Adaptive control method and system for device operating parameters

Abstract

The application provides a kind of adaptive control method and system of equipment operating parameter, it is related to the technical field of adaptive control oilfield operating parameter, the application is by obtaining structural displacement, process state parameter, component analysis data and fluid performance parameter and so on Multi-source data, to evaluate the current process stable state, and output a state quantitative value;Then, combined with component analysis data, the state quantitative value and the current value of fluid performance parameter, analysis and decision are carried out using the pre-trained decision model, to generate the expected target value that fluid performance parameter should reach;Finally, with the expected value as the leading target, and real-time sensing process state parameter feedback, the operating parameters of key process actuators such as drilling rig and mud pump are dynamically adjusted through adaptive control algorithm, which can realize the change from passive response to active optimization in the drilling process, effectively improve the wellbore stability and drilling efficiency.

Description

Technical Field

[0001] This application relates to the technical field of equipment operation control, and in particular to an adaptive control method and system for equipment operating parameters. Background Technology

[0002] In complex industrial process control, achieving adaptive control of equipment operating parameters is a core challenge for improving production efficiency and ensuring process safety and stability. Therefore, by adjusting the operating parameters of process actuators in real time, it is possible to effectively cope with complex and ever-changing working conditions, and the development prospects of its control strategy in terms of precision and intelligence are broad.

[0003] Currently, among existing equipment operating parameter control technologies, some technologies adjust operating parameters through a single parameter and empirical threshold; others focus on monitoring and storing multiple process data streams, providing only data display for operators; still others rely on empirical formulas and adjust equipment operating parameters based on simplified empirical formulas or single equipment status feedback.

[0004] However, these methods rely on only one or a few data sources, resulting in insufficient foresight and a single decision-making dimension when facing complex and dynamic operating conditions. They also struggle to achieve a dynamic balance between multiple objectives such as maintaining process stability, ensuring equipment safety, and pursuing operational efficiency, ultimately leading to inefficient control of equipment operating parameters. Summary of the Invention

[0005] The purpose of this application is to provide an adaptive control method and system for equipment operating parameters to solve the problems of low control accuracy caused by insufficient control foresight and single decision-making dimensions in the prior art.

[0006] To address the aforementioned technical problems, in a first aspect, this application provides an adaptive control method for equipment operating parameters, comprising:

[0007] Acquire multi-source data associated with industrial processes, including current values ​​of structural displacements, process state parameters, component analysis data, and fluid performance parameters;

[0008] The process stability state is evaluated based on the structural displacement and process state parameters to obtain a state quantification value;

[0009] Based on the component analysis data, the state quantification value, and the current values ​​of the fluid performance parameters, the expected values ​​of the fluid performance parameters are determined through a pre-established decision model.

[0010] Based on the expected values ​​of the fluid performance parameters and the process state parameters, the operating parameters of the process actuator are dynamically and adaptively controlled through an adaptive control algorithm.

[0011] Optionally, determining the expected values ​​of the fluid performance parameters based on the component analysis data, the state quantification value, and the current values ​​of the fluid performance parameters using a pre-established decision model includes:

[0012] Based on the component analysis data, feature vectors representing environmental characteristics are determined through the feature extraction layer of the decision model.

[0013] Based on the feature vector, the state quantization value, and the current values ​​of the fluid performance parameters, the following is executed through the instruction generation layer of the decision model:

[0014] The feature vector is matched with multiple preset adjustment modes, wherein each adjustment mode is associated with a set of performance parameters;

[0015] Based on the state quantization value and the current value of the fluid performance parameters, the parameters in the matched performance parameter group are optimized collaboratively to generate the expected value of the fluid performance parameters.

[0016] Optionally, the step of collaboratively optimizing each parameter in the matched performance parameter group based on the state quantization value and the current value of the fluid performance parameters to generate expected values ​​of the fluid performance parameters includes:

[0017] Based on the quantization level corresponding to the state quantization value, the current value of the fluid performance parameter, and the baseline value of each parameter in the performance parameter group, calculate the correction requirement weight of each parameter.

[0018] Based on the corresponding corrected requirement weights, the baseline values ​​of each parameter in the performance parameter group are weighted and fused respectively.

[0019] Boundary constraints are applied to the weighted and fused parameter values ​​to form the expected values ​​of the fluid performance parameters.

[0020] Optionally, the step of dynamically and adaptively controlling the operating parameters of the process actuator based on the expected values ​​of the fluid performance parameters and the process state parameters using an adaptive control algorithm includes:

[0021] Establish a model predictive control system with the control objective of keeping process state parameters within a preset safety window range and the control object of the process actuator;

[0022] The expected values ​​of the fluid performance parameters are used as feedforward variables, and the process state parameters are used as feedback variables, both of which are input into the model prediction and control system.

[0023] The dynamic response model in the model prediction control system is used to predict the trajectory of process state parameters within multiple preset control cycles based on the feedforward and feedback variables.

[0024] The adjustment amount of the operating parameters of the process actuator is determined based on the change trajectory, and the corresponding operating parameters are controlled according to the adjustment amount.

[0025] Optionally, determining the adjustment amount of the operating parameters of the process actuator based on the change trajectory includes:

[0026] Construct a multi-objective optimization function that includes the tracking error of process state parameters and the adjustment range of operating parameters;

[0027] Based on the boundary between the change trajectory and the preset safety window range, the first constraint condition of the process state parameters is determined;

[0028] Based on the operating range of the operating parameters, establish a second constraint condition;

[0029] By solving the optimal solution of the multi-objective optimization function under the conditions of the first constraint and the second constraint, the optimal combination of adjustment amounts containing multiple operating parameters is obtained.

[0030] Optionally, the step of evaluating the process steady state based on the structural displacement and process state parameters to obtain a state quantification value includes:

[0031] Calculate the instantaneous displacement of the structure relative to the reference value, as well as the rate of change of the displacement, and simultaneously calculate the safety margin between the process state parameters and the preset safety reference value.

[0032] Based on the offset, rate of change, and safety margin, the state quantification value is evaluated using a preset stability evaluation model.

[0033] Optionally, after determining the expected values ​​of the fluid performance parameters through a pre-established decision model, the method further includes:

[0034] Based on the expected values ​​and current values ​​of the fluid performance parameters, a dosing scheme for process additives is determined.

[0035] Based on the aforementioned dosing scheme, the additive dosing system is driven to perform the corresponding operations.

[0036] Secondly, this application provides an adaptive control system for equipment operating parameters, comprising:

[0037] The acquisition module is used to acquire multi-source data associated with industrial processes, including: structural displacement, process state parameters, component analysis data, and current values ​​of fluid performance parameters;

[0038] The evaluation module is used to evaluate the process stability based on the structural displacement and process state parameters, and obtain a state quantification value.

[0039] The determination module is used to determine the expected value of the fluid performance parameters based on the component analysis data, the state quantification value, and the current value of the fluid performance parameters through a pre-established decision model.

[0040] The control module is used to dynamically and adaptively control the operating parameters of the process actuator based on the expected values ​​of the fluid performance parameters and the process state parameters through an adaptive control algorithm.

[0041] Thirdly, this application provides an electronic device, comprising:

[0042] Memory, used to store computer programs;

[0043] A processor, used to implement the adaptive control method for device operating parameters as described in the first aspect above when executing the computer program.

[0044] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the adaptive control method for device operating parameters as described in the first aspect above.

[0045] The adaptive control method for equipment operating parameters provided in this application acquires multi-source data associated with an industrial process, including: structural displacement, process state parameters, component analysis data, and current values ​​of fluid performance parameters; assesses the process stability state based on the structural displacement and process state parameters to obtain a state quantification value; determines the expected value of the fluid performance parameters through a pre-established decision model based on the component analysis data, the state quantification value, and the current value of the fluid performance parameters; and dynamically and adaptively controls the operating parameters of the process actuator through an adaptive control algorithm based on the expected value of the fluid performance parameters and the process state parameters.

[0046] The technical solution of this application has the following beneficial effects:

[0047] This application firstly achieves a comprehensive and three-dimensional perception of process status and environmental characteristics by acquiring multi-source data associated with industrial processes, laying a data foundation for accurate decision-making. Secondly, by transforming the complex stable state of the process into a quantifiable and calculable clear indicator, it enables accurate assessment and early warning of potential risks. Subsequently, by integrating process environmental characteristics, current status, and existing performance, it intelligently generates forward-looking fluid performance parameter optimization targets, transforming the control strategy from passive response to active guidance. Finally, by combining performance targets with real-time operating conditions, it forms a feedforward and feedback closed-loop control, thereby dynamically and accurately adjusting the operating parameters of the process actuators to ensure that the production process always operates towards a stable and efficient goal.

[0048] Furthermore, this application uses the feature extraction layer of the decision model to extract feature vectors representing environmental characteristics from the component analysis data; subsequently, the instruction generation layer matches this feature vector with a preset regulation mode library and finds a set of benchmark performance parameters; finally, it combines the current state quantification value and fluid performance parameters to perform collaborative optimization and correction on key parameters such as various parameters in the benchmark performance parameter set, thereby generating a scientific and matching expected target value of fluid performance parameters.

[0049] This application integrates environmental feature recognition, pattern matching, and multi-parameter collaborative optimization, making the setting of fluid performance parameters no longer an isolated numerical adjustment, but a systematic decision-making process that is deeply adapted to the process environment and operating status, thereby improving the scientific nature of parameter setting and its adaptability to complex working conditions. Attached Figure Description

[0050] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 A flowchart illustrating an adaptive control method for device operating parameters provided in an embodiment of this application;

[0052] Figure 2 A schematic diagram illustrating a specific implementation of an adaptive control method for equipment operating parameters provided in this application embodiment;

[0053] Figure 3 A schematic diagram of the structure of an adaptive control system for equipment operating parameters provided in this application embodiment;

[0054] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0055] To achieve precise control of the drilling process, existing methods rely on only one or a few data sources. This results in insufficient foresight and a single decision-making dimension when the system faces complex and dynamic working conditions. Furthermore, it is difficult to achieve a dynamic balance between multiple objectives such as maintaining process stability, ensuring equipment safety, and pursuing operational efficiency, ultimately leading to low efficiency in controlling equipment operating parameters.

[0056] To address the aforementioned limitations, this application proposes an adaptive control method for equipment operating parameters. The core of this method lies in first constructing an intelligent decision-making system capable of comprehensively perceiving the process environment and state by introducing multi-source data such as structural displacement and component analysis data; secondly, comprehensively evaluating the process stability state and using a decision model to proactively generate optimization targets for fluid performance parameters based on real-time component analysis data and current operating conditions; subsequently, combining this target with real-time process state parameters to drive the process actuators to adaptively adjust. This approach, by integrating process environment characteristics and real-time operating condition data, elevates the control mode from a passive "post-event response" to an active "pre-event guidance," effectively solving the core problems of insufficient control foresight and a single decision-making dimension.

[0057] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0058] The core of this application is to provide an adaptive control method for equipment operating parameters, and a flowchart of one specific implementation is shown below. Figure 1 As shown, the method includes:

[0059] S101. Acquire multi-source data associated with the industrial process, including: structural displacement, process state parameters, component analysis data, and current values ​​of fluid performance parameters.

[0060] In the above scheme, an industrial process can refer to a series of continuous or intermittent operations carried out in industrial production to achieve a specific physical or chemical transformation goal. In this process, there are monitorable state variables, controllable process actuators, and process conditions that need to be maintained. The industrial process can be a process in various scenarios, such as chemical production processes, metallurgical processing processes, material synthesis reactions, energy production processes, oil and gas production, etc.

[0061] To facilitate understanding of the method provided in this application, the following embodiments of this application will be described using oil and gas production as an example. Other processes will be described similarly and will not be repeated for the sake of brevity.

[0062] In the context of oil and gas production, structural displacement can refer to wellbore displacement, which is the deformation or movement of the rock surrounding the wellbore during drilling. It is used to directly determine whether the wellbore is stable and whether there is a risk of collapse. In the context of oil and gas production, process state parameters can refer to annular equivalent density, which measures the comprehensive pressure generated by the drilling fluid circulating in the wellbore. Component analysis data in the context of oil and gas production can refer to cuttings composition, which refers to the mineral composition and content of rock fragments that are broken by the drill bit and returned to the surface with the drilling fluid. Fluid performance parameters in the context of oil and gas production can refer to drilling fluid performance. The current value of drilling fluid performance refers to the current key parameters of the drilling fluid, which mainly include density, viscosity, and water loss.

[0063] In this embodiment, key data from the drilling site are collected by rationally deploying various specialized monitoring and detection devices. Displacement sensors are used to capture wellbore displacement data, density detectors are responsible for acquiring annular equivalent density data, cuttings component analyzers are used to analyze the composition of cuttings, and specialized detection devices collect the current performance indicators of the drilling fluid in real time. These devices continuously and in real time capture relevant data, ensuring that the information obtained truly reflects the actual working conditions of the drilling. The collected multi-source data will serve as the core foundation for subsequent evaluation of wellbore stability, determination of expected drilling fluid performance values, and dynamic control of oil and gas equipment operating parameters, providing data support for the entire adaptive control process.

[0064] In this embodiment of the invention, by collecting key and comprehensive real-time data from the industrial process, a reliable data foundation is provided for subsequent process stability assessment, determination of expected values ​​of fluid performance parameters, and adaptive control of equipment, ensuring the accuracy and effectiveness of the entire control process.

[0065] S102. Evaluate the process stability based on the structural displacement and process state parameters to obtain a state quantification value.

[0066] Among them, the state quantification value refers to the specific numerical value output by the model, which is used to intuitively reflect the level of process stability. The magnitude of the value determines the difference in the degree of stability. In the oil and gas production process, this process stability state can refer to the wellbore stability state.

[0067] In one specific implementation, step S102 includes:

[0068] Step 1021: Calculate the offset of the instantaneous value of the structural displacement relative to the reference value, and the rate of change of the offset, and at the same time calculate the safety margin between the process state parameters and the preset safety reference value.

[0069] In the oil and gas production process, the benchmark value refers to the wellbore displacement benchmark value preset before drilling based on formation conditions and design requirements, which serves as a reference for judging whether the actual displacement is abnormal. The safety reference value in the oil and gas production process refers to the preset equivalent density based on the underground formation pore pressure, which is used to measure the pressure inside the formation and is a key reference for judging whether the drilling fluid pressure can balance the formation pressure.

[0070] In step 1021, the pre-set wellbore displacement benchmark value is first determined, and then the collected instantaneous wellbore displacement value is used to calculate the offset formula. Calculate the offset, where Δs is the offset. This represents the instantaneous value of the wellbore displacement. The wellbore displacement is used as the baseline value; then, based on the offset at different times, and using the rate of change formula... Calculate the rate of change, where v is the rate of change. This is the offset at the current moment. This represents the offset from the previous moment. The time interval between adjacent moments is determined by combining the collected annular equivalent density and the preset equivalent density, using the safety margin formula. Calculate the safety margin, where, For safety margin, For the toroidal equivalent density, This is the preset equivalent density.

[0071] For example, in oil and gas production scenarios, fiber optic displacement sensors are used to collect wellbore displacement data to obtain instantaneous wellbore displacement values. The value is 6.5mm, and the preset wellbore displacement reference value is retrieved. The value is 4.5mm. Then, the offset is calculated using the offset formula. =6.5mm - 4.5mm = 2.0mm; then retrieve the offset from the previous moment. The offset is 1.4 mm, and the time interval Δt between adjacent moments is 10 min. The rate of change of the offset is calculated using the formula v = 0.06 mm / min. Simultaneously, the annular equivalent density in the process state parameters is measured using a differential pressure density analyzer. The content is 1.22 g / cm³, and the preset equivalent density is adjusted. The value is 1.09 g / cm³. The safety margin is calculated using the safety margin formula, i.e., the safety margin. =0.13g / cm³.

[0072] Step 1022: Based on the offset, rate of change and safety margin, evaluate the state quantification value through a preset stability evaluation model.

[0073] Among them, the stability assessment model refers to a pre-trained mathematical model used in oil and gas production scenarios to quantitatively assess the stability of the wellbore.

[0074] In step 1022, after calculating the above three parameters, a wellbore stability assessment model trained using historical drilling data is used. The training data for this model comes from 300 sets of complete historical data from similar formation drilling operations in the oil and gas field over the past five years. This historical data includes raw monitoring data such as wellbore displacement, annular equivalent density, and formation pore pressure; model input data such as offset, flow rate, and safety parameters; and field-recorded labels of the actual wellbore stability status. These labels are categorized into four types: no instability, slight deformation, critical risk, and severe instability. The model structure adopts a gradient boosting decision tree ensemble model, consisting of 100 decision trees, each with a depth of... The training process is limited to 5 layers to avoid overfitting. The core evaluation metrics are mean absolute error (MAE), root mean square error (RMSE), and classification accuracy. The model takes the offset, rate of change, and safety margin in historical engineering projects as input features and the corresponding labels of the actual stable state of the wellbore as output targets. The splitting nodes and weight parameters of each decision tree are iteratively optimized through the gradient boosting algorithm. Training stops when the validation set MAE ≤ 3.5, RMSE ≤ 4.2, and classification accuracy ≥ 90%. The three key parameters obtained earlier are then input into the trained model. The model integrates the influence weights of each parameter on wellbore stability and outputs a quantitative value of the state to achieve a quantitative evaluation of the wellbore stability.

[0075] For example, after training a stability assessment model using 300 sets of historical data from similar operations in oil and gas fields, the offset of 2.0 mm, the rate of change of 0.06 mm / min, and the safety margin of 0.13 g / cm³ are input into the preset stability assessment model, and the final output state quantization value is 69.

[0076] Based on this, this embodiment can also combine a preset level division rule, namely, below 40 is level one, 40-60 is level two, 60-80 is level three, and above 80 is level four, with level coefficients of 0.3, 0.5, 0.7, and 0.9 respectively, to determine the interval 60-80 in which the state quantification value 69 is located, and then determine its corresponding level three and associated level coefficient 0.7.

[0077] This application integrates key parameters related to structural displacement and process state parameters to accurately present the stable state of the process in a quantitative manner. This can overcome the limitations of traditional qualitative judgment and provide precise decision guidance for subsequent fluid performance parameter adjustment and equipment operation parameter control. In this way, it can predict potential wellbore risks in advance and improve the safety and efficiency of process operations.

[0078] S103. Based on the component analysis data, the state quantification value, and the current value of the fluid performance parameters, determine the expected value of the fluid performance parameters through a pre-established decision model.

[0079] The decision model includes a feature extraction layer and an instruction generation layer. The feature extraction layer uses a combination of principal component analysis (PCA) and a lightweight convolutional neural network (CNN), while the instruction generation layer uses a collaborative architecture of a rule engine and a multi-objective optimization algorithm. Furthermore, the decision model can be trained using a hierarchical training and joint optimization mode before application.

[0080] Specifically: First, collect data from 500 similar wells in oil and gas fields, which include cuttings composition, state quantification value, current value of drilling fluid performance and corresponding expected value label of drilling fluid performance. Then, divide the data into training, validation and test sets according to a 7:2:1 ratio, and then normalize and fill in missing values ​​before use.

[0081] Subsequently, in the feature extraction layer, PCA is used to calculate the covariance matrix using the rock fragment composition data in the training set, and the four principal components with a cumulative variance contribution rate of 92% are retained to complete dimensionality reduction. For example, the input of the feature extraction layer is the original 6-dimensional features of the rock fragment composition, and after dimensionality reduction by PCA, a 4-dimensional intermediate feature vector is output. The intermediate feature vector is then input into a lightweight CNN. The lightweight CNN uses a 3-layer convolutional layer and a 2-layer fully connected structure, with the intermediate feature vector as input, and performs feature extraction training in an unsupervised or self-supervised manner, finally outputting a 4-dimensional stratigraphic feature vector.

[0082] Next, using the formation feature vector, state quantization value, and current drilling fluid performance value from the training set as input, and the expected drilling fluid performance value as the label, a training instruction generation layer is developed through supervised learning. The training objectives are twofold: first, based on the established adjustment pattern library, training the rule engine to obtain corresponding adjustment patterns based on formation feature vectors through clustering or similarity calculations. Each adjustment pattern includes a formation feature vector template and a corresponding set of performance parameters, with each set of parameters containing baseline values ​​for each parameter. The second objective is to train the parameters of the NSGA-II algorithm, enabling it to collaboratively optimize the baseline values ​​of each parameter in the matched performance parameter set based on the state quantization value and the current drilling fluid performance value, generating the expected drilling fluid performance value. The optimization objective is to minimize the error between the predicted value and the label value.

[0083] Then, the feature extraction layer and the instruction generation layer are concatenated for end-to-end joint training. The parameters of the two layers are optimized simultaneously using training data to minimize the loss function between the predicted drilling fluid performance and the true label. During training, the model performance is monitored through the validation set, and training is stopped when the mean absolute error of the prediction is less than 5%.

[0084] After training was stopped, the model performance was evaluated using a test set. The final model achieved 93% accuracy in the fluid performance parameter prediction task, meeting the requirements of field applications.

[0085] In one specific implementation, such as Figure 2 As shown, step S103 includes:

[0086] Step 1031: Based on the component analysis data, determine the feature vector representing the environmental characteristics through the feature extraction layer of the decision model.

[0087] In the oil and gas production process, the eigenvector representing environmental characteristics can refer to the formation eigenvector.

[0088] In step 1031, the feature extraction layer is implemented using a combination of PCA and lightweight CNN. First, PCA is used to reduce the dimensionality of the original multidimensional information of the component analysis data and retain the core related variables to reduce computational complexity. Then, lightweight CNN is used to extract the deep semantic information representing environmental features in the dimensionality-reduced data and finally output the feature vector.

[0089] For example, the rock fragment composition obtained by the rock fragment composition analyzer is 75% sandstone, 12% mudstone, and 13% other debris. These multi-dimensional raw data are 6-dimensional raw features, including: sandstone, mudstone, other minerals, coarse grains, medium grains, and fine grains. The 6-dimensional raw features are first reduced in dimensionality by principal component analysis, and after removing redundant information, four core related variables are retained: main mineral composition index, particle size distribution coefficient, lithological stability index, and formation response coefficient. Then, a lightweight CNN is used to extract the deep features that characterize formation stability from the data, and finally output the environmental feature vector [0.75, 0.12, 0.08, 0.05].

[0090] It should be understood that after step 1031, the instruction generation layer first uses the rule engine to quickly match the environmental feature vector output by the feature extraction layer with multiple preset adjustment modes to determine the initial performance parameter set adapted to the current environment. Then, it calls a multi-objective optimization algorithm, combining the quantization level corresponding to the state quantization value, the current value of the fluid performance parameter, and the benchmark value in the initial performance parameter set, to calculate the correction requirement weights of each parameter and perform weighted fusion on the benchmark value. Subsequently, after boundary constraint processing, the expected value of the fluid performance parameter is generated. Therefore, in step 1032, based on the feature vector, the state quantization value, and the current value of the fluid performance parameter, the instruction generation layer of the decision model performs the following two steps:

[0091] Step 1033: Match the feature vector with multiple preset adjustment modes, wherein each adjustment mode is associated with a set of performance parameters.

[0092] In the context of oil and gas production, the adjustment mode can refer to a pre-set type of drilling fluid performance adjustment scheme for different formation characteristics. Each type corresponds to a set of suitable performance parameters. The performance parameter set refers to the set of fluid performance parameters benchmark parameters associated with the adjustment mode. In the context of oil and gas production, it can include benchmark values ​​for density, viscosity, and water loss.

[0093] For example, the instruction generation layer activates the rule engine and quickly compares the environmental feature vector with five preset adjustment modes. Because the sandstone proportion feature is the most prominent in the feature vector, it is precisely matched with adjustment mode 3 designed for medium sandstone strata. For instance, the baseline values ​​of each parameter in the performance parameter group associated with mode 3 are: the baseline value of density. The viscosity is 1.20 g / cm³, which is the baseline value. for The baseline value of water loss It is 8 mL.

[0094] Step 1034: Based on the state quantization value and the current value of the fluid performance parameters, perform collaborative optimization on each parameter in the matched performance parameter group to generate the expected value of the fluid performance parameters.

[0095] Step 1034 may specifically include the following steps:

[0096] Step a1: Calculate the correction requirement weight of each parameter based on the quantization level corresponding to the state quantization value, the current value of the fluid performance parameter, and the baseline value of each parameter in the performance parameter group.

[0097] Among them, the adjustment demand weight refers to the adjustment priority value of each fluid performance parameter determined based on the process stability and the current performance of the drilling fluid.

[0098] In step a1, the calculation formula for the demand weight is first modified based on the quantization level corresponding to the state quantization value, the current value of the fluid performance parameters, and the baseline value of each parameter in the performance parameter group. Calculate the required correction weights for each parameter, where, The correction weight for the i-th parameter is defined by α and β, which are preset weighting coefficients, with α taking the value of 0.6 and β taking the value of 0.4. Let i be the current value of the i-th parameter. Let r be the baseline value of the i-th parameter, and r be the level coefficient corresponding to the state quantization value.

[0099] For example, the weights of each item are calculated by modifying the demand weight calculation formula, where the modified demand weight for density is... The viscosity correction requires a weighting of 0.6 × 0.05 + 0.28 = 0.03 + 0.28 = 0.31. The required weight for correcting water loss is 0.6 × 3 + 0.28 = 1.8 + 0.28 = 2.08. The result is 0.6 × 2 + 0.28 = 1.2 + 0.28 = 1.48.

[0100] Step a2: According to the corresponding corrected demand weight, the baseline values ​​of each parameter in the performance parameter group are weighted and fused respectively;

[0101] Weighted fusion refers to the process of adjusting the benchmark values ​​of the performance parameter group by combining the corrected demand weights, which can make the results more in line with actual needs.

[0102] In step a2, the weighted fusion calculation formula is used to calculate the weights required for each parameter. The baseline values ​​of each parameter in the performance parameter group are weighted and fused, whereby... is the fused value of the i-th parameter, and k is a preset adjustment coefficient. Its specific value can be set according to the actual situation and is not limited here.

[0103] For example, by calculating the fused values ​​using the weighted fusion formula, the fused density value is 1.20 × (1 + 0.31 × 0.1) = 1.20 × 1.031 = 1.237, and the fused viscosity value is... The total water loss after fusion is 8×(1+1.48×0.1)=8×1.148=9.184mL.

[0104] Step a3: Apply boundary constraints to the weighted fusion values ​​of each parameter to form the expected values ​​of the fluid performance parameters.

[0105] Boundary constraint processing refers to the operation of limiting the range of the weighted fused parameter values, which can ensure that the parameter values ​​are within the safe and reasonable range allowed by drilling operations.

[0106] In step a3, boundary constraints are applied to the weighted fusion values ​​of each parameter to ensure that the values ​​are within the safe range allowed by drilling operations, thereby generating the expected values ​​of the fluid performance parameters.

[0107] For example, considering the performance boundary requirements of drilling fluids in oil and gas fields: density 1.20-1.25 g / cm³, viscosity... The water loss was 8-10 mL. Boundary constraints were applied to these three values ​​to obtain a target density of 1.237 g / cm³ and a target viscosity of [missing value]. The expected value of drilling fluid performance with a target fluid loss of 9.184 mL will be combined with process state parameters, and the operating parameters of the process actuator will be dynamically adjusted through subsequent adaptive control algorithms.

[0108] This application combines formation characteristics, process stability, and current drilling fluid performance, and uses a model to accurately derive the expected values ​​of drilling fluid performance suitable for actual working conditions. This can improve the pertinence and scientific nature of drilling fluid adjustments, provide a reliable basis for subsequent dynamic equipment control, and help ensure the stability and safety of drilling operations.

[0109] S104. Based on the expected values ​​of the fluid performance parameters and the process state parameters, the operating parameters of the process actuator are dynamically and adaptively controlled through an adaptive control algorithm.

[0110] The process execution mechanism is equipment, such as drilling rigs and mud pumps. The operating parameters of the drilling rig include top drive speed and drilling pressure, while the operating parameters of the mud pump include pump stroke rate. The top drive speed refers to the rotational speed of the top drive device of the drilling rig, which can affect drilling efficiency and cuttings breaking effect. The drilling pressure refers to the pressure applied to the drill bit by the drilling rig, which is a key parameter affecting drilling speed and drill bit life. The pump stroke rate refers to the number of strokes of the mud pump per unit time, which determines the circulation flow rate of the drilling fluid and can affect process state parameters and cuttings carrying effect.

[0111] The control law of the dynamic adaptive control algorithm has a core closed loop of real-time sensing, predictive deduction, optimized adjustment, and dynamic correction. Therefore, in a specific implementation, step S104 includes:

[0112] Step 1041: Establish a model predictive control system with the control objective of ensuring that the process state parameters are within a preset safety window range and the process actuator as the control object.

[0113] Among them, the model predictive control system is a specific type of adaptive control algorithm. Moreover, the model predictive control system is designed around the dynamic control needs of industrial processes. Its state variables are selected from process state parameters, fluid performance parameters, and parameters such as offset and rate of change derived from structural displacement, and comprehensively and accurately represent the real-time operating status of the industrial process. Its control variables are clearly defined as the core operating parameters of the process actuator, that is, the operation objects that the system needs to dynamically adjust.

[0114] The model update mechanism adopts a collaborative approach of real-time deviation feedback and periodic iterative optimization. That is, the system continuously compares the actual monitored values ​​of process state parameters with the model prediction values ​​to calculate the deviation. If the deviation exceeds the preset threshold, the system corrects the dynamic response model parameters in the system online using the least squares method based on the latest collected multi-source data such as structural displacement, process state parameters, and fluid performance parameters.

[0115] Conversely, if the deviation is within the allowable range, the model parameters are iteratively updated using accumulated historical operating data according to a preset cycle, thereby ensuring that the model always keeps in line with the dynamic characteristics of the actual industrial process, and thus steadily improving the control accuracy.

[0116] In step 1041, the control objective and control object are first defined. The control objective is to keep the process state parameters within the preset safety window range, and the control object is the drilling rig and mud pump. On this basis, a model predictive control system is established. The dynamic response model in this system can adopt the autoregressive moving average exogenous model type commonly used in industrial process control, and is obtained by training 800 sets of historical operating data of similar wells in oil and gas fields.

[0117] Furthermore, during training, fluid performance parameters and equipment operating parameters were used as input variables, and annular equivalent density and wellbore displacement offset were used as output variables. Before training, the training set and validation set were divided into 8:2 and standardized. During training, the input and output variables were used as the basis, and the least squares method was used to optimize the model parameters. The goodness of fit R² ≥ 0.92 was used as the termination condition. After verification on the validation set that the annular equivalent density prediction MAE ≤ 0.02 g / cm³ and the wellbore displacement offset prediction MAE ≤ 0.1 mm, the model was put into use, thus completing the training.

[0118] For example, when establishing a model predictive control system, the control objective of the system is to keep the annular equivalent density stable within a preset safety window of 1.15 g / cm³ to 1.28 g / cm³. The controlled objects are the core process actuators of drilling operations, namely the drilling rig and mud pump. The dynamic response model in the system has been trained using historical operating data of similar wells in oil and gas fields. The model takes fluid performance parameters and actuator operating parameters as inputs and annular equivalent density as output. After optimization by the least squares method, it can accurately map the dynamic correlation between parameters.

[0119] Step 1042: Input the expected values ​​of the fluid performance parameters as feedforward variables and the process state parameters as feedback variables into the model prediction and control system.

[0120] Among them, feedforward variables refer to input parameters used to predict system changes in advance, which can provide control basis before disturbances affect the system, thereby reducing lag; feedback variables refer to parameters that reflect the current actual operating state of the system, which are collected in real time and fed back to the control system to correct control deviations.

[0121] For example, the expected values ​​of the determined drilling fluid properties are used as feedforward variables, specifically density 1.237 g / cm³, viscosity... The water loss was 9.184 mL, and the annular equivalent density, a current process parameter, was collected in real time using a density meter. =1.22g / cm³ was used as a feedback variable, and then these two types of variables were synchronously input into the established model predictive control system to provide data support for subsequent state prediction and parameter adjustment.

[0122] Step 1043: Using the dynamic response model in the model prediction control system, based on the feedforward and feedback variables, predict the trajectory of process state parameters within multiple preset control cycles in the future.

[0123] Among them, the dynamic response model is the core component of the model predictive control system. It can be used to simulate the system's response law when the input parameters change and predict future state changes. It is trained by historical operating data, using input and output variables as training data, and the model parameters are optimized by least squares method to complete the training. The control cycle refers to the time interval between the control system executing a complete control process, which is used to ensure the timeliness and continuity of control actions. The change trajectory refers to the predicted change trend curve of the target parameter in the future multiple control cycles output by the dynamic response model.

[0124] For example, the model predictive control system calls the built-in dynamic response model and, based on the input feedforward and feedback variables combined with the preset 5-minute control cycle, predicts the trajectory of the annular equivalent density change within the next three control cycles. The final prediction result is as follows: at the end of the first control cycle (t+5min), the annular equivalent density drops to 1.19 g / cm³, at the end of the second cycle (t+10min), it drops to 1.17 g / cm³, and at the end of the third cycle (t+15min), it drops to 1.15 g / cm³. This trajectory shows that although the annular equivalent density is within the safe window, it is showing a continuous downward trend, and intervention by adjusting the actuator parameters is needed to maintain stability.

[0125] Step 1044: Determine the adjustment amount of the operating parameters of the process actuator based on the change trajectory, and control the corresponding operating parameters according to the adjustment amount.

[0126] The operating parameters of the process actuator include: the top drive speed and drilling pressure of the drilling rig, and the pumping rate of the mud pump.

[0127] Step 1044 may specifically include the following steps:

[0128] Step b1: Construct a multi-objective optimization function that includes the tracking error of process state parameters and the adjustment range of operating parameters;

[0129] The multi-objective optimization function refers to a mathematical function constructed by integrating multiple control objectives, used to find the optimal control scheme that takes into account all objectives. This function can adopt a weighted summation formula. In this application embodiment, the specific expression of the formula is not specifically limited and can be set according to the actual situation.

[0130] In this embodiment, a multi-objective optimization function is first constructed. This function comprehensively considers the tracking error of process state parameters and the adjustment range of operating parameters. For example, the expression of this function can be: Where J is the objective value for optimization. and Δu is the preset target weighting coefficient, e is the process state parameter tracking error, and Δu is the adjustment range of the operating parameters.

[0131] For example, first construct the above multi-objective optimization function. The selection of objective weight coefficients is based on the priority of process safety. Since the annular equivalent density is directly related to well control safety, its priority is higher than that of equipment operation economy. It can be 0.6, The value is 0.4, and it can be adaptively adjusted according to real-time operating conditions, such as when the annular equivalent density approaches the safety window boundary. Automatically upgraded to 0.7, and at the same time It is 0.3.

[0132] Step b2: Based on the boundary between the change trajectory and the preset safety window range, determine the first constraint condition of the process state parameters;

[0133] The first constraint condition refers to the restriction condition set based on the safety window of the process state parameters, which is used to ensure that the predicted process state parameters do not exceed the safety range.

[0134] In step b2, the first constraint condition for the process state parameter is determined based on the boundary between the change trajectory and the preset safety window range, that is, the predicted value must be between the upper and lower limits of the safety window.

[0135] Step b3: Based on the operating range of the operating parameters, establish the second constraint condition;

[0136] The second constraint is a limitation set based on the equipment's operating capabilities, used to ensure that the equipment's operating parameters are adjusted within its safe operating range.

[0137] In step b3, a second constraint is established by combining the operating range of the drilling rig's top drive speed, drilling pressure, and mud pump's pumping rate. That is, the adjusted parameters must be within the equipment's allowable operating range.

[0138] Step b4: By solving the optimal solution of the multi-objective optimization function under the first and second constraints, the optimal combination of adjustment amounts containing multiple operating parameters is obtained.

[0139] In b4, by solving the optimal solution of the multi-objective optimization function under the conditions of simultaneously satisfying the first and second constraints, the optimal combination of top drive speed adjustment, drilling pressure adjustment and pump stroke rate adjustment is obtained. Based on the optimal combination, the corresponding equipment operating parameters are controlled, and finally dynamic adaptive control is achieved.

[0140] The first constraint is: the predicted value of the annular equivalent density needs to be... Under the condition that the drilling rig's top drive speed n is between 60-120 r / min, the drilling pressure F is between 0-150 kN, and the mud pump's pumping speed f is between 50-100 strokes / min, the above multi-objective optimization function can be solved using a sequential quadratic programming algorithm to obtain the optimal combination of adjustment amounts.

[0141] The current top drive speed is 100 r / min, and the adjustment is +7 r / min, so the adjusted speed is 107 r / min; the current drilling pressure is 120 kN, and the adjustment is +8 kN, so the adjusted pressure is 128 kN; the current pump stroke rate is 80 strokes / min, and the adjustment is +4 strokes / min, so the adjusted pump stroke rate is 84 strokes / min. These adjusted parameters are all within the allowable operating range of the equipment. Based on these adjustments, the operating parameters of the drilling rig and mud pump are controlled to ensure that the process parameters are stable within the safety window, and at the same time, to provide accurate equipment operating status basis for the formulation of subsequent process additive dosing plans.

[0142] This application combines the expected values ​​of fluid performance parameters with real-time process status parameters. It achieves dynamic adaptive adjustment of equipment operating parameters through model predictive control, while also taking into account control accuracy and equipment operating safety. This effectively maintains process status parameters within a safe range, ensuring continuous and stable progress of drilling operations.

[0143] S105. After determining the expected values ​​of fluid performance parameters through a pre-established decision model, the method further includes:

[0144] Step 1051: Based on the expected value and the current value of the fluid performance parameters, determine the dosing scheme of the process additives.

[0145] Process additives refer to chemical substances added to adjust fluid performance parameters. They can specifically improve key characteristics of drilling fluids such as density, viscosity, and water loss, and meet the needs of drilling operations. Process additives include clay stabilizers and / or viscosity reducers. Clay stabilizers are a type of process additive used to inhibit the expansion and dispersion of clay particles in drilling fluids, thereby maintaining stable fluid performance parameters and protecting the wellbore rock structure. Viscosity reducers are a type of process additive used to reduce the viscosity of drilling fluids, improve their fluidity, and ensure smooth circulation of drilling fluids.

[0146] Alternatively, the process additives may include weighting agents, thickeners, and dehydration reducers.

[0147] In step 1051, based on the previously determined expected values ​​of fluid performance parameters and the current values ​​collected in real time, the differences between each performance parameter are first calculated, and then the dosage is calculated using the dosage calculation formula. Determine the dosage of each additive, where m is the additive dosage and k is the additive dosage coefficient corresponding to a unit performance difference. Let i be the expected value of the performance of the i-th item. Let V be the current value of the i-th performance parameter and V be the total volume of the drilling fluid. Then, based on the adjustment direction of each performance parameter, select the appropriate type of additive, namely clay stabilizer and viscosity reducer, and integrate the type of additive, dosage, and timing of addition to form a complete addition plan.

[0148] For example, in oil and gas production scenarios, continuing with previous fluid performance parameters, the current density... 1.15 g / cm³, viscosity for Water loss For 10 mL, the expected density The viscosity is 1.237 g / cm³. for Water loss The value is 9.184 mL. Based on the total drilling fluid volume V=100 m³ monitored by the drilling fluid circulation system, and the pre-defined unit performance difference and additive dosage coefficient table for similar formations in oil and gas fields, the dosage coefficients of additives corresponding to each parameter are determined, such as the dosage coefficient for density adjustment additives. for Viscosity adjuster additive dosage coefficient for Water loss adjustment additive dosage coefficient for ;

[0149] The required amount of additive for density adjustment can be calculated using the dosage calculation formula. For a volume of 450 kg, the required amount of additive for viscosity adjustment is... For a water loss of 200 kg, the required amount of additive to adjust is as follows: It weighs 240 kg;

[0150] Based on the drilling fluid circulation cycle, which is a full circulation every 30 minutes, the dosing plan is determined as follows: dosing in three evenly, with an interval of 15 minutes between each dosing. The first dosing consists of 150 kg of weighting agent, 65 kg of viscosity enhancer, and 80 kg of fluid loss reducer. The subsequent two dosings are done in the same amount. The timing of the dosing is determined to be the instantaneous point when the drilling fluid circulates to the mixing tank.

[0151] Step 1052: Based on the dosing scheme, drive the additive dosing system to perform the corresponding operation.

[0152] The dosing plan refers to the additive usage plan formulated according to the adjustment requirements of fluid performance parameters. The additive usage plan includes core contents such as additive type, dosage, and timing of dosing. The additive dosing system refers to the special equipment used to automatically execute the dosing plan, which can accurately control the dosage and dosing rate of additives to ensure that the dosing process is stable and controllable.

[0153] Next, in step 1052, the formulated addition plan is input into the additive addition system. The system will automatically start the relevant actuators based on the core parameters such as the addition amount and addition rate in the plan, and add the corresponding additives to the drilling fluid, so as to achieve targeted adjustment of fluid performance parameters.

[0154] For example, after the scheme is input into the additive dosing system, the system first confirms through sensors that the drilling fluid level in the mixing tank is within the normal range, then activates the quantitative conveying module of the corresponding additive storage bin, and starts the screw conveyor in the order of "weighting agent, thickener, and fluid loss reducer".

[0155] The initial injection rate was then set to 10 kg / min to ensure the additive was fully dissolved. During the injection process, drilling fluid performance data at the mixing tank outlet was collected in real time. After the first injection was completed, the system automatically recorded the injection time and counted down for 15 minutes. After reaching the interval node, the injection operation was repeated until all three injections were completed.

[0156] If the drilling fluid density is detected to reach 1.22 g / cm³ ahead of schedule during the dosing period, the control system will automatically reduce the dosing rate of the remaining weighting agent to 5 kg / min according to the preset logic to avoid exceeding the parameter limit. Finally, through the precise execution of the dosing system, combined with the top drive speed of the drilling rig adjusted by S104 to 107 r / min and the drilling pressure to 128 kN, the drilling fluid performance will steadily approach the expected value.

[0157] This application precisely formulates and automatically executes additive dosing plans, specifically adjusting fluid performance parameters to the expected levels, maintaining the comprehensive characteristics of drilling fluid to meet operating conditions, and providing reliable assurance for wellbore stability and efficient equipment operation.

[0158] Figure 3 A schematic diagram of a specific implementation of an adaptive control system for equipment operating parameters provided in this application is shown below. Figure 3 The system may include:

[0159] The acquisition module 31 is used to acquire multi-source data associated with the industrial process, including: structural displacement, process state parameters, component analysis data and current values ​​of fluid performance parameters.

[0160] Evaluation module 32 is used to evaluate the process stability state based on the structural displacement and process state parameters, and obtain a state quantification value.

[0161] The determination module 33 is used to determine the expected value of the fluid performance parameters based on the component analysis data, the state quantification value, and the current value of the fluid performance parameters through a pre-established decision model.

[0162] The control module 34 is used to dynamically and adaptively control the operating parameters of the process actuator based on the expected values ​​of the fluid performance parameters and the process state parameters through an adaptive control algorithm.

[0163] The adaptive control system for equipment operating parameters in this application embodiment is used to implement the aforementioned adaptive control method for equipment operating parameters. Therefore, the specific implementation of the adaptive control system for equipment operating parameters can be found in the embodiment section of the adaptive control method for equipment operating parameters above. The specific implementation can be referred to the description of the corresponding embodiments, which will not be repeated here.

[0164] like Figure 4 As shown, this application also provides an electronic device, including: a memory 41 for storing a computer program; and a processor 42 for executing the computer program to implement the steps of the adaptive control method for device operating parameters described above.

[0165] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the adaptive control method for device operating parameters described above.

[0166] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0167] Embodiments of the present invention also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in the embodiments of the adaptive control method for operating parameters of any of the above-described devices.

[0168] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0169] The above provides a detailed description of the adaptive control method and system for equipment operating parameters provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A method of adaptive control of a device operating parameter, characterized in that, The method comprises: acquiring multi-source data associated with an industrial process, the multi-source data comprising: current values of structural displacement, process state parameters, component analysis data, and fluid performance parameters, the process state parameters being annulus equivalent density, the component analysis data comprising cuttings components, the fluid performance parameters comprising: density, viscosity, and fluid loss; evaluating a process stable state based on the structural displacement and the process state parameters to obtain a state quantization value; determining expected values of the fluid performance parameters based on the component analysis data, the state quantization value, and the current values of the fluid performance parameters through a pre-established decision model; dynamically and adaptively controlling operating parameters of process execution mechanisms based on the expected values of the fluid performance parameters and the process state parameters through an adaptive control algorithm; the process execution mechanisms comprising a drilling rig and a mud pump, the operating parameters of the drilling rig comprising top drive rotating speed and drilling pressure, the operating parameters of the mud pump comprising pump stroke rate; the determining of the expected values of the fluid performance parameters based on the component analysis data, the state quantization value, and the current values of the fluid performance parameters through the pre-established decision model comprises: determining a feature vector representing environmental features through a feature extraction layer of the decision model based on the component analysis data; the feature extraction layer being a combination of principal component analysis (PCA) and a lightweight convolutional neural network (CNN); executing, through an instruction generation layer of the decision model based on the feature vector, the state quantization value, and the current values of the fluid performance parameters: matching the feature vector with a plurality of preset adjustment modes through a rule engine, wherein each adjustment mode is associated with a performance parameter group; performing collaborative optimization on each parameter in the matched performance parameter group based on the state quantization value and the current values of the fluid performance parameters through a multi-objective optimization algorithm to generate the expected values of the fluid performance parameters; the instruction generation layer being a collaborative architecture of the rule engine and the multi-objective optimization algorithm; the dynamically and adaptively controlling of the operating parameters of the process execution mechanisms based on the expected values of the fluid performance parameters and the process state parameters through the adaptive control algorithm comprises: establishing a model predictive control system with the process state parameters being within a preset safe window range as a control target and the process execution mechanisms as control objects; a model updating mechanism adopting a collaborative mode of real-time deviation feedback and periodic iterative optimization; inputting the expected values of the fluid performance parameters as feedforward variables and the process state parameters as feedback variables into the model predictive control system; predicting a change trajectory of the process state parameters in a plurality of preset control periods in the future based on the feedforward variables and the feedback variables through a dynamic response model in the model predictive control system; determining an adjustment amount of the operating parameters of the process execution mechanisms based on the change trajectory and controlling the corresponding operating parameters according to the adjustment amount.

2. The method of claim 1, wherein, the performing of the collaborative optimization on each parameter in the matched performance parameter group based on the state quantization value and the current values of the fluid performance parameters to generate the expected values of the fluid performance parameters comprises: According to the quantization level corresponding to the state quantization value, the current value of the fluid performance parameter and the reference value of each parameter in the performance parameter group, the correction demand weight of each parameter is calculated; According to the corresponding correction demand weight, the reference values of each parameter in the performance parameter group are weighted and fused respectively; The weighted and fused parameter values are subjected to boundary constraint processing to form the expected value of the fluid performance parameter.

3. The method of claim 1, wherein, The adjustment amount of the operation parameter of the process execution mechanism based on the change trajectory includes: A multi-objective optimization function including the tracking error of the process state parameter and the adjustment range of the operation parameter is constructed; Based on the boundary of the change trajectory and the preset safety window range, a first constraint condition of the process state parameter is determined; In combination with the operation range of the operation parameter, a second constraint condition is established; By solving the optimal solution of the multi-objective optimization function under the first constraint condition and the second constraint condition, an optimal combination including the adjustment amount of multiple operation parameters is obtained.

4. The method of claim 1, wherein, The state quantization value is obtained by evaluating the process stable state based on the structure displacement and the process state parameter, including: The offset of the instantaneous value of the structure displacement relative to the reference value and the change rate of the offset are calculated, and the safety margin between the process state parameter and the preset safety reference value is calculated; Based on the offset, the change rate and the safety margin, the state quantization value is evaluated through a preset stability evaluation model.

5. The method of claim 1, wherein, After determining the expected value of the fluid performance parameter through the pre-established decision model, the method further includes: Based on the expected value and the current value of the fluid performance parameter, a dosing scheme of the process additive is determined; Based on the dosing scheme, the additive dosing system is driven to perform corresponding operations.

6. An adaptive control system of device operating parameters, characterized by The adaptive control method of the equipment operation parameter includes: An acquisition module is configured to acquire multi-source data associated with an industrial process, the multi-source data including: structure displacement, process state parameter, component analysis data and current value of fluid performance parameter; An evaluation module is configured to evaluate the process stable state based on the structure displacement and the process state parameter to obtain a state quantization value; A determination module is configured to determine an expected value of the fluid performance parameter through a pre-established decision model based on the component analysis data, the state quantization value and the current value of the fluid performance parameter; A control module is configured to dynamically and adaptively control the operation parameter of the process execution mechanism through an adaptive control algorithm based on the expected value of the fluid performance parameter and the process state parameter.

7. An electronic device, comprising: It includes: A memory is configured to store a computer program; A processor is configured to execute the computer program to implement the steps of the adaptive control method of the equipment operation parameter.

8. A computer-readable storage medium, characterized in that, The computer program is stored in the computer readable storage medium, and the computer program can implement the adaptive control method of the equipment operation parameter when executed by the processor.

Images