Industrial energy consumption collaborative management and control method based on dynamic carbon cost mapping and stochastic optimization

Abstract

The invention discloses an industrial energy consumption collaborative management and control method based on dynamic carbon cost mapping and stochastic optimization, and belongs to the technical field of industrial energy management. The method comprises the following steps: acquiring operation power data of industrial equipment, a power grid carbon emission factor and a carbon transaction price in real time; constructing a target function by taking the minimum total cost as a target, and carrying out optimization solution on the target function by utilizing an improved alternating direction multiplier method; establishing a multi-dimensional stability index system and a stability scoring function, and performing simulation test and stability judgment according to the optimal decision variable; a multi-target collaborative management and control scheme is constructed in combination with real-time collected data, and a control instruction is generated; and the control instruction is remotely issued to each execution equipment end, and the execution effect is continuously monitored, so that dynamic collaborative management and control are realized. The method aims at solving the problems that in traditional energy consumption, the dimensions of carbon emission cost and economic cost are not unified, and multi-target cooperation is difficult, and the control level of industrial energy consumption can be effectively improved.

Description

Technical Field

[0001] This invention relates to the field of industrial energy management technology, and in particular to a method for coordinated control of industrial energy consumption based on dynamic carbon cost mapping and stochastic optimization. Background Technology

[0002] In the field of industrial energy management, energy consumption control needs to balance carbon emission reduction and economic cost control. However, traditional industrial energy consumption control models often focus on optimizing single energy costs, failing to effectively integrate dynamic carbon costs. In existing methods, carbon costs and economic costs are difficult to map collaboratively due to differences in dimensions, and the inconsistent quality of multi-source data (such as power and carbon emission factors) leads to cost calculation biases. Furthermore, traditional static models cannot adapt to dynamic changes in the carbon market, have weak multi-objective collaborative optimization capabilities, lack stability verification mechanisms, and are prone to policy execution fluctuations. Therefore, there is an urgent need for an energy consumption control system that can achieve dynamic carbon cost mapping, multi-objective collaboration, and stability assurance. Summary of the Invention

[0003] Purpose of the invention: To address the above problems, the purpose of this invention is to provide a method for coordinated management and control of industrial energy consumption based on dynamic carbon cost mapping and stochastic optimization.

[0004] Technical solution: The first aspect of this invention, a method for coordinated industrial energy consumption management based on dynamic carbon cost mapping and stochastic optimization, includes the following steps:

[0005] Step 1: Collect real-time data on the operating power of industrial equipment, the carbon emission factor of the power grid, and the carbon trading price, and perform data preprocessing.

[0006] Step 2: Construct a dynamic mapping relationship for the preprocessed data, build an objective function with the goal of minimizing the total cost, and use the improved alternating direction multiplier method to optimize the objective function and obtain the optimal decision variables;

[0007] Step 3: Establish a multi-dimensional stability index system and stability scoring function, and conduct simulation tests and stability assessments based on the optimal decision variables;

[0008] Step 4: Based on the judgment results and combined with real-time collected data, construct a multi-target collaborative management and control scheme and generate control commands;

[0009] Step 5: Remotely send control commands to each execution device and continuously monitor the execution effect. By establishing a real-time feedback loop, trigger the re-execution of the optimized process, guide the system to dynamically adjust parameters, and achieve dynamic collaborative management and control.

[0010] Further, step 2 includes:

[0011] To establish a direct quantitative relationship between carbon costs and instantaneous carbon emissions, the formula is as follows:

[0012] ,

[0013] in, express Instantaneous carbon emissions at any given moment express Continuously run power data, express The carbon emission factor of the power grid at any given time;

[0014] Combined with carbon trading prices The formula for calculating carbon costs is:

[0015] ,

[0016] The constructed objective function is expressed as follows:

[0017] ,

[0018] in, A set of decision variables related to carbon costs; For the set of decision variables related to economic costs, This indicates taking the minimum value. Indicates carbon cost weighting. Indicates the economic cost weight. ; The economic cost function is expressed as follows:

[0019] ,

[0020] in, These are dynamic weighting coefficients, reflecting adjustment factors for economic costs; For energy prices; This refers to the operating power of the equipment.

[0021] Furthermore, the steps for optimizing the objective function using the improved alternating direction multiplier method include:

[0022] Set constraints to ensure that the optimization results meet equipment operating limitations and power grid dispatch requirements. The constraints are expressed as follows:

[0023] ,

[0024] in, and The coefficient matrix represents the carbon cost variables. With economic cost variables The coupling relationship, matrix Focusing on the constraint coefficients of carbon cost-related decision variables, the matrix Focus on the constraint coefficients of decision variables related to economic costs; constraint constant vector Physical limitations that cover the limits of equipment operation;

[0025] Introducing Lagrange multipliers and penalty items Constructing the augmented Lagrange function The constrained problem is transformed into an unconstrained optimization problem, which can be expressed as:

[0026] ,

[0027] The augmented Lagrangian function is solved using an alternating update mechanism until the convergence condition is met, resulting in an objective function. Minimize the optimal solution and use this optimal solution as the optimal decision variable.

[0028] Furthermore, the alternating update mechanism includes:

[0029] First, update the carbon cost variable using the following formula:

[0030] ,

[0031] in, Indicates the number of alternation iterations;

[0032] Then, the economic cost variable is updated using the following formula:

[0033] ,

[0034] Next, update the multipliers using the following formula:

[0035] ,

[0036] Calculate the residual norm and quantify the current solution. , The degree of deviation from the constraints is expressed as:

[0037] ,

[0038] The convergence condition is: .

[0039] Furthermore, the expression for the multi-dimensional stability index system in step 3 is as follows:

[0040] ,

[0041] in, For the system at time The overall stability index, The number of stability dimensions, For the first Each dimension at time The dynamic weights satisfy ; For the first Normalized stability scores for each dimension, with a value range of [0,1]. The definition is as follows:

[0042] The carbon cost stability score is expressed as follows: ,

[0043] The economic cost stability score is expressed as follows: ,

[0044] The algorithm's convergence stability score is represented as follows: ,

[0045] The equipment operation stability score is expressed as follows: ,

[0046] in, Indicates time Carbon cost optimization target value, It is a very small constant to prevent the denominator from being zero; Indicates time Economic cost optimization target value Let the initial residual norm be , This indicates the rated power of the equipment.

[0047] Furthermore, the expression for the stability scoring function in step 3 is:

[0048] ,

[0049] in, For the first The item strategy at time The dynamic weights reflect their importance in the overall strategy; For the first The real-time failure probability of the sub-strategy is calculated by combining historical data with real-time monitoring; The volatility of strategy returns represents the uncertainty of the synergy between the dual objectives of economic efficiency and carbon emission reduction. Robustness scoring is used to assess the policy's adaptability under perturbation conditions; This refers to global environmental factors.

[0050] Furthermore, step 3 includes:

[0051] An adaptive stability threshold is set, and a three-level response mechanism is triggered when the stability score falls below the threshold.

[0052] Furthermore, the rules that trigger the three-level response mechanism include:

[0053] When rating When the value is below the threshold, the first-level response is triggered first. If the stability score still fails to recover to above the threshold after the first-level response, the second-level response is triggered. If the second-level response still fails to restore the system to stability, the third-level response is triggered.

[0054] The first level of response involves adjusting the weighting of carbon costs and economic costs based on real-time data streams, i.e., adjusting the weighting coefficients. and ;

[0055] The second-level response involves dynamically adjusting the response by monitoring the original residuals during the iteration process. value;

[0056] The third-level response is to restart the data collection and preprocessing process.

[0057] Furthermore, the multi-objective collaborative management and control scheme in step 4 includes carbon cost optimization strategy, economic cost optimization strategy, dynamic weight allocation, stability verification results, and continuous real-time data stream.

[0058] The second aspect of this invention provides an industrial energy consumption collaborative management system based on dynamic carbon cost mapping and stochastic optimization, comprising:

[0059] The multi-source data sensing module is used to sense and collect data from the power sensor, carbon emission factor detection unit, and external data interface.

[0060] The data preprocessing module is used to clean and normalize the acquired data.

[0061] A dynamic mapping model of carbon cost and economic cost is used to accurately calculate costs;

[0062] The stability verification model is used to determine the stability of relevant situations and issue corresponding strategies.

[0063] The collaborative management and control module combines the sensing data and stability verification results to generate a collaborative management and control plan;

[0064] The intelligent decision-making center is used to realize remote monitoring functions and generate decision-making solutions.

[0065] Beneficial effects: Compared with the prior art, the significant advantages of this invention are:

[0066] 1. This invention introduces a dynamic penalty parameter adjustment mechanism through an improved alternating direction multiplier method, enabling the system to respond in real time to carbon market fluctuations and changes in equipment operating status. By solving the objective function using the ADMM algorithm, the relationship between carbon cost and economic cost can be accurately calculated based on real-time changes, thereby improving optimization speed.

[0067] 2. This invention ensures the reliability of the strategy under complex operating conditions through stability verification, generates a comprehensive multi-objective collaborative management and control scheme, realizes the balance and optimization of multiple objectives in industrial energy consumption management, and effectively improves the level of industrial energy consumption management.

[0068] 3. This invention achieves a balance between carbon emission reduction and economic costs through a multi-dimensional stability index system, improves collaborative efficiency, and reduces instruction execution deviation; it also reduces the probability of strategy failure and shortens abnormal response time through a stability scoring function, ensuring the stability of the system under disturbances.

[0069] 4. This invention can comprehensively collect multi-source data, solving the problem of incomplete data acquisition in traditional processes. Attached Figure Description

[0070] Figure 1 This is a flowchart of the present invention;

[0071] Figure 2 This is a comparison chart of cost mapping results;

[0072] Figure 3 Flowchart for the optimization of the improved ADMM algorithm;

[0073] Figure 4 A comparison chart of ADMM convergence performance;

[0074] Figure 5 Comparison of results for collaborative optimization of the target;

[0075] Figure 6 This is a comparison chart of stability verification results;

[0076] Figure 7 This is a structural block diagram of the industrial energy consumption collaborative management and control system of the present invention. Detailed Implementation

[0077] The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and not intended to limit the scope of the invention. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the embodiments of the present invention, and not all structures.

[0078] In the following description, specific details such as target system architecture and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of this application with unnecessary detail.

[0079] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0080] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0081] Furthermore, in the description of this application and the appended claims, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0082] References to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include the target features, structures, or characteristics described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0083] Example 1

[0084] The flowchart of the industrial energy consumption collaborative management method based on dynamic carbon cost mapping and stochastic optimization described in this embodiment is as follows: Figure 1 As shown, it includes the following steps:

[0085] Step 1: Collect real-time data on the operating power of industrial equipment, the carbon emission factor of the power grid, and the carbon trading price, and perform data preprocessing.

[0086] In one example, industrial-grade sensors are used with a sampling frequency of 1Hz to ensure real-time data. Power sensors are installed in key energy-consuming equipment (such as motors and compressors) to measure instantaneous power. The collected real-time operating power data of industrial equipment, grid carbon emission factors, and carbon trading prices are cleaned, anomaly detected, and Z-score standardized and normalized to remove invalid (such as power drop) or redundant data and improve data quality.

[0087] Step 2: Construct a dynamic mapping relationship for the preprocessed data, build an objective function with the goal of minimizing the total cost, and use the improved alternating direction multiplier method (ADMM algorithm) to optimize the objective function and obtain the optimal decision variables.

[0088] Based on preprocessed data, a dynamic mapping relationship between carbon cost and economic cost is constructed. An objective function is built with the goal of minimizing total cost. An improved ADMM algorithm is used to solve the multi-objective function, achieving synergistic optimization of carbon cost and economic cost. By introducing a dynamic penalty parameter adjustment mechanism into the traditional ADMM algorithm, combined with a stochastic optimization framework and stability verification closed-loop feedback, rapid convergence and dynamic environment adaptation of carbon cost and economic cost in multi-objective synergistic optimization are achieved.

[0089] Further, step 2 includes:

[0090] To establish a direct quantitative relationship between carbon costs and instantaneous carbon emissions, the formula is as follows:

[0091] ,

[0092] in, express Instantaneous carbon emissions at any given moment express Continuously run power data, express The carbon emission factor of the power grid at any given time;

[0093] Combined with carbon trading prices The formula for calculating carbon costs is:

[0094] ,

[0095] The objective function, which aims to minimize total cost, is expressed as follows:

[0096] ,

[0097] in, The set of decision variables related to carbon costs (such as emission reduction strategies); This is a set of decision variables related to economic costs (such as energy consumption scheduling). This indicates taking the minimum value. Indicates carbon cost weighting. Indicates the economic cost weight. During implementation, adjustments are made dynamically based on historical data training. The economic cost function (such as electricity price cost) is expressed as follows:

[0098] ,

[0099] in, These are dynamic weighting coefficients that reflect adjustment factors of economic costs (such as energy consumption dispatch preferences). Energy prices (such as electricity prices); This refers to the operating power of the equipment.

[0100] In one example, such as Figure 2 As shown, comparing the dynamic mapping relationship constructed in this embodiment with the traditional static model cost mapping relationship, the calculation shows that the uniform error of the traditional mapping relationship is about 12%, and the success rate is 70%-85%. By using the dynamic carbon cost-economic cost mapping relationship in this example, the uniform error is reduced to 0.3%, and the success rate of dual-cost collaborative optimization is increased to 97%. It can be seen that the dynamic mapping relationship constructed by the present invention significantly improves the accuracy.

[0101] Furthermore, the steps for optimizing the objective function using the improved alternating direction multiplier method include:

[0102] Set constraints to ensure that the optimization results meet equipment operating limitations and power grid dispatch requirements. The constraints are expressed as follows:

[0103] ,

[0104] in, and The coefficient matrix (e.g., equipment power constraints) represents the carbon cost variables. With economic cost variables The coupling relationship, matrix Focusing on the constraint coefficients of carbon cost-related decision variables, the matrix Focus on the constraint coefficients of decision variables related to economic costs; constraint constant vector (Such as maximum load) encompasses the physical limitations of the device's operating limits;

[0105] Introducing Lagrange multipliers and penalty items Constructing the augmented Lagrange function The constrained problem is transformed into an unconstrained optimization problem, which can be expressed as:

[0106] ,

[0107] The augmented Lagrangian function is solved using an alternating update mechanism until the convergence condition is met, resulting in an objective function. The optimal solution is minimized, and this optimal solution is used as the optimal decision variable. This optimal solution is the carbon cost variable. and economic cost variables The set of two algorithms, through alternating optimization using the improved ADMM algorithm, jointly minimizes the objective function. .

[0108] In one example, A and B are typically sparse matrices whose dimensions depend on the number of decision variables and the number of device constraints. For example, if It includes three carbon cost-related variables (such as emission reduction strategy options). If there are two economic cost variables (such as energy consumption dispatch options) and two constraints (such as equipment power limits and grid dispatch requirements), then matrix A can be a 2×3 matrix and matrix B can be a 2×2 matrix. Assume the equipment power limit is... The power grid requires carbon cost variables The sum does not exceed a certain threshold. Therefore, matrix A may take the form: Each row corresponds to a constraint, and each column corresponds to... The decision variables are shown in the first row (e.g., the first row represents the coefficients of each variable under the first constraint). Matrix B is similar, for example... Related economic cost variables .

[0109] It is a column vector, where each element corresponds to a boundary value of a constraint. For example, if there are two constraints (such as maximum power and minimum running time of the equipment), then... It can be a 2-dimensional vector: Assuming the maximum power of the device is If the power grid dispatch requires the total carbon cost to not exceed 200 units, then... .

[0110] Furthermore, the alternating update mechanism includes:

[0111] First, update the carbon cost variable using the following formula:

[0112] ,

[0113] in, Indicates the number of alternation iterations;

[0114] Then, the economic cost variable is updated using the following formula:

[0115] ,

[0116] Next, update the multipliers using the following formula:

[0117] ,

[0118] Calculate the residual norm and quantify the current solution. , The degree of deviation from the constraints is expressed as:

[0119] ,

[0120] The convergence condition is: .

[0121] This iterative process continues until the convergence condition is met, when the residual norm is below a threshold of less than 10⁻⁻⁶. 5 When the condition is met, it indicates that the solution fully satisfies the constraints, and the alternating optimization iteration terminates. Finally, the optimal solution that minimizes the objective function J is obtained, achieving synergistic minimization of carbon costs and economic costs, and improving the accuracy and reliability of industrial energy consumption management.

[0122] Industrial energy consumption optimization involves two objectives with different dimensions: carbon cost and economic cost, requiring a unified optimization framework. This example utilizes an improved ADMM algorithm, specifically referring to dynamically adjusting the penalty parameter. (Real-time monitoring based on residual norm), refined convergence condition settings (such as residual threshold) It also triggers a secondary response in conjunction with the stability verification model, and decomposes the original problem into multiple sub-problems through variable splitting technology.

[0123] Figure 3 The diagram shows the overall flowchart of the improved ADMM algorithm. First, relevant data is collected and preprocessed. Then, variables are set based on the relevant data. , , , for parameters , , Alternating updates and convergence checks, the algorithm iteratively converges to obtain the optimal solution. This algorithm decomposes the original problem into a carbon-cost subproblem (updating...). ) and the sub-problem of economic costs (updated) This approach achieves multi-objective collaborative optimization. Residuals and adjusted parameters are performed in each iteration to collaboratively update the multiple objectives. The problem of inconsistent cost dimensions in the carbon economy is addressed by solving subproblems in parallel. The solutions to the subproblems serve as intermediate solutions, which are then integrated into the final decision variables through multiplier updates, ensuring stable convergence of the subproblems during iteration.

[0124] In one example, the improved ADMM algorithm is compared with the original ADMM algorithm. Before the improvement, approximately 28 iterations were required for convergence, while after the improvement, only 11 iterations are needed, representing a 61% reduction in the number of iterations and a 152% improvement in convergence speed (achieving the target in 11 iterations). The residual achievement rate under fluctuating conditions is 99%, further demonstrating that the improved ADMM algorithm in this example has strong convergence guarantees. The comparison results are as follows... Figure 4 As shown.

[0125] Step 3: Establish a multi-dimensional stability index system and stability scoring function, and conduct simulation tests and stability assessments based on the optimal decision variables.

[0126] Furthermore, the expression for the multi-dimensional stability index system in step 3 is as follows:

[0127] ,

[0128] in, For the system at time The overall stability index, This represents the number of stability dimensions; in this example, it is 4. For the first Each dimension at time The dynamic weights satisfy ; For the first Normalized stability scores for each dimension, with a value range of [0,1]. The definition is as follows:

[0129] The carbon cost stability score is expressed as follows:

[0130] ,

[0131] The economic cost stability score is expressed as follows:

[0132] ,

[0133] The algorithm's convergence stability score is represented as follows:

[0134] ,

[0135] The equipment operation stability score is expressed as follows:

[0136] ,

[0137] in, Indicates time Carbon cost optimization target value, It is a very small constant to prevent the denominator from being zero; Indicates time Economic cost optimization target value Let the initial residual norm be , This indicates the rated power of the equipment. Used to quantify real-time carbon costs With the optimization target value The deviation of (ADMM optimal solution); Used to evaluate the economic cost function The actual value and The degree of deviation of the optimized value Used to measure the convergence stability of the optimization process; Let the initial residual norm be denoted as . Used for monitoring device power With the rated value The deviation reflects the physical operating state.

[0138] A multi-dimensional stability index system is used to calculate the system's stability at time t. The comprehensive stability index aims to integrate stability scores from multiple dimensions (such as carbon cost stability, economic cost stability, algorithm convergence stability, and equipment operation stability) to provide a global stability assessment indicator. It quantifies the overall stability of the system through weighted summation. Scoring from different dimensions (Value range [0,1]) are combined into a single index, facilitating rapid determination of the system state. For example, when A value close to 1 indicates a stable system; conversely, a value close to 1 indicates a stable system. If the value falls below the threshold, a stability response mechanism may be triggered. and stability score Both are used for stability assessment, but at different levels. As a high-level indicator, it can trigger a global response; As a base score, it is used to refine strategy adjustments.

[0139] Furthermore, the expression for the constructed stability scoring function is as follows:

[0140] ,

[0141] in, For the first The item strategy at time The dynamic weights reflect their importance in the overall strategy; For the first The real-time failure probability of the sub-strategy is calculated by combining historical data with real-time monitoring; The volatility of strategy returns represents the uncertainty of the synergy between the dual objectives of economic efficiency and carbon emission reduction. Robustness scoring is used to assess the policy's adaptability under perturbation conditions; This refers to global environmental factors.

[0142] Furthermore, step 3 includes:

[0143] Set adaptive stability threshold When rating Below At that time, a three-level response mechanism is triggered.

[0144] In one example, an adaptive stability threshold is set. The threshold is 0.8, and it is not fixed but has the ability to be adjusted to adapt to the real-time state of the system.

[0145] Furthermore, the rules that trigger the three-level response mechanism include:

[0146] When rating When the value is below the threshold, the first-level response is triggered first. If the stability score still fails to recover to above the threshold after the first-level response, the second-level response is triggered. If the second-level response still fails to restore the system to stability, the third-level response is triggered.

[0147] The first level of response involves adjusting the weighting of carbon costs and economic costs based on real-time data streams, i.e., adjusting the weighting coefficients. and For example, if carbon costs fluctuate significantly, the automatic increase... The proportion of carbon emissions reduction targets should be adjusted to strengthen carbon emission reduction goals; conversely, if economic cost pressures increase, adjustments should be prioritized. ;

[0148] The second-level response involves dynamically adjusting the response by monitoring the original residuals during the iteration process. The value should be increased if the convergence rate lags. To strengthen the penalty of constraints; if oscillations are significant, reduce the penalty. To improve stability;

[0149] The third level of response is to re-perform the data collection and preprocessing process, such as using a sliding window algorithm to remove invalid data points (e.g., power spikes), and reconstructing the data distribution through Z-score normalization.

[0150] Step 4: Based on the judgment results and combined with real-time collected data, construct a multi-target collaborative management and control scheme and generate control commands.

[0151] Furthermore, the multi-objective collaborative management and control scheme in step 4 includes carbon cost optimization strategy, economic cost optimization strategy, dynamic weight allocation, stability verification results, and continuous real-time data stream.

[0152] Step 5: Remotely send control commands to each execution device and continuously monitor the execution effect. By establishing a real-time feedback loop, trigger the re-execution of the optimized process, guide the system to dynamically adjust parameters, and achieve dynamic collaborative management and control.

[0153] The system receives dynamic data streams such as power data, grid carbon emission factors, and carbon trading prices, and simultaneously acquires and verifies optimization strategies. Through data alignment and time series fusion technologies, it constructs a unified, multi-dimensional decision-making data foundation, ensuring data consistency and real-time performance, and providing input guarantees for multi-objective collaborative optimization. This multi-dimensional, multi-source information is deeply integrated and centrally monitored to build a unified system operation status view, providing a comprehensive and consistent data foundation for decision-making.

[0154] Control commands are remotely sent to each execution device, and the execution effect is continuously monitored. By establishing a real-time feedback closed loop, when deviations from expectations or fluctuations in stability indicators are detected, it can proactively trigger the re-execution of optimization processes, guiding the system to dynamically adjust parameters, thereby achieving continuous self-optimization of energy consumption management and ensuring the long-term effectiveness of the system in dynamic environments.

[0155] Comparing the multi-objective collaborative optimization in this example with the traditional linear weighted method, the traditional linear weighted method has a collaborative efficiency of approximately 0.68 and an average instruction execution bias of 12.8%. The collaborative optimization method in this example improves the collaborative efficiency to 1.04, representing a 53% increase in multi-objective collaborative efficiency (reaching 1.04), while reducing the instruction execution bias to 3%. This demonstrates a significant efficiency gain in this example. The comparison results are as follows: Figure 5 As shown. In multi-objective optimization, the Pareto front represents a set of "non-dominated solutions," meaning that one objective cannot be improved without worsening another. Figure 5 In the diagram, the Pareto front illustrates the trade-off between carbon costs and economic costs, with points on the front representing the optimal equilibrium solution. The Pareto front validates the effectiveness of the multi-objective coordinated management scheme in this example, further demonstrating that the present invention achieves a dual optimization of carbon and economic costs through the improved ADMM algorithm.

[0156] In one example, the stability scoring function of this invention is compared with traditional stability scoring by incorporating a dynamic optimization mechanism. Traditional methods mostly focus on a single objective (such as optimizing only economic costs), neglecting the dynamic mapping of carbon costs, leading to inconsistent dimensions and difficulties in coordination. Furthermore, the lack of stability verification in traditional methods easily results in performance fluctuations. The comparison results are as follows: Figure 6 As shown, the probability of policy failure using the traditional method is approximately 24.17%, and the anomaly response time is 1 second. The stability scoring function in this example reduces the probability of policy failure to 5.8%, a reduction of 76%, and shortens the anomaly response time to 0.2 seconds.

[0157] Example 2

[0158] like Figure 7 As shown in this embodiment, the industrial energy consumption collaborative management and control system based on dynamic carbon cost mapping and stochastic optimization includes:

[0159] The multi-source data sensing module is used to sense and collect data from the power sensor, carbon emission factor detection unit, and external data interface.

[0160] The data preprocessing module is used to clean and normalize the acquired data.

[0161] A dynamic mapping model of carbon cost and economic cost is used to accurately calculate costs;

[0162] The stability verification model is used to determine the stability of relevant situations and issue corresponding strategies.

[0163] The collaborative management and control module combines the sensing data and stability verification results to generate a collaborative management and control plan;

[0164] The intelligent decision-making center is used to realize remote monitoring functions and generate decision-making solutions.

[0165] During the data acquisition phase, the multi-source data sensing module collects data and synchronously feeds it back to the data preprocessing module and the collaborative management and control module. During the optimization phase, the carbon cost-economic cost dynamic mapping model and the stability verification model perform dynamic mapping and stability verification on the data in sequence. During the decision execution phase, the collaborative management and control module outputs the management and control plan to the equipment end, and the intelligent decision center monitors and adjusts the execution of the plan.

[0166] This invention aims to construct a system integrating multi-source data processing, dynamic cost mapping, multi-objective collaborative optimization, and stability verification. Relying on stability verification and collaborative management models, it ensures the effectiveness of strategies and adaptability to multi-objective needs; an intelligent decision-making center enables remote monitoring and adaptive adjustment, helping industrial enterprises break through traditional management bottlenecks, improve the scientific, accurate, and collaborative nature of energy consumption management, and promote green, low-carbon, and sustainable industrial development.

Claims

1. A method for coordinated management and control of industrial energy consumption based on dynamic carbon cost mapping and stochastic optimization, characterized in that, Includes the following steps: Step 1: Collect real-time data on the operating power of industrial equipment, the carbon emission factor of the power grid, and the carbon trading price, and perform data preprocessing. Step 2: Construct a dynamic mapping relationship for the preprocessed data, build an objective function with the goal of minimizing the total cost, and use the improved alternating direction multiplier method to optimize the objective function and obtain the optimal decision variables; Step 3: Establish a multi-dimensional stability index system and stability scoring function, and conduct simulation tests and stability assessments based on the optimal decision variables; Step 4: Based on the judgment results and combined with real-time collected data, construct a multi-target collaborative management and control scheme and generate control commands; Step 5: Remotely send control commands to each execution device and continuously monitor the execution effect. By establishing a real-time feedback loop, trigger the re-execution of the optimized process, guide the system to dynamically adjust parameters, and achieve dynamic collaborative management and control.

2. The industrial energy consumption collaborative management method based on dynamic carbon cost mapping and stochastic optimization according to claim 1, characterized in that, Step 2 includes: To establish a direct quantitative relationship between carbon costs and instantaneous carbon emissions, the formula is as follows: ， in, express Instantaneous carbon emissions at any given moment express Continuously run power data, express The carbon emission factor of the power grid at any given time; Combined with carbon trading prices The formula for calculating carbon costs is: ， The constructed objective function is expressed as: ， in, A set of decision variables related to carbon costs; For the set of decision variables related to economic costs, This indicates taking the minimum value. Indicates carbon cost weighting. Indicates the economic cost weight. ; The economic cost function is expressed as: ， in, These are dynamic weighting coefficients, reflecting adjustment factors for economic costs; For energy prices; This refers to the operating power of the equipment.

3. The industrial energy consumption collaborative management method based on dynamic carbon cost mapping and stochastic optimization according to claim 2, characterized in that, The steps for optimizing the objective function using the improved alternating direction multiplier method include: Set constraints to ensure that the optimization results meet equipment operating limitations and power grid dispatch requirements. The constraints are expressed as follows: ， in, and The coefficient matrix represents the carbon cost variables. With economic cost variables The coupling relationship, matrix Focusing on the constraint coefficients of carbon cost-related decision variables, the matrix Focus on the constraint coefficients of decision variables related to economic costs; constraint constant vector Physical limitations that cover the limits of equipment operation; Introducing Lagrange multipliers and penalty items Constructing the augmented Lagrange function The constrained problem is transformed into an unconstrained optimization problem, which can be expressed as: ， The augmented Lagrangian function is solved using an alternating update mechanism until the convergence condition is met, resulting in the objective function. The optimal solution is minimized, and this optimal solution is used as the optimal decision variable.

4. The industrial energy consumption collaborative management method based on dynamic carbon cost mapping and stochastic optimization according to claim 1, characterized in that, Alternating update mechanisms include: First, update the carbon cost variable using the following formula: ， in, Indicates the number of alternation iterations; Then, the economic cost variable is updated using the following formula: ， Next, update the multipliers using the following formula: ， Calculate the residual norm and quantify the current solution. , The degree of deviation from the constraints is expressed as: ， The convergence condition is: .

5. The industrial energy consumption collaborative management method based on dynamic carbon cost mapping and stochastic optimization according to claim 1, characterized in that, The expression for the multi-dimensional stability index system in step 3 is: ， in, For the system at time The overall stability index, The number of stability dimensions, For the first Each dimension at time The dynamic weights satisfy ; For the first Normalized stability scores for each dimension, with a value range of [0,1]. The definition is as follows: The carbon cost stability score is expressed as follows: , The economic cost stability score is expressed as follows: , The algorithm's convergence stability score is represented as follows: , The equipment operation stability score is expressed as follows: , in, Indicates time Carbon cost optimization target value, It is a very small constant to prevent the denominator from being zero; Indicates time Economic cost optimization target value Let the initial residual norm be , This indicates the rated power of the equipment.

6. The industrial energy consumption collaborative management method based on dynamic carbon cost mapping and stochastic optimization according to claim 4, characterized in that, The expression for the stability scoring function in step 3 is: ， in, For the first The item strategy at time The dynamic weights reflect their importance in the overall strategy; For the first The real-time failure probability of the sub-strategy is calculated by combining historical data with real-time monitoring; The volatility of strategy returns represents the uncertainty of the synergy between the dual objectives of economic efficiency and carbon emission reduction. Robustness scoring is used to assess the policy's adaptability under perturbation conditions; This refers to global environmental factors.

7. The industrial energy consumption collaborative management method based on dynamic carbon cost mapping and stochastic optimization according to claim 6, characterized in that, Step 3 includes: An adaptive stability threshold is set, and a three-level response mechanism is triggered when the stability score falls below the threshold.

8. The industrial energy consumption collaborative management method based on dynamic carbon cost mapping and stochastic optimization according to claim 7, characterized in that, The rules that trigger the Level 3 response mechanism include: When rating When the value is below the threshold, the first-level response is triggered first. If the stability score still fails to recover to above the threshold after the first-level response, the second-level response is triggered. If the second-level response still fails to restore the system to stability, the third-level response is triggered. The first level of response involves adjusting the weighting of carbon costs and economic costs based on real-time data streams, i.e., adjusting the weighting coefficients. and ; The second-level response involves dynamically adjusting the response by monitoring the original residuals during the iteration process. value; The third-level response is to restart the data collection and preprocessing process.

9. The industrial energy consumption collaborative management method based on dynamic carbon cost mapping and stochastic optimization according to any one of claims 1 to 8, characterized in that, The multi-objective collaborative management and control scheme in step 4 includes carbon cost optimization strategy, economic cost optimization strategy, dynamic weight allocation, stability verification results, and continuous real-time data stream.

10. An industrial energy consumption collaborative management and control system based on dynamic carbon cost mapping and stochastic optimization, characterized in that, include: The multi-source data sensing module is used to sense and collect data from the power sensor, carbon emission factor detection unit, and external data interface. The data preprocessing module is used to clean and normalize the acquired data. A dynamic mapping model of carbon cost and economic cost is used to accurately calculate costs; The stability verification model is used to determine the stability of relevant situations and issue corresponding strategies. The collaborative management and control module combines the sensing data and stability verification results to generate a collaborative management and control plan; The intelligent decision-making center is used to realize remote monitoring functions and generate decision-making solutions.

Images