Industrial equipment operation complete and overall integrated intelligent price decision method

Abstract

The application belongs to the field of industrial intelligent decision-making, and particularly relates to an intelligent parity decision-making method for the overall fusion of industrial equipment operation completion. The method comprises the following steps: constructing a multi-energy flow topology coupling model of industrial equipment, and mapping heterogeneous energy data to a unified reference space; performing effective energy analysis based on the second law of thermodynamics to identify weak links of energy consumption; establishing a multi-dimensional carbon efficiency ratio evaluation system and a dynamic cost accounting model associated with a carbon trading market; using a multi-objective particle swarm optimization algorithm to perform topology optimization on the operation strategy to balance the contradiction among the lowest energy consumption, the highest efficiency and the smallest carbon consumption; and using a feedback loop to realize closed-loop intelligent control. The application eliminates energy data islands, makes environmental costs explicit, and realizes the Pareto optimality of the overall benefit of industrial equipment operation under the premise of ensuring production stability.

Description

Technical Field

[0001] This invention belongs to the field of industrial intelligent decision-making, specifically involving an intelligent price comparison decision-making method that integrates the operation completion of industrial equipment with the overall system. Background Technology

[0002] With the transformation of the global industrial energy structure and the advancement of low-carbon emission reduction goals, the efficient operation of industrial equipment and refined energy management have become key pathways to enhancing the core competitiveness of the manufacturing sector. Traditional industrial production systems are gradually shifting towards a multi-energy complementary model, aiming to build a more intelligent and green energy consumption evaluation system through real-time monitoring and analysis of equipment operating status. In this process, how to integrate multi-dimensional resources in the production process and achieve a leap from single-equipment operation to a systematic and integrated decision-making model is a key focus in the current field of industrial automation.

[0003] Comparative pricing decision-making methods for industrial equipment operation aim to find optimal cost control and resource allocation schemes in complex energy environments. This technical direction requires the system to be able to uniformly model heterogeneous energy sources such as electricity, natural gas, and heat, and to identify the best operating strategy in dynamically changing production scenarios through algorithmic logic. With the volatility of the energy market and increasingly stringent environmental regulations, the scope of comparative pricing decision-making has expanded from traditional economic cost accounting to a comprehensive assessment covering multiple dimensions such as environmental costs and energy conversion efficiency. Its core lies in establishing a collaborative optimization mechanism under multiple criteria.

[0004] Existing price comparison decision-making systems generally suffer from cost black-box problems. The lack of a deep equivalent conversion mechanism across energy media such as electricity, gas, and heat leads to data gaps during the integration of diverse energy flows. Traditional decision-making models rely excessively on a single monetary settlement dimension, failing to incorporate carbon efficiency indicators and link them to real-time carbon trading market prices, resulting in insufficient ability to capture green benefits. When dealing with complex thermodynamic processes and environmental chemical constraints, conventional optimization algorithms struggle to resolve nonlinear contradictions under multi-benchmark topology optimization, failing to achieve a dynamic balance between minimum energy consumption, maximum thermal efficiency, and minimum carbon consumption, thus limiting the overall operational efficiency of industrial equipment during the energy transition period. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent price comparison decision-making method that integrates the operation completion of industrial equipment with the overall system, which can solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A smart price comparison decision-making method for integrating the operation completion and overall status of industrial equipment, characterized in that the method includes the following steps: Step 1: Construct a multi-energy flow topology coupling model for industrial equipment. Real-time data on electricity, natural gas, heat, and carbon emissions are acquired through sensing units deployed on the production site. The collected heterogeneous energy data is then mapped to a unified energy reference space through a preset equivalent conversion protocol. Step 2: Perform an effective energy analysis based on the second law of thermodynamics to calculate the energy utilization quality of industrial equipment during operation, identify weak links in energy consumption within the system, and generate effective energy evaluation indicators that reflect energy conversion losses. Step 3: Establish a multi-dimensional carbon efficiency ratio decision evaluation system, convert thermal efficiency and electrical efficiency into standard carbon emission indicators, and link them to price fluctuation data of the external carbon trading market in real time to form a dynamic cost accounting model that is linked to green benefits. Step 4: Use a multi-objective particle swarm optimization algorithm to perform topology optimization on the operation strategy of industrial equipment. Under preset constraints, balance the nonlinear contradiction between energy consumption, thermal weighted efficiency and carbon consumption to generate the optimal operation decision instruction. Step 5: Monitor the decision execution results in real time, adjust the topology optimization parameters through feedback loops, and realize closed-loop intelligent control that integrates the operating status of industrial equipment with overall benefits.

[0007] Preferably, the construction of the multi-energy flow topology coupling model of industrial equipment in step one includes: Establish an interconnected topology network for the power supply system, gas transmission system, and steam thermal cycle system; Sensing units are configured at both the energy input and output ends. These sensing units include intelligent power meters, ultrasonic flow meters, infrared thermal imaging sensors, and online flue gas monitoring systems. All sensing units are connected to the industrial fieldbus, and a synchronous clock protocol is used to lock the sampling frequency of each sensor to the same reference frequency. The data preprocessing module performs noise reduction, smoothing, and missing value compensation on the collected raw current, voltage, flow rate, pressure, temperature, and flue gas component signals. The noise reduction process eliminates high-frequency harmonic interference from the power grid using a moving average filtering algorithm, and the missing value compensation process fills in the missing values ​​caused by momentary interruptions in sensor communication using linear interpolation technology.

[0008] Preferably, the equivalent conversion protocol includes: Establish a conversion coefficient matrix based on standard coal consumption or standard calorific value to map energy consumption of different dimensions to an energy reference space based on joules; Dynamic correction logic is introduced during the mapping process. For natural gas, the lower heating value of natural gas is corrected online based on real-time data from pressure and temperature sensors, expressed as: ; in, For actual heat output, Based on the calorific value, This is the pressure correction factor. This is a temperature correction factor; For electrical energy, the real-time load curve of the power grid is connected, and the coefficient for converting electrical energy into standard coal is dynamically adjusted according to the current power supply structure of the power grid. An energy quality factor is defined within the energy reference space to distinguish between high-level and low-level energy. Before making a price comparison decision, the enthalpy value under pressure is obtained by querying the built-in water vapor property table, and the latent heat of phase change is included in the calculation logic of the effective utilization rate of thermal energy.

[0009] Preferably, the effective energy analysis based on the second law of thermodynamics in step two includes: Perform a joint check of energy balance and effective energy balance, and obtain the effective energy efficiency of the system by calculating the ratio of the total effective energy of the input system to the useful work of the output system that is effectively utilized; When dealing with a thermodynamic system, the Carnot cycle efficiency is used to weight the heat at different temperature gradients. The formula for the effective energy weight of the heat source is: ; in The effective energy weight of the heat source, The absolute temperature of the environment. The absolute temperature of the heat source; The effective energy weight is used to characterize the quality degradation of heat energy during its transport through heat exchangers and pipelines or during the work process. The identification of weak points in energy consumption within the system is achieved by performing entropy production analysis on each energy conversion node.

[0010] Preferably, the entropy production analysis includes: The irreversible losses of each submodule during operation are calculated using the Clausius inequality principle; these irreversible losses are characterized by the entropy generation of the nodes, and the formula for node entropy generation is: ; in, This refers to the amount of node entropy generated. The total entropy flow at the node's exit. Total entropy flow at node inlet The heat exchanged between the node and the outside world. The absolute temperature of the boundary; The system converts the calculated entropy output into the effective energy evaluation index; when the effective energy loss of a node exceeds the preset loss threshold, the system triggers an early warning signal and marks the node as a key adjustment target on the topology map; For nodes marked as key adjustment targets, the system extracts their physical characteristic parameters and defines them as key variables in subsequent topology optimization.

[0011] Preferably, the multi-dimensional carbon efficiency ratio decision evaluation system introduces the carbon efficiency ratio as a decision weight coefficient; The carbon efficiency ratio is defined as the ratio of the economic value generated per unit time or the total useful work output by the system to the total carbon emissions generated per unit time. For fossil fuels, fuel consumption is calculated by using the complete combustion equation to inversely deduce fuel consumption based on the oxygen content, carbon monoxide concentration, and flue gas temperature in the exhaust gas, and then multiplied by the carbon dioxide emission factor; for electricity consumption, the indirect carbon emissions of electricity are calculated based on the real-time acquisition of the proportion of green electricity. By combining production planning and scheduling, order urgency, and product unit energy consumption standards, a dynamic carbon consumption quota is allocated to each piece of industrial equipment. When the real-time carbon consumption rate of the equipment exceeds the preset rate threshold, the decision-making system will perform an operation to search for a low-carbon operating path.

[0012] Preferably, the dynamic cost accounting model obtains the settlement price of the real-time carbon trading market through a network communication module; Using Monte Carlo simulation to predict the fluctuation distribution of energy prices and carbon prices over a future decision-making cycle; When carbon quota prices are on an upward trend, the weight of carbon efficiency ratio in the overall price ratio function is increased; when carbon prices are in a preset low range, the proportion of reducing direct energy procurement costs in the overall price ratio function is increased. The system introduces a risk value assessment mechanism, which uses Monte Carlo simulation results to assess the maximum economic loss of the decision scheme under a preset confidence level, and ensures that the maximum economic loss is within a controllable range; The calculation process of the standard carbon emission index also incorporates the nonlinear emission characteristics of industrial equipment under different load rates, establishes a correlation function between load rate and carbon emission intensity through historical data, and corrects the real-time emission data.

[0013] Preferably, the fitness function of the multi-objective particle swarm optimization algorithm includes total energy cost, overall system effective energy efficiency, and carbon emission cost calculated based on real-time carbon price, and each sub-indicator is fused through a preset weighted vector; The multi-objective particle swarm optimization algorithm follows velocity and position update formulas during the iteration process; the velocity update formula is: ; in, for Generation, First The updated velocity vector of each particle. for Generation, First The updated velocity vector of each particle. For inertial weights, As the first learning factor, As the second learning factor, and interval Random numbers within, for Generation, First The historical optimal position vector of each particle. for Generation, First The global optimal position vector of each particle. for Generation, First The current position vector of each particle; The position update formula is: ; for Generation, First The updated position vectors of each particle; The algorithm introduces a non-dominated sorting mechanism based on crowding distance, and selects a set of non-dominated solutions after each iteration according to the Pareto front theory.

[0014] Preferably, the topology optimization in step four includes: By adjusting the actuators, the energy flow paths of each process unit in the industrial production process are reconstructed in series and parallel logic; by adjusting the opening of the bypass valve, the frequency of the variable frequency drive, and the circulation flow of the heat recovery system, the energy flow direction inside the system is changed; in the cogeneration scenario, a game model based on Nash equilibrium is introduced to coordinate the conflict of interests between heat demanders and electricity demanders, and to find the balance point that maximizes the overall carbon efficiency ratio of the system. The preset constraints include equipment rated power limits, production process temperature window limits, and total energy supply contract limits. The algorithm processes constraints through a penalty function. If a candidate solution touches any constraint boundary, its fitness value is increased by a maximum penalty term. The system monitors the equipment's vibration, temperature, and cumulative operating hours and assesses the equipment's health index. This health index is used as a correction factor for the thermal efficiency model to prevent excessive wear on the equipment.

[0015] Preferably, the closed-loop intelligent control in step five includes: The decision deviation is calculated by comparing the expected energy efficiency index with the actual observation value of the sensor; when the decision deviation exceeds the preset allowable range, the self-learning mechanism is activated, and the energy conversion coefficient matrix and carbon emission prediction model are corrected by using a long short-term memory neural network model. The system integrates a multi-criteria expert system, and performs a secondary verification of the decision safety based on a historical case library after the algorithm generates the optimal solution; If the environmental parameters are in the historical high-risk range, the expert system will fine-tune the optimization results according to the preset safety operating procedures. The system displays a real-time dynamic topology diagram of energy flow, carbon flow, and cost flow through a visual interactive interface, and supports switching between green priority mode and cost priority mode. The system aggregates historical execution data and uses clustering analysis algorithms to identify typical optimization patterns under different environmental seasons, providing pre-set templates for future operation strategies.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. It solves the cost black box problem faced by industrial equipment during the energy transition period. By constructing a multi-source heterogeneous energy coupling model and equivalent conversion protocol, it realizes deep integration and equivalent price comparison of cross-energy media such as electricity, gas, and heat, and eliminates data silos between different energy dimensions.

[0017] 2. The introduction of effective energy analysis based on the second law of thermodynamics enables a profound insight into the actual losses within the system from the perspective of energy quality, transforming the basis for decision-making from the apparent quantity of energy to the deep value of energy, thereby enhancing the scientific nature of resource allocation.

[0018] 3. A multi-dimensional carbon efficiency ratio decision-making system linked to real-time carbon trading prices has been established, making environmental costs explicit and transforming them into green benefit indicators. This enables the price comparison decision-making of industrial equipment to not only focus on economic costs, but also to capture the dividends of the low-carbon market.

[0019] 4. By applying the energy flow multi-benchmark topology optimization algorithm with good carbon-electricity properties and using an improved multi-objective particle swarm optimization strategy, the complex nonlinear balance problem between minimum energy consumption, maximum thermal efficiency and minimum carbon consumption was successfully solved, achieving Pareto optimality of overall benefits while ensuring stable production.

[0020] 5. Through closed-loop feedback control and expert system verification throughout the entire process, the adaptability and reliability of the price comparison decision-making method are enhanced. It can cope with energy market fluctuations and process environment changes under complex dynamic operating conditions, providing a systematic decision support solution for green industrial production. This is a smart price comparison decision-making method that integrates industrial equipment operation with overall system integration. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the overall technical solution architecture of the present invention; Figure 2 This is a schematic diagram of the core principle framework of intelligent price comparison decision optimization based on multi-objective particle swarm optimization algorithm in this invention; Figure 3 This is a logical flowchart of the multi-energy flow topology coupling modeling and effective energy analysis and evaluation in this invention; Figure 4 This is a logical flowchart of the construction of multi-dimensional carbon efficiency ratio and dynamic cost accounting in this invention; Figure 5 This is a schematic diagram of the multi-level interaction relationship and data flow between decision execution feedback and closed-loop intelligent control in this invention. Detailed Implementation

[0022] Example 1: Please refer to the appendix Figure 1 To be continued Figure 5 To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to specific embodiments.

[0023] In the intelligent price comparison decision-making method that integrates industrial equipment operation with overall system integration, step one involves constructing a multi-energy flow topology coupling model for industrial equipment. This process begins with multi-dimensional modeling of the physical entities at the industrial production site. The system establishes an interconnected topology network for the power supply system, gas transmission system, and steam thermodynamic cycle system. High-precision sensing units are configured at each energy input and output end. These sensing units include, but are not limited to, intelligent power meters, ultrasonic flow meters, infrared thermal imaging sensors, and flue gas online monitoring systems. All sensing units are connected to the central control unit via industrial fieldbuses, such as industrial Ethernet, bus control networks, or wireless sensor networks. To ensure the alignment of multi-source heterogeneous data on the time scale, the system adopts a preset synchronization clock protocol, locking the sampling frequency of all sensors to the same reference frequency. The sampling period is configured differently according to the sensor type: the sampling period for intelligent power meters is 100ms, the sampling period for ultrasonic flow meters and temperature / pressure sensors is 1s, and the sampling period for the flue gas online monitoring system is 5s. All data is resampled and aligned to a unified 1s time reference through linear interpolation, ensuring the time synchronization of multi-source data. The data preprocessing module performs noise reduction, smoothing, and missing value compensation on the acquired raw current, voltage, flow, pressure, temperature, and flue gas composition signals. For example, it uses a moving average filtering algorithm to eliminate the interference of high-frequency harmonics from the power grid on metering accuracy, and uses linear interpolation technology to fill in missing values ​​caused by momentary interruptions in sensor communication.

[0024] Based on the established physical model, the system establishes an equivalent conversion protocol. The core of this protocol is the creation of a conversion coefficient matrix based on standard coal consumption or standard calorific value. Since electrical energy, natural gas, and thermal energy have different dimensions, the system maps them uniformly to an energy reference space based on joules. During the mapping process, the system uses not only static conversion coefficients but also dynamic correction logic. For natural gas, the system performs online correction of the lower heating value of natural gas based on real-time data from pressure and temperature sensors. The dynamic correction logic is executed as follows: At each synchronous sampling cycle, the operating absolute pressure and operating absolute temperature inside the natural gas pipeline are read; Calculate pressure and temperature correction factors based on baseline conditions (101.325 kPa, 293.15 K); The baseline low calorific value is converted into the actual calorific value under operating conditions by using a correction factor; Substituting the corrected actual calorific value into the conversion coefficient matrix completes the energy baseline mapping at that moment. The calculation formula logic is as follows: ; in, This refers to the actual volumetric heat generation under operating conditions. This represents the lower volumetric heat generation under baseline conditions. This is a dimensionless pressure correction factor; the calculation formula is: , For the absolute pressure under operating conditions, Reference absolute pressure; This is a dimensionless temperature correction factor; the calculation formula is: , The absolute temperature under operating conditions. The reference absolute temperature.

[0025] This formula conforms to the physical logic of the ideal gas law: when the pressure increases, the amount of natural gas of the same volume increases, and the heat of combustion increases, so the pressure correction factor is placed in the numerator; when the temperature increases, the amount of natural gas of the same volume decreases, and the heat of combustion decreases, so the temperature correction factor is placed in the denominator. The logical direction of the formula is consistent with the principles of thermodynamics.

[0026] For electrical energy, the system connects to the real-time load curve of the power grid and dynamically adjusts the coefficient for converting electrical energy into standard coal equivalent based on the current power supply structure of the grid. All energy consumption is ultimately converted into a unified rate index of megajoules per hour. Within this benchmark space, the system further defines an energy quality factor to distinguish between high-level energy sources such as electrical energy and low-level energy sources such as low-temperature waste heat. Before executing price comparison decisions, the system obtains the enthalpy value at a specific pressure by consulting the built-in water vapor property table, accurately incorporating the latent heat of phase change into the effective utilization rate calculation logic of thermal energy, eliminating energy balance errors caused by neglecting latent heat.

[0027] In the above method, step two performs an effective energy analysis based on the second law of thermodynamics. This stage is not limited to the quantity balance of energy, but delves into the quality of energy utilization. The system performs a joint check of energy balance and effective energy balance, calculating the ratio of the total effective energy input to the system to the useful work output to the system that is effectively utilized, thus obtaining the system's effective energy efficiency. When processing thermodynamic systems, the system uses Carnot cycle efficiency to weight the heat at different temperature gradients. The formula for the effective energy weight of the heat source is: ; in The effective energy weight of the heat source, The absolute temperature of the environment. This represents the absolute temperature of the heat source. In this way, the system accurately characterizes the quality degradation of heat energy during its transport through heat exchangers and pipelines or during its application in work.

[0028] To identify energy-inefficient bottlenecks, the system performs entropy production analysis on each energy conversion node. Using Clausius's inequality, the irreversible losses of each submodule during operation are calculated. The formula for node entropy production is: ; in, This refers to the amount of node entropy generated. The total entropy flow at the node's exit. Total entropy flow at node inlet The heat exchanged between the node and the outside world. The absolute temperature of the boundary.

[0029] The total entropy flow at the node inlet and outlet is calculated by multiplying the medium mass flow rate by the specific entropy, using the following formula: ; In the formula , These are the medium mass flow rates at the node inlet and outlet, respectively. , These are the specific entropy values ​​for the inlet and outlet media, respectively. The specific entropy of the steam / water working fluid is obtained by consulting the steam property table based on pressure and temperature. The specific entropy of the gaseous working fluid is calculated using the equation of state and the isobaric specific heat capacity. Electrical energy, as a high-grade ordered energy source, has its entropy flow treated as 0. The system converts the calculated entropy output into an efficiency evaluation index. When the efficiency loss of a specific node exceeds a preset loss threshold, the system automatically triggers an early warning signal and marks the node as a key adjustment target in orange or red on the topology map. For example, if heat loss in a section of steam pipe leads to a decrease in efficiency, the system will identify insulation failure or leakage risk at that location and use it as a key variable in subsequent topology optimization.

[0030] In the above method, step three establishes a multi-dimensional carbon efficiency ratio decision-making and evaluation system. The core of this system is the introduction of the carbon efficiency ratio as a decision-making weight coefficient. The carbon efficiency ratio is defined as: the economic value produced per unit time or the total useful work output by the system, divided by the total carbon emissions generated within that period. To achieve this calculation, the system applies principles of environmental chemistry. For fossil fuels such as natural gas, the system uses the complete combustion equation to inversely deduce the actual fuel consumption based on the oxygen content, carbon monoxide concentration, and flue gas temperature in the exhaust gas, and then multiplies it by the carbon dioxide emission factor. For electricity consumption, the system calculates the indirect carbon emissions of electricity based on the real-time acquired proportion of green electricity. The two are added together to form a full life-cycle carbon footprint model.

[0031] When calculating carbon emission indicators, the system fully considers the nonlinear emission characteristics of industrial equipment under different load rates. Through learning from historical data, the system establishes a correlation function between load rate and carbon emission intensity. When equipment operates under low or overload conditions, the carbon emissions per unit of output typically deviate from the standard value; the system corrects this based on real-time power data. Combining production planning and scheduling, the system allocates dynamic carbon consumption quotas to each piece of industrial equipment based on order urgency and product unit energy consumption standards. When the real-time carbon consumption rate of the equipment exceeds a preset threshold, the decision-making system automatically searches for alternative low-carbon operating paths.

[0032] The dynamic cost accounting model obtains settlement prices from the external carbon trading market in real time via a network communication module. The system uses Monte Carlo simulation from financial engineering to predict the fluctuation distribution of energy and carbon prices during future decision-making cycles. When carbon allowance prices are rising, the system automatically increases the weight of carbon efficiency ratio in the overall price comparison function; conversely, when carbon prices are low, the system focuses more on reducing energy procurement costs. To mitigate price volatility risk, the system also incorporates a risk value assessment mechanism to ensure that the maximum economic loss that a decision-making plan may result in is within the company's acceptable range at a 95% confidence level.

[0033] The system introduces a Value at Risk (VaR) assessment mechanism, using Monte Carlo simulation results to evaluate the maximum economic loss of a decision-making scheme under a pre-set confidence level. The specific calculation steps are as follows: Based on Monte Carlo simulation, 10,000 sets of random scenarios for energy and carbon prices within future decision-making cycles were generated; For each price scenario, calculate the total cost corresponding to the current decision-making option: Total cost = Energy procurement cost + Carbon emission cost - Carbon allowance surplus revenue; Sort the total cost of all scenarios from high to low, and take the quantile cost value corresponding to the 95% confidence level; Maximum economic loss = quantile cost value - average expected cost across all scenarios, representing the maximum loss that a decision might incur beyond expectations under a pre-set confidence level.

[0034] And ensure that the maximum economic loss is kept under control; In the above method, step four employs a multi-objective particle swarm optimization algorithm to perform topology optimization on the operation strategy of industrial equipment. Initially, the algorithm randomly generates a swarm of particles within a preset operating parameter space. Each particle represents a specific combination of operating parameters, such as inverter frequency, valve opening, and gas flow setpoint. The algorithm's fitness function is designed as a comprehensive index comprising three sub-items: the first is the total energy cost, including electricity and gas purchase costs; the second is the overall system efficiency after weighting; and the third is the carbon emission cost calculated based on real-time carbon prices. These sub-items are fused using a preset weighted vector, transforming the complex multi-objective problem into a process of minimizing the comprehensive evaluation value.

[0035] During the particle swarm iteration process, the trajectory of each particle follows a strict logic for velocity and position updates. The velocity update formula is: ; in, for Generation, First The updated velocity vector of each particle. for Generation, First The updated velocity vector of each particle. For inertial weights, As the first learning factor, As the second learning factor, and interval Random numbers within, for Generation, First The historical optimal position vector of each particle. for Generation, First The global optimal position vector of each particle. for Generation, First The current position vector of each particle; The position update formula is: ; for Generation, First The updated position vectors of each particle. To ensure the search process does not get trapped in local optima, the algorithm has an adaptive mutation feature. When the swarm fails to find a better Pareto solution in 20 consecutive iterations, the system randomly selects 10% of the particles to undergo position jumps, allowing them to search again in a wider parameter space.

[0036] The constraints during the optimization process are extremely stringent, including equipment rated power limits, temperature window limits required for the production process, and total energy supply contract limits. The algorithm handles these constraints through a penalty function: if a candidate solution touches any constraint boundary, the system adds a significant penalty term to its fitness value, automatically eliminating the solution in the next round of selection. Topology optimization also involves refactoring the process flow logic. The system changes the energy flow direction within the system by adjusting the circulation flow of bypass valves and the heat recovery system. For example, in a combined heat and power (CHP) scenario, the system introduces a Nash equilibrium-based game theory model to coordinate the interests of heat demanders and electricity demanders, finding a balance point that maximizes the overall carbon efficiency while satisfying the needs of both. During the optimization process, the system monitors equipment vibration, temperature, and cumulative operating hours and assesses the equipment health index. The weighted comprehensive evaluation method is used for calculation, and the formula is as follows: ; In the formula: This represents the effective value of the equipment vibration velocity. The maximum permissible vibration velocity under rated operating conditions; This represents the difference between the measured temperature rise and the reference temperature rise of the key component. The maximum allowable temperature rise; H represents the cumulative operating hours of the equipment. Design the mean time between failures (MTBF) for the equipment; For the weighting coefficients, satisfying Typical values ​​are 0.4, 0.3, and 0.3. The health index ranges from 0 to 1, with values ​​closer to 1 indicating better equipment condition. The thermal efficiency correction factor is the product of the rated thermal efficiency and the health index. The equipment health index is used as a correction factor in the thermal efficiency model to prevent excessive equipment wear.

[0037] After the search is completed, the system uses fuzzy multi-criteria decision analysis to select the final operating scheme from the Pareto solution set. The fuzzy multi-criteria decision analysis method employs fuzzy TOPSIS (Approximation of Ideal Solution Ranking), and the execution steps are as follows: Construct a decision matrix, with each Pareto solution as the evaluation object and total energy cost, effective energy efficiency, and carbon emission cost as evaluation indicators; The weights of the indicators are fuzzified using triangular fuzzy numbers, and the fuzzy values ​​of the weights are determined by combining expert experience. The normalized decision matrix is ​​used to determine the fuzzy positive ideal solution and the fuzzy negative ideal solution. Calculate the Euclidean distance from each scheme to the positive and negative ideal solutions to obtain the proximity coefficient; The scheme with the highest similarity coefficient will be selected as the final running scheme.

[0038] The operational decision commands generated by this scheme are distributed to various actuators via an industrial control network. These actuators include variable frequency drives, electric regulating valves, gas proportional valves, and charge / discharge controllers for the energy storage system. The response delay time of the actuators is also incorporated into the model as a reference for calculating decision lead time.

[0039] In the above method, step five monitors the decision execution results in real time and achieves closed-loop intelligent control through a feedback loop. The system calculates the decision deviation by comparing the expected energy efficiency indicators (such as the expected percentage increase in effective energy) with the actual observations from the sensors. When the decision deviation exceeds the allowable range of 5%, the system activates a self-learning mechanism. The system utilizes a long short-term memory neural network model, using historical operating data from the past 24 hours as input, to retrain and correct the energy conversion coefficient matrix and carbon emission prediction model, dynamically improving the accuracy of the decision.

[0040] To ensure system reliability, a multi-criteria expert system is integrated. After the algorithm generates the optimal solution, the expert system performs a secondary verification of the decision's safety based on a historical operational case library. If current environmental parameters (such as abnormal cooling water temperature caused by extreme weather) are in a historically high-risk range, the expert system will intervene and fine-tune the optimization results according to preset safety operating procedures to ensure that the production system does not trip or overheat. The system also provides a visual interactive interface, displaying real-time dynamic topology diagrams of energy flow, carbon flow, and cost flow. Managers can freely switch between green-priority mode and cost-priority mode. By aggregating long-term execution data, the system uses clustering analysis algorithms to identify typical optimization patterns under different seasons and raw material characteristics, providing pre-set control templates for future operational strategies. This method breaks down information barriers between the power and fuel systems, providing comprehensive decision support for enterprises.

[0041] Example 2: In some large chemical industrial parks or comprehensive industrial clusters, the intelligent price comparison decision-making method integrating industrial equipment operation with the overall system demonstrates its scalability through multi-level topology. In this example, the constructed multi-energy flow topology coupling model for industrial equipment is no longer limited to a single device, but encompasses multiple production workshops, distributed energy stations, rooftop distributed photovoltaic systems, and external microgrids. The topology model forms a large-scale graph structure by defining energy interfaces between various subsystems. Under this structure, the equivalent transformation protocol in step one is further upgraded to a distributed computing framework. The sensing units of each sub-workshop perform denoising and normalization of data locally, and upload the aggregated energy feature vector to the park-level decision-making center.

[0042] In the second step of the efficiency analysis, the system introduces the concept of regional entropy production balance. Besides focusing on the losses of individual equipment, it also emphasizes the devaluation of energy when it is cascaded between different workshops. For example, if high-temperature exhaust gas from one workshop is used as a preheating heat source for another, the system identifies thermodynamic imperfections in the pipeline by calculating the efficiency recovery rate of this process. The efficiency evaluation index is linked to the production load plan. When the entire park is in a high-output cycle, the system automatically lowers the warning threshold for low-grade energy recovery nodes in exchange for higher production throughput.

[0043] The multi-dimensional carbon efficiency ratio decision-making system in Step 3, as described in Example 2, possesses stronger financial attributes. The system not only connects to price data from the national carbon trading market but also considers carbon allowance transfer agreements within the industrial park. The dynamic cost accounting model incorporates an extended version of risk value assessment, using Monte Carlo simulation to predict the price trend of carbon emission rights over the next quarter. The calculation of the carbon efficiency ratio includes not only emissions and indirect emissions from electricity generation but also implicit carbon footprints such as raw material transportation. The system dynamically allocates carbon allowance usage rights based on the carbon efficiency performance of each workshop. This carbon efficiency-based quota allocation mechanism promotes the concentration of energy in high-value-added production lines.

[0044] In the optimization logic of step four, the multi-objective particle swarm optimization algorithm adopts a hierarchical optimization strategy. Bottom-level particles are responsible for searching the optimal operating parameters for specific equipment, while top-level particles are responsible for coordinating the output weights of various energy stations within the park. A grid peak-shaving compensation benefit term is added to the fitness function. When the grid issues a demand response command, the system automatically adjusts the charging and discharging state of the energy storage system and the load of non-core production lines through topology optimization, creating additional revenue by utilizing electricity price differences and carbon quota surpluses while meeting constraints. The Nash equilibrium game model is used here to handle the distribution of benefits among different investment entities within the park, ensuring that the operating scheme of each subsystem is at its optimal equilibrium point.

[0045] Step five, the closed-loop intelligent control process, incorporates a long-term trend analysis module. Through deep cluster analysis of hundreds of millions of execution data records, the system can automatically identify the optimal price comparison strategy under summer high temperature and humidity patterns or raw material impurity fluctuation patterns. When environmental parameters are detected to enter a typical range, the system automatically loads a pre-set decision template, shortening the algorithm's optimization time. A multi-level visual interactive interface allows managers to switch between global and single-device views, observing the value-added path of each carbon emission during the conversion process.

[0046] Example 3: In the metallurgical or pharmaceutical industries involving high-energy-consuming phase change processes, the intelligent price comparison decision-making method, which integrates industrial equipment operation with the overall system, achieves an exceptional characterization of thermal energy quality. In step one, the physical modeling module particularly enhances the monitoring of steam quality. Dryness sensors are installed at key nodes, and combined with temperature and pressure data, the system determines the steam state point in real time. When dealing with such complex fluids, the equivalent conversion protocol no longer uses a single calorific value coefficient, but instead employs a dynamic enthalpy correction algorithm based on real-time property queries. The system accesses the local steam property database every 10 seconds, calculating energy benchmark values ​​accurate to the kilojoule level, supporting subsequent price comparison accuracy.

[0047] Step two, the energy efficiency analysis, focuses on the energy degradation during the latent heat release phase of the phase change process. Since latent heat is released during isothermal processes, its energy efficiency depends on the temperature level. The system uses Carnot efficiency weighting to define the waste heat of low-temperature condensate as low-grade energy and assigns it a low weight in the evaluation metrics. Entropy production analysis is precise down to each heat exchanger tube bundle, using temperature distribution maps acquired by infrared thermal imaging sensors to calculate irreversible losses caused by localized heat transfer temperature differences. When heat exchanger fouling leads to a decrease in heat transfer coefficient and an increase in entropy production, the system automatically flags this and suggests reducing the circulation flow rate of that branch to optimize the overall energy distribution topology.

[0048] In step three, the application of environmental chemistry principles is further refined into real-time correction of combustion efficiency. For industrial boilers, the system adjusts the carbon emission factor in real time based on the ratio of carbon dioxide to carbon monoxide in the flue gas. If carbon monoxide increases due to incomplete combustion, the system not only records energy loss but also converts it into additional environmental risk costs. The dynamic cost accounting model incorporates the company's production rhythm. When products are in a high-premium cycle, the decision-making coefficient for carbon efficiency ratio automatically tilts towards maximizing output; while during market downturns, it tilts towards maximizing carbon emission reduction, achieving a flexible hedging between economic benefits and green gains.

[0049] In step four, the multi-objective particle swarm optimization algorithm introduces deep modeling of the actuator response delay. Under frequently fluctuating loads, the adjustment actions of the frequency converter and gas proportional valve exhibit dynamic lag. The system inputs these delay time constants as constraints into the algorithm. The optimization logic not only seeks the static optimal solution but also searches for a robust control sequence over time. Adaptive mutation features play a crucial role here, preventing the algorithm from falling into local oscillations due to rapid load fluctuations. If the total carbon emissions of a certain production cycle are at risk of exceeding the limit, the penalty term will increase exponentially, forcibly guiding the system to switch to electric heating or other low-carbon processes.

[0050] Step five's feedback loop integrates the deep prediction capabilities of a long short-term memory neural network model. The system not only corrects the current conversion coefficient but also predicts the energy efficiency trend over the next hour. During secondary verification, the expert system pays special attention to the safety of the process temperature window, ensuring that price optimization and gas supply adjustments do not cause temperature fluctuations within the reactor to exceed the preset 0.5-degree upper limit. The visualization interface displays the real-time trajectory of the phase transition point and the distribution of energy quality gradients.

[0051] Example 4: In modern industrial parks with large-scale distributed energy access, an intelligent price comparison decision-making method integrating industrial equipment operation with the overall system enhances decision robustness by introducing financial engineering risk assessment. In step one, the sensing unit is extended to the meteorological station and microgrid side around the park. The model not only collects real-time electricity, gas, and heat data but also simultaneously accesses photovoltaic power prediction and wind power prediction curves. The data preprocessing module introduces wavelet analysis to decompose the renewable energy output signal with random fluctuation characteristics into sub-signals of different frequencies for benchmark mapping in the equivalent conversion protocol.

[0052] Step two, the energy efficiency analysis, is used in this scenario to evaluate the charge-discharge efficiency of the energy storage system. The system calculates the energy loss of the battery and thermal energy storage systems during energy throughput. Through entropy production analysis, the irreversible loss curves of the energy storage converter and heat exchanger at different power levels are identified. The energy efficiency evaluation index is fed back to the scheduling layer as a basis for selecting energy storage carriers: electrical energy storage is prioritized during high-quality energy demand periods, while thermal storage tanks are prioritized during low-quality thermal energy demand periods.

[0053] The carbon efficiency ratio decision-making evaluation system established in step three fully utilizes information from financial derivatives in the external carbon market. The dynamic cost accounting model uses Monte Carlo simulation to simulate not only the current carbon price but also the carbon option price fluctuations over the next week. This application of financial engineering gives the decision-making system a forward-looking capability, enabling adjustments to production plans and increasing the proportion of low-carbon energy consumption before a significant increase in carbon prices is anticipated. The carbon quota allocation logic incorporates the carbon labeling requirements of products, ensuring that every batch of industrial products meets the green trade barriers required by the international market.

[0054] Step four of the multi-objective particle swarm optimization algorithm demonstrates high flexibility in topology optimization. The particle swarm optimizes not only operating parameters but also energy procurement combinations. The algorithm's fitness function incorporates a sensitivity term to fluctuations in electricity purchase costs. During the optimization process, the system adjusts the switching ratio between industrial electric furnaces and gas-fired furnaces based on real-time electricity prices. A Nash equilibrium game model is used to simulate power capacity competition when multiple high-power devices operate concurrently within the industrial park. By establishing a virtual price signal, the system guides the devices to operate at off-peak times, minimizing the overall peak-to-valley difference in the park and improving the overall carbon efficiency ratio.

[0055] Step five, closed-loop intelligent control, achieves a high degree of human-machine collaboration through a visual interface. The interface not only displays real-time energy flow but also shows the carbon intensity distribution in each area via a heat map. An expert system monitors grid frequency and voltage levels in real time, ensuring that the power quality of the microgrid is not impacted during price comparison decisions and significant load switching. The feedback loop uses cluster analysis to identify the optimal energy complementarity mode under different light intensities and temperatures. This method, integrating financial risk management and multi-energy complementarity optimization, provides industrial parks with a scientific, systematic, and resilient intelligent price comparison solution in an uncertain energy market environment.

[0056] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent price comparison decision-making method integrating industrial equipment operation completion and overall system integration, characterized in that, The method includes the following steps: Step 1: Construct a multi-energy flow topology coupling model for industrial equipment. Real-time data on electricity, natural gas, heat, and carbon emissions are acquired through sensing units deployed on the production site. The collected heterogeneous energy data is then mapped to a unified energy reference space through a preset equivalent conversion protocol. Step 2: Perform an effective energy analysis based on the second law of thermodynamics to calculate the energy utilization quality of industrial equipment during operation, identify weak links in energy consumption within the system, and generate effective energy evaluation indicators that reflect energy conversion losses. Step 3: Establish a multi-dimensional carbon efficiency ratio decision evaluation system, convert thermal efficiency and electrical efficiency into standard carbon emission indicators, and link them to price fluctuation data of the external carbon trading market in real time to form a dynamic cost accounting model that is linked to green benefits. Step 4: Use a multi-objective particle swarm optimization algorithm to perform topology optimization on the operation strategy of industrial equipment. Under preset constraints, balance the nonlinear contradiction between energy consumption, thermal weighted efficiency and carbon consumption to generate the optimal operation decision instruction. Step 5: Monitor the decision execution results in real time, adjust the topology optimization parameters through feedback loops, and realize closed-loop intelligent control that integrates the operating status of industrial equipment with overall benefits.

2. The intelligent price comparison decision-making method for integrating industrial equipment operation completion with overall system integration as described in claim 1, characterized in that, The step one of constructing the multi-energy flow topology coupling model of industrial equipment includes: Establish an interconnected topology network for the power supply system, gas transmission system, and steam thermal cycle system; Sensing units are configured at both the energy input and output ends. These sensing units include intelligent power meters, ultrasonic flow meters, infrared thermal imaging sensors, and online flue gas monitoring systems. All sensing units are connected to the industrial fieldbus, and a synchronous clock protocol is used to lock the sampling frequency of each sensor to the same reference frequency. The data preprocessing module performs noise reduction, smoothing, and missing value compensation on the collected raw current, voltage, flow rate, pressure, temperature, and flue gas component signals. The noise reduction process eliminates high-frequency harmonic interference from the power grid using a moving average filtering algorithm, and the missing value compensation process fills in the missing values ​​caused by momentary interruptions in sensor communication using linear interpolation technology.

3. The intelligent price comparison decision-making method for integrating industrial equipment operation completion with overall system integration as described in claim 2, characterized in that, The equivalent conversion protocol includes: Establish a conversion coefficient matrix based on standard coal consumption or standard calorific value to map energy consumption of different dimensions to an energy reference space based on joules; Dynamic correction logic is introduced during the mapping process. For natural gas, the lower heating value of natural gas is corrected online based on real-time data from pressure and temperature sensors, expressed as: ； in, For actual heat output, Based on the calorific value, This is the pressure correction factor. This is a temperature correction factor; For electrical energy, the real-time load curve of the power grid is connected, and the coefficient for converting electrical energy into standard coal is dynamically adjusted according to the current power supply structure of the power grid. An energy quality factor is defined within the energy reference space to distinguish between high-level and low-level energy. Before making a price comparison decision, the enthalpy value under pressure is obtained by querying the built-in water vapor property table, and the latent heat of phase change is included in the calculation logic of the effective utilization rate of thermal energy.

4. The intelligent price comparison decision-making method for integrating industrial equipment operation completion with overall system integration as described in claim 3, characterized in that, Step two, which involves performing an effective energy analysis based on the second law of thermodynamics, includes: Perform a joint check of energy balance and effective energy balance, and obtain the effective energy efficiency of the system by calculating the ratio of the total effective energy of the input system to the useful work of the output system that is effectively utilized; When dealing with a thermodynamic system, the Carnot cycle efficiency is used to weight the heat at different temperature gradients. The formula for the effective energy weight of the heat source is: ; in The effective energy weight of the heat source, The absolute temperature of the environment. The absolute temperature of the heat source; The effective energy weight is used to characterize the quality degradation of heat energy during its transport through heat exchangers and pipelines or during the work process. The identification of weak points in energy consumption within the system is achieved by performing entropy production analysis on each energy conversion node.

5. The intelligent price comparison decision-making method for integrating industrial equipment operation completion with overall system integration as described in claim 4, characterized in that, The entropy production analysis includes: The irreversible losses of each submodule during operation are calculated using the Clausius inequality principle; these irreversible losses are characterized by the entropy generation of the nodes, and the formula for node entropy generation is: ； in, This refers to the amount of node entropy generated. The total entropy flow at the node's exit. Total entropy flow at node inlet The heat exchanged between the node and the outside world. The absolute temperature of the boundary; The system converts the calculated entropy output into the effective energy evaluation index; when the effective energy loss of a node exceeds the preset loss threshold, the system triggers an early warning signal and marks the node as a key adjustment target on the topology map; For nodes marked as key adjustment targets, the system extracts their physical characteristic parameters and defines them as key variables in subsequent topology optimization.

6. The intelligent price comparison decision-making method for integrating industrial equipment operation completion with overall system integration as described in claim 5, characterized in that, The multi-dimensional carbon efficiency ratio decision evaluation system introduces the carbon efficiency ratio as a decision weight coefficient. The carbon efficiency ratio is defined as the ratio of the economic value generated per unit time or the total useful work output by the system to the total carbon emissions generated per unit time. For fossil fuels, fuel consumption is calculated by using the complete combustion equation to inversely deduce the fuel consumption based on the oxygen content, carbon monoxide concentration, and flue gas temperature in the exhaust gas, and then multiplied by the carbon dioxide emission factor. For electricity consumption, the indirect carbon emissions of electricity are calculated based on the proportion of green electricity acquired in real time. By combining production planning and scheduling, order urgency, and product unit energy consumption standards, a dynamic carbon consumption quota is allocated to each piece of industrial equipment. When the real-time carbon consumption rate of the equipment exceeds the preset rate threshold, the decision-making system will perform an operation to search for a low-carbon operating path.

7. The intelligent price comparison decision-making method for integrating industrial equipment operation completion with overall system integration as described in claim 6, characterized in that, The dynamic cost accounting model obtains the settlement price of the carbon trading market in real time through the network communication module; Using Monte Carlo simulation to predict the fluctuation distribution of energy prices and carbon prices over a future decision-making cycle; When carbon quota prices are on an upward trend, the weight of carbon efficiency ratio in the overall price ratio function is increased; when carbon prices are in a preset low range, the proportion of reducing direct energy procurement costs in the overall price ratio function is increased. The system introduces a risk value assessment mechanism, which uses Monte Carlo simulation results to assess the maximum economic loss of the decision scheme under a preset confidence level, and ensures that the maximum economic loss is within a controllable range; The calculation process of the standard carbon emission index also incorporates the nonlinear emission characteristics of industrial equipment under different load rates, establishes a correlation function between load rate and carbon emission intensity through historical data, and corrects the real-time emission data.

8. The intelligent price comparison decision-making method for integrating industrial equipment operation completion with overall system integration as described in claim 7, characterized in that, The fitness function of the multi-objective particle swarm optimization algorithm includes total energy cost, overall system effective energy efficiency, and carbon emission cost calculated based on real-time carbon price. Each sub-indicator is fused through a preset weighted vector. The multi-objective particle swarm optimization algorithm follows velocity and position update formulas during the iteration process; the velocity update formula is: ； in, for Generation, First The updated velocity vector of each particle. for Generation, First The updated velocity vector of each particle. For inertial weights, As the first learning factor, As the second learning factor, and For interval Random numbers within, for Generation, First The historical optimal position vector of each particle. for Generation, First The global optimal position vector of each particle. for Generation, First The current position vector of each particle; The position update formula is: ; for Generation, First The updated position vectors of each particle; The algorithm introduces a non-dominated sorting mechanism based on crowding distance, and selects a set of non-dominated solutions after each iteration according to the Pareto front theory.

9. The intelligent price comparison decision-making method for integrating industrial equipment operation completion with overall system integration as described in claim 8, characterized in that, The topology optimization in step four includes: By adjusting the actuators, the energy flow paths of each process unit in the industrial production process are reconstructed in series and parallel logic; by adjusting the opening of the bypass valve, the frequency of the variable frequency drive, and the circulation flow of the heat recovery system, the energy flow direction inside the system is changed; in the cogeneration scenario, a game model based on Nash equilibrium is introduced to coordinate the conflict of interests between heat demanders and electricity demanders, and to find the balance point that maximizes the overall carbon efficiency ratio of the system. The preset constraints include equipment rated power limits, production process temperature window limits, and total energy supply contract limits. The algorithm processes constraints through a penalty function. If a candidate solution touches any constraint boundary, its fitness value is increased by a maximum penalty term. The system monitors the equipment's vibration, temperature, and cumulative operating hours and assesses the equipment's health index. This health index is used as a correction factor for the thermal efficiency model to prevent excessive wear on the equipment.

10. The intelligent price comparison decision-making method for integrating industrial equipment operation completion with overall system integration as described in claim 9, characterized in that, The closed-loop intelligent control in step five includes: The decision deviation is calculated by comparing the expected energy efficiency index with the actual observation value of the sensor; when the decision deviation exceeds the preset allowable range, the self-learning mechanism is activated, and the energy conversion coefficient matrix and carbon emission prediction model are corrected by using a long short-term memory neural network model. The system integrates a multi-criteria expert system, and performs a secondary verification of the decision safety based on a historical case library after the algorithm generates the optimal solution; If the environmental parameters are in the historical high-risk range, the expert system will fine-tune the optimization results according to the preset safety operating procedures. The system displays a real-time dynamic topology diagram of energy flow, carbon flow, and cost flow through a visual interactive interface, and supports switching between green priority mode and cost priority mode. The system aggregates historical execution data and uses clustering analysis algorithms to identify typical optimization patterns under different environmental seasons, providing pre-set templates for future operation strategies.

Images