Adaptive Industrial Control Energy Management Scheduling System

Abstract

The present invention relates to the technical field of energy optimization and scheduling, and is specifically an adaptive industrial control energy management and scheduling system, wherein the system comprises an energy demand prediction module, an energy supply strategy module, a dynamic weight adjustment module, a multi-period allocation module, a multi-objective optimization strategy module, and a real-time adjustment feedback module. The present invention improves the foresight and accuracy of energy allocation by accurately predicting future demand using historical energy consumption data, ensures the stability and economy of energy supply by optimizing the supply strategy, adopts a linear regression model to automatically adjust the energy use weight, enhances the adaptability of industrial production to energy market price fluctuations and environmental changes, adopts a Pareto frontier analysis method to balance cost, energy efficiency and environmental impact targets, realizes multi-objective optimization of energy management, can respond quickly according to real-time data, optimizes energy allocation, improves energy utilization efficiency, reduces energy waste, and reduces operating costs.

Description

Technical Field

[0001] The present invention relates to the technical field of energy optimization scheduling, and in particular to an adaptive industrial control energy management scheduling system. Background Art

[0002] The field of energy optimization and scheduling technology focuses on maximizing energy efficiency through algorithms and system control methods. This technology combines real-time data analysis and adaptive control strategies to reduce energy waste, optimize energy distribution, reduce operating costs and improve energy efficiency. In industrial environments, it includes the management and scheduling of electricity, heat and water resources. Technical implementations usually involve integrated sensor networks, real-time monitoring systems and intelligent decision support systems that can dynamically adjust energy flows based on current energy demand and supply conditions.

[0003] Among them, the adaptive industrial control energy management and scheduling system is an industrial solution designed to optimize energy use in industrial production and operations through adaptive technologies and algorithms. The purpose of the system is to improve energy efficiency, reduce waste, reduce energy costs, and support sustainable development goals by dynamically adjusting and managing energy distribution. The system monitors energy consumption patterns in real time, predicts future energy demand, and adjusts energy allocation strategies accordingly to ensure the optimal match between energy supply and demand. The system is particularly suitable for industrial environments with large changes in energy demand and high energy costs, such as manufacturing, data centers, and large commercial facilities.

[0004] Traditional industrial control energy management and scheduling systems lack sufficient flexibility and adaptability, making it difficult to accurately predict energy demand and responding slowly to changes in energy market prices and environmental policies. This results in inaccurate energy allocation, oversupply or undersupply, increased energy costs and waste, and failure to fully consider environmental protection requirements. For example, energy supply strategies that fail to accurately match actual demand may lead to production interruptions or excessive energy consumption, which in turn affects the company's economic benefits and environmental responsibilities. The lack of an effective multi-objective optimization strategy makes it difficult for traditional systems to strike a balance between cost control, energy efficiency improvement, and reduced environmental impact, limiting the comprehensive performance and effectiveness of energy management. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose an adaptive industrial control energy management scheduling system.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: an adaptive industrial control energy management and scheduling system, the system comprising:

[0007] The energy demand forecasting module predicts the energy demand for future work shifts and production cycles based on the historical energy consumption data of industrial production, and obtains the energy demand forecast results;

[0008] The energy supply strategy module optimizes the supply strategies of multiple types of energy based on the energy demand forecast results to obtain a supply strategy optimization plan;

[0009] The dynamic weight adjustment module is based on the real-time energy usage data of industrial production, combined with the fluctuation of energy demand in the industrial production process, the uncertainty of external energy market prices and environmental impact, and uses a linear regression model to automatically adjust the weights of multiple types of energy in the production process to obtain dynamically adjusted weight coefficients;

[0010] The multi-period allocation module adjusts the allocation of differentiated energy types in multiple time periods based on the dynamically adjusted weight coefficients and the supply strategy optimization scheme to obtain an optimized energy allocation strategy;

[0011] The multi-objective optimization strategy module is based on the above-mentioned optimized energy allocation strategy, combines cost control, energy efficiency improvement and environmental factors, adopts the Pareto frontier analysis method, captures the balance point between objectives, and obtains the multi-objective optimization results;

[0012] The real-time adjustment feedback module dynamically adjusts the energy allocation strategy in the industrial production process based on the multi-objective optimization results to obtain real-time optimization adjustment feedback.

[0013] The improvements of the present invention are that the energy demand forecasting results include the predicted total energy demand, the demand ratio of multiple energy types, and the peak demand time period; the supply strategy optimization plan includes the supply amount, supply cost and supply schedule of multiple energy sources; the dynamically adjusted weight coefficients include energy cost weights, environmental impact weights, and supply stability weights; the optimized energy allocation strategy includes the allocation ratios of differentiated energy types in multiple time periods, priority rankings, expected cost savings and environmental benefit improvement information; the multi-objective optimization results include cost optimization plans, energy efficiency optimization plans, environmental impact optimization plans and trade-off solutions; the real-time optimization adjustment feedback includes the adjusted energy allocation plan and deviation analysis information between expected and actual energy usage.

[0014] The present invention is improved in that the energy supply strategy module includes:

[0015] The supply analysis submodule analyzes the supply capacity of multiple energy sources based on the energy demand forecast results, including evaluating the supply of differentiated energy sources, combining the seasonal changes and supply stability of multiple energy sources, identifying energy sources that can meet the forecast demand, and formulating a matching supply plan for each energy source to generate energy supply analysis results;

[0016] The cost-benefit evaluation submodule calculates the supply cost and benefit of each energy source based on the energy supply analysis results, combines various economic and environmental factors including energy prices, transportation costs, and environmental taxes, and optimizes energy costs by comparing and analyzing the cost-benefit of energy supply plans to generate energy supply benefit evaluation results;

[0017] Based on the energy supply benefit evaluation results, the strategy determination submodule plans and optimizes the supply strategy of differentiated energy sources, matches the predicted energy demand, optimizes cost-effectiveness and supply stability, analyzes the supply volume, supply cost and environmental impact, formulates energy supply strategy, and obtains the supply strategy optimization plan.

[0018] The present invention is improved in that the dynamic weight adjustment module comprises:

[0019] The weight factor analysis submodule analyzes energy demand fluctuations, external energy market price uncertainty and environmental factors based on real-time energy usage data of industrial production, identifies key factors affecting the weights of multiple energy usage, and obtains energy weight factor analysis results;

[0020] The energy weight adjustment submodule sets the rules and parameters for weight adjustment based on the analysis results of the energy weight factors, including energy price sensitivity and environmental impact weight, and generates a weight adjustment strategy;

[0021] Based on the weight adjustment strategy, the weight update submodule adopts a linear regression model, combines real-time energy consumption data, energy cost changes and the availability of renewable energy, updates the usage weights of multiple types of energy in industrial production, and obtains dynamically adjusted weight coefficients.

[0022] The present invention is improved in that the linear regression model is according to the formula:

[0023] ;

[0024] Calculate the energy usage weight value and obtain the dynamically adjusted weight coefficient, where: is the new weight value, is the old weight value, Due to changes in energy costs, For changes in energy availability, is the proportion of renewable energy, For seasonal factors, For time sensitivity, , , , and is the adjustment factor.

[0025] The present invention is improved in that the multi-period allocation module comprises:

[0026] The constraint condition establishment submodule defines the constraint conditions of the nonlinear programming problem based on the dynamically adjusted weight coefficient and the supply strategy optimization scheme, the constraint conditions including the feasibility of the energy supply and demand balance and the allocation scheme, and generates a constraint condition set;

[0027] The planning and solving submodule solves the nonlinear planning problem of the allocation of multiple energy types in differentiated time periods based on the constraint condition set and in combination with the objectives of cost and environmental impact, and generates a basic energy allocation plan;

[0028] The optimization scheme generation submodule refines and optimizes the scheme based on the basic energy allocation scheme, adjusts the scheme to meet the industrial production needs, and obtains an optimized energy allocation strategy.

[0029] The present invention is improved in that the multi-objective optimization strategy module includes:

[0030] The optimization target definition submodule analyzes and defines the targets of multi-objective optimization based on the optimization energy allocation strategy, including minimizing cost, maximizing energy efficiency and minimizing environmental impact, plans the energy use in the industrial production process, and generates a multi-objective optimization target set;

[0031] The Pareto front analysis submodule applies the Pareto front analysis method based on the multi-objective optimization target set to analyze the balance points among multiple optimization targets, the trade-offs and mutual influences among the targets, capture the multi-objective equilibrium solution, and obtain the Pareto front equilibrium solution set;

[0032] The optimization solution selection submodule evaluates the feasibility and benefits of differentiated solutions based on the Pareto front equilibrium solution set, selects the optimization solution that matches the actual application in combination with the actual needs and constraints of industrial production, and obtains the multi-objective optimization result.

[0033] The present invention is improved in that the Pareto frontier analysis method is according to the formula:

[0034] ;

[0035] Calculate the target values ​​of minimizing cost, maximizing energy efficiency and minimizing environmental impact, where: , , The usage of various energy types, , , are the weight coefficients of cost, energy efficiency and environmental impact, , , are the benchmark values ​​for cost, energy efficiency and environmental impact, Function calculation considers the cost weight and the cost optimization target value of the benchmark. Function calculation considers energy efficiency weights and benchmark energy efficiency optimization target values. The function calculates the environmental optimization target value considering the environmental impact weight and benchmark.

[0036] Compared with the prior art, the advantages and positive effects of the present invention are:

[0037] In the present invention, by using historical energy consumption data to accurately predict future demand, the foresight and accuracy of energy distribution are improved, and by optimizing the supply strategy, the stability and economy of energy supply are ensured. A linear regression model is used to automatically adjust the energy use weight, thereby enhancing the adaptability of industrial production to energy market price fluctuations and environmental changes. The Pareto frontier analysis method is used to balance cost, energy efficiency and environmental impact goals, thereby achieving multi-objective optimization of energy management. It can respond quickly according to real-time data, optimize energy distribution, improve energy utilization efficiency, reduce energy waste, and reduce operating costs. At the same time, it supports sustainable development goals and provides a more flexible, economical and environmentally friendly energy management solution for industrial production. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 The module diagram of the adaptive industrial control energy management and scheduling system proposed by the present invention;

[0039] Figure 2 The system framework diagram of the adaptive industrial control energy management and scheduling system proposed by the present invention;

[0040] Figure 3 A schematic diagram of an energy demand prediction module in an adaptive industrial control energy management and scheduling system is proposed for the present invention;

[0041] Figure 4 A schematic diagram of an energy supply strategy module in an adaptive industrial control energy management and scheduling system is proposed for the present invention;

[0042] Figure 5 A schematic diagram of a dynamic weight adjustment module in an adaptive industrial control energy management and scheduling system is proposed by the present invention;

[0043] Figure 6 The present invention proposes a schematic diagram of a multi-period allocation module in an adaptive industrial control energy management and scheduling system;

[0044] Figure 7 A schematic diagram of a multi-objective optimization strategy module in an adaptive industrial control energy management and scheduling system proposed by the present invention;

[0045] Figure 8A schematic diagram of a real-time adjustment feedback module in an adaptive industrial control energy management and scheduling system is provided for the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0047] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating positions or positional relationships, are based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0048] Example

[0049] See also Figure 1 ,The present invention provides a technical solution: an adaptive industrial control energy management and scheduling system, the system includes an energy demand prediction module, an energy supply strategy module, a dynamic weight adjustment module, a multi-period allocation module, a multi-objective optimization strategy module, and a real-time adjustment feedback module;

[0050] The energy demand forecasting module is based on the historical energy consumption data of industrial production, analyzes the seasonal changes of multiple types of energy consumption, historical energy consumption trends and energy use patterns in differentiated time periods, and predicts the energy demand of future work shifts and production cycles to obtain energy demand forecast results;

[0051] The energy supply strategy module analyzes the applicability and cost-effectiveness of differentiated energy sources for industrial production processes based on energy demand forecast results, evaluates the supply situation and cost changes in the current energy market, and optimizes the supply strategy of multiple types of energy, including the forecast of supply volume and the calculation of cost-effectiveness, to obtain the supply strategy optimization plan;

[0052] The dynamic weight adjustment module is based on the real-time energy usage data of industrial production, combined with the fluctuation of energy demand in the industrial production process, the uncertainty of external energy market prices and environmental impact, and uses a linear regression model to automatically adjust the weights of multiple types of energy in the production process to obtain dynamically adjusted weight coefficients;

[0053] The multi-period allocation module is based on dynamically adjusted weight coefficients and supply strategy optimization schemes. By solving nonlinear programming problems, it adjusts the allocation of differentiated energy types in multiple time periods to obtain an optimized energy allocation strategy.

[0054] The multi-objective optimization strategy module is based on optimizing energy allocation strategy, combining cost control, energy efficiency improvement and environmental factors, using the Pareto frontier analysis method to capture the balance point between objectives and obtain multi-objective optimization results;

[0055] The real-time adjustment feedback module monitors energy usage and external environmental factors in real time based on the multi-objective optimization results, dynamically adjusts the energy allocation strategy in the industrial production process, responds to real-time changes in the industrial production process, and obtains real-time optimization adjustment feedback;

[0056] The energy demand forecast results include the predicted total energy demand, the demand ratio of multiple energy types, and the peak demand time period. The supply strategy optimization plan includes the supply volume, supply cost and supply schedule of multiple energy sources. The dynamically adjusted weight coefficients include energy cost weight, environmental impact weight, and supply stability weight. The optimized energy allocation strategy includes the allocation ratio of differentiated energy types in multiple time periods, priority ranking, expected cost savings and environmental benefit improvement information. The multi-objective optimization results include cost optimization plan, energy efficiency optimization plan, environmental impact optimization plan and trade-off solution. Real-time optimization adjustment feedback includes the adjusted energy allocation plan and deviation analysis information between expected and actual energy use.

[0057] See also Figure 2 and Figure 3 ,The energy demand prediction module includes a time series analysis submodule, a demand pattern recognition submodule, and a cycle prediction submodule;

[0058] The time series analysis submodule conducts time series analysis based on the historical energy consumption data of industrial production, analyzes the seasonal changes, historical trends and cyclical changes of the data, identifies the basic patterns and abnormal points of energy consumption, and generates the specific process of time series analysis results as follows;

[0059] The time series analysis submodule is based on the historical energy consumption data of industrial production. It adopts the autoregressive difference moving average model and uses the statsmodels library in Python to perform time series analysis on the data. The model parameters are configured to automatically identify seasonal differences and autoregressive terms. The lag order of the model is determined through ACF and PACF diagrams. The model is applied to analyze seasonal changes, historical trends, and cyclical changes, identify basic patterns and anomalies in energy consumption, and generate time series analysis results.

[0060] The demand pattern recognition submodule analyzes the energy usage patterns in differentiated time periods based on the time series analysis results, identifies the energy demand patterns in the target work shifts and production cycles, and generates the specific process of energy demand pattern recognition results as follows;

[0061] The demand pattern recognition submodule is based on the results of time series analysis and adopts the K-means clustering algorithm. Using the scikit-learn library in Python, the number of clusters is set to the optimal number determined according to the business logic. By calculating the Euclidean distance of energy usage data in different time periods, the data is divided into several categories, the energy demand patterns in different work shifts and production cycles are identified, and the energy demand pattern recognition results are generated.

[0062] The cycle prediction submodule predicts the energy demand of future work shifts and production cycles based on the energy demand pattern recognition results. The specific process of obtaining the energy demand prediction results is as follows;

[0063] The cycle prediction submodule adopts the random forest algorithm and the gradient boosting tree algorithm based on the results of energy demand pattern recognition. The scikit-learn library in Python is used to fuse the two algorithms. The number of trees in the random forest is set to 100, and the learning rate of the gradient boosting tree is set to 0.1. The prediction results of the two algorithms are integrated through the weighted average of the model to predict the energy demand of future work shifts and production cycles, and the energy demand prediction results are obtained.

[0064] See also Figure 2 and Figure 4 ,The energy supply strategy module includes a supply quantity analysis submodule, a cost-benefit evaluation submodule, and a strategy determination submodule;

[0065] The supply analysis submodule analyzes the supply capacity of multiple energy sources based on the energy demand forecast results, including evaluating the supply of differentiated energy sources, combining the seasonal changes and supply stability of multiple energy sources, identifying energy sources that can meet the forecast demand, and formulating matching supply plans for each energy source. The specific process of generating energy supply analysis results is as follows;

[0066] The supply analysis submodule analyzes the supply capacity of differentiated energy sources based on the energy demand forecast results. It uses the linear programming method and the PuLP library in Python to set the objective function of the linear programming model to maximize the supply that meets energy demand. The constraints include the maximum supply capacity of each energy source and seasonal variation factors. By solving the linear programming problem, it identifies the energy sources that can meet the forecast demand, and formulates a matching supply plan for each energy source to generate the energy supply analysis results.

[0067] The cost-benefit evaluation submodule calculates the supply cost and benefit of each energy source based on the energy supply analysis results. It combines various economic and environmental factors including energy prices, transportation costs, and environmental taxes and fees, and optimizes energy costs by comparing and analyzing the cost-benefit of energy supply plans. The specific process of generating energy supply benefit evaluation results is as follows:

[0068] The cost-benefit evaluation submodule calculates the supply cost and benefit of each energy source based on the energy supply analysis results. It adopts the cost-benefit analysis method, sets up the operation using pivot tables and formulas in Excel, collects data on economic and environmental factors including energy prices, transportation costs, and environmental taxes, and screens out the energy supply plan with the best cost and the lowest environmental impact by comparing and analyzing the cost-effectiveness of different energy supply plans, generating energy supply benefit evaluation results.

[0069] The strategy determination submodule plans and optimizes the supply strategy of differentiated energy sources based on the energy supply benefit evaluation results, matches and predicts energy demand, optimizes cost-effectiveness and supply stability, analyzes supply volume, supply cost and environmental impact, formulates energy supply strategy, and obtains the specific process of supply strategy optimization plan as follows;

[0070] The strategy determination submodule is based on the energy supply benefit evaluation results, adopts the mixed integer dynamic programming method, and uses the Gurobi optimizer to configure and solve in the Python environment. The model goal is set to optimize cost-effectiveness and supply stability while meeting the predicted energy demand. The constraints cover supply quantity, supply cost, and environmental impact restrictions. By adjusting the energy supply ratio and selecting economical and environmentally friendly energy types, an economical and environmentally friendly energy supply strategy is formulated, and a supply strategy optimization plan is obtained.

[0071] See also Figure 2 and Figure 5 ,The dynamic weight adjustment module includes a weight factor analysis submodule, an energy weight adjustment submodule, and a weight updating submodule;

[0072] The weight factor analysis submodule analyzes energy demand fluctuations, external energy market price uncertainty and environmental factors based on real-time energy usage data of industrial production, identifies key factors affecting the weights of multiple energy usage, and obtains the specific process of energy weight factor analysis results as follows;

[0073] The weight factor analysis submodule is based on the real-time energy usage data of industrial production and uses data analysis technology, specifically the Pandas library in Python for data collation and preliminary analysis. By calculating the standard deviation of energy demand fluctuations, the trend line of energy market price changes, and the impact of environmental policy changes on energy use, it conducts a comprehensive analysis of energy demand fluctuations, external energy market price uncertainties, and environmental influencing factors to generate energy weight factor analysis results.

[0074] The energy weight adjustment submodule sets the rules and parameters for weight adjustment based on the analysis results of energy weight factors, including energy price sensitivity and environmental impact weights. The specific process of generating the weight adjustment strategy is as follows;

[0075] The energy weight adjustment submodule is based on the analysis results of energy weight factors and adopts a decision tree algorithm. Through the scikit-learn library in Python, it builds a decision tree model to automatically formulate weight adjustment rules and parameters according to energy price sensitivity, environmental policy impact score and energy supply stability index. The depth of the decision tree is set to the optimal depth automatically determined according to the data set to generate an adaptive weight adjustment strategy.

[0076] The weight update submodule is based on the weight adjustment strategy, adopts a linear regression model, combines real-time energy consumption data, energy cost changes and the availability of renewable energy, updates the weights of multiple types of energy in industrial production, and obtains the specific process of dynamically adjusted weight coefficients.

[0077] The weight update submodule is based on an adaptive weight adjustment strategy. It combines real-time energy consumption data, energy cost changes, and the availability of renewable energy. It uses a linear regression model and the NumPy library in Python to update the weights of multiple energy types in industrial production. It calculates the weight adjustment coefficient of each energy type relative to cost changes and availability, determines the new weight value through linear regression analysis, and obtains a list of dynamically adjusted weight coefficients.

[0078] Linear regression model, according to the formula:

[0079] ;

[0080] Calculate the energy usage weight value and obtain the dynamically adjusted weight coefficient, where: is the new weight value, representing the updated energy usage weight, is the old weight value, representing the energy usage weight before the update, Energy cost changes reflect the impact of cost changes on energy use weights. For energy availability changes, indicate the impact of availability changes on energy use weights, Renewable energy ratio, showing the proportion of renewable energy in the total energy mix, calculated based on the proportion of renewable energy in total energy consumption. As a seasonal factor, the impact of seasonal changes on energy demand is considered. Based on the impact of seasonal changes on energy demand, it is concluded through historical data analysis. Time sensitivity reflects the urgency of energy demand in different time periods. , , , and It is an adjustment coefficient determined based on historical data and expert experience to balance the role of various influencing factors in weight updating.

[0081] The specific implementation process of the improved formula is as follows:

[0082] Based on the adaptive weight adjustment strategy, combined with real-time energy consumption data, energy cost changes and the availability of renewable energy, the linear regression model is implemented using the NumPy library. The calculation not only considers the direct impact of energy cost and availability, but also introduces the proportion of renewable energy. Reflects the proportion of renewable energy in the energy mix, seasonal factors to consider the impact of seasonal changes on energy demand, as well as time sensitivity To reflect the urgency of energy demand in different time periods, by analyzing historical data to determine , , , and The specific value of the new weight is determined by the linear regression model to ensure that the new weight value more accurately reflects the current energy usage and external environment changes, so as to achieve more refined energy management and allocation. , dynamically adjust the weights of various types of energy used in industrial production, obtain an economical and environmentally friendly energy use strategy, and ensure that energy allocation is both efficient and meets environmental protection requirements.

[0083] See also Figure 2 and Figure 6 ,The multi-period allocation module includes a constraint condition establishment submodule, a planning ,solution submodule, and an optimization solution generation submodule;

[0084] The constraint condition establishment submodule defines the constraint conditions of the nonlinear programming problem based on the dynamically adjusted weight coefficient and the supply strategy optimization scheme. The constraint conditions include the balance of energy supply and demand and the feasibility of the allocation scheme. The specific process of generating the constraint condition set is as follows;

[0085] The constraint establishment submodule is based on dynamically adjusted weight coefficients and supply strategy optimization schemes, uses linear programming methods, and adopts the PuLP library in Python to define the constraints of nonlinear programming problems. The conditions are generated by setting energy supply and demand balance constraints and feasibility constraints of allocation schemes, including the maximum supply limit of each energy, minimum usage requirements, and supply and demand matching rules within a time period, to ensure that all formulated constraints meet both the continuity requirements of industrial production and the actual situation of energy supply, thereby generating a constraint set.

[0086] The planning and solving submodule solves the nonlinear planning problem of the allocation of multiple energy types in differentiated time periods based on the set of constraints, combined with the goals of cost and environmental impact, and generates the specific process of the basic energy allocation plan as follows;

[0087] The planning and solving submodule is based on a set of constraints, uses a mixed integer linear programming method, and adopts the Gurobi optimizer for solving. The objective function is set to minimize the cost and environmental impact. By defining the allocation variables of various energy types in differentiated time periods, the mixed integer linear programming algorithm is applied to find the solution that minimizes the cost and environmental impact under given constraints, ensuring that the solution process is efficient and accurate, and generating a basic energy allocation plan.

[0088] The optimization scheme generation submodule refines and optimizes the scheme based on the basic energy allocation scheme, adjusts the scheme to meet the needs of industrial production, and obtains the specific process of optimizing the energy allocation strategy as follows;

[0089] The optimization scheme generation submodule is based on the basic energy allocation scheme, adopts the sensitivity analysis method, and applies the SciPy library in Python to refine and adjust the optimization scheme. By analyzing the marginal costs and environmental benefits of different energy types, the energy proportion and usage time period in the original allocation scheme are adjusted. The optimization strategy takes into account the changing needs of industrial production and the dynamic changes of the energy market, ensuring that the optimization scheme not only meets the needs of industrial production but also has high cost-effectiveness and low environmental impact, thereby obtaining an optimized energy allocation strategy.

[0090] See also Figure 2 and Figure 7 ,The multi-objective optimization strategy module includes the optimization objective definition submodule, the Pareto front analysis submodule, and the optimization solution selection submodule;

[0091] The optimization target definition submodule analyzes and defines the multi-objective optimization targets based on the optimization energy allocation strategy. The targets include minimizing costs, maximizing energy efficiency, and minimizing environmental impact. It plans the energy use in the industrial production process and generates the specific process of the multi-objective optimization target set.

[0092] The optimization target definition submodule is based on the optimization energy allocation strategy and uses decision analysis methods. Through the ExcelSolver plug-in, it sets goals including minimizing costs, maximizing energy efficiency, and minimizing environmental impact. It uses existing energy usage data and expected goals to define and solve multi-objective optimization problems through Solver, sets target weights to balance the importance of different goals, and constructs an objective function that can reflect the energy usage optimization needs in the industrial production process, generating a multi-objective optimization target set.

[0093] The Pareto frontier analysis submodule is based on the multi-objective optimization target set. It applies the Pareto frontier analysis method to analyze the balance points between multiple optimization targets, analyze the trade-offs and mutual influences between targets, capture the multi-objective equilibrium solution, and obtain the specific process of the Pareto frontier equilibrium solution set.

[0094] The Pareto front analysis submodule is based on a multi-objective optimization target set and uses the Pareto front analysis method. It is implemented through a multi-objective optimization toolbox in MATLAB or Python, such as the PyMOO library. The algorithm is configured as NSGA-II for the goals of minimizing costs, maximizing energy efficiency, and minimizing environmental impact. It simulates multi-generation genetic processes by initializing the population and defining crossover and mutation operations, gradually approaching the Pareto frontier. In the process, a non-dominated solution set is maintained to ensure that the newly generated solution is added to the solution set if it does not dominate the existing solution. Through the iterative update solution process, the trade-off relationship between multiple objectives is effectively explored to obtain the Pareto frontier equilibrium solution set.

[0095] Pareto frontier analysis method, according to the formula:

[0096]

[0097] Calculate the target values ​​of minimizing cost, maximizing energy efficiency and minimizing environmental impact, where: , , The usage of various energy types, including electricity, natural gas, renewable energy, etc. , , are the weight coefficients of cost, energy efficiency and environmental impact, respectively, which are used to adjust the relative importance of each objective in the overall optimization process. , , The benchmark values ​​for cost, energy efficiency and environmental impact correspond to the target baselines of minimizing cost, maximizing energy efficiency and minimizing environmental impact. They can be set based on historical data, industry standards or predetermined targets. Function calculation considers the cost weight and the cost optimization target value of the benchmark. Function calculation considers energy efficiency weights and benchmark energy efficiency optimization target values. The function calculates the environmental optimization target value considering the environmental impact weight and benchmark.

[0098] The specific implementation process of the improved formula is as follows:

[0099] By initializing the decision variables , , , the variables represent the expected usage of different energy types, and set the weight coefficient for each optimization objective , , and benchmark value , , , coefficients and benchmark values ​​are determined based on historical data, expert opinions or industry standards. By applying the NSGA-II algorithm, the decision variables are iteratively updated, and crossover and mutation operations are performed simultaneously to simulate the genetic process to approach the Pareto frontier of multi-objective optimization. In each iteration, the optimized value of each objective is calculated based on the current weight coefficient and benchmark value to ensure that the solution set gradually approaches the optimal Pareto frontier and generates a non-dominated solution set, which represents a set of solutions that can simultaneously meet the optimization goals of cost, energy efficiency and environmental impact under given weights and benchmark values.

[0100] The optimization solution selection submodule evaluates the feasibility and benefits of differentiated solutions based on the Pareto frontier equilibrium solution set, and selects the optimization solution that matches the actual application in combination with the actual needs and constraints of industrial production. The specific process of obtaining the multi-objective optimization results is as follows;

[0101] The optimization solution selection submodule is based on the Pareto front equilibrium solution set and uses the simulated annealing algorithm. It is optimized through the SimAnneal library in Python, sets the initial temperature and cooling rate to simulate the annealing process, evaluates the feasibility and benefits of differentiated solutions, and finds the optimization solution that can meet the actual needs of industrial production and achieve the best balance between cost, energy efficiency and environmental impact through continuous probabilistic search. The constraints and target priorities of actual industrial production are taken into account to obtain multi-objective optimization results.

[0102] See also Figure 2 and Figure 8 ,The real-time adjustment feedback module includes a real-time monitoring submodule, an adjustment strategy formulation submodule, and an execution feedback update submodule;

[0103] The real-time monitoring submodule monitors the energy usage and external environmental factors in the industrial production process in real time based on the multi-objective optimization results, including energy consumption, energy supply status and market price changes. The specific process of generating the real-time monitoring data set is as follows;

[0104] The real-time monitoring submodule is based on the results of multi-objective optimization, uses the Internet of Things technology and time series database, and adopts the MQTT protocol and InfluxDB to collect and store real-time data. It monitors energy usage and external environmental factors in the industrial production process, including deploying sensors to collect energy consumption, using energy management systems to obtain energy supply conditions, and using API interfaces to capture market price changes. By setting the data collection frequency and storage structure, the real-time and accuracy of the data is ensured, and a real-time monitoring data set is generated.

[0105] The adjustment strategy formulation submodule implements multi-objective optimization results based on real-time monitoring data sets, dynamically adjusts energy allocation in the industrial production process, matches real-time changes in the industrial production process, and obtains the specific process of dynamic adjustment strategy as follows;

[0106] The adjustment strategy formulation submodule is based on the real-time monitoring data set, applies the dynamic programming algorithm, and uses the SciPy library in Python to dynamically adjust the energy allocation strategy according to the real-time data, taking into account the real-time changes in energy consumption, energy supply status, and market price fluctuations. Through the algorithm optimization process, the initial state, decision rules, and objective function are set to match the real-time changes in the industrial production process, ensuring that the adjustment strategy is consistent with the multi-objective optimization results and adapts to real-time changes, thereby obtaining a dynamic adjustment strategy.

[0107] The execution feedback update submodule updates the energy allocation strategy in the industrial production process in real time based on the dynamic adjustment strategy, adjusts the usage ratio and allocation plan of differentiated energy types, optimizes the effectiveness and flexibility of the energy allocation strategy, and obtains the specific process of real-time optimization adjustment feedback as follows;

[0108] The execution feedback update submodule is based on dynamic adjustment strategies, uses rule engines and script automation tools, and adopts Node-RED for process control. It updates the energy allocation strategy in the industrial production process in real time according to the dynamic adjustment strategy, adjusts the usage ratio and allocation plan of differentiated energy types, and achieves rapid response and adjustment of energy allocation by writing automation rules and scripts, ensuring that the optimized energy allocation strategy can be reflected in production activities in real time, improving the efficiency and flexibility of energy utilization, and obtaining real-time optimization and adjustment feedback.

[0109] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. Adaptive industrial control energy management and scheduling system, characterized by: The system comprises: The energy demand forecasting module predicts the energy demand for future work shifts and production cycles based on the historical energy consumption data of industrial production, and obtains the energy demand forecast results; The energy supply strategy module optimizes the supply strategies of multiple types of energy based on the energy demand forecast results to obtain a supply strategy optimization plan; The dynamic weight adjustment module is based on the real-time energy usage data of industrial production, combined with the fluctuation of energy demand in the industrial production process, the uncertainty of external energy market prices and environmental impact, and uses a linear regression model to automatically adjust the weights of multiple types of energy in the production process to obtain dynamically adjusted weight coefficients; The dynamic weight adjustment module includes: The weight factor analysis submodule analyzes energy demand fluctuations, external energy market price uncertainty and environmental factors based on real-time energy usage data of industrial production, identifies key factors affecting the weights of multiple energy usage, and obtains energy weight factor analysis results; The energy weight adjustment submodule sets the rules and parameters for weight adjustment based on the analysis results of the energy weight factors, including energy price sensitivity and environmental impact weight, and generates a weight adjustment strategy; The weight update submodule updates the weights of multiple types of energy in industrial production based on the weight adjustment strategy, adopts a linear regression model, combines real-time energy consumption data, energy cost changes, and the availability of renewable energy, and obtains dynamically adjusted weight coefficients; The linear regression model is based on the formula: ； Calculate the energy usage weight value and obtain the dynamically adjusted weight coefficient, where: is the new weight value, is the old weight value, Due to changes in energy costs, For changes in energy availability, is the proportion of renewable energy, For seasonal factors, For time sensitivity, , , , and is the adjustment factor; The multi-period allocation module adjusts the allocation of differentiated energy types in multiple time periods based on the dynamically adjusted weight coefficients and the supply strategy optimization scheme to obtain an optimized energy allocation strategy; The multi-period allocation module comprises: The constraint condition establishment submodule defines the constraint conditions of the nonlinear programming problem based on the dynamically adjusted weight coefficient and the supply strategy optimization scheme, the constraint conditions including the feasibility of the energy supply and demand balance and the allocation scheme, and generates a constraint condition set; The planning and solving submodule solves the nonlinear planning problem of the allocation of multiple energy types in differentiated time periods based on the set of constraints and in combination with the objectives of cost and environmental impact, and generates a basic energy allocation plan; The optimization scheme generation submodule refines and optimizes the scheme based on the basic energy allocation scheme, adjusts the scheme to meet the industrial production needs, and obtains an optimized energy allocation strategy; The multi-objective optimization strategy module is based on the above-mentioned optimized energy allocation strategy, combines cost control, energy efficiency improvement and environmental factors, adopts the Pareto frontier analysis method, captures the balance point between objectives, and obtains the multi-objective optimization results; The multi-objective optimization strategy module includes: The optimization target definition submodule analyzes and defines the targets of multi-objective optimization based on the optimization energy allocation strategy, including minimizing cost, maximizing energy efficiency and minimizing environmental impact, plans the energy use in the industrial production process, and generates a multi-objective optimization target set; The Pareto front analysis submodule applies the Pareto front analysis method based on the multi-objective optimization target set to analyze the balance points among multiple optimization targets, the trade-offs and mutual influences among the targets, capture the multi-objective equilibrium solution, and obtain the Pareto front equilibrium solution set; The optimization solution selection submodule evaluates the feasibility and benefits of differentiated solutions based on the Pareto front equilibrium solution set, selects the optimization solution that matches the actual application in combination with the actual needs and constraints of industrial production, and obtains the multi-objective optimization result; The Pareto frontier analysis method is based on the formula: ； Calculate the target values ​​of minimizing cost, maximizing energy efficiency and minimizing environmental impact, where: , , The usage of various energy types, , , are the weight coefficients of cost, energy efficiency and environmental impact, , , are the benchmark values ​​for cost, energy efficiency and environmental impact, Function calculation considers the cost weight and the cost optimization target value of the benchmark. Function calculation considers energy efficiency weights and benchmark energy efficiency optimization target values. The function calculates the environmental optimization target value considering the environmental impact weight and benchmark; The real-time adjustment feedback module dynamically adjusts the energy allocation strategy in the industrial production process based on the multi-objective optimization results to obtain real-time optimization adjustment feedback.

2. The adaptive industrial control energy management and scheduling system according to claim 1, characterized in that: The energy demand forecast results include the predicted total energy demand, the demand ratio of multiple energy types, and the peak demand time period; the supply strategy optimization plan includes the supply volume, supply cost and supply schedule of multiple energy sources; the dynamically adjusted weight coefficients include energy cost weight, environmental impact weight, and supply stability weight; the optimized energy allocation strategy includes the allocation ratio of differentiated energy types in multiple time periods, priority ranking, expected cost savings and environmental benefit improvement information; the multi-objective optimization results include cost optimization plan, energy efficiency optimization plan, environmental impact optimization plan and trade-off solution; the real-time optimization adjustment feedback includes the adjusted energy allocation plan and deviation analysis information between expected and actual energy usage.

3. The adaptive industrial control energy management and scheduling system according to claim 1, characterized in that: The energy supply strategy module includes: The supply analysis submodule analyzes the supply capacity of multiple energy sources based on the energy demand forecast results, including evaluating the supply of differentiated energy sources, combining the seasonal changes and supply stability of multiple energy sources, identifying energy sources that can meet the forecast demand, and formulating a matching supply plan for each energy source to generate energy supply analysis results; The cost-benefit evaluation submodule calculates the supply cost and benefit of each energy source based on the energy supply analysis results, combines various economic and environmental factors including energy prices, transportation costs, and environmental taxes, and optimizes energy costs by comparing and analyzing the cost-benefit of energy supply plans to generate energy supply benefit evaluation results; Based on the energy supply benefit evaluation results, the strategy determination submodule plans and optimizes the supply strategy of differentiated energy sources, matches the predicted energy demand, optimizes cost-effectiveness and supply stability, analyzes the supply volume, supply cost and environmental impact, formulates energy supply strategy, and obtains the supply strategy optimization plan.

Images