Energy efficiency optimization system for training and optimizing according to actual data

Abstract

The invention discloses an energy efficiency optimization system for training and optimizing according to actual data, and relates to the technical field of energy efficiency optimization. Comprising an energy consumption prediction model, a temperature optimization module, an adaptive control module, a data collection module, a model training module, a variable optimization module, a prediction maintenance module, a collaborative optimization module, a user operation module and an algorithm updating module. According to the energy efficiency optimization system for training and optimizing according to the actual data, the prediction model of the system is trained and optimized by using the actual data, and when the actual energy use condition changes, adjustment can be made in time to achieve the purpose of dynamic adjustment, so that the limitation of a fixed model and an algorithm is avoided, and the applicability of the system is improved; the method is particularly suitable for kitchen application scenes, and meanwhile, more input variables are continuously introduced in combination with actual situations to further improve the accuracy and reliability of the prediction model, so that the implementation effect of energy-saving measures is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy efficiency optimization, and in particular to an energy efficiency optimization system that is trained and optimized based on actual data. Background Art

[0002] Energy efficiency, in short, refers to the ratio of useful energy produced by energy during its use to the energy consumed. It is an important criterion for measuring the level of energy utilization. It not only reflects the physical and economic efficiency of energy utilization, but also reflects the technical and management level of energy utilization. In today's society, with the continuous growth of energy demand and the increasing awareness of environmental protection, improving energy efficiency has become a global consensus and the key to promoting sustainable economic and social development. From a physical point of view, by improving the physical efficiency of energy, more energy output can be obtained under the same energy consumption, thereby improving energy utilization and reducing energy waste. From an economic point of view, optimizing energy efficiency pays more attention to the economic benefits that energy use can bring. By optimizing the industrial structure, developing low-energy consumption and high-value-added industries, and restricting the development of high-energy consumption and high-pollution industries, the energy intensity of the entire economic system can be reduced and energy utilization efficiency can be improved. Under this premise, the energy efficiency optimization system came into being.

[0003] The energy efficiency optimization system is a system that integrates multiple technologies and strategies to improve the efficiency of energy utilization in the production, transmission, distribution and consumption processes. It is widely used in various fields, including industrial manufacturing, construction, transportation, data centers, etc., and with the continuous development of information technology, it is gradually applied to more fields. For example, in recent years, the system has been applied to kitchens to optimize the energy efficiency of kitchens. In kitchen applications, the system can monitor the energy consumption of different equipment and food in the kitchen in real time, help identify energy waste links, propose improvement measures, and reduce production costs.

[0004] However, in actual use, since the existing energy efficiency optimization systems are usually designed and implemented based on fixed models and algorithms, when the actual energy usage changes, the adaptability of the model is limited, and it is impossible to make timely adjustments. In particular, the energy consumption of different kitchen scenarios varies greatly, with different energy usage characteristics and needs, and low regularity. This further leads to the system being unable to adapt well to the actual needs of different kitchens. In addition, due to the lack of training and optimization of actual data, the optimization effect of the system is easily limited. In many cases, the system will be unable to identify the true energy-saving potential, resulting in limited implementation of energy-saving measures.

[0005] Therefore, there is an urgent need to improve this shortcoming. The present invention studies and improves the existing technology and its shortcomings, and provides an energy efficiency optimization system that is trained and optimized based on actual data. Summary of the invention

[0006] The object of the present invention is to provide an energy efficiency optimization system that is trained and optimized based on actual data to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: an energy efficiency optimization system trained and optimized according to actual data, comprising: An energy consumption prediction model, wherein the energy consumption prediction model uses a polynomial regression (capable of capturing the nonlinear relationship between input variables and output variables) model to predict energy consumption, and uses temperature (under different temperature conditions, the energy consumption characteristics of the equipment will be different. Taking temperature as an input variable helps the model to more accurately predict energy consumption) and cooking time (the length of cooking time will directly affect the operating time and energy consumption level of the equipment. Taking cooking time as an input variable helps the model to more comprehensively consider energy consumption factors) as input variables, and aims to predict future energy consumption based on historical data and current conditions; A temperature optimization module, which is responsible for iteratively searching within a given temperature range using a temperature optimization algorithm such as a genetic algorithm or a particle swarm algorithm to find an optimal temperature setting that meets cooking requirements and minimizes energy consumption, and uses this optimal temperature as a subsequent target temperature; An adaptive control module, which is responsible for dynamically adjusting the heating power according to the target temperature and the current temperature using a PID (proportional-integral-derivative) control algorithm, while considering historical error data to improve control accuracy; A data collection module, which is responsible for real-time monitoring and recording of energy consumption data of different equipment and food in the kitchen through sensors and measurement and control equipment to collect actual operation data, including energy consumption characteristics of different foods and different equipment; A model training module, which is responsible for periodically retraining the model to adapt to equipment aging and seasonal changes, and improving the accuracy and reliability of predictions by continuously updating and optimizing the model; A variable optimization module, which is responsible for combining actual data to introduce more input variables into the energy consumption prediction model, such as humidity, food weight, preheating time, etc., to promote the model to capture nonlinear relationships; A predictive maintenance module, which is responsible for analyzing abnormalities in energy consumption patterns, predicting possible equipment failures, and recommending optimal maintenance times based on usage; A collaborative optimization module, which is responsible for considering the energy use of the entire kitchen, coordinating the operation of multiple devices, achieving peak load management, and avoiding simultaneous high-power operation; A user operation module, wherein the user operation module provides an intuitive user interface, allows chefs to easily set parameters and view suggestions, and generates detailed energy usage reports to help managers make decisions; implementation method: the user operation module uses graphical user interface (GUI) technology to develop an intuitive and easy-to-use operation interface, and generates detailed energy usage reports through a report generation tool; Algorithm update module, which is responsible for uploading data to the cloud for in-depth analysis, and pushing algorithm updates through the cloud to continuously improve system performance; implementation method: the algorithm update module realizes remote analysis and updates through cloud computing and data analysis technology, specifically using the data storage and analysis capabilities of the cloud computing platform to conduct in-depth analysis of the collected data to discover potential energy-saving opportunities and optimization points, and at the same time regularly pushes the latest algorithms and optimization strategies to the local system through the cloud push algorithm to continuously improve system performance and adapt to new application scenarios.

[0008] Furthermore, the construction process of the energy consumption prediction model is as follows: Data collection and preprocessing: Collect historical data, including energy consumption, temperature, cooking time, etc., and clean and preprocess the data, such as removing outliers and filling missing values, and then divide the data into training sets and test sets for model training and verification; Polynomial regression model construction: Select polynomial regression as the prediction model and construct polynomial features based on the input variables (temperature and cooking time). Then train the model using the training set data and adjust the model parameters to minimize the prediction error. Model optimization: Evaluate the performance of the model through methods such as cross-validation and select the optimal model parameters. At the same time, use regularization techniques (such as L1 regularization, L2 regularization, etc.) to prevent model overfitting.

[0009] Further, the temperature optimization algorithm includes but is not limited to a genetic algorithm and a particle swarm algorithm. The genetic algorithm has strong global search capability and robustness, but has a slow convergence speed and requires more parameters to be adjusted. The particle swarm algorithm has a fast convergence speed and relatively simple parameter adjustment, but may fall into a local optimal solution when dealing with complex problems. When selecting an optimization algorithm, a trade-off and selection are made based on the characteristics and requirements of the problem. Specifically, a genetic algorithm is selected for the optimization of controller parameters and a particle swarm algorithm is selected for the optimal arrangement of sensor measuring points. Specifically, a particle swarm algorithm is selected for the need to quickly find the optimal solution and have high computational efficiency, and a genetic algorithm is selected for the need to more comprehensively search the solution space and avoid falling into a local optimum. A variety of algorithms can be combined or improved to further improve the performance and effect of the temperature optimization module. For example, a genetic algorithm and a particle swarm algorithm can be combined to use the global search capability of the genetic algorithm and the rapid convergence of the particle swarm algorithm to jointly solve the problem, or the algorithm can be improved according to the needs of a specific problem, such as adjusting the algorithm parameters and optimizing the algorithm structure. The application process of the genetic algorithm (GA) in the temperature optimization module is as follows: Initialize the population: randomly generate a set of temperature settings within a predetermined temperature range as the initial population; Fitness assessment: Based on cooking needs and energy consumption, each temperature setting is assessed for fitness, that is, the degree to which it meets cooking needs and the amount of energy consumption are calculated; Selection: Based on the fitness evaluation results, select the temperature setting with higher fitness as the parent generation to generate the next generation; Crossover: Perform a crossover operation on the selected parent generation to generate new child generation temperature settings; Mutation: Randomly mutate the offspring temperature settings to increase the diversity of the population; Iteration: Repeat the above steps until the predetermined number of iterations is reached or the optimal solution that meets the conditions is found; The application process of the particle swarm algorithm (PSO) in the temperature optimization module is as follows: Initialize the particle swarm: randomly generate a set of particles within a predetermined temperature range (i.e., temperature setting), each particle has a velocity and position; Fitness evaluation: The fitness of each particle position (i.e., temperature setting) is evaluated based on cooking requirements and energy consumption; Update speed and position: Update the speed and position of particles according to the current position of each particle, the best historical position (pbest), and the best historical position (gbest) of the entire particle group; Iteration: Repeat the above steps until the predetermined number of iterations is reached or the optimal solution that meets the conditions is found.

[0010] Furthermore, the adaptive control module includes a PID controller, and the PID controller measures the difference between the target temperature and the current temperature, and calculates the corresponding control quantity to adjust the system output, i.e., the heating power, based on the control effects of the three links of proportion, integration and differentiation, so as to eliminate deviations and ensure stable operation of the system.

[0011] Furthermore, the data collection module monitors and records data in real time through sensors or measurement and control equipment installed on the equipment, such as measuring the equipment's energy consumption, temperature, humidity and other parameters, and transmits the data to the data collection module for storage and analysis, while obtaining more comprehensive data through user input or interfaces with other systems.

[0012] Furthermore, the model training module periodically obtains the latest actual operation data from the data collection module, and uses these data to retrain the polynomial regression model. During the training process, appropriate optimization algorithms (such as gradient descent, stochastic gradient descent, etc.) and loss functions (such as mean square error, cross entropy, etc.) are used to minimize the prediction error and adjust the parameters of the model. At the same time, the impact of equipment aging and seasonal changes on energy consumption is considered, and corresponding adjustment factors or characteristic variables are introduced into the model to improve the accuracy and adaptability of the prediction.

[0013] Furthermore, the variable optimization module specifically includes the following submodules: Data preprocessing submodule: used to clean (remove duplicate data, process missing values, correct erroneous data, etc.) and convert raw data (convert data into a format suitable for model training, such as standardization and normalization). The processed data includes new variables such as humidity, food weight, preheating time, and traditional variables such as temperature and cooking time. Feature selection submodule: It is used to determine which features have a significant impact on the prediction results by using correlation analysis, feature importance evaluation (such as using models such as random forest to evaluate the importance of features) or feature selection algorithms (such as recursive feature elimination, Lasso regression, etc.), and to select features that have a significant impact on energy consumption prediction from the preprocessed data, i.e., input variables; Variable introduction submodule: It is used to use the newly selected features as new input variables, introduce them into the original energy consumption prediction model, and obtain an accurate prediction model after model training.

[0014] Furthermore, the predictive maintenance module identifies abnormal changes in energy consumption patterns through data analysis algorithms. Specifically, by analyzing abnormal changes in historical energy consumption data (such as sudden increases or decreases), possible equipment failures or performance degradations are predicted and early warnings are issued, thereby avoiding the impact of sudden equipment failures on production operations. Combined with the equipment's usage and maintenance history, the module recommends the best maintenance time, maintenance content, and maintenance methods to extend equipment life and reduce maintenance costs.

[0015] Furthermore, the collaborative optimization module coordinates the operation plans of multiple devices through an optimization algorithm, specifically integrating and analyzing the operation data from multiple kitchen devices, and using optimization algorithms (such as genetic algorithms, particle swarm algorithms, etc.) to coordinate the operation plans of multiple devices to achieve optimal energy use in the entire kitchen. By avoiding simultaneous high-power operation of devices and reasonably distributing loads, peak loads can be reduced and energy consumption can be reduced.

[0016] The present invention provides an energy efficiency optimization system that is trained and optimized based on actual data, which has the following beneficial effects: The present invention uses actual data to train and optimize the system's prediction model. When the actual energy usage changes, timely adjustments can be made to achieve the purpose of dynamic adjustment. Compared with traditional energy efficiency optimization systems, it not only avoids fixed model and algorithm limitations and improves the applicability of the system, making it better suitable for energy efficiency optimization in various enterprises and industries, especially for kitchen application scenarios. At the same time, it can also continuously introduce more input variables based on actual conditions according to different energy usage characteristics and needs, so as to further improve the accuracy and reliability of the prediction model, thereby helping the system to accurately identify the real energy-saving potential and improve the implementation effect of energy-saving measures. In addition, the system is also designed with algorithm update logic, which can upload data to the cloud for in-depth analysis to obtain cloud-pushed algorithm updates and continuously improve system performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The overall architecture diagram of an energy efficiency optimization system for training and optimization based on actual data of the present invention; Figure 2 The present invention is a schematic diagram of the operation flow of a variable optimization module of an energy efficiency optimization system that is trained and optimized based on actual data. DETAILED DESCRIPTION

[0018] The following embodiments of the present invention are described in further detail in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0019] like Figure 1-Figure 2As shown, an energy efficiency optimization system trained and optimized according to actual data includes an energy consumption prediction model, a temperature optimization module, an adaptive control module, a data collection module, a model training module, a variable optimization module, a predictive maintenance module, a collaborative optimization module, a user operation module and an algorithm update module; Energy consumption prediction model: a polynomial regression model (capable of capturing the nonlinear relationship between input variables and output variables) is used to predict energy consumption, and temperature (under different temperature conditions, the energy consumption characteristics of the equipment will be different. Taking temperature as an input variable helps the model to more accurately predict energy consumption) and cooking time (the length of cooking time will directly affect the operating time and energy consumption level of the equipment. Taking cooking time as an input variable helps the model to more comprehensively consider energy consumption factors) are used as input variables, aiming to predict future energy consumption based on historical data and current conditions. In this embodiment, by analyzing the relationship between temperature and cooking time in historical data and energy consumption, a prediction model is established using polynomial regression technology to achieve prediction of future energy consumption; In this embodiment, the construction process of the energy consumption prediction model is as follows: Data collection and preprocessing: Collect historical data, including energy consumption, temperature, cooking time, etc., and clean and preprocess the data, such as removing outliers and filling missing values, and then divide the data into training sets and test sets for model training and verification; Polynomial regression model construction: Select polynomial regression as the prediction model and construct polynomial features based on the input variables (temperature and cooking time). Then train the model using the training set data and adjust the model parameters to minimize the prediction error. Model optimization: Evaluate the performance of the model through methods such as cross-validation and select the optimal model parameters. At the same time, use regularization techniques (such as L1 regularization, L2 regularization, etc.) to prevent the model from overfitting. Temperature optimization module: Use temperature optimization algorithms such as genetic algorithm and particle swarm algorithm to search iteratively within a given temperature range to find the optimal temperature setting that meets cooking needs and minimizes energy consumption, and use this optimal temperature as the subsequent target temperature; Among them, temperature optimization algorithms include but are not limited to genetic algorithm and particle swarm algorithm. Genetic algorithm has strong global search ability and robustness, but slow convergence speed and more parameters need to be adjusted. Particle swarm algorithm has fast convergence speed and relatively simple parameter adjustment, but it may fall into local optimal solution when dealing with complex problems; When selecting an optimization algorithm, weigh and select according to the characteristics and needs of the problem, and select a specific one based on the characteristics of the problem. Specifically, the genetic algorithm is selected for the optimization of controller parameters, and the particle swarm algorithm is selected for the optimal arrangement of sensor measuring points. Specifically, the particle swarm algorithm is selected for the need to quickly find the optimal solution and have high computational efficiency, and the genetic algorithm is selected for the need to search the solution space more comprehensively and avoid falling into the local optimum. Multiple algorithms can be combined or improved to further improve the performance and effect of the temperature optimization module. For example, the genetic algorithm and the particle swarm algorithm can be combined to use the global search ability of the genetic algorithm and the fast convergence of the particle swarm algorithm to jointly solve the problem, or the algorithm can be improved according to the needs of the specific problem, such as adjusting the algorithm parameters, optimizing the algorithm structure, etc. 1) The application process of genetic algorithm (GA) in temperature optimization module is as follows: Initialize the population: randomly generate a set of temperature settings within a predetermined temperature range as the initial population; Fitness assessment: Based on cooking needs and energy consumption, each temperature setting is assessed for fitness, that is, the degree to which it meets cooking needs and the amount of energy consumption are calculated; Selection: Based on the fitness evaluation results, select the temperature setting with higher fitness as the parent generation to generate the next generation; Crossover: Perform a crossover operation on the selected parent generation to generate new child generation temperature settings; Mutation: Randomly mutate the offspring temperature settings to increase the diversity of the population; Iteration: Repeat the above steps until the predetermined number of iterations is reached or the optimal solution that meets the conditions is found; 2) The application process of particle swarm optimization (PSO) in the temperature optimization module is as follows: Initialize the particle swarm: randomly generate a set of particles within a predetermined temperature range (i.e., temperature setting), each particle has a velocity and position; Fitness evaluation: The fitness of each particle position (i.e., temperature setting) is evaluated based on cooking requirements and energy consumption; Update speed and position: Update the speed and position of particles according to the current position of each particle, the best historical position (pbest), and the best historical position (gbest) of the entire particle group; Iteration: Repeat the above steps until the predetermined number of iterations is reached or the optimal solution that meets the conditions is found; Adaptive control module: Use PID (proportional-integral-differential) control algorithm to dynamically adjust the heating power according to the target temperature and current temperature, and consider historical error data to improve control accuracy; the adaptive control module includes a PID controller, and the PID controller measures the difference between the target temperature and the current temperature, and calculates the corresponding control amount according to the control effects of the three links of proportion, integration and differentiation to adjust the system output, that is, the heating power, to eliminate deviations and make the system run stably; Data collection module: monitors and records the energy consumption data of different equipment and food in the kitchen in real time through sensors and measurement and control equipment to collect actual operation data, including energy consumption characteristics of different foods and different equipment; in this embodiment, the data collection module monitors and records data in real time through sensors or measurement and control equipment installed on the equipment, such as measuring the energy consumption, temperature, humidity and other parameters of the equipment, and transmits the data to the data collection module for storage and analysis, and obtains more comprehensive data through user input or interface with other systems; Model training module: retrain the model regularly to adapt to equipment aging and seasonal changes, and improve the accuracy and reliability of prediction by continuously updating and optimizing the model; in this embodiment, the model training module regularly obtains the latest actual operation data from the data collection module, and uses these data to retrain the polynomial regression model. During the training process, appropriate optimization algorithms (such as gradient descent method, stochastic gradient descent method, etc.) and loss functions (such as mean square error, cross entropy, etc.) are used to minimize the prediction error and adjust the parameters of the model. At the same time, the impact of equipment aging and seasonal changes on energy consumption is considered, and corresponding adjustment factors or characteristic variables are introduced into the model to improve the accuracy and adaptability of the prediction; Variable optimization module: Combined with actual data, more input variables are introduced into the energy consumption prediction model, such as humidity, food weight, preheating time, etc., to promote the model to capture nonlinear relationships; this module specifically includes the following sub-modules: Data preprocessing submodule: cleaning (removing duplicate data, processing missing values, correcting erroneous data, etc.) and converting raw data (converting data into a format suitable for model training, such as standardization and normalization). The processed data includes new variables such as humidity, food weight, preheating time, and traditional variables such as temperature and cooking time. Feature selection submodule: using correlation analysis, feature importance evaluation (such as using a model such as random forest to evaluate the importance of features) or feature selection algorithms (such as recursive feature elimination, Lasso regression, etc.) to determine which features have a significant impact on the prediction results, and screen out features that have a significant impact on energy consumption prediction from the preprocessed data, i.e., input variables; in this embodiment, a feature importance evaluation method is used, and the specific process is: calculating the correlation between each feature and energy consumption, and using a model such as random forest to evaluate the importance of each feature, and then selecting features that have a significant impact on the prediction results based on the importance evaluation results; Variable introduction submodule: The newly selected features are used as new input variables and introduced into the original energy consumption prediction model. After model training, an accurate prediction model is obtained; Predictive maintenance module: analyzes abnormalities in energy consumption patterns, predicts possible equipment failures, and recommends optimal maintenance time based on usage. In this embodiment, the predictive maintenance module identifies abnormal changes in energy consumption patterns through a data analysis algorithm. Specifically, by analyzing abnormal changes in historical energy consumption data (such as sudden increases or decreases), the module predicts possible equipment failures or performance degradation and issues early warnings, thereby avoiding the impact of sudden equipment failures on production operations. In addition, the module recommends optimal maintenance time, maintenance content, and maintenance methods based on equipment usage and maintenance history records, thereby extending equipment life and reducing maintenance costs. Collaborative optimization module: considers the energy usage of the entire kitchen, coordinates the operation of multiple devices, realizes peak load management, and avoids simultaneous high-power operation; in this embodiment, the collaborative optimization module coordinates the operation plans of multiple devices through an optimization algorithm, specifically integrating and analyzing the operation data from multiple kitchen devices, and using an optimization algorithm (such as a genetic algorithm, a particle swarm algorithm, etc.) to coordinate the operation plans of multiple devices to achieve the optimization of the energy usage of the entire kitchen. By avoiding the simultaneous high-power operation of the devices and reasonably distributing the load, the peak load can be reduced and the energy consumption can be reduced; User operation module: provides an intuitive user interface, allowing chefs to easily set parameters and view suggestions, and generates detailed energy usage reports to help managers make decisions; in this embodiment, the user operation module specifically uses graphical user interface (GUI) technology to develop an intuitive and easy-to-use operation interface, and generates detailed energy usage reports through a report generation tool; Algorithm update module: upload data to the cloud for in-depth analysis, and push algorithm updates through the cloud to continuously improve system performance; in this embodiment, the algorithm update module implements remote analysis and updates through cloud computing and data analysis technology, specifically using the data storage and analysis capabilities of the cloud computing platform to conduct in-depth analysis of the collected data to discover potential energy-saving opportunities and optimization points, and at the same time regularly push the latest algorithms and optimization strategies to the local system through the cloud push algorithm to continuously improve system performance and adapt to new application scenarios.

[0020] The embodiments of the present invention are given for the purpose of illustration and description, and are not intended to be exhaustive or to limit the invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

Claims

1. An energy efficiency optimization system trained and optimized based on actual data, comprising an energy consumption prediction model, wherein the energy consumption prediction model uses a polynomial regression model to predict energy consumption and uses temperature and cooking time as input variables, characterized in that: It also includes a temperature optimization module, an adaptive control module, a data collection module, a model training module, and a variable optimization module; The temperature optimization module is responsible for using the temperature optimization algorithm to perform iterative search within a given temperature range to find the best temperature setting that meets cooking requirements and minimizes energy consumption, and uses this best temperature as the subsequent target temperature; The adaptive control module is responsible for using a PID control algorithm to dynamically adjust the heating power according to the target temperature and the current temperature; The data collection module is responsible for real-time monitoring and recording of energy consumption data through sensors and measurement and control equipment to collect actual operation data, including energy consumption characteristics of different foods and different equipment; The model training module is responsible for periodically retraining the model to adapt to equipment aging and seasonal changes; The variable optimization module is responsible for combining actual data to introduce more input variables into the energy consumption prediction model to promote the model to capture nonlinear relationships; the variable optimization module specifically includes the following submodules: Data preprocessing submodule: used to clean and convert raw data, including newly added variables and traditional variables; Feature selection submodule: used to select features that have a significant impact on energy consumption prediction from preprocessed data using correlation analysis, feature importance evaluation or feature selection algorithm; Variable introduction submodule: It is used to use the newly selected features as new input variables, introduce them into the original energy consumption prediction model, and obtain an accurate prediction model after model training.

2. The energy efficiency optimization system according to claim 1, characterized in that: The system further comprises: A predictive maintenance module, which is responsible for analyzing abnormalities in energy consumption patterns, predicting possible equipment failures, and recommending optimal maintenance times based on usage; A collaborative optimization module, which is responsible for considering the energy use of the entire kitchen, coordinating the operation of multiple devices, achieving peak load management, and avoiding simultaneous high-power operation; A user operation module, which provides a user interface, allows chefs to easily set parameters and view suggestions, and generates detailed energy usage reports to help managers make decisions; The algorithm update module is responsible for uploading data to the cloud for in-depth analysis and pushing algorithm updates through the cloud to continuously improve system performance.

3. The energy efficiency optimization system according to claim 1, characterized in that: The construction process of the energy consumption prediction model is as follows: Data collection and preprocessing: Collect historical data, clean and preprocess the data, and then divide the data into training sets and test sets; Polynomial regression model construction: Select polynomial regression as the prediction model and build polynomial features based on the input variables. Then use the training set data to train the model and adjust the model parameters to minimize the prediction error. Model optimization: The performance of the model is evaluated through cross-validation, and the optimal model parameters are selected. Regularization techniques are used to prevent the model from overfitting.

4. The energy efficiency optimization system according to claim 1, characterized in that: The temperature optimization algorithm includes but is not limited to a genetic algorithm and a particle swarm algorithm. When selecting an optimization algorithm, a trade-off and selection are made based on the characteristics and requirements of the problem, and different algorithms can be combined or improved. The application process of the genetic algorithm in the temperature optimization module is as follows: Initialize the population: randomly generate a set of temperature settings within a predetermined temperature range as the initial population; Fitness assessment: Based on cooking needs and energy consumption, each temperature setting is assessed for fitness, that is, the degree to which it meets cooking needs and the amount of energy consumption are calculated; Selection: Based on the fitness evaluation results, select the temperature setting with higher fitness as the parent generation to generate the next generation; Crossover: Perform a crossover operation on the selected parent generation to generate new child generation temperature settings; Mutation: Randomly mutate the offspring temperature settings to increase the diversity of the population; Iteration: Repeat the above steps until the predetermined number of iterations is reached or the optimal solution that meets the conditions is found; The application process of the particle swarm algorithm in the temperature optimization module is as follows: Initialize the particle swarm: randomly generate a group of particles within a predetermined temperature range, each particle has a speed and position; Fitness evaluation: The fitness of each particle position is evaluated based on cooking requirements and energy consumption; Update speed and position: Update the speed and position of particles according to the current position of each particle, the best historical position, and the best historical position of the entire particle group; Iteration: Repeat the above steps until the predetermined number of iterations is reached or the optimal solution that meets the conditions is found.

5. The energy efficiency optimization system for training and optimizing based on actual data according to claim 4, characterized in that: The selection logic of the optimization algorithm includes: According to the characteristics of the problem: for the optimization of controller parameters, the genetic algorithm is selected; for the optimal arrangement of sensor measurement points, the particle swarm algorithm is selected; According to the needs: If you need to find the optimal solution quickly and have high computational efficiency, choose the particle swarm algorithm; if you need to search the solution space more comprehensively and avoid falling into the local optimum, choose the genetic algorithm.

6. The energy efficiency optimization system according to claim 1, characterized in that: The adaptive control module includes a PID controller, and the PID controller measures the difference between the target temperature and the current temperature, and calculates the corresponding control quantity to adjust the system output, that is, the heating power, according to the control effects of the three links of proportion, integration and differentiation, so as to eliminate deviations and make the system run stably.

7. The energy efficiency optimization system according to claim 1, characterized in that: The data collection module monitors and records data in real time through sensors or measurement and control equipment installed on the equipment, and transmits the data to the data collection module for storage and analysis, while obtaining more comprehensive data through user input or interfaces with other systems.

8. The energy efficiency optimization system for training and optimizing based on actual data according to claim 1, characterized in that: The model training module periodically obtains the latest actual operation data from the data collection module, and uses this data to retrain the polynomial regression model. During the training process, an optimization algorithm and a loss function are used to minimize the prediction error and adjust the parameters of the model. At the same time, the impact of equipment aging and seasonal changes on energy consumption is considered, and corresponding adjustment factors or characteristic variables are introduced into the model to improve the accuracy and adaptability of the prediction.

9. The energy efficiency optimization system for training and optimizing based on actual data according to claim 2, characterized in that: The predictive maintenance module identifies abnormal changes in energy consumption patterns through data analysis algorithms. Specifically, by analyzing abnormal changes in historical energy consumption data, it predicts possible equipment failures or performance degradation and provides early warnings. It also recommends the best maintenance time, maintenance content, and maintenance methods based on the equipment's usage and maintenance history.

10. The energy efficiency optimization system for training and optimization based on actual data according to claim 2, characterized in that: The collaborative optimization module coordinates the operation plans of multiple devices through an optimization algorithm, specifically integrating and analyzing the operation data from multiple kitchen devices, and using the optimization algorithm to coordinate the operation plans of multiple devices to achieve optimal energy use in the entire kitchen.

Images