Conduction oil boiler system optimization control method and system based on adaptive algorithm

Abstract

The invention provides a conduction oil boiler system optimization control method and system based on an adaptive algorithm. The method comprises the steps of data acquisition, data preprocessing, model training, parameter optimization and real-time control. The system comprises a sensor network, a data processing module, a model training module, an optimization calculation module and a control execution module. Boiler operation data are collected in real time through a sensor network, a combustion boiler natural gas dynamic prediction model and a circulating pump efficiency optimization model are constructed after preprocessing, parameters such as the number of boilers, the circulating pump frequency and the oil supply temperature are optimized through an adaptive differential evolution (ADE) algorithm, and finally closed-loop adjustment is achieved through a control execution module. According to the method, the data-driven modeling technology and the self-adaptive optimization capability are utilized, so that system parameters can quickly respond to process demand changes, the heat efficiency is remarkably improved, the energy consumption is reduced, the temperature fluctuation is reduced, the operation and maintenance cost is optimized, and the operation performance and the economical efficiency of the heat conduction oil boiler system are comprehensively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of optimizing the control of heat transfer oil boilers. Specifically, it relates to an optimized control method and system for a heat transfer oil boiler system based on an adaptive algorithm, which is applicable to the operation optimization of heat transfer oil boilers in the coating process. Background Art

[0002] As a key surface treatment technology, the coating process is widely used in many industries such as printing, packaging, textiles, and electronics. Its core lies in applying a specific coating on the surface of the substrate to improve the performance of the substrate, such as enhancing wear resistance, waterproofness, and corrosion resistance, endowing a decorative or functional appearance, etc., to meet the diverse, personalized, and high-performance requirements of different industries for materials. Taking the electronics industry as an example, insulation or conductive functions can be achieved through coatings. The flexibility and versatility of this technology make it an indispensable part of modern industrial production.

[0003] In the coating process, the heat transfer oil boiler plays a crucial role. It heats the heat transfer oil to a specific temperature and uses the high heat capacity and excellent heat transfer performance of the heat transfer oil to provide a stable high-temperature environment for the coating material and related equipment. It also plays a decisive role in the drying and curing stages after coating, ensuring that the coating material adheres evenly to the substrate and has a good curing effect, which is directly related to the quality and performance stability of the final product. Whether it is film coating, fabric coating, or surface treatment of electronic components, it is inseparable from it.

[0004] However, the current heat transfer oil boiler technology has many significant defects, which seriously limit the improvement and development of the overall efficiency of the coating process. On the one hand, the thermal efficiency is low. Due to heat dissipation losses and system design limitations during the heat transfer process, some energy is not fully utilized, resulting in energy waste. On the other hand, the temperature control accuracy is insufficient. In precision coating application scenarios that are extremely sensitive to temperature, such as the manufacturing of precision electronic components, a small temperature fluctuation may cause unstable coating quality or even product defects. At the same time, the high energy consumption further increases the production cost burden, especially in the current situation of continuously rising energy prices, posing a challenge to the economic benefits of enterprises. In addition, in terms of maintenance costs, on-site workers often need to manually adjust the heat transfer oil temperature and circulation flow rate according to the real-time production situation and personal experience. This manual operation mode not only has a high labor intensity but also lacks consistency in temperature adjustment due to the limitations of human judgment, affecting the stability of coating quality. Moreover, manual adjustment requires frequent monitoring and intervention, prolonging the equipment maintenance time and increasing labor costs.

[0005] Through the retrieval of patent documents, it is found that the invention patent with the publication number CN105318554A discloses a control system and method for a heat transfer oil boiler. The system includes: a coal consumption monitoring module for counting and recording the coal consumption of the heat transfer oil boiler within a set time period; a heat transfer oil operating parameter monitoring module for obtaining the temperature, pressure, and viscosity of the heat transfer oil at each set collection point in the heat transfer oil boiler and the heat transfer oil pipeline network. This patent focuses on monitoring coal consumption and heat transfer oil parameters and lacks depth in improving system performance and economy.

[0006] In summary, in view of the problems of the above-mentioned existing technologies, it has become a key task that needs to be solved urgently to research an optimized control method and system for a heat transfer oil boiler system based on an adaptive algorithm. Summary of the Invention

[0007] Aiming at the defects in the existing technologies, the purpose of the present invention is to provide an optimized control method and system for a heat transfer oil boiler system based on an adaptive algorithm.

[0008] An optimized control system for a heat transfer oil boiler system based on an adaptive algorithm according to the present invention includes:

[0009] A sensor network for collecting real-time operation data of the heat transfer oil boiler system, and the operation data includes the entire process of heat source, heat transfer, and circulation;

[0010] A data processing module connected to the sensor network for preprocessing and storing the operation data;

[0011] A model training module connected to the data processing module for constructing a dynamic prediction model for natural gas in the combustion boiler and an optimization model for the efficiency of the circulating oil pump;

[0012] An optimization calculation module connected to the model training module for running an adaptive differential evolution algorithm to perform parameter optimization and generate optimal control parameters;

[0013] A control execution module connected to the optimization calculation module for converting the optimal control parameters into control signals and driving the actuator;

[0014] A human-computer interaction module providing a human-computer interaction interface for monitoring the system status and supporting the adjustment of operation parameters and the optimization of control strategies;

[0015] The sensor network, data processing module, model training module, optimization calculation module, control execution module, and human-computer interaction module achieve data interaction through an industrial communication network.

[0016] Preferably, the sensor network includes:

[0017] A temperature sensor, deployed on the heat-conducting oil pipeline, is used to collect the outlet temperature, inlet temperature, total pipeline oil outlet temperature, and return oil temperature of the boiler;

[0018] A hot air temperature sensor, deployed at the outlet of the preheater, is used to collect the hot air temperature;

[0019] An electricity quantity monitoring unit, including an electricity meter deployed on the combustion boiler, is used to collect the heat load of the burner;

[0020] A pump monitoring unit, including an inverter, an electricity meter, and a flowmeter deployed on the circulating oil pump, is used to collect the pump frequency, power, and flow rate.

[0021] Preferably, the preprocessing performed by the data processing module includes:

[0022] Using the Kalman filter algorithm to smooth the temperature data and flow data;

[0023] Based on statistical methods, abnormal data is eliminated;

[0024] Synchronize multi-source data through time series alignment technology;

[0025] Store the preprocessed data in a time series database.

[0026] Preferably, the natural gas dynamic prediction model for the combustion boiler is based on the principle of energy conservation. By analyzing the correlation relationships among the heat-conducting oil temperature difference, hot air temperature, and flow rate, and combining heat calculation and regression analysis, the prediction relationship for natural gas consumption is derived;

[0027] The circulating oil pump efficiency optimization model is based on the principles of fluid mechanics and electric energy conversion. By analyzing the relationships between the pump frequency, opening state, and system resistance coefficient and the flow rate and power, the efficiency prediction relationship is established.

[0028] Preferably, the calculation formula for establishing the prediction relationship of natural gas consumption is as follows:

[0029]

[0030] Among them, Q is the heat, TOut is the oil outlet temperature, F is the flow rate, TIn is the return oil temperature, C p is the specific heat capacity of the heat-conducting oil, p is the constant pressure, ρ is the density, t and s are time and unit conversion factors respectively. Combining the combustion efficiency η and the calorific value of natural gas Hv, the model further derives the prediction relationship of natural gas consumption,

[0031]

[0032] Among them, Gas is the natural gas consumption, and the regression coefficients k1 and k2 are determined by fitting historical data;

[0033] The calculation formula for establishing the performance prediction relationship is as follows:

[0034] F = a·Hz 2 +b·Hz + c·E = d·Hz·F + e

[0035] Where E is power, and a, b, c, d, and e are fitting parameters.

[0036] Preferably, the natural gas dynamic prediction model of the combustion boiler is based on the random forest algorithm. The training process improves the prediction accuracy through the integration of multiple decision trees and introduces a regularization term to control the model complexity.

[0037] The training process of the circulating oil pump performance optimization model uses the support vector regression algorithm, uses the radial basis function kernel, and optimizes the hyperparameters through grid search to ensure the generalization ability of the model.

[0038] Preferably, the parameter optimization performed by the optimization calculation module includes:

[0039] Based on the trained model, the adaptive differential evolution algorithm uses the main pipe return oil temperature, circulating oil pump frequency, and boiler opening state as optimization variables. By constructing a multi-objective fitness function including thermal efficiency, energy consumption cost, and temperature control accuracy, and the constraint conditions include the flow safety range and process requirements. The adaptive differential evolution algorithm generates mutant individuals through the differential evolution mechanism and achieves global optimization by adaptively adjusting the mutation factor and crossover probability, and outputs the optimal control parameters.

[0040] Preferably, the control execution module includes:

[0041] An execution unit, including a programmable logic controller, for driving the frequency converter to adjust the circulating pump frequency, solenoid valve to control the oil supply temperature, and relay to switch the number of boilers;

[0042] A feedback unit, for monitoring the deviation between the actual operating state and the prediction result;

[0043] A safety control unit, for switching to a preset safe operating mode under abnormal conditions.

[0044] The present invention also provides an optimization control method for a heat transfer oil boiler system based on an adaptive algorithm. Based on the above optimization control system for a heat transfer oil boiler system based on an adaptive algorithm, it includes the following steps:

[0045] A data acquisition step, for real-time collecting the operating data of the heat transfer oil boiler system through a sensor network;

[0046] A data preprocessing step, for preprocessing and storing the operating data;

[0047] Model training step: constructing and training a dynamic prediction model for natural gas in a combustion boiler and an optimization model for the efficiency of a circulating oil pump using the preprocessed data;

[0048] Parameter optimization step: based on the trained dynamic prediction model for natural gas in a combustion boiler and the optimization model for the efficiency of a circulating oil pump, using the adaptive differential evolution algorithm to calculate the optimal control parameters;

[0049] Real-time control step: converting the optimal control parameters into execution instructions through a control execution module, and adjusting the number of boilers, the frequency of the circulating pump, and the oil supply temperature of the heat transfer oil boiler system;

[0050] Human-machine interaction step: providing a human-machine interaction interface for monitoring the system status and supporting the adjustment of operating parameters and the optimization of control strategies.

[0051] Preferably, the adaptive differential evolution algorithm includes the following sub-steps:

[0052] Step a: setting the population size, defining the optimization variables, initializing the mutation factor, crossover probability, and maximum number of iterations, and generating the initial population;

[0053] Step b: constructing a multi-objective fitness function including energy consumption cost, temperature deviation, and flow safety index;

[0054] Step c: through iterative optimization, outputting the optimal control parameter combination.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] 1. Aiming at the technical problems such as low thermal efficiency, insufficient temperature control accuracy, high energy consumption, and high maintenance cost commonly existing in heat transfer oil boilers in the coating process, the present invention significantly improves the operating performance of the system through real-time data acquisition, intelligent modeling, and adaptive optimization technology.

[0057] 2. The present invention uses data-driven modeling technology to construct a high-precision prediction model, combines the dynamic optimization ability of the adaptive algorithm, and realizes the precise adjustment of key parameters such as the number of boilers, the frequency of the circulating pump, and the oil supply temperature. The optimization process can be seamlessly integrated into the control system to form a closed-loop control mechanism, enabling the system to quickly respond to changes in process requirements, thereby achieving the improvement of system thermal efficiency, the reduction of energy consumption, and the optimization of operation and maintenance costs.

[0058] 3. The present invention realizes global parameter optimization through a differential evolution mechanism, effectively avoiding being trapped in local optimal solutions and significantly enhancing the comprehensiveness of optimization; by adaptively adjusting the mutation factor and crossover probability, the robustness and adaptability of the algorithm under dynamic working conditions are enhanced; a multi-objective optimization design is adopted to intelligently balance the energy consumption cost, temperature control accuracy, and flow safety requirements, ensuring the optimal comprehensive performance of the system; it has the characteristics of high computational efficiency and good real-time performance, and the time-consuming for a single optimization meets the requirements of the industrial control cycle; the system architecture has good scalability and can flexibly add optimization variables or adjust optimization objectives. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] By reading the detailed description of the non-restrictive embodiments with reference to the following drawings, other features, objectives, and advantages of the present invention will become more apparent:

[0060] Figure 1 Schematic diagram of an optimization control system for a heat transfer oil boiler system based on an adaptive algorithm in an embodiment of the present invention;

[0061] Figure 2 Flowchart of an optimization control method for a heat transfer oil boiler system based on an adaptive algorithm in an embodiment of the present invention;

[0062] Figure 3 Schematic diagram of constructing a dynamic prediction model for natural gas in a combustion boiler in an embodiment of the present invention;

[0063] Figure 4 Schematic diagram of constructing an optimization model for the efficiency of a circulating oil pump in an embodiment of the present invention;

[0064] Figure 5 Flowchart of the adaptive differential evolution (ADE) algorithm in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0065] The present invention will be described in detail below with reference to specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that those of ordinary skill in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0066] In view of the problems existing in the heat-conducting oil boiler in the coating process, such as low thermal efficiency, insufficient temperature control accuracy, high energy consumption, and high maintenance costs, the present invention proposes an optimized control method and system for a heat-conducting oil boiler system based on an adaptive algorithm. The method includes steps such as data acquisition, data preprocessing, model training, parameter optimization, and real-time control. The system includes a sensor network, a data processing module, a model training module, an optimization calculation module, and a control execution module. The operation data of the boiler is collected in real time through the sensor network, and after preprocessing, a dynamic prediction model of natural gas for the combustion boiler and an optimization model of the efficiency of the circulating oil pump are constructed. The adaptive differential evolution (ADE) algorithm is used to optimize parameters such as the number of boilers, the frequency of the circulating pump, and the oil supply temperature. Finally, the closed-loop regulation is realized through the control execution module. The present invention utilizes data-driven modeling technology and adaptive optimization capabilities, enabling the system parameters to quickly respond to changes in process requirements, thereby significantly improving the thermal efficiency, reducing energy consumption, reducing temperature fluctuations, and optimizing operation and maintenance costs, comprehensively improving the operation performance and economy of the heat-conducting oil boiler system.

[0067] Embodiment 1:

[0068] Figure 1 It is a schematic diagram of an optimized control system for a heat-conducting oil boiler system based on an adaptive algorithm in an embodiment of the present invention.

[0069] As Figure 1 shown, this embodiment provides an optimized control system for a heat-conducting oil boiler system based on an adaptive algorithm, including:

[0070] A sensor network for collecting real-time operation data of the heat-conducting oil boiler system. The operation data includes the entire process of heat source, heat transfer, and circulation, and also supports temporary recording under abnormal conditions.

[0071] Specifically, the sensor network includes:

[0072] Temperature sensors deployed on the heat-conducting oil pipeline for collecting the outlet temperature, inlet temperature, total pipe outlet oil temperature (TOut), and return oil temperature (TIn) of the boiler;

[0073] Hot air temperature sensors deployed at the outlet of the preheater for collecting the hot air temperature (Tha);

[0074] An electric quantity monitoring unit including an electric meter deployed on the combustion boiler for collecting the heat load (TE) of the burner;

[0075] A pump monitoring unit including an inverter, an electric meter, and a flow meter deployed on the circulating oil pump for collecting the pump frequency (Hz), power (E), and flow rate (F).

[0076] The sensor network of this embodiment is deployed based on the actual structure and operating characteristics of the thermal oil boiler system. By placing temperature sensors at key nodes such as the inlet and outlet of the thermal oil pipeline, the combustion boiler, the burner combustion chamber, and the circulating pump, the boiler outlet temperature, inlet temperature, main pipe oil outlet temperature (TOut), and return oil temperature (TIn) are collected in real time. At the same time, a hot air temperature sensor is deployed at the preheater outlet to collect the hot air temperature (Tha). An electricity meter is installed on the combustion boiler to monitor the burner heat load (TE). The circulating oil pump is equipped with a frequency converter, electricity meter, and flow meter to collect the oil pump frequency (Hz), power (E), and flow rate (F), respectively.

[0077] All sensor data is centrally collected and managed by a programmable logic controller (PLC). High-frequency sampling ensures that dynamic changes in the system's operating status are captured. The collected temperature parameters (TOut, TIn, Tha) reflect the state of heat transfer, load parameters (TE) characterize combustion intensity, and operating parameters (Hz, E, F) describe the state of the circulating pump, collectively forming a multidimensional data matrix. The system also features an abnormal operating condition recording function. In the event of a burner failure or sudden change in pump flow, the system automatically records the changing trends of various parameters before and after the failure, providing data support for subsequent analysis and optimization adjustments.

[0078] Data processing module, connected to the sensor network, for pre-processing and storing operation data;

[0079] Specifically, the preprocessing performed by the data processing module includes:

[0080] The Kalman filter algorithm is used to smooth the temperature data and flow data, and the noise interference is eliminated through state estimation to improve the data smoothness.

[0081] Abnormal data is eliminated based on statistical methods (such as standard deviation analysis) to ensure data reliability and avoid deviations in model training.

[0082] Multi-source data are synchronized through time series alignment technology. Specifically, multi-source data are aligned through timestamps to eliminate timing deviations caused by differences in sensor response times and ensure data consistency.

[0083] The preprocessed data is stored in a time series database, which supports historical data accumulation and real-time access. The database design takes into account high concurrent reading and writing requirements.

[0084] The preprocessing process not only improves data usability but also provides a standardized input format for subsequent steps. In addition, preprocessing parameters (such as filter window size) can be adjusted according to process requirements to adapt to different operating scenarios.

[0085] The model training module, connected to the data processing module, is used to construct a dynamic prediction model for natural gas in a combustion boiler and an optimization model for the efficiency of a circulating oil pump, and describe the dynamic characteristics of the combustion and circulation processes.

[0086] Figure 3 It is a schematic diagram of constructing a dynamic prediction model for natural gas in a combustion boiler in an embodiment of the present invention.

[0087] As Figure 3 shown, the dynamic prediction model for natural gas in a combustion boiler is based on the principle of energy conservation. By analyzing the correlation relationship between the temperature difference of the heat transfer oil (TOut - TIn), the temperature of the hot air (Tha), and the flow rate (F), combined with the calculation of heat (Q) and nonlinear regression analysis, the prediction relationship of the natural gas consumption is deduced.

[0088] Furthermore, the calculation formula for the prediction relationship of the natural gas consumption is as follows:

[0089]

[0090] Where Q is the heat, TOut is the outlet oil temperature, F is the flow rate, TIn is the return oil temperature, C p is the specific heat capacity of the heat transfer oil, p is the constant pressure, ρ is the density, t and s are time and unit conversion factors respectively. Combining the combustion efficiency η and the calorific value of natural gas Hv, the model further deduces the prediction relationship of the natural gas consumption,

[0091]

[0092] Where Gas is the natural gas consumption, and the regression coefficients k1 and k2 are determined by fitting historical data.

[0093] The training process uses data-driven modeling techniques to fit the model parameters through regression analysis. In this embodiment, the dynamic prediction model for natural gas in a combustion boiler is based on the random forest algorithm. The training process improves the prediction accuracy through the integration of multiple decision trees, and introduces a regularization term to control the model complexity and prevent overfitting. At the same time, the tree depth and number are adjusted through cross-validation to ensure the robustness of the model.

[0094] Figure 4 It is a schematic diagram of constructing an optimization model for the efficiency of a circulating oil pump in an embodiment of the present invention.

[0095] As Figure 4 shown, the optimization model for the efficiency of a circulating oil pump is based on the principles of fluid mechanics and electric energy conversion. By analyzing the relationship between the pump frequency (Hz), the opening state (State), and the system resistance coefficient (k) and the flow rate (F) and power (E), the support vector regression algorithm is used to establish the efficiency prediction relationship.

[0096] Furthermore, the calculation formula for establishing the efficiency prediction relationship is:

[0097] F = a·Hz 2 + b·Hz + c·E = d·Hz·F + e

[0098] Wherein, E is power, and a, b, c, d, and e are fitting parameters.

[0099] Considering the dynamic changes in system resistance, the model supports online update of parameters to adapt to pipeline aging or resistance fluctuations during long-term operation. The training process of the circulating oil pump efficiency optimization model uses the Support Vector Regression (SVR) algorithm, with the Radial Basis Function (RBF) kernel, and optimizes hyperparameters through grid search to ensure the generalization ability of the model and ensure the accuracy and generalization ability of the model.

[0100] The above model is trained with historical data and provides predictions of the operating characteristics of the system for the ADE algorithm, and specific variables are dynamically adjusted according to the actual working conditions.

[0101] The optimization calculation module is connected to the model training module and is used to run the Adaptive Differential Evolution (ADE) algorithm to perform parameter optimization and generate optimal control parameters.

[0102] Specifically, the parameter optimization performed by the optimization calculation module includes:

[0103] Based on the trained model, the ADE algorithm uses the main return oil temperature (TIn), circulating oil pump frequency (Hz), and boiler on / off state (State) as optimization variables. By constructing a multi-objective fitness function including thermal efficiency, energy consumption cost, and temperature control accuracy, and with constraints including flow safety range and process requirements, the ADE algorithm generates mutant individuals through the differential evolution mechanism and achieves global optimization by adaptively adjusting the mutation factor and crossover probability, and outputs the optimal control parameters. <s

[0104] In this embodiment, the optimization calculation module directly obtains the real-time operating parameters (including TIn temperature value, Hz frequency, etc.) collected by the sensor network through the industrial Ethernet and uses them as the input reference for dynamic optimization calculation. Based on the trained natural gas dynamic prediction model of the combustion boiler and the circulating oil pump efficiency optimization model, the Adaptive Differential Evolution (ADE) algorithm is used to perform global optimization in the multi-dimensional parameter space composed of the number of boilers, circulating pump frequency, and oil supply temperature. The algorithm generates the optimal solution that maximizes thermal efficiency, minimizes energy consumption, and meets the temperature control accuracy through dynamically adjusted mutation factors and crossover probabilities, combined with real-time working condition feedback data, and outputs the current optimal parameter combination after filtering by safety constraints.

[0105] Figure 5 This is the flow chart of the Adaptive Differential Evolution (ADE) algorithm in the embodiment of the present invention.

[0106] In this embodiment, the ADE algorithm plays a core role in the parameter optimization stage of optimal control. Its task is to calculate the control parameter combinations that meet multi-objective optimization based on real-time data and the output of the prediction model, including minimizing energy consumption, maximizing temperature control accuracy, and ensuring flow safety. Compared with traditional control methods (such as fixed parameter tables or manual experience adjustment), the ADE algorithm can efficiently solve non-linear and multi-constrained optimization problems through the differential evolution mechanism and adaptive parameter adjustment, ensuring that the system reaches the global optimal state under dynamic operating conditions, which directly determines the effectiveness of the control strategy.

[0107] As Figure 5 shown, the ADE algorithm includes the following sub-steps:

[0108] Step a, set the population size, define the optimization variables including the supply oil temperature (TIn), the circulation pump frequency (Hz), and the number of operating boilers (State), and initialize the mutation factor F, the crossover probability CR, and the maximum number of iterations. Randomly generate the initial population through a uniform distribution, and each individual represents a set of parameter combinations to ensure that the population diversity covers the parameter space.

[0109] Step b, construct a multi-objective fitness function:

[0110] Fitness = w1·Cost + w2·|T set - TIn| + w3·max(0, F min - F)

[0111] where Cost is the energy consumption cost (the sum of gas and electricity costs), T set is the target return oil temperature, F min is the minimum flow demand, and w1, w2, w3 are weight coefficients, which respectively reflect the priorities of energy consumption, temperature deviation, and flow safety. The fitness function uses the prediction model to calculate the performance indicators of each parameter combination, and the weights can be adjusted according to process requirements. For example, w1 is increased in energy consumption sensitive scenarios. The constraint conditions (such as temperature range, flow lower limit) are incorporated into the fitness calculation through penalty functions to ensure the feasibility of the solution.

[0112] Step c, iterative optimization:

[0113] Step c1, for the current optimal individual and a random individual, use the differential evolution strategy to generate a mutant individual:

[0114] v i = x best + F·(x r1 - x r2 )

[0115] where x best is the current optimal individual, x r1 and xr2 is a randomly selected individual, and F is a mutation factor used to control the mutation amplitude. The mutation operation introduces population diversity through the differential vector to prevent the algorithm from falling into local optimality.

[0116] Step c2, based on the mutated individual, generate a trial individual through the crossover probability:

[0117]

[0118] where, v i,j is the j-th dimensional parameter of the mutated individual v i , x i,j is the j-th dimensional parameter of the current target individual x i , rand is a random number uniformly generated in the interval [0, 1], which determines whether to adopt the parameters of the mutated individual, and CR is the crossover probability, which adjusts the fusion ratio of the old and new parameters. The crossover operation balances the exploration and exploitation capabilities and ensures the stability of the optimization process.

[0119] Step c3, compare the fitness values of the trial individual u i and the original individual x i , and select the one with better fitness to enter the next iteration.

[0120] Step c4, dynamically adjust the mutation factor F and the crossover probability CR according to the optimization process:

[0121] In the initial stage of iteration, if the convergence speed is slow, increase F to enhance the global search ability;

[0122] In the later stage of iteration, if the convergence tends to be stable, decrease F to improve the local search accuracy;

[0123] CR is adjusted according to the population diversity. If the diversity decreases, increase CR to maintain the exploration ability of the algorithm.

[0124] The adaptive mechanism improves the adaptability of the algorithm to different working conditions.

[0125] Step c5, the iteration process continues until the maximum iteration number is reached or the optimal fitness value tends to be stable, and output the optimal parameter combination.

[0126] The iteration termination condition can be adjusted according to the real-time requirement, such as setting a convergence threshold or a time limit.

[0127] To improve the calculation efficiency, parallel computing technology can be used to accelerate the population iteration process.

[0128] Through the in-depth application of the ADE algorithm in the optimization control link, this embodiment realizes the efficient optimization of the parameters of the heat transfer oil boiler system. Its global optimization ability, adaptive characteristics, and multi-objective balance design significantly improve the thermal efficiency, temperature control accuracy, and economy of the system, and have high technological innovation and practical value. The system design takes into account scalability and robustness and can be extended to other similar industrial scenarios.

[0129] The control execution module, connected to the optimization calculation module, is used to convert the optimal control parameters into control signals and drive the actuator.

[0130] Specifically, the control execution module includes:

[0131] The execution unit includes a programmable logic controller (PLC) and is used to drive the frequency converter to adjust the frequency of the circulation pump, control the oil supply temperature with a solenoid valve, and switch the number of boilers with a relay;

[0132] The feedback unit is used to monitor the deviation between the actual operating state and the prediction result and dynamically adjust the model parameters to maintain long-term stability. The main monitored parameters include key operating data such as the return oil temperature, circulation pump frequency, oil supply pressure, output current of the frequency converter, and valve opening. When it is detected that the temperature deviation exceeds the set threshold or the pressure fluctuation is greater than the set threshold, the system automatically adjusts the model weight coefficient and regression parameters to ensure the control accuracy. At the same time, the cumulative operating time and abnormal state of the equipment are recorded to provide data support for system maintenance. All parameters are collected in real time through the PLC, and the sampling period is 1 second to ensure the timeliness and accuracy of control.

[0133] The safety control unit is used to switch to a preset safe operating mode under abnormal working conditions and maintain the basic operation of the system.

[0134] The optimal control parameters are used to drive the actuator (such as a frequency converter, valve) through the PLC to adjust the system operating parameters in real time. The deviation between the actual operating state and the prediction result is continuously monitored through the feedback mechanism, and the model parameters are dynamically adjusted according to the deviation to maintain the long-term stability and optimization effect of the system, forming a closed-loop control mechanism.

[0135] In this embodiment, the control execution module applies the optimal control parameters to the optimization control, including: the algorithm runs on an industrial control computer, connects to the PLC through an industrial communication protocol, and drives the actuators (including the frequency converter for adjusting the circulating pump frequency, the solenoid valve for controlling the oil supply temperature, and the relay for switching the number of boilers). The response time of each mechanism is short, supporting rapid adjustment. During the control process, the sensor data is regularly read, the optimal control parameters generated by the prediction model and the ADE algorithm are called, and the control instructions are issued through the PLC. When the deviation between the return oil temperature and the target value exceeds the set range, the ADE algorithm is automatically triggered for re-optimization, and the optimization period can be configured according to the process requirements. By monitoring the difference between the real-time operating state and the prediction result, the model parameters (such as regression coefficients and weights) are dynamically adjusted, and the deviation trend is recorded for performance analysis. In abnormal conditions such as sensor failures, it automatically switches to the safe mode to maintain basic operation while recording the event log.

[0136] In summary, the sensor network transmits the real-time operating data to the data processing module through the information transmission network. The data processing module preprocesses and stores the data; the model training module calls the processed data, uses the data-driven modeling method to construct the natural gas dynamic prediction model of the combustion boiler and the optimization model of the circulating oil pump efficiency, and transmits the trained models to the optimization calculation module; while receiving the models, the optimization calculation module combines the real-time operating parameters collected by the sensor network, uses the ADE algorithm for parameter optimization, so that the system thermal efficiency, energy consumption, and temperature control accuracy reach the comprehensive optimal values.

[0137] The human-machine interaction module provides a human-machine interaction interface for monitoring the system status and supporting the adjustment of operating parameters and the optimization of control strategies.

[0138] In this embodiment, a touch-screen human-machine interaction interface is provided, which displays the operating data in real time (including temperature parameters such as TOut, TIn, Tha, etc., and optimization parameters such as Hz, State, etc.) and the system status (operating mode, optimization progress). The interface supports flexible switching between manual and automatic modes: in the manual mode, the operator can input the target parameters (such as the set temperature), and in the automatic mode, the system regularly performs optimization and updates the parameters. At the same time, it has the following functions:

[0139] Historical data recording: Stores the parameter combinations and fitness values of each optimization, supporting performance analysis and strategy adjustment.

[0140] Safety monitoring: When abnormalities such as temperature overrun are detected, an alarm prompt is immediately popped up and the event is recorded.

[0141] Parameter adjustment: Allows the operator to update the control strategy or model parameters online to ensure the flexible operation of the system.

[0142] The sensor network, data processing module, model training module, optimization calculation module, control execution module, and human-machine interaction module achieve data interaction through an industrial communication network (Ethernet) to ensure real-time performance. Among them, the optimization calculation module is a core component, and the ADE algorithm realizes the optimization adjustment of the number of boilers, the frequency of the circulating pump, and the fuel supply temperature in this module, directly affecting the system thermal efficiency, energy consumption, and temperature control accuracy.

[0143] Embodiment 2:

[0144] This embodiment provides an optimization control method for a heat-conducting oil boiler system based on an adaptive algorithm, which is implemented on the optimization control system of the heat-conducting oil boiler system based on the adaptive algorithm in the above embodiment. That is, those skilled in the art can understand the optimization control method for the heat-conducting oil boiler system based on the adaptive algorithm as the operation mode of the optimization control system of the heat-conducting oil boiler system based on the adaptive algorithm.

[0145] Figure 2 It is a flowchart of an optimization control method for a heat-conducting oil boiler system based on an adaptive algorithm in an embodiment of the present invention.

[0146] As Figure 2 shown, the optimization control method for the heat-conducting oil boiler system based on the adaptive algorithm includes the following steps:

[0147] Data acquisition step, real-time collecting the operation data of the heat-conducting oil boiler system through the sensor network;

[0148] Data preprocessing step, preprocessing and storing the operation data;

[0149] Model training step, using the preprocessed data to construct and train a combustion boiler natural gas dynamic prediction model and a circulating oil pump efficiency optimization model;

[0150] Parameter optimization step, based on the trained combustion boiler natural gas dynamic prediction model and circulating oil pump efficiency optimization model, using the adaptive differential evolution (ADE) algorithm to calculate the optimal control parameters. By calculating the optimal control parameters, the comprehensive optimization of system thermal efficiency improvement, energy consumption minimization, and temperature control accuracy is realized, while meeting the operation safety constraints.

[0151] Figure 5 It is a flowchart of the adaptive differential evolution (ADE) algorithm in an embodiment of the present invention.

[0152] In this embodiment, the ADE algorithm plays a core role in the parameter optimization stage of the optimal control. Its task is to calculate the control parameter combinations that meet the multi-objective optimization based on the real-time data and the output of the prediction model, including minimizing energy consumption, maximizing the temperature control accuracy, and ensuring the flow safety. Compared with traditional control methods (such as fixed parameter tables or manual experience adjustment), the ADE algorithm can efficiently solve non-linear and multi-constrained optimization problems through the differential evolution mechanism and adaptive parameter adjustment, ensuring that the system reaches the global optimal state under dynamic working conditions, which directly determines the effectiveness of the control strategy.

[0153] As Figure 5 shown, the ADE algorithm includes the following sub-steps:

[0154] Step a, set the population size, define the optimization variables including the supply oil temperature (TIn), the circulating pump frequency (Hz), the number of operating boilers (State), and initialize the mutation factor F, the crossover probability CR, and the maximum number of iterations. Randomly generate the initial population through a uniform distribution, and each individual represents a set of parameter combinations to ensure that the population diversity covers the parameter space.

[0155] Step b, construct the multi-objective fitness function:

[0156] Fitness = w1·Cost + w2·|T set - TIn| + w3·max(0, F min - F)

[0157] where, Cost is the energy consumption cost (the sum of gas and electricity costs), T set is the target return oil temperature, F min is the minimum flow demand, and w1, w2, w3 are weight coefficients, which respectively reflect the priorities of energy consumption, temperature deviation, and flow safety. The fitness function uses the prediction model to calculate the performance indicators of each parameter combination, and the weights can be adjusted according to the process requirements. For example, w1 is increased in the energy consumption sensitive scenario. The constraint conditions (such as temperature range, flow lower limit) are incorporated into the fitness calculation through penalty functions to ensure the feasibility of the solution.

[0158] Step c, iterative optimization:

[0159] Step c1, for the current optimal individual and the random individual, adopt the differential evolution strategy to generate the mutant individual:

[0160] v i = x best + F·(x r1 - x r2 )

[0161] where, x best is the current optimal individual, x r1 and xr2 is a randomly selected individual, and F is a mutation factor used to control the mutation amplitude. The mutation operation introduces population diversity through the differential vector to prevent the algorithm from falling into local optima.

[0162] Step c2, based on the mutated individual, generate a trial individual through the crossover probability:

[0163]

[0164] where v i,j is the j-th dimensional parameter of the mutated individual v i , x i,j is the j-th dimensional parameter of the current target individual x i , rand is a random number uniformly generated in the interval [0, 1] that determines whether to adopt the parameter of the mutated individual, and CR is the crossover probability that adjusts the fusion ratio of the old and new parameters. The crossover operation balances the exploration and exploitation capabilities and ensures the stability of the optimization process.

[0165] Step c3, compare the fitness values of the trial individual u i and the original individual x i , and select the one with better fitness to enter the next iteration.

[0166] Step c4, dynamically adjust the mutation factor F and the crossover probability CR according to the optimization process:

[0167] In the initial stage of iteration, if the convergence speed is slow, increase F to enhance the global search ability;

[0168] In the later stage of iteration, if the convergence tends to be stable, decrease F to improve the local search accuracy;

[0169] CR is adjusted according to the population diversity. If the diversity decreases, increase CR to maintain the exploration ability of the algorithm.

[0170] The adaptive mechanism improves the adaptability of the algorithm to different working conditions.

[0171] Step c5, the iteration process continues until the maximum iteration number is reached or the optimal fitness value tends to be stable, and then output the optimal parameter combination.

[0172] The iteration termination condition can be adjusted according to real-time requirements, such as setting a convergence threshold or a time limit.

[0173] To improve the calculation efficiency, parallel computing technology can be used to accelerate the population iteration process.

[0174] In this embodiment, through the in-depth application of the ADE algorithm in the optimization control link, the efficient optimization of the parameters of the heat transfer oil boiler system is realized. Its global optimization ability, adaptive characteristics and multi-objective balance design significantly improve the thermal efficiency, temperature control accuracy and economy of the system, and have high technological innovation and practical value. The system design takes into account scalability and robustness and can be extended to other similar industrial scenarios.

[0175] Real-time control step: Through the control execution module, the optimal control parameters are converted into execution instructions to adjust the number of boilers, the frequency of the circulating pump and the oil supply temperature of the heat transfer oil boiler system.

[0176] Human-computer interaction step: Provide a human-computer interaction interface for monitoring the system status and supporting the adjustment of operation parameters and the optimization of control strategies.

[0177] In this embodiment, a touch-screen human-computer interaction interface is provided to display the operation data (including temperature parameters such as TOut, TIn, Tha, etc., and optimization parameters such as Hz, State, etc.) and the system status (operation mode, optimization progress) in real time. The interface supports flexible switching between manual and automatic modes: in the manual mode, the operator can input target parameters (such as the set temperature), and in the automatic mode, the system periodically performs optimization and updates the parameters. At the same time, it has the following functions:

[0178] Historical data recording: Store the parameter combinations and fitness values of each optimization, and support performance analysis and strategy adjustment.

[0179] Safety monitoring: When anomalies such as temperature overrun are detected, an alarm prompt will be immediately popped up and the event will be recorded.

[0180] Parameter adjustment: Allow the operator to update the control strategy or model parameters online to ensure the flexible operation of the system.

[0181] Those skilled in the art know that in addition to implementing the system and its various devices, modules, and units provided by the present invention in the form of pure computer-readable program code, the method steps can be logically programmed to enable the system and its various devices, modules, and units provided by the present invention to be in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers to achieve the same function. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a kind of hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as the structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as either software modules for implementing the method or structures within the hardware component.

[0182] The specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. Without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. An optimized control system for a heat transfer oil boiler system based on an adaptive algorithm, characterized in that, Including: A sensor network for collecting real-time operation data of the heat-conducting oil boiler system, where the operation data includes the whole process of heat source, heat transfer, and circulation; A data processing module connected to the sensor network for preprocessing and storing the operation data; A model training module connected to the data processing module for constructing a dynamic prediction model of natural gas for the combustion boiler and an optimization model for the efficiency of the circulating oil pump; An optimization calculation module connected to the model training module for running an adaptive differential evolution algorithm to perform parameter optimization and generate optimal control parameters; A control execution module connected to the optimization calculation module for converting the optimal control parameters into control signals and driving the actuator; A human-machine interaction module providing a human-machine interaction interface for monitoring the system status and supporting the adjustment of operation parameters and the optimization of control strategies; The sensor network, the data processing module, the model training module, the optimization calculation module, the control execution module, and the human-machine interaction module achieve data interaction through an industrial communication network.

2. The optimized control system of a heat transfer oil boiler system based on an adaptive algorithm according to claim 1, wherein, The sensor network includes: Temperature sensors deployed on the heat-conducting oil pipeline for collecting the boiler outlet temperature, inlet temperature, main pipe oil outlet temperature, and return oil temperature; Hot air temperature sensors deployed at the outlet of the preheater for collecting the hot air temperature; An electric quantity monitoring unit including an ammeter deployed on the combustion boiler for collecting the heat load of the burner; A pump monitoring unit including a frequency converter, an ammeter, and a flowmeter deployed on the circulating oil pump for collecting the pump frequency, power, and flow rate.

3. The optimized control system of a heat transfer oil boiler system based on an adaptive algorithm according to claim 1, characterized in that, The preprocessing performed by the data processing module includes: Using the Kalman filter algorithm to smooth the temperature data and flow data; Eliminating abnormal data based on statistical methods; Synchronizing multi-source data through time series alignment technology; Storing the preprocessed data in a time series database.

4. The optimized control system of a heat transfer oil boiler system based on an adaptive algorithm according to claim 1, characterized in that, The dynamic prediction model of natural gas for the combustion boiler is based on the principle of energy conservation. By analyzing the correlation relationship between the heat-conducting oil temperature difference, hot air temperature, and flow rate, and combining heat calculation and regression analysis, the prediction relationship of natural gas consumption is deduced; The optimization model for the efficiency of the circulating oil pump is based on the principles of fluid mechanics and electric energy conversion. By analyzing the relationship between the pump frequency, opening state, and system resistance coefficient and the flow rate and power, the efficiency prediction relationship is established.

5. The optimized control system of a heat transfer oil boiler system based on an adaptive algorithm according to claim 4, characterized in that, The calculation formula for deducing the prediction relationship of natural gas consumption is as follows: where Q is the heat, TOut is the outlet oil temperature, F is the flow rate, TIn is the return oil temperature, C p is the specific heat capacity of the heat transfer oil, p is the constant pressure, ρ is the density, t and s are the time and the unit conversion factor respectively. Combining the combustion efficiency η and the calorific value Hv of natural gas, the model further derives the prediction relationship of the natural gas consumption. Where Gas is the natural gas consumption, and the regression coefficients k1 and k2 are determined by fitting historical data; The calculation formula for establishing the efficiency prediction relationship is: F = a·Hz 2 + b·Hz + c·E = d·Hz·F + e Where E is the power, and a, b, c, d, and e are fitting parameters.

6. The optimized control system of a heat transfer oil boiler system based on an adaptive algorithm according to claim 4, characterized in that, The dynamic prediction model of natural gas for the combustion boiler is based on the random forest algorithm. The training process improves the prediction accuracy through the integration of multiple decision trees and introduces a regularization term to control the model complexity; The training process of the optimization model for the efficiency of the circulating oil pump adopts the support vector regression algorithm, uses the radial basis function kernel, and optimizes the hyperparameters through grid search to ensure the generalization ability of the model.

7. The optimized control system of a heat transfer oil boiler system based on an adaptive algorithm according to claim 1, characterized in that, The parameter optimization performed by the optimization calculation module includes: Based on the trained model, the adaptive differential evolution algorithm is used with the main pipe return oil temperature, the frequency of the circulating oil pump, and the boiler opening state as optimization variables. By constructing a multi-objective fitness function including thermal efficiency, energy consumption cost, and temperature control accuracy, the constraint conditions include the flow safety range and process requirements. The adaptive differential evolution algorithm generates mutant individuals through the differential evolution mechanism and achieves global optimization by adaptively adjusting the mutation factor and crossover probability, and outputs the optimal control parameters.

8. The optimized control system of a heat transfer oil boiler system based on an adaptive algorithm according to claim 1, characterized in that, The control execution module includes: An execution unit, including a programmable logic controller, used to drive the frequency converter to adjust the circulating pump frequency, control the oil supply temperature with a solenoid valve, and switch the number of boilers with a relay; A feedback unit, used to monitor the deviation between the actual operating state and the prediction result; A safety control unit, used to switch to a preset safe operating mode under abnormal conditions.

9. An optimization control method for a heat transfer oil boiler system based on an adaptive algorithm, based on the optimization control system for a heat transfer oil boiler system based on an adaptive algorithm according to any one of claims 1-8, characterized in that, It includes the following steps: A data acquisition step, in which the operating data of the heat transfer oil boiler system is collected in real time through a sensor network; A data preprocessing step, in which the operating data is preprocessed and stored; A model training step, in which the preprocessed data is used to construct and train a natural gas dynamic prediction model for the combustion boiler and an optimization model for the circulating oil pump efficiency; A parameter optimization step, based on the trained natural gas dynamic prediction model for the combustion boiler and the optimization model for the circulating oil pump efficiency, using the adaptive differential evolution algorithm to calculate the optimal control parameters; A real-time control step, in which the optimal control parameters are converted into execution instructions through the control execution module to adjust the number of boilers, the circulating pump frequency, and the oil supply temperature of the heat transfer oil boiler system; A human-computer interaction step, providing a human-computer interaction interface for monitoring the system status and supporting the adjustment of operating parameters and the optimization of control strategies.

10. The optimized control method for a heat transfer oil boiler system based on an adaptive algorithm according to claim 9, characterized in that, The adaptive differential evolution algorithm includes the following sub-steps: Step a, set the population size, define the optimization variables, initialize the mutation factor, crossover probability, and maximum number of iterations, and generate the initial population; Step b, construct a multi-objective fitness function including energy consumption cost, temperature deviation, and flow safety index; Step c, through iterative optimization, output the optimal control parameter combination.

Images