Temperature control system and method based on multi-point heating

Abstract

The application relates to the technical field of automatic control, in particular to a temperature control system and method based on multi-point heating, which is composed of a nonlinear control fusion module, an adaptive weight optimization module, a distributed control design module, a predictive control strategy module, an intelligent scheduling strategy module, a dynamic heating distribution module and an adaptive fault detection module.In the application, the nonlinear control fusion technology is used to integrate PID control, fuzzy logic and neural network, so that precise and stable temperature control is realized; the adaptive weight optimization module makes the system more flexible and dynamically responds to working condition changes; the distributed design is adopted, the processing efficiency of large-scale temperature control is improved, parallelism and cooperation are strengthened; the predictive control strategy can optimize and control in advance when facing uncertainty; the intelligent scheduling strategy realizes intelligent optimization while ensuring heating speed and energy saving; and the adaptive fault detection guarantees the stability of the system.

Description

Technical Field

[0001] This invention relates to the field of automatic control technology, and in particular to a temperature control system and method based on multi-point heating. Background Technology

[0002] Automatic control technology is the field that studies how to design, implement, and optimize automatic control systems. It involves multiple technical aspects such as sensors, actuators, control algorithms, and data processing, aiming to improve the performance, stability, and efficiency of the system.

[0003] Multi-point heating-based temperature control systems are systems that utilize automatic control technology to control the temperature of multiple heating points to meet specific requirements and objectives. This system precisely controls the heating power of multiple heating points to achieve the target temperature. Its main purpose is to achieve precise control of the multi-point heating process, maintaining a stable and uniform distribution of the set target temperature across different heating points to meet specific process requirements or application needs. Examples include the need for temperature control in multiple areas during electronic device manufacturing, or the need for heating control in multiple areas within industrial heating equipment. To achieve this goal, multi-point heating-based temperature control systems typically acquire temperature data through sensors, use automatic control algorithms such as PID control and fuzzy logic control to calculate control signals, and adjust the heating power through actuators to achieve fine control of multiple heating points and stable regulation of the target temperature. Such systems provide accurate, stable, and balanced heating point temperature control to meet the temperature control needs of various application fields.

[0004] In existing temperature control systems based on multi-point heating, the temperature control algorithms are mostly relatively simple and cannot meet the nonlinear temperature control requirements of multi-point heating in complex environments. Furthermore, existing systems typically use static settings for weight adjustments, failing to adapt dynamically to actual conditions and lacking sufficient flexibility. For large-scale control tasks, the processing power and efficiency of existing systems are relatively low, and the response speed is slow. When facing uncertainties, the predictive and control capabilities of existing systems are weak, failing to effectively adjust the system's operating state and potentially leading to significant deviations. When handling anomalies and faults, existing systems usually adopt a passive response approach, lacking early warning and automatic diagnostic mechanisms, which hinders timely problem-solving and ensures stable system operation. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the existing technology and to propose a temperature control system and method based on multi-point heating.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: the temperature control system based on multi-point heating is composed of a nonlinear control fusion module, an adaptive weight optimization module, a distributed control design module, a predictive control strategy module, an intelligent scheduling strategy module, a dynamic heating allocation module, and an adaptive fault detection module;

[0007] The nonlinear control fusion module combines PID control algorithm with fuzzy logic control and neural network control to achieve nonlinear temperature control and generate a temperature control strategy.

[0008] The adaptive weight optimization module is based on a temperature control strategy and uses a genetic algorithm and a particle swarm algorithm to achieve dynamic weight adjustment and generate a weight optimization scheme.

[0009] The distributed control design module, based on a weighted optimization scheme, designs the temperature control task for multi-point heating to be allocated to multiple control nodes and generates a control node allocation strategy.

[0010] The predictive control strategy module is based on the control node allocation strategy, and uses model predictive control and model reference adaptive control to achieve optimized control and generate predictive control strategy output.

[0011] The intelligent scheduling strategy module, based on the output of predictive control strategy, combines optimization algorithms and artificial intelligence technology to predict and optimize temperature change trends, and generate an optimized scheduling scheme.

[0012] The dynamic heating allocation module generates a dynamic heating scheme by setting a dynamic allocation algorithm for heating power based on an optimized scheduling scheme.

[0013] The adaptive fault detection module is based on a dynamic heating scheme and combines machine learning and statistical methods to perform anomaly detection and fault diagnosis, generating a fault diagnosis report.

[0014] As a further aspect of the present invention, the nonlinear control fusion module includes a PID submodule, a fuzzy logic submodule, and a neural network submodule;

[0015] The adaptive weight optimization module includes a genetic algorithm submodule, a particle swarm algorithm submodule, and a weight dynamic adjustment submodule.

[0016] The distributed control design module includes a task allocation submodule, a parallel control submodule, and a cooperative control submodule.

[0017] The predictive control strategy module includes a model prediction submodule, an adaptive control submodule, and a strategy optimization submodule.

[0018] The intelligent scheduling strategy module includes an artificial intelligence technology submodule, a trend prediction submodule, and an optimization strategy submodule.

[0019] The dynamic heating distribution module includes a power adjustment submodule, a temperature feedback submodule, and a load demand submodule.

[0020] The adaptive fault detection module includes a machine learning submodule, a statistical detection submodule, and a real-time diagnosis submodule.

[0021] As a further embodiment of the present invention, the PID submodule uses the proportional-integral-derivative control method to adjust the temperature deviation in real time and outputs a basic temperature control command as the PID control output.

[0022] The fuzzy logic submodule is based on PID control output, uses fuzzy logic control to process uncertainty and nonlinear characteristics, and uses triangular membership functions and Gaussian membership functions to determine output rules and generate fuzzy control output.

[0023] The neural network submodule is based on fuzzy control output. It learns and optimizes the temperature control mode by training the neural network, applies the backpropagation algorithm to optimize the weights, and generates a neural network control strategy.

[0024] As a further aspect of the present invention, the genetic algorithm submodule is based on a neural network control strategy, uses a genetic algorithm to initially optimize the weights, and uses crossover, mutation, and selection operations to find the optimal weights generation by generation to generate a genetic weight scheme.

[0025] The particle swarm algorithm submodule is based on the genetic weight scheme. It uses the particle swarm algorithm to search and update the particle velocity and position in the weight space, find the optimal solution, and establish a particle swarm optimization weight scheme.

[0026] The weight dynamic adjustment submodule is based on the particle swarm optimization weight scheme. It dynamically adjusts the weight of each heating point according to real-time feedback to achieve optimized control and generate dynamic weight adjustment results.

[0027] As a further aspect of the present invention, the task allocation submodule uses a greedy algorithm to allocate temperature control tasks to different control nodes based on the dynamic weight adjustment results, thereby generating task allocation results.

[0028] The parallel control submodule enables multiple control nodes to work in parallel based on the task allocation results, forming a parallel working mode;

[0029] The collaborative control submodule is based on a parallel working mode and uses data synchronization and sharing mechanisms to achieve collaborative work between nodes and generate collaborative control strategies.

[0030] As a further aspect of the present invention, the model prediction submodule is based on a cooperative control strategy and uses a model prediction control algorithm to predict and control the temperature, and outputs the model prediction temperature result.

[0031] The adaptive control submodule, based on the model-predicted temperature results, adopts a model reference adaptive control strategy to adaptively adjust the temperature and generate an adaptive adjustment scheme.

[0032] The strategy optimization submodule optimizes the strategy based on an adaptive adjustment scheme, combining model reference and feedback control to generate predictive control strategy output.

[0033] As a further aspect of the present invention, the artificial intelligence technology submodule, based on the output of the predictive control strategy, combines deep learning and machine learning technologies to intelligently predict temperature changes and generate intelligent prediction results.

[0034] The predicted trend submodule uses time series analysis based on the intelligent prediction results to predict the trend changes in temperature and generate a temperature trend prediction report.

[0035] The optimization strategy submodule optimizes the scheduling strategy based on the temperature trend prediction report and combines it with the optimization algorithm to generate an optimized scheduling scheme.

[0036] As a further aspect of the present invention, the power adjustment submodule is based on an optimized scheduling scheme and uses a dynamic programming algorithm to adjust the heating power in real time, and outputs a power adjustment scheme.

[0037] The temperature feedback submodule collects temperature feedback from each heating point based on the power adjustment scheme and generates a temperature feedback summary report.

[0038] The load demand submodule uses linear programming to adjust and allocate the heating load based on the temperature feedback summary report, generating a dynamic heating scheme.

[0039] As a further aspect of the present invention, the machine learning submodule is based on a dynamic heating scheme, and applies support vector machine and decision tree algorithms to perform fault detection on temperature data and output preliminary fault detection results.

[0040] The statistical detection submodule uses chi-square test and T test to perform statistical analysis on the data based on the preliminary fault detection results, confirms outliers, and generates an anomaly detection report.

[0041] The real-time diagnostic submodule diagnoses faults based on anomaly detection reports, combined with real-time and historical data, and generates a fault diagnosis report.

[0042] The temperature control method based on multi-point heating is implemented based on the above-mentioned temperature control system based on multi-point heating, and includes the following steps:

[0043] S1: Based on the proportional-integral-derivative control method, fuzzy logic and neural networks are used to optimize temperature control and generate a preliminary neural network control strategy;

[0044] S2: Based on the preliminary neural network control strategy, the weights are optimized using a genetic algorithm and a particle swarm optimization algorithm to generate dynamic weight adjustment results;

[0045] S3: Based on the dynamic weight adjustment results, a greedy algorithm is used to allocate tasks and coordinate work to generate a collaborative control strategy;

[0046] S4: Based on the aforementioned collaborative control strategy, a model predictive control algorithm is used to predict and adaptively adjust the temperature, generating a predictive control strategy output;

[0047] S5: Based on the output of the predictive control strategy, intelligent prediction and heating power adjustment are performed using deep learning and time series analysis to generate a dynamic heating scheme;

[0048] S6: Based on the dynamic heating scheme, support vector machine, decision tree algorithm and chi-square test are used for fault detection and diagnosis to generate a fault diagnosis report.

[0049] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0050] This invention utilizes a nonlinear control algorithm fusion method, combining PID control, fuzzy logic control, and neural network control to achieve more precise and stable temperature control, meeting the temperature regulation needs in complex environments. An adaptive weight optimization module dynamically adjusts system weights, making the system more flexible and effectively adapting to various operating conditions, ensuring optimal system operation. A distributed control design is adopted, significantly improving processing speed and efficiency when handling large-scale heating point temperature control tasks. Parallel and collaborative control also aligns better with modern computing models, contributing to improved system response speed. The introduction of a predictive control strategy enables predictive and optimized control under uncertainties, allowing for proactive adjustments and more stable system operation. An intelligent scheduling strategy module achieves intelligent optimized scheduling while ensuring heating speed, considering the influence between heating points, and saving energy, fully demonstrating the system's intelligent characteristics. An adaptive fault detection module monitors and identifies anomalies and faults in the system in real time, handling them promptly to ensure stable system operation. Attached Figure Description

[0051] Figure 1 This is a system flowchart of the present invention;

[0052] Figure 2 This is a system block diagram of the present invention;

[0053] Figure 3 This is a flowchart of the nonlinear control fusion module of the present invention;

[0054] Figure 4 This is a flowchart of the adaptive weight optimization module of the present invention;

[0055] Figure 5 This is a flowchart of the distributed control design module of the present invention;

[0056] Figure 6 This is a flowchart of the predictive control strategy module of the present invention;

[0057] Figure 7 This is a flowchart of the intelligent scheduling strategy module of the present invention;

[0058] Figure 8 This is a flowchart of the dynamic heating distribution module of the present invention;

[0059] Figure 9 This is a flowchart of the adaptive fault detection module of the present invention;

[0060] Figure 10 This is a schematic diagram of the working steps of the present invention. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0062] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0063] Example 1

[0064] Please see Figure 1 The present invention provides a technical solution: a temperature control system based on multi-point heating is composed of a nonlinear control fusion module, an adaptive weight optimization module, a distributed control design module, a predictive control strategy module, an intelligent scheduling strategy module, a dynamic heating allocation module, and an adaptive fault detection module;

[0065] The nonlinear control fusion module combines PID control algorithm with fuzzy logic control and neural network control to achieve nonlinear temperature control and generate temperature control strategy.

[0066] The adaptive weight optimization module is based on a temperature control strategy and uses genetic algorithms and particle swarm optimization algorithms to achieve dynamic weight adjustment and generate a weight optimization scheme.

[0067] The distributed control design module, based on a weighted optimization scheme, designs the temperature control task for multi-point heating to be distributed to multiple control nodes and generates a control node allocation strategy.

[0068] The predictive control strategy module is based on the control node allocation strategy, and uses model predictive control and model reference adaptive control to achieve optimized control and generate predictive control strategy output.

[0069] The intelligent scheduling strategy module, based on the output of predictive control strategy, combines optimization algorithms and artificial intelligence technology to predict and optimize temperature change trends, and generate optimized scheduling schemes.

[0070] The dynamic heating allocation module generates a dynamic heating scheme by setting a dynamic allocation algorithm for heating power based on an optimized scheduling scheme.

[0071] The adaptive fault detection module is based on a dynamic heating scheme and combines machine learning and statistical methods to detect anomalies and diagnose faults, generating a fault diagnosis report.

[0072] By integrating nonlinear control algorithms with fuzzy logic control and neural network control, the system achieves more precise temperature regulation, improving control system performance. The adaptive weight optimization module dynamically adjusts weights based on the temperature control strategy, further optimizing control effects. The distributed control design module distributes tasks across multiple nodes, improving parallelism and processing power. The predictive control strategy module utilizes model prediction and adaptive control methods to improve prediction and regulation accuracy. The intelligent scheduling strategy module combines optimization algorithms and artificial intelligence technology to predict and optimize temperature change trends, improving energy efficiency. The dynamic heating allocation module adjusts heating power in real time, optimizing energy distribution and stability. The adaptive fault detection module performs anomaly detection and fault diagnosis, improving system reliability.

[0073] Please see Figure 2 The nonlinear control fusion module includes a PID submodule, a fuzzy logic submodule, and a neural network submodule.

[0074] The adaptive weight optimization module includes a genetic algorithm submodule, a particle swarm optimization submodule, and a dynamic weight adjustment submodule;

[0075] The distributed control design module includes a task allocation submodule, a parallel control submodule, and a cooperative control submodule;

[0076] The predictive control strategy module includes a model prediction submodule, an adaptive control submodule, and a strategy optimization submodule;

[0077] The intelligent scheduling strategy module includes an artificial intelligence technology submodule, a trend prediction submodule, and an optimization strategy submodule;

[0078] The dynamic heating distribution module includes a power adjustment submodule, a temperature feedback submodule, and a load demand submodule;

[0079] The adaptive fault detection module includes a machine learning submodule, a statistical detection submodule, and a real-time diagnosis submodule.

[0080] The PID submodule enables traditional proportional, integral, and derivative control, providing fast response and stable control. The fuzzy logic submodule utilizes fuzzy inference rules to handle fuzziness and uncertainty, offering flexible control strategies. The neural network submodule, based on a trained neural network model, can learn nonlinear mapping relationships to achieve accurate temperature control.

[0081] The adaptive weight optimization module, through its genetic algorithm and particle swarm optimization sub-modules, can automatically optimize the weights of different control strategies to adapt to various working scenarios and requirements. The dynamic weight adjustment sub-module dynamically adjusts the weights based on real-time feedback and temperature change trends, improving the flexibility and accuracy of control.

[0082] The distributed control design module distributes tasks to multiple nodes through the task allocation submodule, enabling parallel control and improving the processing power and efficiency of the control system. The parallel control submodule can handle multiple temperature control tasks simultaneously, enhancing the system's parallelism. The collaborative control submodule enables coordinated work between different nodes, ensuring the stability and consistency of the overall system.

[0083] The predictive control strategy module predicts temperature changes through a model prediction submodule, and combines this with an adaptive control submodule and a strategy optimization submodule to achieve intelligent adjustment and control. The model prediction submodule provides accurate temperature predictions based on historical data and the system model. The adaptive control submodule dynamically adjusts the control strategy according to the prediction results, achieving adaptive adjustment. The strategy optimization submodule uses optimization algorithms to optimize the control strategy, improving energy efficiency and control performance.

[0084] The intelligent scheduling strategy module analyzes historical data and trend information through artificial intelligence technology sub-modules to predict temperature change trends and provide guidance for temperature control. The trend prediction sub-module can accurately predict long-term temperature trends and short-term fluctuations, providing intelligent decision support. The optimization strategy sub-module combines optimization algorithms to optimize temperature scheduling strategies to achieve optimal utilization of energy and resources, improving system efficiency and economy.

[0085] The dynamic heating distribution module adjusts the heating power in real time according to demand through the power adjustment submodule, achieving dynamic energy distribution. The temperature feedback submodule monitors and controls the heating process in real time based on temperature feedback information. The load demand submodule optimizes and schedules according to system load demand, achieving rational allocation and utilization of heating resources.

[0086] The adaptive fault detection module performs real-time anomaly detection and fault diagnosis through machine learning and statistical detection submodules. The machine learning submodule learns the system's normal and abnormal operating modes, providing accurate fault diagnosis results. The statistical detection submodule uses statistical analysis methods to detect system data, providing rapid anomaly detection and fault determination. The real-time diagnosis submodule can respond to and handle system faults promptly, improving system reliability and security.

[0087] Please see Figure 3 The PID submodule uses proportional-integral-derivative control to adjust the temperature deviation in real time and outputs basic temperature control commands as PID control outputs.

[0088] The fuzzy logic submodule is based on PID control output. It uses fuzzy logic control to handle uncertainty and nonlinearity, and uses triangular membership functions and Gaussian membership functions to determine output rules and generate fuzzy control output.

[0089] The neural network submodule is based on fuzzy control output. It learns and optimizes the temperature control mode by training the neural network, applies the backpropagation algorithm to optimize the weights, and generates a neural network control strategy.

[0090] The PID submodule adjusts the temperature deviation in real time using proportional, integral, and derivative control methods, and generates basic temperature control commands as the PID control output. The fuzzy logic submodule uses fuzzy logic control methods to handle uncertainties and nonlinear characteristics. It uses triangular membership functions and Gaussian membership functions to define output rules and generate fuzzy control outputs. This allows for flexible adjustment of the temperature control strategy through fuzzy rule reasoning.

[0091] The neural network submodule further enhances temperature control performance. It uses fuzzy control output as input and leverages a trained neural network model to learn and optimize the temperature control pattern. By adjusting network weights through backpropagation, the neural network submodule generates a more precise control strategy, further improving the accuracy and stability of temperature control.

[0092] Please see Figure 4 The genetic algorithm submodule is based on a neural network control strategy. It uses a genetic algorithm to initially optimize the weights and uses crossover, mutation, and selection operations to find the best weights generation by generation, generating a genetic weight scheme.

[0093] The particle swarm optimization submodule is based on the genetic weight scheme. It uses the particle swarm optimization algorithm to search and update the particle velocity and position in the weight space, find the optimal solution, and establish a particle swarm optimization weight scheme.

[0094] The dynamic weight adjustment submodule is based on the particle swarm optimization weight scheme. It dynamically adjusts the weight of each heating point according to real-time feedback to achieve optimized control and generate dynamic weight adjustment results.

[0095] The genetic algorithm submodule uses a genetic algorithm to initially optimize the weights, iteratively searching for the optimal weight combination through operations such as crossover, mutation, and selection. This leverages the global search capability of the genetic algorithm to quickly find a good initial weight scheme, providing a solid starting point for subsequent optimization.

[0096] The particle swarm optimization (PSO) submodule uses a weight scheme generated by a genetic algorithm to search and update the weight space. By continuously updating the particle velocity and position, the system can find a better solution in the weight space, obtaining a PSO-optimized weight scheme. The PSO algorithm effectively utilizes global and local search capabilities, improving the efficiency and accuracy of weight optimization.

[0097] The dynamic weight adjustment submodule uses a particle swarm optimization (PSO) weighting scheme to dynamically adjust the weights of each heating point based on real-time feedback. By monitoring temperature and control performance in real time, the system can adjust the weights of different heating points as needed to achieve a more optimized control strategy. This dynamic adjustment improves the system's adaptability and robustness, enabling flexible control based on actual conditions.

[0098] Please see Figure 5 The task allocation submodule uses a greedy algorithm to allocate temperature control tasks to different control nodes based on the dynamic weight adjustment results, and generates task allocation results.

[0099] Based on the task allocation results, the parallel control submodule enables multiple control nodes to work in parallel, forming a parallel working mode;

[0100] The collaborative control submodule is based on a parallel working mode. Through data synchronization and sharing mechanisms, it enables collaborative work between nodes and generates collaborative control strategies.

[0101] The task allocation submodule, based on the results of dynamic weight adjustment, uses a greedy algorithm to distribute temperature control tasks across different control nodes. By considering the load and capabilities of each node, a reasonable task allocation result is generated. This achieves dynamic balancing of tasks among multiple control nodes, improving the overall system efficiency and response speed.

[0102] Based on task allocation results, the parallel control submodule enables multiple control nodes to perform temperature control tasks simultaneously. Through parallel operation, the system can handle multiple temperature control tasks concurrently, improving its processing power and efficiency. Parallel control allows temperature control tasks to be completed in a shorter time, enhancing system response and real-time performance.

[0103] The collaborative control submodule, based on a parallel operating mode, implements a data synchronization and sharing mechanism among multiple nodes, enabling collaborative work between them. Through real-time data interaction and sharing, nodes can communicate and coordinate their work to generate collaborative control strategies. This ensures the consistency and stability of temperature control across all nodes and achieves better overall control performance.

[0104] Please see Figure 6 The model prediction submodule is based on a collaborative control strategy and uses model predictive control algorithms to predict and control temperature, and outputs the model predicted temperature results.

[0105] The adaptive control submodule uses the model reference adaptive control strategy to adaptively adjust the temperature based on the model's predicted temperature results, generating an adaptive adjustment scheme.

[0106] The strategy optimization submodule optimizes the strategy based on the adaptive adjustment scheme, combining model reference and feedback control to generate predictive control strategy output.

[0107] The model prediction submodule, based on a cooperative control strategy, utilizes model predictive control algorithms to predict and control temperature. By establishing a dynamic model of the system and combining it with real-time input and state information, the future temperature trend can be predicted. The model prediction submodule outputs the predicted temperature results, providing a foundation for subsequent adaptive control and strategy optimization.

[0108] The adaptive control submodule uses a model reference adaptive control strategy to adaptively adjust the temperature based on model-predicted temperature results. By comparing the predicted temperature with the actual temperature feedback, the adaptive control submodule can dynamically adjust the control strategy to achieve more accurate temperature control. By continuously adjusting the control parameters and weights, adaptive control can adapt to different operating states and changing conditions, improving the robustness and performance of temperature control.

[0109] The strategy optimization submodule is based on an adaptive adjustment scheme, combining model reference and feedback control for strategy optimization. By considering the trade-offs between model reference and actual feedback, the strategy optimization submodule can find the optimal trade-off point in the predictive control strategy. By optimizing control parameters and adjustment strategies, the accuracy and performance of temperature control can be further improved, resulting in a faster and more stable system response.

[0110] Please see Figure 7 The artificial intelligence technology submodule, based on the output of the predictive control strategy, combines deep learning and machine learning technologies to intelligently predict temperature changes and generate intelligent prediction results.

[0111] The trend prediction submodule uses time series analysis based on intelligent prediction results to predict temperature trend changes and generate a temperature trend prediction report.

[0112] The optimization strategy submodule optimizes the scheduling strategy based on the temperature trend prediction report and combines it with optimization algorithms to generate an optimized scheduling scheme.

[0113] The artificial intelligence technology submodule, based on the output of a predictive control strategy, combines deep learning and machine learning techniques to intelligently predict temperature changes. By learning from historical temperature data and training models, the AI ​​submodule can learn complex patterns and norms of temperature. In this way, the system can intelligently predict future temperature changes based on current input data, generating intelligent prediction results.

[0114] The trend prediction submodule uses time series analysis to predict temperature trends based on intelligent forecasting results. By analyzing historical trends and periodicity in temperature data, the trend prediction submodule can predict future temperature trends. This allows the system to generate a temperature trend prediction report, providing information about future temperature directions.

[0115] The optimization strategy submodule optimizes the scheduling strategy based on temperature trend prediction reports and combined with optimization algorithms. By analyzing the temperature trend prediction reports and the current control environment, the optimization strategy submodule can determine the optimal scheduling scheme. In this way, the system can improve the performance and energy efficiency of temperature control by adjusting control parameters and optimizing the scheduling strategy. The optimization strategy submodule provides more intelligent and accurate decision-making, enabling the system to respond more effectively to temperature changes and achieve optimized control.

[0116] Please see Figure 8 The power adjustment submodule, based on an optimized scheduling scheme, adjusts the heating power in real time through a dynamic programming algorithm and outputs the power adjustment scheme.

[0117] The temperature feedback submodule collects temperature feedback from each heating point based on the power adjustment scheme and generates a temperature feedback summary report.

[0118] The load demand submodule uses linear programming to adjust and allocate heating loads based on temperature feedback summary reports, generating dynamic heating schemes.

[0119] The power adjustment submodule, based on an optimized scheduling scheme, adjusts the heating power in real time using a dynamic programming algorithm. According to the optimal power allocation strategy determined in the optimized scheduling scheme, the power adjustment submodule can adjust the power of each heating point in real time to achieve precise temperature control. Through the dynamic programming algorithm, the system can weigh the power adjustments between different heating points to meet temperature control requirements.

[0120] The temperature feedback submodule, based on the power adjustment scheme, collects temperature feedback from each heating point and generates a summary temperature feedback report. By monitoring and collecting temperature feedback from the heating points in real time, the temperature feedback submodule can obtain accurate data on the actual temperature conditions. This temperature feedback data can be used for subsequent analysis and decision-making to further optimize the temperature control strategy.

[0121] The load demand submodule adjusts and allocates heating loads based on the temperature feedback summary report using linear programming. By analyzing the temperature data and system load demand in the temperature feedback summary report, the load demand submodule can optimize the allocation of heating loads using linear programming to achieve more efficient energy utilization. Through dynamic adjustment and allocation of heating loads, the system can quickly adapt to real-time demand and implement dynamic heating schemes.

[0122] Please see Figure 9 The machine learning submodule is based on a dynamic heating scheme and uses support vector machine and decision tree algorithms to perform fault detection on temperature data and output preliminary fault detection results.

[0123] Based on the preliminary fault detection results, the statistical detection submodule uses chi-square test and T test to perform statistical analysis on the data, identify outliers, and generate an anomaly detection report.

[0124] The real-time diagnostic submodule diagnoses faults based on anomaly detection reports, combined with real-time and historical data, and generates a fault diagnosis report.

[0125] The machine learning submodule, based on a dynamic heating scheme, applies machine learning algorithms such as support vector machines and decision trees to perform fault detection on temperature data. By training the model and using historical data, the machine learning submodule can learn normal and abnormal patterns in the temperature data. Based on the temperature data from the dynamic heating scheme, the machine learning submodule can perform preliminary fault detection on the temperature data and output the possible fault detection results.

[0126] Based on the preliminary fault detection results, the statistical detection submodule employs statistical analysis methods such as chi-square test and T-test to further detect and analyze anomalies in the data. By comparing with the normal pattern and analyzing the statistical characteristics of outliers, the statistical detection submodule can identify existing anomalies and generate an anomaly detection report. This eliminates the possibility of accidental false alarms and improves the accuracy and reliability of fault detection.

[0127] The real-time diagnostic submodule diagnoses faults based on anomaly detection reports, combined with real-time and historical data. By analyzing and comparing anomalies, the real-time diagnostic submodule can trace the causes of anomalies and generate fault diagnosis reports by combining historical fault data and model training results. This provides maintenance personnel with detailed fault information, enabling timely remedial measures and ensuring the stability and reliability of the system.

[0128] Please see Figure 10 The temperature control method based on multi-point heating is executed based on the above-mentioned temperature control system based on multi-point heating, and includes the following steps:

[0129] S1: Based on the proportional-integral-derivative control method, fuzzy logic and neural networks are used to optimize temperature control and generate a preliminary neural network control strategy;

[0130] S2: Based on the preliminary neural network control strategy, the weights are optimized using genetic algorithm and particle swarm algorithm to generate dynamic weight adjustment results;

[0131] S3: Based on the dynamic weight adjustment results, a greedy algorithm is used to allocate tasks and coordinate work to generate a collaborative control strategy;

[0132] S4: Based on the cooperative control strategy, the model predictive control algorithm is used to predict and adaptively adjust the temperature, and generate the predictive control strategy output;

[0133] S5: Based on the output of the predictive control strategy, deep learning and time series analysis are used to perform intelligent prediction and heating power adjustment to generate a dynamic heating scheme;

[0134] S6: Based on the dynamic heating scheme, support vector machine, decision tree algorithm and chi-square test are used for fault detection and diagnosis, and fault diagnosis report is generated.

[0135] First, by employing proportional-integral-derivative (PI-DI) control combined with fuzzy logic and neural networks for optimization, the accuracy and stability of temperature control can be improved. Second, by optimizing control parameters, such as adjusting weights using genetic algorithms and particle swarm optimization, the performance of the temperature control system can be improved. Furthermore, through a greedy algorithm for task allocation and collaborative work, coordinated operation among multiple heating points is achieved, further enhancing the overall balance and stability of temperature control. Additionally, by using model predictive control algorithms for temperature prediction and adaptive adjustment, the system can dynamically adjust based on real-time information, achieving more accurate temperature control. Further application of deep learning and time series analysis techniques enables intelligent prediction and heating power adjustment, improving energy efficiency and temperature control stability. Finally, using support vector machines, decision tree algorithms, and chi-square tests for fault detection and diagnosis allows for timely fault identification and resolution, improving system reliability and maintenance efficiency.

[0136] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A temperature control system based on multi-point heating, characterized in that: The temperature control system based on multi-point heating consists of a nonlinear control fusion module, an adaptive weight optimization module, a distributed control design module, a predictive control strategy module, an intelligent scheduling strategy module, a dynamic heating allocation module, and an adaptive fault detection module. The nonlinear control fusion module combines PID control algorithm with fuzzy logic control and neural network control to achieve nonlinear temperature control and generate a temperature control strategy. The adaptive weight optimization module is based on a temperature control strategy and uses a genetic algorithm and a particle swarm algorithm to achieve dynamic weight adjustment and generate a weight optimization scheme. The distributed control design module, based on a weighted optimization scheme, designs the temperature control task for multi-point heating to be allocated to multiple control nodes and generates a control node allocation strategy. The predictive control strategy module is based on the control node allocation strategy, and uses model predictive control and model reference adaptive control to achieve optimized control and generate predictive control strategy output. The intelligent scheduling strategy module, based on the output of predictive control strategy, combines optimization algorithms and artificial intelligence technology to predict and optimize temperature change trends, and generate an optimized scheduling scheme. The dynamic heating allocation module generates a dynamic heating scheme by setting a dynamic allocation algorithm for heating power based on an optimized scheduling scheme. The adaptive fault detection module is based on a dynamic heating scheme and combines machine learning and statistical methods to perform anomaly detection and fault diagnosis, generating a fault diagnosis report.

2. The temperature control system based on multi-point heating according to claim 1, characterized in that: The nonlinear control fusion module includes a PID submodule, a fuzzy logic submodule, and a neural network submodule; The adaptive weight optimization module includes a genetic algorithm submodule, a particle swarm algorithm submodule, and a weight dynamic adjustment submodule. The distributed control design module includes a task allocation submodule, a parallel control submodule, and a cooperative control submodule. The predictive control strategy module includes a model prediction submodule, an adaptive control submodule, and a strategy optimization submodule. The intelligent scheduling strategy module includes an artificial intelligence technology submodule, a trend prediction submodule, and an optimization strategy submodule. The dynamic heating distribution module includes a power adjustment submodule, a temperature feedback submodule, and a load demand submodule. The adaptive fault detection module includes a machine learning submodule, a statistical detection submodule, and a real-time diagnosis submodule.

3. The temperature control system based on multi-point heating according to claim 2, characterized in that: The PID submodule uses proportional-integral-derivative control to adjust the temperature deviation in real time and outputs basic temperature control commands as PID control output. The fuzzy logic submodule is based on PID control output, uses fuzzy logic control to process uncertainty and nonlinear characteristics, uses triangular membership function and Gaussian membership function to determine output rules, and generates fuzzy control output; The neural network submodule is based on fuzzy control output. It learns and optimizes the temperature control mode by training the neural network, and applies the backpropagation algorithm to optimize the weights to generate a neural network control strategy.

4. The temperature control system based on multi-point heating according to claim 2, characterized in that: The genetic algorithm submodule is based on a neural network control strategy. It uses a genetic algorithm to initially optimize the weights and uses crossover, mutation, and selection operations to find the optimal weights generation by generation, generating a genetic weight scheme. The particle swarm algorithm submodule is based on the genetic weight scheme. It uses the particle swarm algorithm to search and update the particle velocity and position in the weight space, find the optimal solution, and establish a particle swarm optimization weight scheme. The weight dynamic adjustment submodule is based on the particle swarm optimization weight scheme. It dynamically adjusts the weight of each heating point according to real-time feedback to achieve optimized control and generate dynamic weight adjustment results.

5. The temperature control system based on multi-point heating according to claim 2, characterized in that: The task allocation submodule uses a greedy algorithm to allocate temperature control tasks to different control nodes based on the dynamic weight adjustment results, and generates task allocation results. The parallel control submodule enables multiple control nodes to work in parallel based on the task allocation results, forming a parallel working mode; The collaborative control submodule is based on a parallel working mode and uses data synchronization and sharing mechanisms to achieve collaborative work between nodes and generate collaborative control strategies.

6. The temperature control system based on multi-point heating according to claim 2, characterized in that: The model prediction submodule is based on a cooperative control strategy and uses a model predictive control algorithm to predict and control temperature, and outputs the model predicted temperature results. The adaptive control submodule, based on the model-predicted temperature results, adopts a model reference adaptive control strategy to adaptively adjust the temperature and generate an adaptive adjustment scheme. The strategy optimization submodule optimizes the strategy based on an adaptive adjustment scheme, combining model reference and feedback control to generate predictive control strategy output.

7. The temperature control system based on multi-point heating according to claim 2, characterized in that: The artificial intelligence technology submodule, based on the output of the predictive control strategy, combines deep learning and machine learning technologies to intelligently predict temperature changes and generate intelligent prediction results. The predicted trend submodule uses time series analysis based on the intelligent prediction results to predict the trend changes in temperature and generate a temperature trend prediction report. The optimization strategy submodule optimizes the scheduling strategy based on the temperature trend prediction report and combines it with the optimization algorithm to generate an optimized scheduling scheme.

8. The temperature control system based on multi-point heating according to claim 2, characterized in that: The power adjustment submodule is based on an optimized scheduling scheme and uses a dynamic programming algorithm to adjust the heating power in real time and output the power adjustment scheme. The temperature feedback submodule collects temperature feedback from each heating point based on the power adjustment scheme and generates a temperature feedback summary report. The load demand submodule uses linear programming to adjust and allocate the heating load based on the temperature feedback summary report, generating a dynamic heating scheme.

9. The temperature control system based on multi-point heating according to claim 2, characterized in that: The machine learning submodule is based on a dynamic heating scheme and uses support vector machine and decision tree algorithms to perform fault detection on temperature data, and outputs preliminary fault detection results. The statistical detection submodule uses chi-square test and T test to perform statistical analysis on the data based on the preliminary fault detection results, confirms outliers, and generates an anomaly detection report. The real-time diagnostic submodule diagnoses faults based on anomaly detection reports, combined with real-time and historical data, and generates a fault diagnosis report.

10. A temperature control method based on multi-point heating, characterized in that, The temperature control method based on multi-point heating is executed based on the temperature control system based on multi-point heating according to any one of claims 1-9, and includes the following steps: Based on the proportional-integral-derivative control method, fuzzy logic and neural networks are used to optimize temperature control and generate a preliminary neural network control strategy. Based on the aforementioned preliminary neural network control strategy, the weights are optimized using a genetic algorithm and a particle swarm optimization algorithm to generate dynamic weight adjustment results. Based on the dynamic weight adjustment results, a greedy algorithm is used to allocate tasks and coordinate work, generating a collaborative control strategy. Based on the aforementioned collaborative control strategy, a model predictive control algorithm is used to predict and adaptively adjust the temperature, generating a predictive control strategy output. Based on the output of the predictive control strategy, deep learning and time series analysis are used to perform intelligent prediction and heating power adjustment to generate a dynamic heating scheme. Based on the aforementioned dynamic heating scheme, support vector machine, decision tree algorithm, and chi-square test are used for fault detection and diagnosis to generate a fault diagnosis report.

Images