlet of the quench boiler, multiple temperature sensors deployed in the reaction section of the cracking furnace, and a collaborative control server connected to these sensors. Adjacent pressure sensors are spaced a set distance apart. The specific implementation steps are as follows:

[0064] Pressure and temperature sensors collect pressure and temperature data during the coordinated operation of the quench boiler and cracking furnace to generate a pressure state dataset. The pressure sensor monitors pressure changes at different locations in the quench boiler in real time, while the temperature sensor collects the temperature of the cracking furnace reaction zone in real time.

[0065] The collaborative control server receives real-time data from multiple pressure and temperature sensors and constructs a pressure state matrix. The server integrates and processes the received data, storing the real-time pressure and temperature state information in the form of a matrix.

[0066] The distribution characteristics of pressure balance are determined based on the pressure state dataset and the real-time data from each sensor. This process includes processing the pressure state matrix, extracting pressure features based on the pressure state dataset, and predicting pressure distribution based on pressure gradient changes and flow path information. The collaborative control model outputs the spatiotemporal distribution characteristics of pressure balance, and then updates the pressure state dataset based on these spatiotemporal distribution characteristics.

[0067] Based on the distribution characteristics, the resources of the quench boiler and cracking furnace are dynamically controlled. Based on the determined pressure balance distribution characteristics, the resources in the system are rationally allocated and adjusted to achieve pressure balance and stable operation of the system.

[0068] Example 1: When determining the distribution characteristics of pressure balance, the pressure state matrix needs to be processed to extract the pressure intensity distribution, flow delay characteristics and pressure change trends. First, the pressure state matrix is ​​standardized to eliminate the influence of data of different dimensions and make the data comparable. Then, the pressure hotspot areas in the matrix are intercepted through a sliding window. These areas are often locations where pressure changes drastically or are critical to the operation of the system. Noise filtering is performed on them, and appropriate filtering algorithms are used to remove noise mixed in during data acquisition and transmission to ensure data accuracy. The processed data is then analyzed using a differential decomposition algorithm to calculate the pressure change trend, so as to understand the law of pressure change over time.

[0069] When calculating the spatial correlation characteristics of the pressure state matrix, the correlation between data at different locations in the matrix is ​​analyzed to determine the interference intensity between units, the pipeline stability coefficient, and the pressure blind zone index. Based on these characteristics, a feature fusion network is constructed. This network can comprehensively process various feature information. Through network calculations, flow delay characteristics are obtained, thereby understanding the delay of fluid flow in the system.

[0070] The time and frequency domain features of the data collected by each pressure sensor are extracted. Time domain features reflect pressure variations over time, such as the amplitude and period of pressure fluctuations. Frequency domain features reveal the frequency components of pressure fluctuations, helping to analyze the causes of pressure fluctuations. The sensor's pressure feature vector is calculated based on the phase difference between the time and frequency domain features. This vector comprehensively characterizes the pressure characteristics at the sensor's location. Based on the pressure feature vector, feature matching is performed on pressure sensors at different locations. This matching analysis analyzes the pressure correlations between different sensors and allows the pressure trend to be calculated.

[0071] The pressure intensity modeling is performed on the pressure state matrix based on the pressure intensity distribution and flow delay characteristics. Pressure intensity sampling points are extracted from each frame of data. These sampling points are key information points that represent the pressure intensity in that frame of data. The sampling points are correlated with the flow delay characteristics and mapped, combining the pressure intensity with the flow delay characteristics of the fluid to generate a pressure intensity map. The pressure intensity maps collected by multiple sensors are spatially aligned to ensure spatial consistency and comparability of data from different sensors. This allows the pressure intensity distribution of each unit to be calculated, providing a clear understanding of the pressure intensity within each unit.

[0072] An interference threshold is set, which is used to determine whether a unit has a pressure blind spot. The interference source is located based on the pressure intensity value of the multi-frame pressure state matrix to determine the source of the interference. The interference intensity difference is calculated. If the difference is greater than or equal to the interference threshold, it indicates that the unit has a pressure blind spot. At this time, the flow model constraint compensation is performed on the current unit, and the pressure intensity distribution of the current unit is iteratively corrected based on the flow resistance loss model corresponding to the current unit. Through continuous iteration, the pressure intensity distribution is made more consistent with the actual situation. The intensity compensation value of the blind spot pressure is calculated based on the correction result to compensate for the impact of the pressure blind spot on the system operation. Finally, the pressure state matrix is ​​pressure-intensity modeled according to the pressure intensity distribution, and a model that accurately describes the system pressure intensity is established. The pressure model of the unit is time-delayed and annotated through the flow delay characteristics so that the model can reflect the time delay characteristics of the pressure change.

[0073] To calculate the pressure intensity gradient based on the location of pressure sensors, multiple sets of pressure coverage data are first processed to extract pressure intensity change points. These points represent locations where significant pressure changes occur and are therefore of great analytical value. Based on the sensor deployment locations, these change points are mapped to a unified coordinate framework, allowing analysis and processing of change points at different locations within the same coordinate system. Spatial interpolation algorithms are used to fit these change points to generate a pressure field model for the cell, which provides a visual representation of the pressure distribution within the cell.

[0074] Sampling is performed at equal intervals along the flow path of the pressure field model to obtain pressure data at each point along the path. Based on the sampling results, the path's pressure attenuation rate, interference fluctuation index, and pressure change slope are calculated. These parameters reflect the varying characteristics of pressure along the flow path. Pressure change parameters are calculated based on these parameters, laying the foundation for pressure intensity gradient calculations. Based on the deployment parameters and acquisition accuracy of the pressure sensors, the distribution characteristics of pressure changes in each frame of data are projected onto the pressure field model. The pressure field model is partitioned along the flow direction according to the number of sensors. The pressure changes within each partition are analyzed to identify patterns in pressure variation within the partition. Based on these patterns, the pressure distribution characteristics are calculated to understand the distribution of pressure in different regions.

[0075] Finally, the pressure intensity gradient is calculated based on the pressure change parameters and pressure distribution characteristics. The specific process is as follows: Based on the coordinate position range from the first pressure sensor to the last pressure sensor, spatial coordinate points are selected in the sensor deployment direction. The product of the pressure field intensity characteristic weight value and the pressure distribution characteristic weight value within the spatial resolution range is cumulatively calculated. At the same time, the influence of the sensor acquisition frequency on the pressure intensity change rate is added to obtain an accurate pressure intensity gradient.

[0076] When calculating the pressure prediction for each subunit, the pressure field model's main flow path is used as the baseline, and the peak position of the pressure change in each frame of data is used as the reference point. The pressure offset is calculated, and a pressure distribution curve is plotted according to the coordinates. Based on the pressure intensity gradient, the growth rate and direction of the pressure change trend are corrected to make the prediction more realistic. Starting from the most recent pressure distribution point, the distribution curve is drawn based on the corrections to the growth rate and direction, generating the pressure distribution points for the next period until the distribution points cover the entire target unit, thus generating the pressure prediction information.

[0077] A collaborative control model is constructed, consisting of an input layer, a feature fusion layer, and a resource allocation layer. The input layer organizes pressure prediction information into spatially distributed data and standardizes it to a uniform format for subsequent processing. The feature fusion layer processes the spatially distributed data to extract unit-related pressure features and construct dependencies between physical units, thereby understanding the mutual influence between them. The resource allocation layer integrates the correlation between pressure balances at the spatial level and generates resource control strategies, providing guidance for the dynamic regulation of system resources.

[0078] Based on the spatiotemporal distribution characteristics of pressure balance output by the collaborative control model, the subunit identifiers are mapped to these characteristics, enabling accurate identification of the spatiotemporal pressure balance characteristics of each subunit. The unit data in the pressure state dataset is reorganized according to the spatiotemporal characteristics, generating a unit distribution map sorted by pressure balance strength to visually display the pressure balance status of each unit. Based on the reorganized unit distribution map, the optimized pressure balance distribution characteristics are output.

[0079] Example 2: When processing the pressure state matrix, the matrix needs to be normalized. Because pressure sensors are deployed at different locations at the inlet and outlet of the quench boiler, the dimensions of the collected data may differ. Normalization can eliminate the impact of different dimensions on the data, making data from sensors at different locations comparable and providing a unified data foundation for subsequent analysis. For example, mapping the real-time data from each pressure sensor to the same numerical range ensures that dimensionality issues will not cause analytical bias during subsequent processing.

[0080] After normalization, a sliding window technique is used to capture pressure hotspots within the matrix. The sliding window size is determined based on the set distance between adjacent pressure sensors and the frequency of system pressure changes. Typically, a window size that covers data from at least two adjacent sensors is selected to ensure that the captured area contains sufficient pressure variation information. These hotspots are typically locations with significant pressure gradient variations, such as the quench boiler inlet and outlet pipe elbows. Pressure variations at these locations significantly impact the overall system pressure balance and require focused analysis.

[0081] A combination of median filtering and Kalman filtering is used to filter noise in the captured hotspots. Median filtering effectively removes impulse noise and avoids data anomalies caused by transient sensor interference. Kalman filtering, based on the system state equation and observation equation, optimizes the pressure data, further smoothing the data and reducing the impact of random noise.

[0082] When calculating pressure trends using the differential decomposition algorithm, the time series pressure data is decomposed into a trend term, a cycle term, and a random term. The trend term reflects the overall direction of pressure change, the cycle term reflects the periodic fluctuations of pressure over time, and the random term represents unpredictable, random fluctuations.

[0083] When calculating the spatial correlation characteristics of the pressure state matrix, the correlation coefficients of the data in different rows and columns of the matrix are analyzed. For example, for multiple pressure sensors deployed on the same pipeline section, the correlation of their collected data should be high. If the correlation is abnormally reduced, it may indicate problems such as pipeline leakage or blockage near the section. Based on these correlation characteristics, the interference intensity between units is calculated. This intensity value reflects the degree of mutual influence of pressure changes in adjacent units. At the same time, the pipeline stability coefficient is calculated to evaluate the stability performance of the pipeline under pressure fluctuations; and the pressure blind zone index is used to identify areas in the system where insufficient pressure monitoring may exist.

[0084] When constructing the feature fusion network, a multi-layer perceptron structure was employed, incorporating spatial correlation features, interference intensity, pipeline stability coefficient, and pressure blind zone index as input layer nodes. Nonlinear feature fusion was achieved through weight matrix operations in the hidden layer. The network's output layer nodes correspond to flow delay features. This feature continuously adjusts network weights based on training data, ensuring that the output flow delay features accurately reflect the actual flow delay of the fluid in the pipeline, providing key parameters for subsequent pressure intensity modeling.

[0085] For the data collected by each pressure sensor, time domain and frequency domain features are extracted. Time domain feature extraction involves calculating the mean, variance, peak value, rise time, and other parameters of the pressure data. These parameters can intuitively reflect the changing characteristics of pressure in the time dimension, such as the amplitude and frequency of pressure fluctuations. Frequency domain feature extraction converts the time domain signal into a frequency domain signal through fast Fourier transform to obtain the main frequency components of the pressure fluctuations. For example, if a certain frequency component in the frequency domain accounts for a significant proportion, it may indicate the presence of a periodic pressure disturbance source in the system, such as the operating frequency of a pump.

[0086] When calculating the sensor's pressure characteristic vector based on the phase difference between the time and frequency domain characteristics, the corresponding frequency domain phases of each characteristic point in the time domain signal (such as peaks and zero crossings) are first determined. The statistics of these phase differences (such as mean and standard deviation) are calculated and combined with key parameters of the time and frequency domain characteristics to form a multidimensional vector. This vector contains the time, frequency, and phase information of the pressure change at the sensor's location, fully characterizing the pressure characteristics at that location.

[0087] When matching the features of pressure sensors at different locations, the cosine similarity algorithm is used to calculate the similarity of each sensor's pressure feature vector. Sensors with high similarity indicate similar pressure variation characteristics at their respective locations and can be classified as belonging to the same pressure feature region. Sensors with low similarity may be located in different pressure-affected regions or exhibit abnormalities. This matching analysis can be used to divide the system into multiple pressure feature regions, providing a basis for subsequent pressure distribution prediction and resource regulation.

[0088] When calculating pressure trends, we comprehensively consider regional pressure changes after matching the characteristics of each sensor. For sensors within the same characteristic region, we take the average of their pressure trends as the pressure trend for that region. For different regions, we analyze the differences and correlations between their trends to obtain a comprehensive picture of the pressure trend for the entire system.

[0089] Example 3: When modeling the pressure intensity of the pressure state matrix, it is first necessary to extract pressure intensity sampling points in each frame of data based on the pressure intensity distribution. Since pressure sensors are deployed at different locations of the quench boiler inlet, outlet, and pipeline, each frame of data will show pressure intensity values ​​at different locations. By setting a sampling threshold, points with significant changes in pressure intensity are selected as sampling points. These points can effectively characterize the pressure characteristics of the frame data. For example, in the pressure sensor data at the quench boiler inlet, when the pressure intensity exceeds 10% of the upper and lower limits of the normal operating range, the point is marked as a sampling point.

[0090] The extracted sampling points are mapped to the flow delay characteristics. The flow delay characteristics are calculated by analyzing the spatial correlation characteristics of the pressure state matrix and calculating the feature fusion network. They reflect the time delay of the fluid flowing from one location to another in the pipeline. By establishing a mapping relationship between the pressure intensity value of the sampling point and the flow delay time at the corresponding location, a pressure intensity map is generated. This map uses spatial position as the horizontal axis and pressure intensity as the vertical axis. The depth of color in the map represents the length of the flow delay time, thus intuitively linking these two key parameters, pressure intensity and flow delay, in the spatial dimension.

[0091] The pressure intensity maps collected by multiple sensors are spatially aligned. Because the sensors are deployed in different locations, a unified spatial coordinate system needs to be established based on the physical structure of the quench boiler and cracking furnace. For example, with the inlet of the cracking furnace reaction section as the coordinate origin, the x-axis along the material flow direction, and the y- and z-axes perpendicular to the flow direction, the pressure intensity maps of each sensor are converted to this coordinate system to ensure consistency and comparability in spatial position of the data from different sensors. The pressure intensity distribution of the unit is then calculated, clarifying the spatial distribution of pressure intensity within each subunit.

[0092] When setting the interference threshold, it's important to consider the pressure fluctuation range during normal system operation and historical interference data. By statistically analyzing the pressure data during normal system operation, the mean and standard deviation of the pressure intensity are calculated. The interference threshold is typically set to the mean plus twice the standard deviation, serving as a benchmark for determining whether a unit has a pressure blind spot. When locating interference sources based on the pressure intensity values ​​in a multi-frame pressure state matrix, the approximate location of the interference source is determined by comparing the magnitude and temporal sequence of pressure intensity changes at different locations.

[0093] To calculate the interference intensity difference, for adjacent subcells, the pressure intensity change values ​​within the same time period are obtained. The difference is then calculated by subtracting the smaller change from the larger change. If this difference is greater than or equal to the interference threshold, it indicates a possible pressure blind spot in the cell—meaning that the pressure in that area is not being effectively monitored or that an abnormal pressure distribution exists. Flow model constraint compensation is then applied to the current cell. Based on the corresponding flow resistance loss model, which describes the pressure loss caused by pipe resistance, elbows, and valves when the fluid flows within the cell, the pressure intensity distribution of the current cell is iteratively corrected. First, a theoretical pressure loss is calculated based on the initial pressure intensity distribution and the flow resistance loss model. This is then compared with the actual measured pressure difference. If the difference is significant, the pressure intensity distribution parameters are adjusted and recalculated until the error between the theoretical calculation and the actual measurement is within the allowable range (typically set to no more than 5%). Based on the corrected pressure intensity distribution, a pressure intensity compensation value is calculated for the blind spot pressure. This compensation value is used to subsequently correct the pressure in the blind spot to ensure the accuracy of the system pressure balance analysis.

[0094] Based on the pressure intensity distribution, the pressure state matrix is ​​modeled for pressure intensity. The pressure intensity values ​​in each frame of data are organized according to spatial position and time sequence, forming a model that can reflect the dynamic changes in the system's pressure intensity. Simultaneously, the unit pressure model is time-delayed using the flow delay feature. A corresponding flow delay time label is added to each pressure intensity data point in the model. This allows the model to reflect not only the spatial distribution of pressure intensity but also the time delay characteristics of pressure changes, thereby more accurately describing the system's pressure dynamic behavior.

[0095] Example 4: When calculating the pressure intensity gradient based on the position of the pressure sensor, multiple sets of pressure coverage data need to be processed. These data come from pressure sensors deployed at the inlet, outlet and different positions of the quenching boiler. For example, in a chemical production system, 2 pressure sensors are deployed at the inlet of the quenching boiler, 3 pressure sensors are deployed at the outlet, and 1 pressure sensor is deployed every 5 meters on the pipeline. The distance between two adjacent pressure sensors is 5 meters. When the system is running, these sensors will collect pressure data in real time to form multiple sets of pressure coverage data. From these data, the pressure intensity change point, that is, the position where the pressure value changes significantly, is extracted. For example, when the difference between the pressure value of a sensor at two adjacent sampling moments exceeds 15% of the normal fluctuation range, the pressure point at that moment is marked as a change point.

[0096] Based on the sensor deployment locations, the change points are mapped to a unified coordinate framework. With the center of the cracking furnace reaction section as the coordinate origin, the direction of material flow from the cracking furnace to the quench boiler is set as the positive x-axis, and the positive z-axis is perpendicular to the ground and upward, establishing a three-dimensional rectangular coordinate system. For example, the coordinates of the first pressure sensor at the inlet are (0, 0, 1.5), the second is (0, 0, 2.0), and the coordinates of the first sensor at the outlet are (10, 0, 1.8). The locations of all pressure intensity change points are mapped to this coordinate system based on the coordinates of the corresponding sensors, allowing change points at different locations to be analyzed within the same coordinate system.

[0097] A spatial interpolation algorithm is used to fit the change points and generate a pressure field model for the unit. Kriging interpolation is used, which accounts for the spatial correlation of the data. Based on the coordinates and pressure values ​​of the change points, an estimated pressure value is calculated for any location within the entire unit space, thereby constructing a continuous pressure field model. For example, for the piping unit between the quench boiler and the cracking furnace, the interpolation method is used to fit the pressure distribution surface within the pipeline, visually displaying the pressure distribution within the unit, such as the changing trend of the pressure distribution at the pipeline elbow.

[0098] Sampling is performed at equal intervals along the flow path of the pressure field model. Assuming the flow path is the material flow trajectory from the outlet of the cracking furnace reaction section to the inlet of the quench boiler, pressure data is obtained at sampling points every 0.5 meters. Based on the sampling results, the path's pressure decay rate (the pressure drop per unit length) is calculated. For example, in a 5-meter-long pipeline with an inlet pressure of 1.2 MPa and an outlet pressure of 1.1 MPa, the pressure decay rate is (1.2-1.1) / 5 = 0.02 MPa / m. The interference fluctuation index (IFI) is calculated by analyzing the amplitude and frequency of pressure fluctuations at the sampling points to assess the impact of interference on pressure. For example, if pressure fluctuates frequently and significantly within a short period of time, the IFI is high. The pressure change slope (the rate of change of pressure with distance) is calculated to reflect the speed of pressure change along the flow path. For example, if the pressure drops from 1.0 MPa to 0.8 MPa in a 4-meter section, the slope is (0.8-1.0) / 4 = -0.05 MPa / m. The pressure change parameter is calculated based on these parameters, and the pressure attenuation rate, interference fluctuation index and pressure change slope are weighted and summed according to certain weights (such as 0.4, 0.3, and 0.3 respectively) to obtain a comprehensive pressure change parameter, which is used to characterize the pressure change characteristics on the flow path.

[0099] Based on the deployment parameters and acquisition accuracy of the pressure sensors, for example, a certain pressure sensor model is deployed at a height of 1.5 meters above the ground with an acquisition accuracy of ±0.01 MPa, the distribution characteristics of pressure changes in each frame of data are projected onto the pressure field model. Each frame of data contains the pressure values ​​of each sensor at the same moment. These pressure values ​​are mapped to corresponding positions in the pressure field model to form a projection of the pressure distribution at that moment. The pressure field model is partitioned along the flow direction based on the number of sensors. If a total of 10 sensors are deployed along the flow direction, the flow path is divided into 10 intervals, each corresponding to the monitoring range of a sensor. The pressure variation patterns within the intervals are analyzed. For example, within a certain interval, the pressure fluctuates periodically over time, with a fluctuation period of 5 minutes and an amplitude of 0.05 MPa. Based on this pattern, the pressure distribution characteristics are calculated to determine the average pressure level and fluctuation range within the interval.

[0100] The pressure intensity gradient is calculated based on pressure variation parameters and pressure distribution characteristics. Based on the coordinate range from the first to the last pressure sensor, for example, the first sensor's coordinates are (0, 0, 1.5) and the last's are (15, 0, 1.7), several spatial coordinate points are selected along the sensor deployment direction (x-axis), for example, one point every 1 meter. For each coordinate point, the product of the pressure field intensity characteristic weight and the pressure distribution characteristic weight within the spatial resolution range (e.g., a spherical area centered at the point with a radius of 0.5 meters) is calculated. The pressure field intensity characteristic weight is determined by the magnitude of the pressure gradient within the area, while the pressure distribution characteristic weight is determined by the stability of the pressure distribution within the area. The effect of the sensor acquisition frequency on the rate of change of pressure intensity is also added. For example, if the sensor acquisition frequency is 10 Hz, a higher acquisition frequency can reflect pressure changes more promptly and has a greater impact on the rate of change; a lower acquisition frequency has a smaller impact. By accumulating these products and influence values, the pressure intensity gradient is obtained, which reflects the intensity and direction of pressure changes in space. For example, in the x-axis direction, the pressure intensity gradient is 0.03 MPa / m, which means that the pressure increases by an average of 0.03 MPa for every meter along the x-axis.

[0101] During the calculation process, the actual situation of sensor deployment must be fully considered, such as whether the sensor locations are evenly spaced and whether the data collection accuracy is consistent. For example, if sensors are densely deployed in a certain area, the data collected can more accurately reflect pressure changes, and this area should be given a higher weight when calculating the pressure intensity gradient. If the data collection accuracy of a certain sensor is low, its data needs to be appropriately corrected during the calculation. This calculation method based on actual deployment parameters and data characteristics ensures that the calculated pressure intensity gradient accurately reflects the actual pressure distribution of the system, providing a key basis for subsequent calculation of sub-unit pressure prediction information, building collaborative control models, and dynamic resource regulation.

[0102] Example 5: When dynamically regulating the resources of the quench boiler and the cracking furnace, it is first necessary to map the unit identification to the unit of the distribution characteristic of the pressure balance one by one. Taking a chemical production system as an example, the system divides the quench boiler and the cracking furnace into 5 units according to the physical structure, namely the cracking furnace reaction section unit, the quench boiler inlet pipe unit, the quench boiler heat exchange tube unit, the quench boiler outlet pipe unit and the gas-liquid separation unit. Each unit is given a unique identification, such as U1, U2, U3, U4, and U5. Through the previously determined pressure balance distribution characteristics, the pressure balance state of each unit is clarified. For example, the pressure intensity distribution of the U2 unit at a certain moment presents a gradient characteristic of high at the inlet and low at the outlet. This characteristic is associated with the identification of the U2 unit and mapped to establish a clear corresponding relationship so that the pressure situation of each unit can be accurately regulated later.

[0103] Based on the spatiotemporal distribution of pressure balance, control resource actions, including valve opening adjustment, flow regulation, and coolant distribution, are executed. When the pressure balance in a target unit reaches a preset intensity threshold, an instruction to increase the opening of adjacent valves is triggered. For example, if the pressure intensity of unit U3 (the quench boiler heat exchanger unit) exceeds the preset threshold of 1.5 MPa, the system automatically determines that the unit may be at risk of overpressure. At this time, an instruction to increase the opening of valve V1 at the inlet of unit U3 is triggered, raising the valve opening from 50% to 60%, thereby increasing fluid flow and reducing the pressure in the unit.

[0104] The system dynamically combines available fluid resources based on the flow regulation strategy to generate a coolant allocation vector. Assume that the system's available coolant resources include circulating water and chilled water, with a flow regulation range of 100-300 m³ / h for circulating water and 50-150 m³ / h for chilled water. When the pressure balance distribution at unit U4 (the quench boiler outlet piping unit) indicates high outlet temperatures and large pressure fluctuations, the system, based on the preset flow regulation strategy, calculates the required combination of 200 m³ / h of circulating water and 100 m³ / h of chilled water, generating a coolant allocation vector of [200, 100]. Based on this vector, the pipeline parameters of the target unit's transmission nodes are adjusted, such as adjusting the valve opening of the corresponding pipeline, so that circulating water and chilled water enter unit U4 at the set flow rates, cooling the unit and stabilizing the pressure.

[0105] Resources are dynamically allocated to corresponding physical units based on the spatial distribution of pressure balance and pre-set control strategies. For example, pressure field model analysis revealed a pressure blind spot in the central region of unit U1 (the cracking furnace reaction section). Pressure in this area was not effectively transmitted to the sensor, resulting in inadequate pressure monitoring. Based on the pre-set control strategy, the coolant supply to this area needed to be increased to balance the pressure. The system automatically allocated an additional 50 m³ / h of circulating water to the central region of unit U1, injecting it through the cooling water pipeline in this area to adjust the pressure distribution there and improve the pressure blind spot.

[0106] When adjusting valve opening, it's important to consider both the valve's adjustment accuracy and response time. For example, if valve V1 has an adjustment accuracy of ±1% and a response time of 5 seconds, the system will monitor the actual valve opening in real time after issuing an opening increase command to ensure the valve operates accurately according to the command. If, after 5 seconds, the valve opening is less than 60%, the system will issue a follow-up adjustment command until the valve opening reaches the target value.

[0107] When regulating flow, the supply capacity of each fluid resource and the system's pressure balance requirements must be comprehensively considered. If the system requires flow regulation for multiple units simultaneously, such as when both U2 and U3 require increased flow, the system will allocate flow appropriately based on the pressure balance priority of each unit and the total supply of fluid resources. For example, if U2 presents a higher risk of pressure imbalance and is set to priority 1, while U3 is set to priority 2 and the total circulating water supply is 400 m³ / h, U2's required 200 m³ / h will be met first, and the remaining 200 m³ / h will be allocated to U3.

[0108] During coolant distribution, the impact of coolant temperature and pressure on the system must be considered. The circulating water temperature is 25°C and the pressure is 0.4 MPa, while the chilled water temperature is 5°C and the pressure is 0.6 MPa. When distributing coolant, ensure that the coolant temperature and pressure entering each unit meet the unit's process requirements to avoid adverse effects on the system due to abnormal coolant parameters. For example, the upper limit of the coolant pressure allowed by the heat exchange tubes of unit U3 is 0.5 MPa. When distributing chilled water, the chilled water pressure must be reduced from 0.6 MPa to 0.45 MPa using a pressure regulating valve before injection into the unit.

[0109] During dynamic resource control, the system monitors changes in the pressure balance of each unit in real time and adjusts control strategies based on the latest pressure distribution characteristics. For example, after increasing the valve opening and distributing coolant to unit U3, the system continuously monitors changes in the unit's pressure intensity. If the pressure gradually drops to 1.2 MPa within 10 minutes, which is within the normal range, the current control state is maintained. If the pressure drop is not significant or continues to rise, a backup control plan is activated, such as further increasing the valve opening or increasing the coolant supply.

[0110] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0111] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A pressure balance control method for the coordinated operation of a quenching boiler and a cracking furnace, applied to a chemical production control system, comprising a plurality of pressure sensors deployed at the inlet and outlet of the quenching boiler, a plurality of temperature sensors deployed in the reaction section of the cracking furnace, and a coordinated control server connected to the pressure sensors and the temperature sensors, wherein two adjacent pressure sensors are spaced a set distance apart, characterized in that: The method comprises: The pressure sensor and the temperature sensor are used to collect pressure data and temperature data during the coordinated operation of the quenching boiler and the cracking furnace to generate a pressure state data set; Receive real-time data from multiple pressure sensors and multiple temperature sensors through the collaborative control server to construct a pressure state matrix; Determining distribution characteristics of pressure balance based on the pressure state data set and real-time data from each sensor, wherein determining the distribution characteristics of pressure balance includes: processing a pressure state matrix, extracting pressure characteristics based on the pressure state data set, and predicting pressure distribution based on pressure gradient changes and flow path information, outputting spatiotemporal distribution characteristics of pressure balance through a collaborative control model, and updating the pressure state data set based on the spatiotemporal distribution characteristics; Dynamically regulating the resources of the quench boiler and the cracking furnace according to the distribution characteristics; The determination of the distribution characteristics of the pressure balance includes: Processing the pressure state matrix to extract pressure intensity distribution, flow delay characteristics and pressure change trends; Performing pressure intensity modeling on the pressure state matrix according to the pressure intensity distribution and flow delay characteristics, dividing the quench boiler and the cracking furnace into a plurality of subunits and marking the unit identifiers, performing correlation matching with the pressure state data set according to the pressure intensity of the subunits, and marking the unit identifiers in the pressure state data set; Calculating a pressure intensity gradient according to the position of the pressure sensor, predicting the distribution of pressure balance according to the pressure intensity gradient and the pressure change trend, and calculating pressure prediction information for each subunit; Constructing a collaborative control model, taking the pressure prediction information as an input parameter of the collaborative control model, performing spatial correlation modeling on the pressure prediction information through the collaborative control model, and outputting spatiotemporal distribution characteristics of pressure balance; The pressure state data set is updated according to the spatiotemporal distribution characteristics to obtain the distribution characteristics of the pressure balance.

2. The pressure balance control method for the coordinated operation of a quenching boiler and a cracking furnace according to claim 1, characterized in that: The processing of the pressure state matrix includes: The pressure state matrix is ​​normalized, the pressure hotspot area in the matrix is ​​intercepted by a sliding window, the hotspot area is noise filtered, and the pressure change trend is calculated by a differential decomposition algorithm; Calculating the spatial correlation characteristics of the pressure state matrix, calculating the interference intensity between units, the pipeline stability coefficient and the pressure blind zone index based on the spatial correlation characteristics, constructing a feature fusion network, and calculating the flow delay characteristics through the feature fusion network; The time domain features and frequency domain features collected by each pressure sensor are extracted, and the pressure feature vector of the sensor is calculated based on the phase difference between the time domain features and the frequency domain features. Based on the pressure feature vector, feature matching is performed on pressure sensors at different positions to calculate the pressure change trend.

3. The pressure balance control method for the coordinated operation of a quenching boiler and a cracking furnace according to claim 2, characterized in that: The pressure intensity modeling of the pressure state matrix includes: According to the pressure intensity distribution, pressure intensity sampling points are extracted from each frame of data, and the sampling points are correlated and mapped with flow delay characteristics to generate a pressure intensity map. The pressure intensity maps collected by multiple sensors are spatially aligned to calculate the pressure intensity distribution of the unit; Set the interference threshold value, locate the interference source according to the pressure intensity value of the multi-frame pressure state matrix, and calculate the interference intensity difference. If the interference intensity difference is greater than or equal to the interference threshold value, it means that there is a pressure blind spot in the unit. Perform flow model constraint compensation on the current unit, iteratively correct the pressure intensity distribution of the current unit according to the flow resistance loss model corresponding to the current unit, and calculate the intensity compensation value of the blind spot pressure based on the correction result; The pressure intensity model is performed on the pressure state matrix according to the pressure intensity distribution, and the pressure model of the unit is marked with a time delay through the flow delay feature.

4. The pressure balance control method for the coordinated operation of a quenching boiler and a cracking furnace according to claim 3, characterized in that: Calculating the pressure intensity gradient according to the position of the pressure sensor includes: Based on multiple sets of pressure coverage data, the pressure intensity change points are extracted and mapped to a unified coordinate framework according to the deployment location of the sensor. The change points are fitted using a spatial interpolation algorithm to generate a unit pressure field model. Performing sampling at equal intervals along the flow path of the pressure field model, calculating the pressure decay rate, interference fluctuation index, and pressure change slope of the path based on the sampling results, and calculating the pressure change parameter based on the pressure decay rate, interference fluctuation index, and pressure change slope; Based on the deployment parameters and acquisition accuracy of the pressure sensors, the distribution characteristics of the pressure changes in each frame of data are projected onto the pressure field model. The pressure field model is partitioned along the flow direction according to the number of sensors. The changing patterns of the pressure changes within the partitions are analyzed, and the pressure distribution characteristics are calculated based on these changing patterns. The pressure intensity gradient is calculated according to the pressure change parameter and the pressure distribution characteristics. The calculation process of the pressure intensity gradient includes: based on the coordinate position range from the first pressure sensor to the last pressure sensor, selecting spatial coordinate points in the sensor deployment direction, accumulating the product of the pressure field strength characteristic weight value and the pressure distribution characteristic weight value within the spatial resolution range, and superimposing the influence value of the sensor acquisition frequency on the pressure intensity change rate.

5. The pressure balance control method for the coordinated operation of a quenching boiler and a cracking furnace according to claim 4, characterized in that: The calculating of the pressure prediction information of each subunit includes: Using the main flow path of the pressure field model as a baseline and the peak position of the pressure change in each frame of data as a reference point, the pressure offset is calculated and a pressure distribution curve is drawn according to the coordinates; Correcting the growth rate and direction of the pressure change trend according to the pressure intensity gradient; Starting from the most recent pressure distribution point, the distribution curve is continuously drawn according to the correction results of the growth rate and direction to generate the pressure distribution points for the next period until the distribution points cover the entire target unit and the pressure prediction information is generated.

6. The pressure balance control method for the coordinated operation of a quenching boiler and a cracking furnace according to claim 1, characterized in that: The constructing of the collaborative control model includes: The input layer is used to organize the pressure prediction information into spatial distribution data and perform normalization processing; The feature fusion layer is used to extract the unit-related features of pressure by processing spatial distribution data and construct the dependency relationship between physical units; The resource allocation layer is used to integrate the correlation between pressure balance in spatial units and generate resource regulation strategies.

7. The pressure balance control method for the coordinated operation of a quenching boiler and a cracking furnace according to claim 1, characterized in that: The obtaining of the distribution characteristics of the pressure balance includes: According to the spatiotemporal distribution characteristics of the pressure balance output by the collaborative control model, the identification of the subunit is matched with the spatiotemporal distribution characteristics; The unit data in the pressure state dataset are reorganized according to the spatiotemporal characteristics to generate a unit distribution map sorted by the pressure balance intensity; According to the reorganized unit distribution map, the optimized pressure balance distribution characteristics are output.

8. The pressure balance control method for the coordinated operation of a quenching boiler and a cracking furnace according to claim 1, characterized in that: The dynamic regulation of the resources of the quench boiler and the cracking furnace includes: Mapping unit identifications to units of pressure balance distribution characteristics one by one; According to the temporal and spatial distribution characteristics of pressure balance, the execution actions of control resources are controlled, including valve opening adjustment, flow regulation and coolant distribution operations; Resources are dynamically allocated to corresponding physical units based on the spatial distribution of pressure balance and preset control strategies.

9. A pressure balance control method for coordinated operation of a quenching boiler and a cracking furnace according to claim 8, characterized in that: The execution action of the control resource includes: When the pressure balance in the target unit reaches a preset intensity threshold, an opening increase instruction of the adjacent valve is triggered; Dynamically combine available fluid resources according to the flow regulation strategy to generate a coolant allocation vector; A pipeline parameter of a target unit transmission node is adjusted based on the coolant distribution vector.

Images