Self-adaptive energy consumption optimization method and system for vacuum isothermal forging system production line

Abstract

The invention relates to the field of manufacturing industry, in particular to a self-adaptive energy consumption optimization method and system for a vacuum isothermal forging system production line. The method comprises the following steps: acquiring real-time data of production line equipment, and generating a real-time data sequence; preprocessing, denoising and standardizing the real-time data sequence to obtain clear data for feature extraction; key features are extracted by using an AI model and are trained, and a pre-trained energy efficiency prediction model is generated; and inputting the real-time data into a pre-trained model, generating an optimization decision, adjusting equipment operation parameters, and forming a feedback control system. Through real-time data acquisition and AI optimization, dynamic optimization of energy efficiency is realized, production efficiency is improved, energy consumption is reduced, system adaptability is enhanced, and green manufacturing requirements are met.

Description

Technical Field

[0001] The present invention relates to the field of manufacturing industry, and in particular to a self-adaptive energy consumption optimization method and system for a vacuum isothermal forging system production line. Background Art

[0002] As an efficient and high-precision metal hot working process, vacuum isothermal forging is widely used in the fields of aviation, aerospace, and gas turbines. However, the vacuum isothermal forging process is usually accompanied by high energy consumption, mainly concentrated on the energy consumption of key equipment such as heating furnaces and forging presses. In the prior art, the energy consumption management of the production line mainly relies on fixed parameter settings and static optimization methods. This method is difficult to cope with the real-time changing needs and environmental conditions in the production process, resulting in energy waste and low production efficiency.

[0003] Specifically, traditional energy consumption optimization methods usually make rough adjustments to equipment energy consumption based on historical data or preset process parameters, and lack the ability to accurately monitor and dynamically respond to the real-time operating status of the production line. For example, during the production process, equipment parameters such as load and temperature will fluctuate with changes in production rhythm and workpiece characteristics, but it is difficult for existing systems to capture the changes in real time and make corresponding energy consumption adjustments. This not only affects the effective use of energy, but may also lead to reduced production efficiency and unstable product quality.

[0004] In addition, with the development of Industry 4.0 and intelligent manufacturing, the demand for intelligent, automated and efficient energy management in production lines is increasing. The existing energy management system has significant deficiencies in intelligence and adaptive capabilities, and it is difficult to meet the requirements of modern manufacturing for green and sustainable development. Therefore, there is an urgent need for a method and system that can dynamically adjust the energy consumption of equipment based on real-time data of production equipment using artificial intelligence technology to achieve adaptive energy consumption optimization of the vacuum isothermal forging system production line, so as to improve energy utilization efficiency, reduce production costs, enhance the intelligence level of the production line, and meet the needs of modern manufacturing for efficient and green production. Summary of the invention

[0005] The present invention provides an adaptive energy consumption optimization method and system for a vacuum isothermal forging system production line, so as to solve the problem of how to dynamically adjust the energy consumption of equipment based on the real-time data of production equipment by using an AI model to achieve adaptive energy consumption optimization of the vacuum isothermal forging system production line.

[0006] In order to solve the above technical problems, the present invention provides an adaptive energy consumption optimization method for a vacuum isothermal forging system production line, comprising: Obtain real-time data from production line equipment and generate real-time data sequences; Preprocessing the real-time data sequence, performing denoising and standardization to obtain clear data for feature extraction; Based on the clear data, the AI ​​model is used to extract key features and perform training to generate a pre-trained energy efficiency prediction model; The real-time data is input into the pre-trained energy efficiency prediction model to generate optimization decisions and adjust the operating parameters of the production line equipment to form a feedback control system.

[0007] Furthermore, the acquisition of real-time data of production line equipment also includes: Obtain the temperature (T), power (P), workload (L) and equipment status (S) data of production line equipment, and collect the data in real time through sensors and monitoring systems; The data is transmitted to a data acquisition terminal via an industrial control network (such as a SCADA system).

[0008] Furthermore, the preprocessing of the real-time data sequence further includes: De-noising the real-time data sequence, using a filtering algorithm to eliminate sensor noise and external interference; The denoised data is standardized to adjust the data to a uniform scale range.

[0009] Furthermore, the generating of the pre-trained energy efficiency prediction model further includes: Extract key features such as energy consumption patterns and equipment operation rules from the clear data; The key features are used to train the AI ​​model and optimize the model parameters.

[0010] Furthermore, the generating of optimization decisions and adjusting the operating parameters of the production line equipment to form a feedback control system also includes: Based on the optimization decision, determining optimal operating parameters of the production line equipment; The optimal operating parameters are fed back to the production line control system to adjust the operating status of the equipment.

[0011] Furthermore, the real-time data is collected through temperature sensors, current sensors, pressure sensors, etc.

[0012] Furthermore, the denoising process uses a low-pass filter or a Kalman filter algorithm.

[0013] Furthermore, the key features include statistical features such as the mean, standard deviation, maximum value, and minimum value of the data.

[0014] Furthermore, the optimization decision is generated through linear programming, genetic algorithm and particle swarm optimization algorithm.

[0015] Furthermore, an adaptive energy consumption optimization system for a vacuum isothermal forging system production line comprises: A data acquisition module is used to obtain real-time data of production line equipment, including temperature (T), power (P), workload (L) and equipment status (S); A data processing module, used for preprocessing the real-time data, wherein the preprocessing includes denoising and standardization; A feature extraction and model training module, used to extract key features based on the preprocessed clear data and train using an AI model to generate a pre-trained energy efficiency prediction model; The optimization decision and feedback control module is used to input the real-time data into the pre-trained energy efficiency prediction model, generate optimization decisions and adjust the operating parameters of the production line equipment to form a feedback control system.

[0016] The key innovations of the present invention are: (1) Integration of real-time data and AI models: The present invention combines the real-time data collection of production line equipment with the AI ​​energy efficiency prediction model. By inputting the pre-processed clear data into the AI ​​model, accurate prediction of energy efficiency and optimized decision-making can be achieved, significantly improving the intelligent level of energy consumption management.

[0017] (2) Closed-loop feedback control system: By feeding back the optimization decision to the production line control system, the equipment operating parameters are dynamically adjusted to form a closed-loop control with continuous optimization. This mechanism ensures the real-time optimization of the production line energy consumption and the system’s adaptive ability to quickly respond to changes in the production process.

[0018] (3) Pre-trained energy efficiency prediction model: Use historical data and real-time data to train the AI ​​model to generate a highly accurate pre-trained energy efficiency prediction model. This model can predict energy efficiency levels in real time, provide a reliable basis for optimization decisions, and ensure that the energy consumption of the production line is optimized under different working conditions.

[0019] The following are its main beneficial effects: (1) Accurate energy efficiency optimization: By acquiring key data such as temperature, power, workload and equipment status of production line equipment in real time, and performing denoising and standardization processing, the accuracy and reliability of the data can be ensured. Using the pre-trained AI energy efficiency prediction model, the energy efficiency level under the current production conditions can be accurately predicted, and optimization decisions can be generated, significantly improving the accuracy of energy consumption management.

[0020] (2) Strong system adaptability: The present invention forms a closed-loop feedback control system by inputting real-time data into a pre-trained model to dynamically adjust the operating parameters of the production line equipment. The system can respond to changes in the production process in real time, automatically optimize the operating status of the equipment, improve the flexibility and adaptability of the production line, and adapt to changes in different production needs and environmental conditions.

[0021] (3) Improvement of production efficiency and quality: While optimizing energy efficiency, precise parameter adjustment is performed to ensure that production line equipment operates in the best condition, thereby improving production efficiency and product quality. This reduces production instability caused by energy consumption fluctuations and ensures product consistency and high quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A schematic flow chart of an adaptive energy consumption optimization method for a vacuum isothermal forging system production line provided in an embodiment of the present application; Figure 2 A structural block diagram of an adaptive energy consumption optimization system for a vacuum isothermal forging system production line provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of this application; the terms used in the specification of the application herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "" and "having" in the specification and claims of this application and the above-mentioned drawings and any variations thereof are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of this application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0024] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0025] Example 1: Reference Figure 1 , is a flow chart of an adaptive energy consumption optimization method for a vacuum isothermal forging system production line provided by an embodiment of the present invention. The flow chart includes at least steps S100-S400: S100, acquiring real-time data of production line equipment and generating a real-time data sequence; S200, preprocessing the real-time data sequence, performing denoising and standardization to obtain clear data for feature extraction; S300, based on the clear data, using an AI model to extract key features and perform training to generate a pre-trained energy efficiency prediction model; S400, inputting real-time data into the pre-trained energy efficiency prediction model, generating optimization decisions and adjusting operating parameters of production line equipment to form a feedback control system.

[0026] Step S100 at least includes steps S110-S130: S110: Obtain real-time data of production line equipment and collect data in real time through sensors and monitoring systems.

[0027] Specifically, first, the key parameters of the production line equipment are collected in real time through the sensors and monitoring systems installed on the equipment. The parameters are: Temperature (T): real-time temperature data of each component (heating furnace, forging die, etc.) during equipment operation. Power (P): real-time power consumption data of production line equipment (motor, heating system, etc.). Workload (L): data reflecting the current load level of the equipment, the load of the robot arm during processing, the load data fed back by the pressure sensor, etc. Equipment status (S): data reflecting the operating status of the production line equipment, such as the start and stop status of the equipment, fault alarm information, etc.

[0028] The data is collected by sensors (temperature sensors, current sensors, pressure sensors, etc.) in various links of the production line. Specifically, the sensors monitor the operation of production equipment in real time and transmit the data to the data acquisition terminal through the industrial control network (such as SCADA system). The data collected by the sensors are sent to the data processing system in real time through interfaces and protocols (such as Modbus, OPC, etc.) for further processing.

[0029] The real-time data provides basic data support for subsequent energy efficiency analysis and optimization decisions, and is transmitted to the data processing module as input data.

[0030] S120: The collected real-time data is processed into a time series and organized into a data sequence in chronological order.

[0031] Furthermore, the real-time data collected by S110 is processed in a time series. Specifically, the collected data does not exist independently on the time axis, but is a dynamic process that changes over time. Therefore, it is necessary to sort the data according to the time sequence of collection, and combine the collected data at different time points into a time series for subsequent analysis.

[0032] Time series ( ): Organize all collected data (such as temperature T, power P, load L, etc.) in chronological order to generate an ordered data set that represents the change of device status over time. For example, assuming that data is collected once per second, the temperature at a certain time t can be expressed as , the power is , the workload is After sorting all collected data, a time series is obtained: ={( , , ),( , , ),…,( , , )} in, Represents the status data of the device at time point t, including temperature, power, load and other data.

[0033] After time series conversion, the data series can provide structured input data for subsequent feature extraction and AI model training. By organizing the time series, it can be ensured that the time variation of the equipment operation status is taken into account in the subsequent data analysis and model training process.

[0034] S130: The sorted data sequence is transmitted to the data processing module as input for subsequent preprocessing and feature extraction.

[0035] In this step, the time-series data ( ) is transmitted to the data processing module for subsequent preprocessing and feature extraction. Specifically, the sorted data sequence will be used as input data and enter the data processing module. Input data sequence ( ) will be sent to the data processing module for further cleaning, denoising, standardization and other operations to ensure the quality and consistency of the data.

[0036] The data processing module will select the appropriate denoising algorithm (such as mean filtering, Kalman filtering, etc.) according to actual needs, smooth the data, and eliminate the noise caused by sensor errors or external environmental changes. In the data standardization step, all input data will be converted to the same standard range to ensure that the data distribution of each feature is consistent. For example, all data can be transformed into a distribution with a mean of 0 and a standard deviation of 1 through Z-score standardization.

[0037] Connection description: The real-time data obtained in step S110 provides the most basic original input for subsequent data processing and analysis. The data is sorted in chronological order in step S120 to ensure that the dynamic changes in equipment operation can be reflected in the analysis.

[0038] The time series data organized in S120 will be passed as input to S130 and enter the data processing module for preprocessing and feature extraction to ensure data quality and applicability, providing data support for AI model training and energy efficiency optimization.

[0039] This data transmission and processing process ensures seamless connection between each step, and provides clear and effective input data for subsequent AI model training and optimization decisions.

[0040] Step S200 at least includes steps S210-S230: S210: De-noising the real-time data sequence, using a filtering algorithm to eliminate sensor noise and external interference.

[0041] First, the real-time data sequence collected by S100 is denoised. Specifically, since the data collection process is affected by factors such as sensor errors, environmental changes, and external interference, the raw data needs to be filtered to eliminate noise and ensure the purity of the data. Denoising can be done by using a filtering algorithm, for example: Low-pass filter (LPF): used to remove high-frequency noise signals, especially high-frequency interference signals that occur during equipment operation. The filtered data can be obtained by the formula: Where D(t) is the original data sequence, h(t−τ) is the impulse response of the filter, (t) is the data after low-pass filtering.

[0042] Kalman filter: By combining the measured data and the state model of the system, the estimated value of the data is optimized, which is suitable for processing dynamic system data with noise. Kalman filter can be implemented by the following recursive formula: = + ( −H ) in, is the estimated value at the current moment, is the Kalman gain, is the measurement value, and H is the measurement matrix.

[0043] After denoising, the obtained data sequence is clearer, the interference caused by external noise and sensor errors is removed, and the reliability of the data is ensured.

[0044] S220: performing standardization processing on the denoised data to adjust the data to a uniform scale range.

[0045] Furthermore, after denoising, the obtained clear data may have different dimensions (for example, temperature in degrees Celsius, power in watts, load in kilograms, etc.), which will affect subsequent data analysis and feature extraction. Therefore, the denoised data needs to be standardized to eliminate the impact of the dimension so that different feature data can be in the same scale range.

[0046] Specifically, the Z-score standardization method can be used to process the data. This method eliminates the scale differences between features by converting the data into a standard normal distribution with a mean of 0 and a standard deviation of 1. The standardization formula is as follows: = in, is the original data value; is the mean of the data; σ is the standard deviation of the data; is the standardized data value.

[0047] During the standardization process, all data features such as temperature, power, load, etc. will be converted into unitless standard scores to ensure that they have the same dimension, laying the foundation for subsequent feature extraction and AI model training.

[0048] S230: Send the standardized data to the feature extraction module to generate a clear data sequence for subsequent analysis and training of the AI ​​model.

[0049] After the normalization process, the resulting data will be passed as input to the feature extraction module. The task of the feature extraction module is to extract features that can effectively reflect the energy efficiency status of the equipment from the normalized data and generate feature data for AI model training.

[0050] Specifically, feature extraction can be performed in the following steps: Time domain feature extraction: For example, extracting statistical features such as maximum value, minimum value, mean value, standard deviation, etc. from data series such as temperature, power, and load; Frequency domain feature extraction: Perform Fourier transform on the standardized time series data to extract frequency features in the data, such as power spectrum density; Sliding window method: By performing sliding window processing on the data, the feature values ​​within each window are calculated to generate multiple subsequence features for further analysis.

[0051] Assume that the statistical features extracted from the three features of temperature, power and load are: ={ , , , } ={ , , , } ={ , , , } Among them, μ and σ are the mean and standard deviation respectively, max and min are the maximum and minimum values ​​respectively, and T, P, and L represent the characteristics of temperature, power, and load.

[0052] After the features are extracted, a new feature data set is generated, which becomes the input data for AI model training and is further used for the model's energy efficiency prediction and production line optimization decisions.

[0053] Description of the connection between the previous and next steps: In S210, the original data is subjected to denoising processing by a filtering algorithm to obtain a pure signal, and the denoised data is passed as input to S220 for standardization processing.

[0054] In S220 , the denoised data is standardized to eliminate dimensional differences so that different features can be compared within a uniform scale range. The standardized data is then transferred to S230 for feature extraction.

[0055] The feature data extracted from S230 will become the input of the AI ​​model and provided to subsequent AI models for training, further promoting the energy efficiency optimization of the system.

[0056] Step S300 at least includes steps S310-S330: S310: Based on clear data sequences, key feature parameters are extracted, such as energy consumption patterns, equipment operation rules, etc., as inputs to the AI ​​model.

[0057] In this step, firstly, based on the clear data obtained in step S200, key features related to energy efficiency are extracted. Specifically, clear data refers to relatively pure production line equipment data, temperature, power, load and other real-time monitoring data that have eliminated dimension effects after denoising, standardization and feature extraction. In order to effectively improve the accuracy of energy efficiency prediction, it is necessary to extract key features that can reflect the energy efficiency of the production line from the data.

[0058] The key features are: Energy consumption pattern: Energy consumption trend information extracted based on power and load changes.

[0059] Equipment operation rules: Regular features extracted through statistical analysis of equipment operation status (such as on / off cycles, load fluctuations, etc.).

[0060] Time series characteristics: statistical characteristics of the data such as mean, standard deviation, maximum value, minimum value, and frequency domain characteristics such as power spectral density.

[0061] For example, from power ( ) and load ( ) The features extracted from the data sequence can be expressed as: ={ , , , } ={ , , , } in, and are the mean values ​​of power and load respectively, and is the standard deviation of power and load, and is the maximum value of power and load, and is the minimum value of power and load.

[0062] The features serve as input to the AI ​​model and provide key information for model training.

[0063] S320: Use historical data and clear data to train the AI ​​model and optimize model parameters.

[0064] In this step, the AI ​​model is trained using the clean data extracted from S200 and S310 and historical data. Historical data contains various equipment data and energy efficiency performance during the past operation of the production line, and is an important basis for training the AI ​​model. Specifically, by inputting historical data and clean data into the AI ​​model, the model learns the regularity in the data to optimize its parameters and improve the accuracy of energy efficiency prediction.

[0065] During the training process, supervised learning algorithms are usually used to optimize the model parameters by minimizing the error between the predicted value and the true value. The commonly used loss function is the mean square error (MSE): in, is the real energy efficiency value; Based on the input features and the energy efficiency value predicted by the model parameters θ; N is the number of training samples.

[0066] The model parameters θ are continuously adjusted through the back-propagation algorithm until the loss function L(θ) is minimized, thereby improving the prediction accuracy of the model. At this point, the AI ​​model has been trained with historical data and clear data, and can make relatively accurate predictions on the energy efficiency of production line equipment, and is ready to make real-time optimization decisions.

[0067] S330: Generate a trained energy efficiency prediction model and output model parameters that can be used for real-time optimization decision-making.

[0068] After the model training is completed, the generated AI energy efficiency prediction model can be used to optimize the energy efficiency management of the production line in real time. Specifically, the task of this step is to apply the trained AI model to the real-time data stream to generate model parameters for production line optimization decisions. The trained and optimized model will have a high prediction accuracy, can accurately predict the energy efficiency of production line equipment, and adjust the production process according to the prediction results.

[0069] For example, when real-time data streams are input into a trained AI model, the model generates energy efficiency prediction results based on the input real-time data. , which is the predicted energy efficiency value of the current production line. Based on this predicted value, the system can further generate optimization decisions and adjust the operating parameters of the production line equipment, such as power and load, through a feedback mechanism to achieve energy efficiency optimization.

[0070] The specific output process can be expressed as: =f( ; ) in, For real-time data input, are the model parameters after training and optimization, is the predicted energy efficiency value.

[0071] Energy efficiency prediction results output by the model It will be used to adjust the energy efficiency configuration of the production line in real time.

[0072] Explanation of the connection between the previous and next steps: The feature data extracted in S310, as the input of the AI ​​model, will be trained together with the historical data in S320 to optimize the model parameters.

[0073] The AI ​​model trained and optimized in S320 improves the accuracy of energy efficiency prediction by minimizing the loss function. After training, the model is used in S330 to generate accurate energy efficiency prediction results and output model parameters for real-time optimization decision-making.

[0074] Step S400 at least includes steps S410-S430: S410: Input the real-time collected production line data into the pre-trained energy efficiency prediction model to generate energy efficiency prediction results for the current production status.

[0075] First, the real-time collected production line data (temperature, power, equipment load, etc.) will be used as input to the pre-trained AI model that has been trained previously. (For example, the power at the current moment ,temperature ,load ) is combined with the features extracted in the previous training (such as energy consumption patterns, equipment operation rules, etc.) as the input of the AI ​​model. Based on the input data, the AI ​​model generates energy efficiency prediction results for the current production status.

[0076] The prediction results It indicates the predicted energy efficiency level, which can accurately reflect the energy usage of the current equipment. Its calculation formula is: =f( ; ) in, is the predicted energy efficiency value at the current time t; Real-time collection of production data, including temperature, power, load, etc. The AI ​​model parameters have been optimized through training. It will serve as the basis for subsequent optimization decisions.

[0077] S420: Generate optimization decisions based on the prediction results and production requirements to determine the optimal parameters for equipment operation.

[0078] In this step, based on the energy efficiency prediction result generated in S410 , combined with actual production requirements (such as production efficiency, product quality, energy consumption limit, etc.), to generate optimization decisions. The core task of this decision is to determine the optimal operating parameters of the production line equipment based on the predicted energy efficiency level and the set production requirements.

[0079] Specifically, the optimization decision is generated through the following steps: Determine the objective function: Set an objective function J(t) that takes into account factors such as production efficiency, product quality, and energy consumption. For example, the objective function can be expressed as: J(t)= · + · + · in, Energy efficiency prediction results ; Production efficiency, usually determined by factors such as production rate and product quality; for energy costs or production costs; , , is the weight coefficient, which is used to balance various objectives.

[0080] Optimization solution: Based on the set objective function J(t), use optimization algorithms (such as linear programming, genetic algorithm, particle swarm optimization, etc.) to determine the optimal equipment operating parameters, such as power ,temperature ,load The goal of the optimization process is to minimize or maximize the objective function to ensure that the production line can achieve the optimal energy efficiency level while meeting production needs.

[0081] Generate optimized parameters: Through the optimization process, the optimal parameters for equipment operation are obtained. For example, suppose the optimal power is obtained through the optimization algorithm and temperature ,but: =argmin(J(t)) =argmin(J(t)) The optimal parameters will serve as decision input in the feedback control system.

[0082] S430: Feedback the optimization decision to the production line control system, adjust the equipment operating parameters, and continuously optimize the model based on the feedback information to form a dynamic feedback closed-loop control.

[0083] In this step, the decision results (such as the optimal power Optimal temperature The real-time feedback will be used to optimize the adjustment process and form a dynamic feedback closed-loop control system through continuous monitoring and data collection.

[0084] Specifically: Execution of the feedback control system: The generated optimization decision is passed to the equipment through the production line control system to adjust the operating parameters of the equipment. Sent to the power regulator Sent to the temperature control system to ensure that the operating status of the equipment is consistent with the optimized parameters.

[0085] Continuous data collection and feedback: The data after equipment operation adjustment (such as new power, load, temperature, etc.) will be collected in real time and returned to the AI ​​model as new input. At this time, the system continuously optimizes the AI ​​model based on the new data feedback to further improve the accuracy of prediction and decision-making.

[0086] Model update and iteration: Based on new feedback data, the parameters of the AI ​​model are adjusted. Continuous optimization can be achieved through incremental learning, online learning, etc. For example, the training set can be updated with new real-time data, and the model can be retrained regularly to enable it to adapt to changes in the production line environment, thus forming a dynamic feedback control closed loop.

[0087] Description of the connection between the previous and next steps: In S410, after the real-time data is input into the pre-trained model, the energy efficiency prediction results generated This provides a basis for the optimization decision in S420. The optimization decision generates the optimal equipment operating parameters according to the prediction results and production requirements.

[0088] The optimal operating parameters generated in S420 (such as and ) will be fed back to the production line control system, and the operating status of the equipment will be actually adjusted in S430, and the AI ​​model will be continuously optimized through real-time feedback to form a closed-loop control. Each step is closely connected to ensure the continuity and efficiency of data flow and decision-making process.

[0089] The key innovations of the present invention are: (1) Integration of real-time data and AI models: The present invention combines the real-time data collection of production line equipment with the AI ​​energy efficiency prediction model. By inputting the pre-processed clear data into the AI ​​model, accurate prediction of energy efficiency and optimized decision-making can be achieved, significantly improving the intelligent level of energy consumption management.

[0090] (2) Closed-loop feedback control system: By feeding back the optimization decision to the production line control system, the equipment operating parameters are dynamically adjusted to form a closed-loop control with continuous optimization. This mechanism ensures the real-time optimization of the production line energy consumption and the system’s adaptive ability to quickly respond to changes in the production process.

[0091] (3) Pre-trained energy efficiency prediction model: Use historical data and real-time data to train the AI ​​model to generate a highly accurate pre-trained energy efficiency prediction model. This model can predict energy efficiency levels in real time, provide a reliable basis for optimization decisions, and ensure that the energy consumption of the production line is optimized under different working conditions.

[0092] The following are its main beneficial effects: (1) Accurate energy efficiency optimization: By acquiring key data such as temperature, power, workload and equipment status of production line equipment in real time, and performing denoising and standardization processing, the accuracy and reliability of the data can be ensured. Using the pre-trained AI energy efficiency prediction model, the energy efficiency level under the current production conditions can be accurately predicted, and optimization decisions can be generated, significantly improving the accuracy of energy consumption management.

[0093] (2) Strong system adaptability: The present invention forms a closed-loop feedback control system by inputting real-time data into a pre-trained model to dynamically adjust the operating parameters of the production line equipment. The system can respond to changes in the production process in real time, automatically optimize the operating status of the equipment, improve the flexibility and adaptability of the production line, and adapt to changes in different production needs and environmental conditions.

[0094] (3) Improvement of production efficiency and quality: While optimizing energy efficiency, precise parameter adjustment is performed to ensure that production line equipment operates in the best condition, thereby improving production efficiency and product quality. This reduces production instability caused by energy consumption fluctuations and ensures product consistency and high quality.

[0095] Embodiment 2: Figure 2 FIG. 2 shows a block diagram of a self-adaptive energy consumption optimization system for a vacuum isothermal forging system production line according to an embodiment of the present invention. Figure 2 As shown, the system can: The data acquisition module 10 is used to obtain real-time data from production line equipment, including temperature (T), power (P), workload (L) and equipment status (S). Through high-precision sensors and monitoring systems installed in key locations, key parameters of the equipment operation process are collected in real time. The sensors transmit data to the data acquisition terminal through an industrial control network (such as a SCADA system) to ensure the extensiveness and effectiveness of the data.

[0096] The data processing module 20 is used to pre-process the collected real-time data sequence, including denoising and standardization. First, the filtering algorithm is used to eliminate sensor noise and external interference to ensure the purity of the data; then, the denoised data is adjusted to a uniform scale range to eliminate the influence of the dimension and make it suitable for further feature extraction. The processed clear data sequence is transmitted to the feature extraction module for subsequent analysis and training of the AI ​​model.

[0097] The feature extraction and model training module 30 uses the AI ​​model to extract key features and train based on clear data to generate a pre-trained energy efficiency prediction model. Specifically, key features such as energy consumption patterns and equipment operation rules are extracted from the clear data sequence as input to the AI ​​model. The AI ​​model is trained in combination with historical data to optimize model parameters to ensure high accuracy of the model in energy efficiency prediction. After training, the generated energy efficiency prediction model can be used for real-time optimization decision-making.

[0098] The optimization decision and feedback control module 40 inputs real-time data into the pre-trained energy efficiency prediction model, generates optimization decisions and adjusts the operating parameters of the production line equipment to form a feedback control system. Specifically, the real-time collected production line data is input into the pre-trained AI model to generate energy efficiency prediction results for the current production status; based on the prediction results and production requirements, an optimization decision is generated to determine the optimal parameters for equipment operation, such as temperature, power, etc.; the optimization decision is fed back to the production line control system, the equipment operating parameters are adjusted, and the model is continuously optimized based on the feedback information to form a dynamic feedback closed-loop control.

[0099] The adaptive energy consumption optimization system of the vacuum isothermal forging system production line of the embodiment of the present invention realizes the dynamic optimization of the energy consumption of the production line through the coordinated work of modules such as data collection, data processing, feature extraction and model training, and optimization decision and feedback control. The specific beneficial effects are as follows: (1) Accurate energy efficiency optimization: By collecting and processing key data in real time and combining it with accurate predictions from AI models, efficient optimization of production line energy consumption is achieved, ensuring optimal energy use.

[0100] (2) Strong system adaptability: The system has dynamic adjustment capabilities and can respond to changes in the production environment in real time, automatically optimize equipment operating parameters, and improve the flexibility and adaptability of the production line.

[0101] (3) Improved production efficiency and quality: While optimizing energy consumption, the system ensures that production efficiency and product quality are not affected, and in some cases are even improved to meet high production standards.

[0102] (4) Energy conservation and emission reduction: effectively reducing energy consumption of the production line, complying with the requirements of green manufacturing and sustainable development, reducing environmental impact, and enhancing the corporate social responsibility image.

[0103] (5) Intelligent management: The introduction of AI technology has improved the intelligence level of the production process, reduced human intervention, and improved management efficiency and accuracy.

[0104] Obviously, the embodiments described above are only some embodiments of the present application, rather than all embodiments. The preferred embodiments of the present application are given in the accompanying drawings, but they do not limit the patent scope of the present application. The present application can be implemented in many different forms. On the contrary, the purpose of providing the embodiments is to make the understanding of the disclosure of the present application more thorough and comprehensive. Although the present application is described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific implementation methods, or to replace some of the technical features therein with equivalents. Any equivalent structure made using the contents of the specification and drawings of this application, directly or indirectly used in other related technical fields, is similarly within the scope of patent protection of this application.

Claims

1. An adaptive energy consumption optimization method for a vacuum isothermal forging system production line, characterized in that: The steps include: Obtain real-time data from production line equipment and generate real-time data sequences; Preprocessing the real-time data sequence, performing denoising and standardization to obtain clear data for feature extraction; Based on the clear data, the AI ​​model is used to extract key features and perform training to generate a pre-trained energy efficiency prediction model; The real-time data is input into the pre-trained energy efficiency prediction model to generate optimization decisions and adjust the operating parameters of the production line equipment to form a feedback control system.

2. The method according to claim 1, characterized in that The acquisition of real-time data of production line equipment also includes: Obtain the temperature (T), power (P), workload (L) and equipment status (S) data of production line equipment, and collect the data in real time through sensors and monitoring systems; The data is transmitted to a data acquisition terminal via an industrial control network (such as a SCADA system).

3. The method according to claim 2, characterized in that The preprocessing of the real-time data sequence further comprises: De-noising the real-time data sequence, using a filtering algorithm to eliminate sensor noise and external interference; The denoised data is standardized to adjust the data to a uniform scale range.

4. The method according to claim 3, characterized in that The generating of the pre-trained energy efficiency prediction model further includes: Extract key features such as energy consumption patterns and equipment operation rules from the clear data; The key features are used to train the AI ​​model and optimize the model parameters.

5. The method according to claim 4, characterized in that The generating of optimization decisions and adjusting the operating parameters of the production line equipment to form a feedback control system also includes: Based on the optimization decision, determining optimal operating parameters of the production line equipment; The optimal operating parameters are fed back to the production line control system to adjust the operating status of the equipment.

6. The method according to claim 1, characterized in that The real-time data is collected through temperature sensors, current sensors, pressure sensors, etc.

7. The method according to claim 1, characterized in that The denoising process uses a low-pass filter or a Kalman filter algorithm.

8. The method according to claim 1, characterized in that: The key features include statistical features such as the mean, standard deviation, maximum value, and minimum value of the data.

9. The method according to claim 1, characterized in that: The optimization decisions are generated through linear programming, genetic algorithm and particle swarm optimization algorithm.

10. An adaptive energy consumption optimization system for a vacuum isothermal forging system production line, characterized in that: include: A data acquisition module is used to obtain real-time data of production line equipment, including temperature (T), power (P), workload (L) and equipment status (S); A data processing module, used for preprocessing the real-time data, wherein the preprocessing includes denoising and standardization; A feature extraction and model training module, used to extract key features based on the preprocessed clear data and train using an AI model to generate a pre-trained energy efficiency prediction model; The optimization decision and feedback control module is used to input the real-time data into the pre-trained energy efficiency prediction model, generate optimization decisions and adjust the operating parameters of the production line equipment to form a feedback control system.

Images