Molding equipment large inertia dynamic flow balance and control method based on knowledge graph

Abstract

The invention belongs to the technical field of industrial automation and control, and particularly relates to a large-inertia dynamic flow balance and control method for forming equipment based on a knowledge graph, which comprises the following steps: firstly, collecting real-time sensor data and historical operation records, denoising by adopting a moving average method, and carrying out normalization processing; a complex association model based on a multi-level knowledge graph is constructed, and intelligent reasoning and optimization decision making are supported; then, a reinforcement learning model is constructed, a state space, an action space and a reward function are defined, and the training efficiency is improved in combination with experience playback and a target network mechanism; and finally, according to the real-time state of the equipment, the system generates an optimal control signal, and dynamically adjusts control parameters by means of a knowledge graph reasoning mechanism to ensure that the equipment works in the optimal operation state. The method can solve the problems that in an existing industrial control system, the flow control precision is not high in the face of large-inertia dynamic flow balance and precise control scenes, and the control is not stable under the large-inertia load.

Description

Technical Field

[0001] The present invention belongs to the field of industrial automation and control technology, and in particular to a large-inertia dynamic flow balancing and control method for molding equipment based on a knowledge graph. Background Art

[0002] In traditional industrial control systems, the regulation of parameters such as flow and pressure usually relies on classical control algorithms, such as PID control and fuzzy control. These methods work well in simple linear systems, but when faced with complex nonlinear systems or large inertia loads, the control accuracy and response speed are often greatly limited. Traditional control algorithms rely on preset rules and fixed parameters, lack adaptive capabilities, and have difficulty coping with dynamic changes in equipment status. Especially in complex industrial environments, these algorithms often cannot quickly adapt to changes in equipment load or working conditions, resulting in reduced efficiency or even equipment out of control. In addition, traditional methods still require manual intervention and lack intelligent decision-making and optimization capabilities, which is a significant bottleneck for modern automation systems.

[0003] In recent years, reinforcement learning, as a technology that continuously optimizes decisions based on interaction with the environment, has gradually been applied to control systems. Unlike traditional control methods, reinforcement learning can dynamically adjust control strategies based on real-time feedback, making the system adaptive and able to automatically adjust operations based on device status and environmental changes. Through a reward mechanism, reinforcement learning guides the intelligent agent to find the optimal control action, thereby improving system efficiency and stability. This technology is particularly suitable for dynamic and complex environments and helps to continuously improve control decisions.

[0004] Although reinforcement learning has made some progress in control systems, it is still challenging to rely solely on its exploration mechanism when facing large-scale and complex systems. In particular, during the training process, the cost of trial and error is high and the learning progress is slow. To solve this problem, research combining knowledge graphs with reinforcement learning has attracted attention. Knowledge graphs integrate a large amount of heterogeneous data through graph structures, provide relationships and rules between devices, sensors, and control strategies, thereby reducing unnecessary exploration, accelerating the learning process, and improving decision accuracy. In control systems, knowledge graphs can not only store the relationship between devices and control signals, but also provide real-time decision support for reinforcement learning models through reasoning mechanisms and optimize control strategies.

[0005] Focusing on large inertia dynamic flow balance and precise control scenarios, such scenarios are widely present in many industrial fields such as heavy machinery manufacturing and large-scale chemical production. In these fields, the equipment has huge operating inertia, and the dynamic balance control of the flow is extremely demanding. Slight flow deviations may cause product quality problems or even equipment failures. Traditional control methods are unable to cope with such stringent requirements. Although reinforcement learning has potential, it is limited by its own defects and cannot take on the heavy responsibility. Although the research on the combination of knowledge graphs and reinforcement learning has started, the actual integration of the two and their application in such complex scenarios still faces many technical difficulties.

[0006] Based on this, this solution closely combines reinforcement learning with knowledge graph reasoning to design a new method to ensure that the system can dynamically optimize control signals even under complex and changeable working conditions, and quickly and adaptively adjust control strategies according to the real-time status of the equipment, thereby comprehensively improving the flexibility and stability of the system and filling the current technological gaps. Summary of the invention

[0007] The technical problem solved by the present invention is to provide a large-inertia dynamic flow balancing and control method for molding equipment based on a knowledge graph, so as to solve the problems of low flow control accuracy and unstable control under large-inertia loads in existing industrial control systems when facing large-inertia dynamic flow balancing and precise control scenarios.

[0008] The basic solution provided by the present invention: a large inertia dynamic flow balance and control method for forming equipment based on knowledge graph, including: S1: collect real-time operation data and historical operation data of the molding equipment, and pre-process the real-time operation data to generate pre-processed real-time operation data; S2: Based on the preprocessed real-time operation data and historical operation data, call the database to build a knowledge graph model based on entity nodes and relationships; S3: constructing a reinforcement learning model, in which the state space is defined by the real-time operation data of the molding equipment, the action space is defined by the control signal of the molding equipment, and the reward function is defined by the preset molding equipment performance standard; S4: The reinforcement learning model interacts with the large inertia dynamic flow balance and precise control scenario of the molding equipment, initializes the state according to the state space, selects actions according to the current state of the molding equipment, optimizes the strategy through reward function feedback, and provides control strategy guidance for the initial stage of the reinforcement learning model training through the knowledge graph model, so as to train the reinforcement learning model to generate the optimal control signal and obtain a trained reinforcement learning model; S5: Deploy the knowledge graph model and the trained reinforcement learning model to the control system of the molding equipment. The control system uses the knowledge graph model to infer the control strategy of the reinforcement learning model, and calls the reinforcement learning model to generate action control signals according to the inferred control strategy to control the operating state of the molding equipment. The reinforcement learning model then continuously optimizes the control strategy of the molding equipment based on the feedback from the molding equipment.

[0009] The principle and advantage of the present invention are as follows: In the technical solution of the present application, the real-time operation data and historical operation data of the molding equipment are first collected, wherein the preprocessing of the real-time operation data helps to provide high-quality data support for the subsequent model input, and then, a knowledge graph model is constructed based on the preprocessed real-time operation data, historical operation data and NE04J database. At this time, the construction of the knowledge graph model provides efficient storage and query of data, and provides a basis for subsequent intelligent reasoning and optimization decision-making; In the subsequent control strategy design, reinforcement learning technology was used to generate a reinforcement learning model, define the state space, action space and reward function, and use the knowledge graph model to provide control strategy guidance in the initial training stage of the reinforcement learning model, solving the inefficiency problem of traditional reinforcement learning relying on a large number of trial and error, thereby accelerating the convergence to the optimal control strategy and quickly obtaining a trained reinforcement learning model. Finally, when the model is deployed to the control system of the molding equipment, the reinforcement learning model and knowledge graph reasoning are deeply combined, so that the control system still has high flexibility and intelligence in the large-inertia dynamic flow balance and precise control scenarios of the molding equipment. On the one hand, it can optimize the control signal according to the real-time status of the equipment, and on the other hand, it can continuously optimize the control strategy through feedback, thereby significantly improving the accuracy and adaptability of the molding equipment operation.

[0010] Further, the S1 includes: S1-1: Collect real-time operation data and historical operation data of molding equipment; S1-2: Use the moving average operation to perform denoising and smoothing on the collected real-time running data. The expression is:

[0011] in, Indicates at time The moving average of Indicates the window size, Represents the time series data data points, Indicates the current moment; S1-3: The real-time running data that has been denoised and smoothed is normalized and scaled to a specific range In the expression:

[0012] in, represents the normalized data points, Represents a data point in real-time running data. Indicates the minimum value in real-time running data. Indicates the maximum value in real-time running data.

[0013] Beneficial effects: The real-time operation data of the collected molding equipment may contain noise, that is, random errors or meaningless fluctuations. These noises come from hardware failures, environmental changes or other factors, and may cause the equipment control system to make misleading decisions, thereby affecting the training and reasoning results of the model. Therefore, the moving average operation is used for denoising and smoothing, which can effectively reduce short-term fluctuations in the data, capture long-term trends, and ensure that the control signal of the equipment is more stable; and the normalization operation is aimed at the possible differences in the measurement units and magnitudes of the data source, which will cause certain features to dominate the model training, so the differences are eliminated through the normalization operation.

[0014] Further, the S2 includes: S2-1: Extract the historical operation parameters, standard production process parameters, case experience, physical laws, control theory, and manufacturing site processing records and experience notes from the pre-processed real-time operation data and historical operation data to obtain data features; S2-2: Call the NEO4J database, convert data features into a graph structure, and build a knowledge graph model based on entity nodes and relationships; the entity nodes represent key entities, and the relationships represent the relationships between entities through different types of edges.

[0015] Beneficial effects: By constructing a knowledge graph model and using the graph structure, the association between equipment status, control parameters and historical data can be efficiently stored and queried, supporting subsequent reasoning and decision-making.

[0016] Further, in said S2-2, the entity nodes include equipment nodes, sensor nodes, control signal nodes and historical data nodes; The relationship in S2-2 represents the relationship between entities through different types of edges: Equipment nodes and sensor nodes are connected through HAS relationships; The sensor nodes and control signal nodes are connected through the INFLUENCES relationship; The control signal node is connected to the historical data node through the BASED_ON relationship.

[0017] Beneficial effects: By using edges to represent the relationships between entities, the graph can comprehensively and accurately represent the complex relationships between equipment, sensors, control signals, and historical data, facilitating rapid query and reasoning in the subsequent reasoning process, automatically selecting appropriate control signals, and thus optimizing control strategies.

[0018] Further, the S3 includes: S3-1: Building a reinforcement learning model based on the Deep Q-Learning algorithm; S3-2: Determine the state space. Define the construction state space with the real-time operation data of the molding equipment. The expression is:

[0019] in, Indicates time status, , and Respectively represent various parameters of the real-time operation of the equipment; S3-3: Determine the action space, and define the action space with the control signal of the molding equipment. The expression is:

[0020] in, express The action of the moment, , , , Respectively represent each control signal parameter; S3-4: Determine a reward function, and define a reward function based on a preset molding equipment performance standard, wherein the reward function includes a positive reward and a negative reward; S3-5: Call the Deep Q-Learning algorithm to approximate the Q function, and process the state space and action space through the deep neural network to continuously optimize the Q value to generate the optimal control strategy for the equipment. The expression is:

[0021] in, Indicates in status Next action of value, Indicates immediate reward, represents the discount factor, Represents the target network parameters.

[0022] Beneficial effects: By constructing multi-dimensional states in the state space, the reinforcement learning model can fully capture the dynamic characteristics of equipment operation and make corresponding control decisions based on the current state; the definition of the action space can ensure that the reinforcement learning model can select the optimal control strategy under different working environments; positive and negative rewards are set in the reward function, so that the reinforcement learning model can adjust the control strategy according to real-time feedback and optimize the control effect of the equipment; the Deep Q-Learning algorithm approximates the Q function, so that the reinforcement learning model can continuously adjust the control signal to adapt to the dynamic changes of the equipment and ensure the optimized operation of the model.

[0023] Further, the S4 includes: S4-1: Initial training of the reinforcement learning model using historical operation data and control strategies in the knowledge graph; S4-2: The reinforcement learning model that has completed initial training interacts in real time with the control system of the large inertia dynamic flow balance and precise control scenario of the molding equipment. During the real-time interaction, the state is initialized according to the state space, the action is selected according to the current state of the molding equipment, and the optimization strategy is fed back through the reward function; S4-3: Call the Deep Q-Learning algorithm and The value is updated, The value update formula is:

[0024] in, Indicates the current status and actions Corresponding value, represents the immediate reward obtained from the environment, represents the discount factor, represents the target network parameters, Represents the learning rate.

[0025] Beneficial effects: The reinforcement learning model can be initially trained through historical data and control strategies in the graph reasoning of the knowledge graph model to quickly adapt to the real-time status and changing working conditions of the equipment, thereby avoiding the trial and error process in traditional reinforcement learning methods; and by continuously updating value, the reinforcement learning model can select the control signal that best suits the current state, thereby optimizing the control efficiency and response speed of the equipment.

[0026] Further, the S5 includes: S5-1: Deploy the knowledge graph model and the trained reinforcement learning model to the control system of the molding equipment. The reinforcement learning model generates the optimal control signal based on the real-time collected molding equipment operation data. The control system is controlled by The strategy selects the optimal control action, which is expressed as:

[0027] in, Indicates the current time Select the control action, Indicates the current equipment status. represents the target network parameters, The value represents the expected reward of the action in the current state; S5-2: Then, based on the knowledge graph model reasoning, the learning decision of the reinforcement learning model is supported to adjust the molding equipment control signal in real time; S5-3: Calculate the immediate reward based on the feedback of the formed equipment and update it through the Bellman equation The value is used to optimize the control signal, and the update equation is:

[0028] in, Indicates the current status and actions Corresponding value, represents the immediate reward obtained from the environment, represents the discount factor, represents the target network parameters, represents the learning rate, The maximum value of the next state value.

[0029] Beneficial effects: In the process of dynamic control of equipment, by dynamically generating control signals and adjusting key equipment parameters in real time, the operating efficiency and stability of the system under complex working conditions are guaranteed. The deep combination of reinforcement learning model and knowledge graph reasoning makes the system highly flexible and intelligent. It can not only optimize control signals according to the real-time status of equipment, but also continuously optimize control strategies through feedback, thereby significantly improving the accuracy and adaptability of equipment operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 is a flowchart of an embodiment of the present invention; Figure 2 is a logic block diagram of an embodiment of the present invention; Figure 3 Schematic diagram of the graph structure of the knowledge graph model in an embodiment of the present invention; Figure 4 Schematic diagram of the Deep Q-Learning algorithm flow in an embodiment of the present invention. DETAILED DESCRIPTION

[0031] The following is further described in detail through specific implementation methods: The embodiment is basically as shown in the attached Figure 1 and Figure 2 Shown: A large inertia dynamic flow balance and control method for forming equipment based on knowledge graph, including: S1: collecting real-time operation data and historical operation data of the molding equipment, and preprocessing the real-time operation data to generate preprocessed real-time operation data; wherein S1 includes: S1-1: Collect real-time operation data and historical operation data of molding equipment; S1-2: Use the moving average operation to perform denoising and smoothing on the collected real-time running data. The expression is:

[0032] in, Indicates at time The moving average of Indicates the window size, Represents the time series data data points, Indicates the current moment; S1-3: The real-time running data that has been denoised and smoothed is normalized and scaled to a specific range In the expression:

[0033] in, represents the normalized data points, Represents a data point in real-time running data. Indicates the minimum value in real-time running data. Indicates the maximum value in real-time running data.

[0034] In this embodiment, the real-time operation data of the equipment collected is the key parameters such as flow, pressure and temperature of the equipment monitored in real time by sensors, and the historical operation data collected at the same time includes historical records of key parameters, fault records and previous control strategies.

[0035] In the collection of real-time operation data, because of real-time monitoring through sensors, the sensor data may contain noise, that is, random errors or meaningless fluctuations. These noises may come from hardware failures, environmental changes or other factors, and may cause the equipment control system to make misleading decisions, thereby affecting the training and reasoning results of the model. Therefore, the use of moving average operations for denoising and smoothing can effectively reduce short-term fluctuations in the data, capture long-term trends, and ensure that the control signals of the equipment are more stable; and the normalization operation is aimed at the possible differences in the measurement units and magnitudes of the data source, which will cause certain features to dominate the model training, so the differences are eliminated through normalization operations.

[0036] S2: Based on the pre-processed real-time operation data and historical operation data, the NEO4J database is called to build a knowledge graph model based on entity nodes and relationships; S2 includes: S2-1: Extract the historical operation parameters, standard production process parameters, case experience, physical laws, control theory, and manufacturing site processing records and experience notes from the pre-processed real-time operation data and historical operation data to obtain data features; S2-2: Call the NEO4J database, convert data features into a graph structure, and build a knowledge graph model based on entity nodes and relationships; the entity nodes represent key entities, and the relationships represent the relationships between entities through different types of edges.

[0037] In this embodiment, the data used to construct the knowledge graph model includes historical operating parameters, standard production process parameters, physical laws, control theory, manufacturing thread processing records and experience notes. In addition, it also includes case experience in the actual production process. Case experience can avoid the difficulty of reasoning like case reasoning when facing new problems, and avoid the inability of rule reasoning to cope with abnormal situations.

[0038] Based on the above data features, the NEO4J database is called to build a knowledge graph model based on entity nodes and relationships, and then the data features are converted into a graph structure. Specifically, entity nodes represent key entities, such as equipment nodes, sensor nodes, control signal nodes, and historical data nodes. Relationships represent the relationships between entities through different types of edges, such as "collection", "influence", "based on", etc. Then, on the key entity nodes, the relationship is specifically manifested as: Equipment nodes and sensor nodes are connected through HAS relationships; The sensor nodes and control signal nodes are connected through the INFLUENCES relationship; The control signal node is connected to the historical data node through the BASED_ON relationship.

[0039] The graph structure of the generated knowledge graph model is as follows Figure 3 As shown, therefore, the above-mentioned graph structure design of this scheme can comprehensively and accurately represent the complex relationship between equipment, sensors, control signals and historical data, so that the system can quickly query and reason in the subsequent reasoning process, automatically select the appropriate control signal, and thus optimize the control strategy.

[0040] S3: constructing a reinforcement learning model, in which the state space is defined by the real-time operation data of the molding equipment, the action space is defined by the control signal of the molding equipment, and the reward function is defined by the preset molding equipment performance standard; wherein S3 includes: S3-1: Building a reinforcement learning model based on the Deep Q-Learning algorithm; S3-2: Determine the state space. Define the construction state space with the real-time operation data of the molding equipment. The expression is:

[0041] in, Indicates time status, , and Respectively represent the parameters of the real-time operation of the equipment; and in this embodiment, Indicates flow rate, Indicates pressure, Indicates temperature. In other embodiments of this embodiment, other equipment parameters may also be included.

[0042] S3-3: Determine the action space, and define the action space with the control signal of the molding equipment. The expression is:

[0043] in, express The action of the moment, , , , Respectively represent the control signal parameters; in this embodiment, Indicates the action of increasing flow. Indicates flow reduction action. Indicates an action of increasing pressure. It represents the action of reducing pressure. In addition, it also includes actions such as increasing temperature and decreasing temperature, which is not limited in the embodiments of the present application.

[0044] S3-4: Determine the reward function, and define the reward function based on the preset molding equipment performance standard. The reward function includes positive rewards and negative rewards. In this embodiment, the reward function is used to evaluate the system performance of the reinforcement learning model after taking a certain action in the current state. Therefore, the reward function guides the learning process of the reinforcement learning model by feeding back the system performance. Specifically, in this solution, the reward function is associated with the equipment performance, such as flow control accuracy, pressure stability, etc., and the corresponding reward function expression is:

[0045] Therefore, the above reward function expression includes positive rewards and negative rewards. Positive rewards indicate that the equipment operates well, while negative rewards indicate that the system state is unstable or the efficiency is low. Therefore, by designing positive rewards and negative rewards, the model can adjust the control strategy in real time and optimize the control effect of the equipment.

[0046] S3-5: Call the Deep Q-Learning algorithm to approximate the Q function, and process the state space and action space through the deep neural network to continuously optimize the Q value to generate the optimal control strategy for the equipment. The expression is:

[0047] in, Indicates in status Next action of value, Indicates immediate reward, represents the discount factor, Represents the target network parameters.

[0048] In this embodiment, the Deep Q-Learning algorithm flow chart is as follows: Figure 4 As shown in the figure, the Deep Q-Learning algorithm is used to enable the model to continuously adjust the control signal to adapt to the dynamic changes of the equipment and ensure the optimized operation of the system.

[0049] S4: The reinforcement learning model interacts with the large inertia dynamic flow balance and precise control scenario of the molding equipment, initializes the state according to the state space, selects the action according to the current state of the molding equipment, feeds back the optimization strategy through the reward function, and provides control strategy guidance for the initial stage of the training of the reinforcement learning model through the knowledge graph model, so as to train the reinforcement learning model to generate the optimal control signal and obtain a trained reinforcement learning model; wherein, S4 includes: S4-1: Initial training of the reinforcement learning model through historical operating data and control strategies in the knowledge graph; In this embodiment, the reinforcement learning model can be initially trained through historical data and control strategies in graph reasoning, quickly adapting to the real-time status and changing working conditions of the equipment, thereby avoiding the trial and error process in traditional reinforcement learning methods. The historical data and rules provided by graph reasoning help the intelligent agent avoid unnecessary exploration and quickly converge to the optimal control strategy, enhancing the adaptability and flexibility of the system.

[0050] S4-2: The reinforcement learning model that has completed initial training interacts in real time with the control system of the large-inertia dynamic flow balance and precise control scenario of the molding equipment. During the real-time interaction, the state is initialized according to the state space, the action is selected according to the current state of the molding equipment, and the optimization strategy is fed back through the reward function. In addition, in order to improve the learning effect and stability, the reinforcement learning model adopts experience replay and target network mechanisms to learn from historical experience and reduce the variance in the update process.

[0051] S4-3: Call the Deep Q-Learning algorithm and The value is updated, The value update formula is:

[0052] in, Indicates the current status and actions Corresponding value, represents the immediate reward obtained from the environment, represents the discount factor, represents the target network parameters, Represents the learning rate; by continuously updating value, the reinforcement learning model can select the control signal that best suits the current state, thereby optimizing the control efficiency and response speed of the equipment.

[0053] S5: Deploy the knowledge graph model and the trained reinforcement learning model to the control system of the molding equipment. The control system uses the knowledge graph model to infer the control strategy of the reinforcement learning model, and calls the reinforcement learning model to generate action control signals according to the inferred control strategy to control the operating state of the molding equipment. The reinforcement learning model then continuously optimizes the control strategy of the molding equipment according to the feedback from the molding equipment. Among them, S5 includes: S5-1: Deploy the knowledge graph model and the trained reinforcement learning model to the control system of the molding equipment. The reinforcement learning model generates the optimal control signal based on the real-time collected molding equipment operation data. The control system is controlled by The strategy selects the optimal control action, which is expressed as:

[0054] in, Indicates the current time Select the control action, Indicates the current equipment status. represents the target network parameters, The value represents the expected reward of the action in the current state; S5-2: Then, based on the knowledge graph model reasoning, the learning decision of the reinforcement learning model is supported to adjust the molding equipment control signal in real time; S5-3: Calculate the immediate reward based on the feedback of the formed equipment and update it through the Bellman equation The value is used to optimize the control signal, and the update equation is:

[0055] in, Indicates the current status and actions Corresponding value, represents the immediate reward obtained from the environment, represents the discount factor, represents the target network parameters, represents the learning rate, The maximum value of the next state value.

[0056] In this embodiment, the reinforcement learning model can be combined with historical data and reasoning rules in the knowledge graph to provide optimized decisions on control signals through graph reasoning. Specifically, based on historical equipment operation data and equipment performance, the graph reasoning engine can derive control strategies in advance under similar working conditions, thereby accelerating the training process of the reinforcement learning model and improving the convergence speed and stability of the model.

[0057] Therefore, in this solution, the knowledge graph-based large-inertia dynamic balancing and control method for forming equipment is proposed. With the help of intelligent means, it ensures the precise control of key parameters of forming equipment, ensuring that the equipment can maintain efficient and stable operation even under complex and changeable working conditions.

[0058] First, the data collection and preprocessing process is extremely sophisticated. The system captures the flow, pressure, temperature and other data covered by the equipment sensors in real time, as well as the precious historical operation data, and preprocesses them using professional means. On the one hand, the moving average method is used to cleverly eliminate short-term fluctuations, just like brushing off the dust on the surface of the data, making the data curve smoother; on the other hand, the normalization technology is used to resolve the troubles caused by the differences in units and magnitudes of different sensors, laying a solid and stable data foundation for subsequent modeling work, ensuring that every piece of data can "speak" accurately.

[0059] Secondly, the construction of knowledge graphs is of great significance. Based on the pre-processed data, the system builds a multi-level knowledge graph, just like weaving a large intelligent information network for equipment, closely linking equipment, sensors, control signals and historical data. Relying on the powerful NEO4J graph structure, integrating historical operating parameters, standard process parameters, physical laws and rich production experience, it not only realizes high-speed storage and convenient query of data, but more importantly, its intelligent reasoning and decision support functions are activated, injecting a "smart brain" into the system operation, and can provide scientific and reasonable decision-making guidance at critical moments.

[0060] Third, the reinforcement learning model is exquisitely designed. In order to achieve the ultimate goal of precise control, the system carefully builds a reinforcement learning model based on Deep Q-Learning. This model accurately defines the real-time status of the equipment, such as flow and pressure, as the state space, and the control signals, such as flow regulation and pressure regulation, become the action space. At the same time, the introduction of the reward function provides a clear optimization direction for improving the stability and efficiency of the equipment. During the training process, the model continuously refines the control strategy through deep interaction with the environment, and innovatively uses experience playback and target network mechanisms, which is like pressing the "accelerator" for the learning process, cleverly avoiding the time and resource waste caused by a large number of trial and error in traditional reinforcement learning.

[0061] Fourth, deployment and optimization work together. The trained reinforcement learning model is successfully installed in the control system, generating control signals in real time and accurately, and works perfectly with the knowledge graph reasoning function to dynamically adjust the operating status based on equipment feedback. Whether it is flow or pressure, the control signal can be flexibly changed according to the real-time needs of the equipment and continuously optimized, making the equipment like a well-trained athlete, always maintaining the best operating state.

[0062] In summary, the excellent dynamic control capability demonstrated by the system under complex working conditions is remarkable, which greatly improves the intelligence level of equipment, brings control accuracy to a new level, and significantly enhances operational flexibility. It undoubtedly has extremely high application and promotion value in the industrial field and is expected to reshape the new control pattern of molding equipment.

[0063] The above are only embodiments of the present invention. Common knowledge such as the known specific structures and characteristics in the scheme is not described in detail here. Ordinary technicians in the relevant field know all the common technical knowledge in the technical field to which the invention belongs before the application date or priority date, can obtain all the existing technologies in the field, and have the ability to apply conventional experimental means before that date. Ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the scope of protection of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A large inertia dynamic flow balance and control method for forming equipment based on knowledge graph, characterized by: include: S1: collect real-time operation data and historical operation data of the molding equipment, and pre-process the real-time operation data to generate pre-processed real-time operation data; S2: Based on the preprocessed real-time operation data and historical operation data, call the database to build a knowledge graph model based on entity nodes and relationships; S3: constructing a reinforcement learning model, in which the state space is defined by the real-time operation data of the molding equipment, the action space is defined by the control signal of the molding equipment, and the reward function is defined by the preset molding equipment performance standard; S4: The reinforcement learning model interacts with the large inertia dynamic flow balance and precise control scenario of the molding equipment, initializes the state according to the state space, selects actions according to the current state of the molding equipment, optimizes the strategy through reward function feedback, and provides control strategy guidance for the initial stage of the reinforcement learning model training through the knowledge graph model, so as to train the reinforcement learning model to generate the optimal control signal and obtain a trained reinforcement learning model; S5: Deploy the knowledge graph model and the trained reinforcement learning model to the control system of the molding equipment. The control system uses the knowledge graph model to infer the control strategy of the reinforcement learning model, and calls the reinforcement learning model to generate action control signals according to the inferred control strategy to control the operating state of the molding equipment. The reinforcement learning model then continuously optimizes the control strategy of the molding equipment based on the feedback from the molding equipment.

2. The method for balancing and controlling large inertia dynamic flow of forming equipment based on knowledge graph according to claim 1 is characterized in that: The S1 includes: S1-1: Collect real-time operation data and historical operation data of molding equipment; S1-2: Use the moving average operation to perform denoising and smoothing on the collected real-time running data. The expression is: in, Indicates at time The moving average of Indicates the window size, Represents the time series data data points, Indicates the current moment; S1-3: The real-time running data that has been denoised and smoothed is normalized and scaled to a specific range In the expression: in, represents the normalized data points, Represents a data point in real-time running data. Indicates the minimum value in real-time running data. Indicates the maximum value in real-time running data.

3. The large inertia dynamic flow balance and control method of forming equipment based on knowledge graph according to claim 2 is characterized by: The S2 includes: S2-1: Extract the historical operation parameters, standard production process parameters, case experience, physical laws, control theory, and manufacturing site processing records and experience notes from the pre-processed real-time operation data and historical operation data to obtain data features; S2-2: Call the NEO4J database, convert data features into a graph structure, and build a knowledge graph model based on entity nodes and relationships; where entity nodes represent key entities, and relationships represent the relationships between entities through different types of edges.

4. The large inertia dynamic flow balance and control method of forming equipment based on knowledge graph according to claim 3 is characterized by: In said S2-2, the entity nodes include equipment nodes, sensor nodes, control signal nodes and historical data nodes; The relationship in S2-2 represents the relationship between entities through different types of edges: Equipment nodes and sensor nodes are connected through HAS relationships; The sensor nodes and control signal nodes are connected through the INFLUENCES relationship; The control signal node is connected to the historical data node through the BASED_ON relationship.

5. The method for balancing and controlling large inertia dynamic flow of forming equipment based on knowledge graph according to claim 4 is characterized in that: The S3 includes: S3-1: Building a reinforcement learning model based on the Deep Q-Learning algorithm; S3-2: Determine the state space. Define the construction state space with the real-time operation data of the molding equipment. The expression is: in, Indicates time status, , and They respectively represent the parameters of the real-time operation of the equipment; S3-3: Determine the action space, and define the action space with the control signal of the molding equipment. The expression is: in, express The action of the moment, , , , Respectively represent each control signal parameter; S3-4: Determine a reward function, and define a reward function based on a preset molding equipment performance standard, wherein the reward function includes a positive reward and a negative reward; S3-5: Call the Deep Q-Learning algorithm to approximate the Q function, and process the state space and action space through the deep neural network to continuously optimize the Q value to generate the optimal control strategy for the equipment. The expression is: in, Indicates in status Next action of value, Indicates immediate reward, represents the discount factor, Represents the target network parameters.

6. The knowledge graph-based large inertia dynamic flow balance and control method for forming equipment according to claim 5 is characterized by: The S4 includes: S4-1: Initial training of the reinforcement learning model using historical operation data and control strategies in the knowledge graph; S4-2: The reinforcement learning model that has completed initial training interacts in real time with the control system of the large inertia dynamic flow balance and precise control scenario of the molding equipment. During the real-time interaction, the state is initialized according to the state space, the action is selected according to the current state of the molding equipment, and the optimization strategy is fed back through the reward function; S4-3: Call the Deep Q-Learning algorithm and The value is updated, The value update formula is: in, Indicates the current status and actions Corresponding value, represents the immediate reward obtained from the environment, represents the discount factor, represents the target network parameters, Represents the learning rate.

7. The knowledge graph-based large inertia dynamic flow balance and control method for forming equipment according to claim 6 is characterized by: The S5 includes: S5-1: Deploy the knowledge graph model and the trained reinforcement learning model to the control system of the molding equipment. The reinforcement learning model generates the optimal control signal based on the real-time collected molding equipment operation data. The control system is controlled by The strategy selects the optimal control action, expressed as: in, Indicates the current time Select the control action, Indicates the current equipment status. represents the target network parameters, The value represents the expected reward of the action in the current state; S5-2: Then, based on the knowledge graph model reasoning, the learning decision of the reinforcement learning model is supported to adjust the molding equipment control signal in real time; S5-3: Calculate the immediate reward based on the feedback of the formed equipment and update it through the Bellman equation The value is used to optimize the control signal, and the update equation is: in, Indicates the current status and actions Corresponding value, represents the immediate reward obtained from the environment, represents the discount factor, represents the target network parameters, represents the learning rate, The maximum value of the next state value.

Images