A method and system for optimizing energy saving of water-cooled air conditioning in a manufacturing plant

Abstract

The application discloses a kind of energy-saving optimization method and system of manufacturing factory water-cooled air conditioner, it is related to intelligent energy-saving optimization technical field, including collection multimodal data, construct initial gene bank, form cluster gene bank by spectrum decomposition and K-means, using genetic algorithm to update cluster gene bank, output configuration parameter table, set target function and heat balance constraint condition, convert into quadratic binary optimization problem, using quantum annealing algorithm parallel search, by discretization mapping and screening, output initial operation optimization parameter, carry out use and calculate deviation rate, update GP model parameter, output operation optimization parameter.The application forms configuration parameter table by exergy analysis combined with advanced algorithm, improves the accuracy and adaptability of system initial configuration, introduces quantum annealing and GP model dynamic optimization operating parameter, improves the dynamic adaptability and optimization efficiency of operating parameter.

Description

Technical Field

[0001] This invention relates to the field of intelligent energy-saving optimization technology, and in particular to an energy-saving optimization method and system for water-cooled air conditioning in manufacturing plants. Background Technology

[0002] With the acceleration of industrialization, manufacturing plants are consuming more and more energy during production, especially in the operation of air conditioning and refrigeration systems, where energy consumption and resource waste are becoming increasingly apparent. Traditional water-cooled air conditioning systems suffer from low efficiency and unstable operation in practical applications. The collection and analysis of multimodal data, existing algorithms, and GP prediction models are gradually being applied, which can reduce energy consumption to a certain extent and achieve dynamic optimization of the system.

[0003] Existing energy-saving optimization methods for water-cooled air conditioners still have shortcomings. Existing algorithms only analyze multimodal data and lack in-depth consideration of exergy analysis, resulting in the inability of the system configuration to achieve efficient operation. The GP prediction model fails to combine quantum optimization for dynamic prediction, resulting in low efficiency and inability to adapt to dynamically changing operating environments. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides an energy-saving optimization method and system for water-cooled air conditioning in manufacturing plants, which solves the problems of existing advanced methods that only analyze multimodal data and lack in-depth consideration of exergy analysis, resulting in the inability of the system configuration to achieve efficient operation, and the failure of the GP prediction model to combine quantum optimization for dynamic prediction, leading to low efficiency and inability to adapt to dynamically changing operating environments.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides an energy-saving optimization method for water-cooled air conditioning in a manufacturing plant, comprising: collecting multimodal data, constructing an initial gene library, screening using the Pareto front, constructing an exergy coupling weight matrix using FCM, forming clustered gene libraries using spectral decomposition and K-means, updating the clustered gene libraries using a genetic algorithm, outputting a configuration parameter table, predicting using a GP model, setting an objective function and thermal balance constraints, converting the problem into a quadratic binary optimization problem, using a quantum annealing algorithm for parallel search, outputting initial running optimization parameters through discretization mapping and screening, using the parameters and calculating the deviation rate, updating the GP model parameters, outputting running optimization parameters, adjusting the frequency of the optimized equipment based on the running optimization parameters, performing real-time monitoring and setting an alarm mechanism, and storing the collected and analyzed multimodal data.

[0008] As a preferred embodiment of the energy-saving optimization method for water-cooled air conditioning in manufacturing plants described in this invention, the following steps are described: collecting multimodal data, constructing an initial gene library, screening using the Pareto front, constructing an exergy coupling weight matrix using FCM, forming clustered gene libraries using spectral decomposition and K-means, updating the clustered gene libraries using a genetic algorithm, and outputting a configuration parameter table, including:

[0009] The multimodal data includes infrared images of the equipment's exterior, ambient temperature and humidity, the equipment's current operating power, cooling tower outlet water temperature, chilled water supply and return temperatures, and chilled water flow rate.

[0010] Based on infrared images, calculate the real-time exergy efficiency factor and exergy cost;

[0011] The current operating power, chilled water supply and return temperatures, chilled water flow rate, rated power, design cooling capacity, and exergy cost of the equipment are defined as gene units;

[0012] Use systematic sampling to extract N sets of equipment configurations, and calculate the total exergy cost and energy efficiency ratio of each set of equipment configurations.

[0013] By setting a total exergy cost threshold based on historical operating data, equipment configurations with total exergy cost ≤ the total exergy cost threshold are screened. Using the Pareto front extraction method, based on non-dominated screening and crowding calculation, crowding is sorted in ascending order, and the top H group configurations with low crowding are screened to form an initial gene pool.

[0014] The membership degree of the device configuration in the initial gene pool was calculated using the FCM method;

[0015] The distance from each group of equipment to the centroid is calculated using the Euclidean formula, and the coupling weights are obtained by combining the membership degree and arranging them to form a coupling weight matrix.

[0016] The exergy coupling weight matrix is ​​transformed into a similarity matrix through cosine similarity. A normalized Laplace matrix is ​​constructed, and the Laplace matrix is ​​subjected to spectral decomposition. The equipment configuration is clustered using the K-means algorithm, and the equipment configuration is divided into 3 clusters. The average exergy coupling weight of each cluster is calculated.

[0017] The cluster with the highest average fire coupling weight is selected as the high coupling group, and a cluster gene library is generated. n groups are randomly selected as the parent generation, and the offspring coupling groups are generated through crossover mutation.

[0018] Calculate the total exergy cost, energy efficiency ratio, and temperature deviation for the high-coupling group and the offspring coupling groups generated by cross-mutation;

[0019] Calculate the ratio of actual cooling capacity to design cooling capacity to obtain the load rate;

[0020] The threshold for total exergy cost minus energy efficiency ratio is set based on the total exergy cost and energy efficiency ratio. The threshold for temperature deviation is set based on historical temperature deviation data analysis. The maximum and minimum load rates are set based on the safety boundary method.

[0021] The system filters equipment configurations that meet the thresholds of total cost of pyrolysis minus energy efficiency ratio ≤ total cost minus energy efficiency ratio, temperature deviation ≤ temperature deviation threshold, and load rate between maximum and minimum load rates. It then inputs the clustered gene library for updates, stops updating when the maximum number of updates is reached, outputs the latest gene library, and converts it into a configuration parameter table through structured mapping.

[0022] As a preferred embodiment of the energy-saving optimization method for water-cooled air conditioning in manufacturing plants described in this invention, the following steps are taken: Prediction is performed using a GP model, setting the objective function and thermal balance constraints, transforming it into a quadratic binary optimization problem, using a quantum annealing algorithm for parallel search, and outputting initial running optimization parameters through discretization mapping and filtering. These parameters are then used and the deviation rate is calculated, the GP model parameters are updated, and the running optimization parameters are output, including:

[0023] Extract the total exergy cost, energy efficiency ratio, load rate, chilled water flow rate, and current operating power of the equipment from the configuration parameter table, and extract the ambient temperature and humidity from the multimodal data;

[0024] A GP model was constructed, and historical environmental temperature and humidity, equipment operating power, and chilled water flow rate were extracted from historical operating data to train the GP model.

[0025] The ambient temperature and humidity, the current operating power of the equipment, and the chilled water flow rate are input into the GP model. The predicted mean and predicted variance are calculated through the posterior distribution. The predicted mean and predicted variance are formed into a predicted distribution through the Gaussian regression process. The predicted ambient temperature and humidity, equipment operating power, and chilled water flow rate for future times are output.

[0026] Define the Wasserstein uncertainty set, use the steady-state thermal model to calculate the thermal equilibrium, set the thermal equilibrium threshold based on statistical analysis, if the deviation between the left and right sides of the thermal equilibrium is greater than the thermal equilibrium threshold, adjust the GP model parameters by maximizing the marginal likelihood, recalculate the predicted mean and predicted variance through the posterior distribution, update the predicted distribution, and stop when the deviation between the left and right sides of the thermal equilibrium is less than or equal to the thermal equilibrium threshold.

[0027] Based on the predicted ambient temperature and humidity at future times, the PMV score is calculated using the PMV model.

[0028] Define the objective function and set the heat balance formula as the constraint condition, transform it into a quadratic bivariate optimization problem, and define the Hamiltonian formula.

[0029] The Hamiltonian is obtained by discretizing the equipment operating power, chilled water flow rate and PMV score for predicting future time into qubit states and substituting them into the Hamiltonian formula.

[0030] The Hamiltonian is saved as a matrix, the number of samplings and annealing time are set, and the quantum annealing algorithm is used to predict the combination of equipment operating power and chilled water flow rate in the future through parallel search of quantum superposition states. The energy distribution is obtained by the number of samplings, the energy mean is calculated by weighted average, and solutions with less than 50% energy are screened out. Based on discretization calculation, the results are mapped back to continuous variables to generate candidate parameter groups.

[0031] Based on the verification of thermal equilibrium and PMV, the parameter groups that pass the verification are selected from the candidate parameter groups, the objective function value is calculated, and the parameter with the smallest objective value is selected as the initial running optimization parameter.

[0032] The initial operation optimization parameters are used to obtain real-time chilled water flow rate, equipment operating power and PMV score, and the deviation rate is calculated.

[0033] If a deviation rate threshold is set based on statistical significance testing, and the deviation rate is greater than the deviation rate threshold, the real-time collected chilled water supply and return water temperature difference, power, and PMV score will be input into the GP model to update the model parameters. Optimization will stop when the deviation rate is less than or equal to the deviation rate threshold, and the operating optimization parameters will be output, including the predicted chilled water flow rate, equipment operating power, and PMV score.

[0034] As a preferred embodiment of the energy-saving optimization method for water-cooled air conditioning in manufacturing plants according to the present invention, the step of adjusting and optimizing the equipment frequency based on operating optimization parameters includes:

[0035] The operating optimization parameters will be used to adjust and optimize the water pump frequency, cooling tower fan speed, and chiller unit frequency converter frequency.

[0036] As a preferred embodiment of the energy-saving optimization method for water-cooled air conditioning in manufacturing plants according to the present invention, the step of performing real-time monitoring and setting an alarm mechanism includes:

[0037] Every c minutes, the energy efficiency ratio, PMV score, and total power are recalculated. Thresholds for energy efficiency ratio, PMV score, and total efficiency are set. If the energy efficiency ratio is greater than the preset threshold, the chiller frequency coefficient is adjusted. If the PMV score is less than the preset threshold, the cooling tower fan speed coefficient is adjusted. If the total power exceeds the total power threshold, an alarm is triggered, and staff are notified to conduct an inspection.

[0038] As a preferred embodiment of the energy-saving optimization method for water-cooled air conditioning in manufacturing plants according to the present invention, the multimodal data is first preprocessed;

[0039] The preprocessing includes size standardization, pixel normalization, mean filtering for noise reduction, CLAHE enhancement of image features, and edge detection to extract thermal features of the image, as well as processing of ambient temperature, ambient humidity, current operating power of the equipment, chilled water supply and return temperatures, and chilled water flow rate.

[0040] As a preferred embodiment of the energy-saving optimization method for water-cooled air conditioning in manufacturing plants according to the present invention, the storage, collection, and analysis of the generated multimodal data includes:

[0041] The collected multimodal data, along with the operational optimization parameters and adjustment frequency data generated from the analysis, are used to generate a complete evaluation report through data visualization and then uploaded to the central database.

[0042] Secondly, the present invention provides an energy-saving optimization system for water-cooled air conditioning in a manufacturing plant, comprising:

[0043] The collection and configuration module is used to collect multimodal data, build an initial gene library, screen it using the Pareto front, construct the pyrokinetic coupling weight matrix through FCM, form clustered gene libraries through spectral decomposition and K-means, update the clustered gene libraries using a genetic algorithm, and output a configuration parameter table.

[0044] The prediction and optimization module is used to make predictions using the GP model. It sets the objective function and thermal balance constraints, transforms the problem into a quadratic binary optimization problem, uses the quantum annealing algorithm for parallel search, and outputs the initial running optimization parameters through discretization mapping and filtering. It then uses these parameters to calculate the deviation rate, updates the GP model parameters, and outputs the running optimization parameters.

[0045] The monitoring module is adjusted to optimize equipment frequency based on operational optimization parameters, perform real-time monitoring, and set alarm mechanisms.

[0046] The storage and analysis module is used to store the collected and analyzed multimodal data.

[0047] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the energy-saving optimization method for water-cooled air conditioning in a manufacturing plant as described in the first aspect of the present invention.

[0048] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the energy-saving optimization method for water-cooled air conditioning in a manufacturing plant as described in the first aspect of the present invention.

[0049] The beneficial effects of this invention are as follows: This invention collects multimodal data, constructs an initial gene pool, uses Pareto fronts for screening, constructs an pyrokinetic coupling weight matrix using FCM, forms clustered gene pools using spectral decomposition and K-means, updates the clustered gene pools using a genetic algorithm, outputs a configuration parameter table, performs prediction using a GP model, sets the objective function and thermal equilibrium constraints, transforms it into a quadratic binary optimization problem, uses a quantum annealing algorithm for parallel search, outputs initial running optimization parameters through discretization mapping and screening, uses the parameters and calculates the deviation rate, updates the GP model parameters, and outputs running optimization parameters; thus improving the accuracy and adaptability of the initial system configuration and enhancing the dynamic adaptability and optimization efficiency of the running parameters. Attached Figure Description

[0050] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a flowchart of the energy-saving optimization method for water-cooled air conditioning in a manufacturing plant in Example 1.

[0052] Figure 2 This is a schematic diagram of the water-cooled air conditioning energy-saving optimization system in the manufacturing plant in Example 1.

[0053] Figure 3 This is a flowchart of the quantum optimization in Example 1.

[0054] Figure 4 This is a schematic diagram of real-time monitoring and early warning in Example 1. Detailed Implementation

[0055] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0056] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0057] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0058] Example 1, referring to Figures 1 to 4 This is the first embodiment of the present invention, which provides an energy-saving optimization method for water-cooled air conditioning in a manufacturing plant, comprising the following steps:

[0059] S1. Collect multimodal data, construct an initial gene pool, use Pareto front for screening, construct the pyrokinetic coupling weight matrix through FCM, form clustered gene pools through spectral decomposition and K-means, update the clustered gene pools using a genetic algorithm, and output a configuration parameter table.

[0060] Specifically, multimodal data is collected and preprocessed using smart sensors, including:

[0061] The intelligent sensors include temperature and humidity sensors, infrared thermal imagers, smart meters, and electromagnetic flow meters.

[0062] The multimodal data includes infrared images of the equipment's exterior, ambient temperature and humidity, the equipment's current operating power, cooling tower outlet water temperature, chilled water supply and return temperatures, and chilled water flow rate.

[0063] The preprocessing includes size standardization, pixel normalization, mean filtering for noise reduction, CLAHE enhancement of image features (Contrast Limited Adaptive Histogram Equalization), and edge detection to extract thermal features of the image, as well as ambient temperature, ambient humidity, current operating power of the equipment, chilled water supply and return temperatures, and chilled water flow rate.

[0064] By preprocessing multimodal data, especially infrared images and environmental data, the accuracy of fault detection and equipment monitoring can be significantly improved. After reasonable preprocessing, different types of data can provide clearer and more accurate information for equipment energy efficiency optimization and fault early warning. The beneficial effects of this process not only improve the stability of equipment operation, but also enable early fault warning and more efficient energy management.

[0065] Furthermore, multimodal data were collected to construct an initial gene pool, which was then screened using the Pareto front. An exergy coupling weight matrix was constructed using FCM, and clustered gene pools were formed through spectral decomposition and K-means. Genetic algorithms were used to update the clustered gene pools, and a configuration parameter table was output, including:

[0066] The system obtains equipment parameters, including rated power, design temperature, historical operating data, and design cooling capacity, through the CMMS system (Computerized Maintenance Management System) and PLC interface (Programmable Logic Controller Interface).

[0067] The real-time pyroefficiency factor is calculated based on infrared images using the following formula:

[0068] ,

[0069] in, For exergy efficiency factor, The surface temperature of the device (obtained via infrared imaging). For the design temperature, The ambient reference temperature (set based on the energy-saving design standards for industrial buildings).

[0070] The exergy cost is calculated based on the real-time exergy efficiency factor, using the following formula:

[0071] ,

[0072] in, For the first The exergy cost of the equipment For the first The current operating power of the equipment For the first Rated power of the equipment Standard exergy cost (set according to industry benchmarks). This is the equipment aging and degradation coefficient (obtained from historical operating data).

[0073] The current operating power, chilled water supply and return temperatures, chilled water flow rate, rated power, design cooling capacity, and exergy cost of the equipment are defined as gene units;

[0074] Based on gene units, N sets of equipment configurations are extracted using systematic sampling (defined by empirical rules, each configuration includes a chiller, cooling tower, and water pump). The total exergy cost and energy efficiency ratio of each equipment configuration are calculated using the following formula:

[0075] ,

[0076] ,

[0077] in, Total cost of fire, For the total number of devices, For energy efficiency ratio, It is the actual cooling capacity (the product of water's specific heat capacity, chilled water flow rate, and the chilled water supply and return temperature difference). This is the sum of the current operating power of the three devices;

[0078] By setting a total exergy cost threshold based on historical operating data, equipment configurations with total exergy cost ≤ the total exergy cost threshold are selected. The exergy cost and energy efficiency ratio of each selected equipment are extracted using the Pareto front method. Through non-dominated screening and crowding calculation, the crowding is sorted in ascending order, and the top H group configurations with low crowding (set according to computational resource constraints) are selected to form an initial gene pool.

[0079] The membership degree of the device configuration in the initial gene pool is calculated using the FCM method (Fuzzy C-Means Clustering Method), and the membership degree is updated using the membership degree update formula.

[0080] The distance from each group of devices to the centroid is calculated using the Euclidean formula. Combined with membership degrees, the exergy coupling weights are obtained and arranged to form a coupling weight matrix. The formula is as follows:

[0081] ,

[0082] in, For fire coupling weights, Configure the equipment to the center of mass distance, It is a constant. Configure the equipment Belongs to cluster Membership degree;

[0083] The coupling weight matrix is ​​transformed into a similarity matrix using cosine similarity, and a normalized Laplace matrix is ​​constructed.

[0084] The Laplace matrix is ​​subjected to spectral decomposition, and the equipment configuration is clustered using the K-means algorithm (interpreted as K-means algorithm). The equipment configuration is divided into 3 clusters (set according to business requirements), and the average exergy coupling weight of each cluster is calculated.

[0085] The cluster with the highest average pyrolysis coupling weight is selected as the high coupling group, and a cluster gene library is generated. n groups are randomly selected as the parent generation, and the offspring coupling groups are generated through crossover mutation. The process stops when the set number of iterations is reached (based on the fixed number of iterations method).

[0086] Calculate the total exergy cost, energy efficiency ratio, and temperature deviation (the difference between the actual temperature and the set temperature) of the offspring coupled groups generated by the high coupling group and the cross-mutation.

[0087] Calculate the ratio of actual cooling capacity to design cooling capacity to obtain the load rate;

[0088] The threshold for total exergy cost minus energy efficiency ratio is set based on the total exergy cost and energy efficiency ratio. The threshold for temperature deviation is set based on historical temperature deviation data analysis. The maximum and minimum load rates are set based on the safety boundary method.

[0089] The system filters equipment configurations that meet the thresholds of total cost of pyrotechnics minus energy efficiency ratio (rQR) ≤ total cost minus energy efficiency ratio, temperature deviation ≤ temperature deviation threshold, and load rate between maximum and minimum load rates. It then inputs the clustered gene library for updates, stopping updates when the maximum number of updates is reached (set via fatigue detection). The system outputs the latest gene library, which is then converted into a configuration parameter table through structured mapping.

[0090] By calculating the exergy efficiency factor and exergy cost in real time, the operating status of equipment can be monitored in real time, and changes in equipment status can be responded to quickly, avoiding energy waste caused by low operating efficiency. The Pareto frontier method not only considers the balance between cost and benefit, but also optimizes equipment configuration in all aspects under different constraints through multi-objective optimization, effectively avoiding inefficient choices in terms of performance and cost. After constructing the exergy coupling weight matrix, the equipment is classified by the K-means clustering algorithm, which reduces the complexity and redundancy of the equipment configuration scheme and improves the optimization efficiency. The membership degree of the equipment configuration is dynamically updated by the FCM algorithm, so that the equipment can be continuously adjusted according to the changes in its performance indicators during actual operation, thereby optimizing the overall system efficiency. The genetic algorithm and crossover mutation method are used to update the equipment configuration, so that the optimized configuration of the equipment can gradually approach the optimal solution in multiple iterations and find more creative configuration schemes.

[0091] S2. Predict using the GP model, set the objective function and thermal balance constraints, transform it into a quadratic binary optimization problem, use the quantum annealing algorithm for parallel search, through discretization mapping and filtering, output initial running optimization parameters, use them and calculate the bias rate, update the GP model parameters, and output running optimization parameters, including:

[0092] Extract the total exergy cost, energy efficiency ratio, load rate, chilled water flow rate, and current operating power of the equipment from the configuration parameter table, and extract the ambient temperature and humidity from the multimodal data;

[0093] The GP model (Gaussian Process Model) is constructed, including an input layer, a kernel function layer, a mean function layer, a noise layer, and an output layer.

[0094] Historical environmental temperature and humidity, equipment operating power, and chilled water flow rate were extracted from historical operating data to train the GP model;

[0095] The ambient temperature and humidity, the current operating power of the equipment, and the chilled water flow rate are input into the GP model. The predicted mean and predicted variance are calculated through the posterior distribution. The predicted mean and predicted variance are formed into a predicted distribution through the Gaussian regression process. The predicted ambient temperature and humidity, equipment operating power, and chilled water flow rate for future times are output.

[0096] The Wasserstein Ambiguity Set is defined based on the prediction of environmental temperature and humidity in the future.

[0097] The steady-state thermal model is used to calculate the thermal equilibrium. The thermal equilibrium threshold is set based on statistical analysis. If the deviation between the left and right sides of the thermal equilibrium is greater than the thermal equilibrium threshold, the GP model parameters are adjusted by maximizing the marginal likelihood. The predicted mean and predicted variance are recalculated using the posterior distribution, and the predicted distribution is updated until the deviation between the left and right sides of the thermal equilibrium is less than or equal to the thermal equilibrium threshold.

[0098] Based on the predicted future ambient temperature and humidity, worker labor metabolic rate (set according to ISO 7730 standard), tool insulation (set according to ISO 9920 standard), and air velocity (set according to working condition setting method), the PMV score is calculated using the PMV model (Predicted Mean Vote).

[0099] Define the objective function and set the heat balance formula as the constraint condition. The formula is as follows:

[0100] ,

[0101] in, These are adaptive weighting coefficients (derived from economic analysis of real-world scenarios). Let be the set of Wasserstein uncertainties at time t. For the predicted distribution of time t, The difference between the PMV score and the target PMV score (the target PMV score is set based on the comfort analysis of the ISO 7730 standard). This is the set of current operating power for the three devices;

[0102] The objective function and the heat balance formula are transformed into a quadratic bivariate optimization problem. The Hamiltonian formula is defined as follows:

[0103]

[0104] Where H is the Hamiltonian, and These are the penalty coefficients for the objective function and the thermal balance constraint (set based on the constraint importance method). The specific heat capacity of water, For the first The projected operating power of the equipment in the future. To predict the chilled water flow rate in the future;

[0105] The Hamiltonian is obtained by discretizing the equipment operating power, chilled water flow rate and PMV score for predicting future time into qubit states and substituting them into the Hamiltonian formula.

[0106] The Hamiltonian is stored in matrix format. The number of samplings and annealing time are set through experiments and incremental methods. The quantum annealing algorithm is used to predict the combination of equipment operating power and chilled water flow rate in the future through parallel search of quantum superposition states. The energy distribution is obtained by sampling number. The energy mean is calculated by weighted average and the solution with less than 50% energy is selected (based on the median solution method).

[0107] The selected solutions are discretized and mapped back to continuous variables to generate candidate parameter sets.

[0108] Based on the verification of thermal equilibrium and PMV, the parameter groups that pass the verification are selected from the candidate parameter groups, the objective function value is calculated, and the parameter with the smallest objective value is selected as the initial running optimization parameter.

[0109] The initial operation optimization parameters are used to obtain real-time chilled water flow rate, equipment operating power and PMV score, and the deviation rate is calculated.

[0110] If a deviation rate threshold is set based on statistical significance testing, and the deviation rate is greater than the deviation rate threshold, the real-time collected chilled water supply and return water temperature difference, power, and PMV score will be input into the GP model to update the model parameters. Optimization will stop when the deviation rate is less than or equal to the deviation rate threshold, and the operating optimization parameters will be output, including the predicted chilled water flow rate, equipment operating power, and PMV score.

[0111] Prediction using the GP model provides more accurate and reliable results. By defining the Wasserstein uncertainty set, the model becomes more robust in the face of high uncertainty, improving its reliability in practical applications. Adjusting GP model parameters based on thermal balance calculations allows for effective management of equipment temperature variations and energy efficiency, ensuring equipment stability and energy savings. The quantum annealing algorithm optimizes equipment operating parameters, enabling rapid searching in multi-dimensional, multi-objective optimization problems, significantly improving optimization efficiency and accuracy while reducing computational resource consumption. Real-time optimization and model updates effectively address the impact of environmental changes and equipment aging on equipment operation, thereby improving equipment stability and efficiency.

[0112] S3. Adjust and optimize the equipment frequency based on the operation optimization parameters, perform real-time monitoring, and set up an alarm mechanism;

[0113] Specifically, adjusting and optimizing equipment frequencies based on operational optimization parameters includes:

[0114] The optimized operating parameters are used, including real-time collection of actual cooling capacity and cooling tower outlet water temperature by smart sensors. These parameters are combined with design cooling capacity, design temperature difference, and predicted temperature and humidity to adjust and optimize the pump frequency (using a PID algorithm), cooling tower fan speed, and chiller unit inverter frequency. The formula is as follows:

[0115] ,

[0116] ,

[0117] in, For the frequency conversion frequency of the chiller unit, This is the adjustment coefficient for the frequency conversion of the chiller unit (set based on engineering experience). To design cooling capacity, This refers to the cooling tower fan speed. The adjustment coefficient for the cooling tower fan speed (set based on on-site commissioning experience). The outlet water temperature of the cooling tower. To predict wet-bulb temperature (calculated by predicting ambient temperature and humidity). Design temperature difference (based on common standard settings for industrial water-cooled air conditioners).

[0118] By combining real-time data collection, PID algorithm adjustment, and environmental data, the operating efficiency of the refrigeration system can be significantly improved, energy waste reduced, and equipment operational stability enhanced. This allows for precise adjustment of equipment operating conditions, extending equipment lifespan, reducing failure rates, and achieving intelligent refrigeration system optimization.

[0119] Furthermore, real-time monitoring and alarm mechanisms should be implemented, including:

[0120] Every c minutes (based on industry practice calculation cycle settings), the energy efficiency ratio, PMV score, and total power are recalculated. Energy efficiency ratio thresholds (based on industry standards), PMV score thresholds (based on human thermal comfort standards), and total efficiency thresholds (set using a fixed threshold method) are set. If the energy efficiency ratio is greater than the preset energy efficiency ratio threshold, the chiller frequency coefficient is adjusted. If the PMV score is less than the preset PMV score threshold, the cooling tower fan speed coefficient is adjusted. If the total power exceeds the total power threshold, an alarm is triggered, and staff are notified to conduct an inspection.

[0121] By real-time monitoring and dynamic adjustment of energy efficiency ratio, PMV score and total power, the energy efficiency and environmental comfort of chiller units and cooling tower systems are improved. At the same time, alarms can be triggered in time when equipment is overloaded or its performance is abnormal, which improves the intelligence level of the system and effectively solves the problems of energy waste, equipment overload and insufficient environmental comfort in traditional technologies, thereby achieving significant energy saving and environmental protection effects.

[0122] S4. Store and analyze the generated multimodal data;

[0123] Specifically, the storage, collection, and analysis of the resulting multimodal data includes:

[0124] The collected multimodal data, along with the operational optimization parameters and adjustment frequency data generated from the analysis, are used to generate a complete evaluation report through data visualization and then uploaded to the central database.

[0125] By integrating and analyzing multimodal data to generate operational optimization parameters and adjustment frequency data, a complete evaluation report is produced. This report comprehensively and accurately reflects the equipment's operating status, enabling personalized and intelligent adjustments to improve energy efficiency and reduce failure rates. It also achieves centralized data management and remote access.

[0126] This embodiment also provides an energy-saving optimization system for water-cooled air conditioning in a manufacturing plant, including:

[0127] The collection and configuration module is used to collect multimodal data, build an initial gene library, screen it using the Pareto front, construct the pyrokinetic coupling weight matrix through FCM, form clustered gene libraries through spectral decomposition and K-means, update the clustered gene libraries using a genetic algorithm, and output a configuration parameter table.

[0128] The prediction and optimization module is used to make predictions using the GP model. It sets the objective function and thermal balance constraints, transforms the problem into a quadratic binary optimization problem, uses the quantum annealing algorithm for parallel search, and outputs the initial running optimization parameters through discretization mapping and filtering. It then uses these parameters to calculate the deviation rate, updates the GP model parameters, and outputs the running optimization parameters.

[0129] The monitoring module is adjusted to optimize equipment frequency based on operational optimization parameters, perform real-time monitoring, and set alarm mechanisms.

[0130] The storage and analysis module is used to store the collected and analyzed multimodal data.

[0131] This embodiment also provides a computer device applicable to the energy-saving optimization method for water-cooled air conditioning in manufacturing plants, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the energy-saving optimization method for water-cooled air conditioning in manufacturing plants as proposed in the above embodiment.

[0132] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0133] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements the energy-saving optimization method for water-cooled air conditioning in manufacturing plants as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0134] In summary, this invention collects multimodal data, constructs an initial gene pool, uses Pareto fronts for screening, constructs an pyrokinetic coupling weight matrix using FCM, forms clustered gene pools through spectral decomposition and K-means, updates the clustered gene pools using a genetic algorithm, outputs a configuration parameter table, performs predictions using a GP model, sets the objective function and thermal equilibrium constraints, transforms it into a quadratic binary optimization problem, uses a quantum annealing algorithm for parallel search, outputs initial running optimization parameters through discretization mapping and screening, uses the data and calculates the deviation rate, updates the GP model parameters, and outputs running optimization parameters; thus improving the accuracy and adaptability of the initial system configuration and enhancing the dynamic adaptability and optimization efficiency of the running parameters.

[0135] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for optimizing energy saving in a water-cooled air conditioning system for a manufacturing plant, characterized in that: include, Multimodal data was collected, an initial gene pool was constructed, and Pareto fronts were used for screening. A fire coupling weight matrix was constructed using FCM. Clustered gene pools were formed using spectral decomposition and K-means. Genetic algorithms were used to update the clustered gene pools, and a configuration parameter table was output, including: The multimodal data includes infrared images of the equipment's exterior, ambient temperature and humidity, the equipment's current operating power, cooling tower outlet water temperature, chilled water supply and return temperatures, and chilled water flow rate. Based on infrared images, calculate the real-time exergy efficiency factor and exergy cost; The current operating power, chilled water supply and return temperatures, chilled water flow rate, rated power, design cooling capacity, and exergy cost of the equipment are defined as gene units; Use systematic sampling to extract N sets of equipment configurations, and calculate the total exergy cost and energy efficiency ratio of each set of equipment configurations. By setting a total exergy cost threshold based on historical operating data, equipment configurations with total exergy cost ≤ the total exergy cost threshold are screened. Using the Pareto front extraction method, based on non-dominated screening and crowding calculation, crowding is sorted in ascending order, and the top H group configurations with low crowding are screened to form an initial gene pool. The membership degree of the device configuration in the initial gene pool was calculated using the FCM method; The distance from each group of equipment to the centroid is calculated using the Euclidean formula, and the coupling weights are obtained by combining the membership degree and arranging them to form a coupling weight matrix. The exergy coupling weight matrix is ​​transformed into a similarity matrix through cosine similarity. A normalized Laplace matrix is ​​constructed, and the Laplace matrix is ​​subjected to spectral decomposition. The equipment configuration is clustered using the K-means algorithm, and the equipment configuration is divided into 3 clusters. The average exergy coupling weight of each cluster is calculated. The cluster with the highest average fire coupling weight is selected as the high coupling group, and a cluster gene library is generated. n groups are randomly selected as the parent generation, and the offspring coupling groups are generated through crossover mutation. Calculate the total exergy cost, energy efficiency ratio, and temperature deviation for the high-coupling group and the offspring coupling groups generated by cross-mutation; Calculate the ratio of actual cooling capacity to design cooling capacity to obtain the load rate; The threshold for total exergy cost minus energy efficiency ratio is set based on the total exergy cost and energy efficiency ratio. The threshold for temperature deviation is set based on historical temperature deviation data analysis. The maximum and minimum load rates are set based on the safety boundary method. The system filters equipment configurations that meet the thresholds of total cost of pyrolysis minus energy efficiency ratio ≤ total cost minus energy efficiency ratio, temperature deviation ≤ temperature deviation threshold, and load rate between maximum and minimum load rate. It then inputs the cluster gene library for updating, stops updating when the maximum number of updates is reached, outputs the latest gene library, and converts it into a configuration parameter table through structured mapping. Prediction is performed using the GP model. The objective function and thermal equilibrium constraints are set, transforming the problem into a quadratic binary optimization problem. A parallel search using the quantum annealing algorithm is employed. Through discretization mapping and filtering, initial optimization parameters are output. These parameters are then used to calculate the bias rate, update the GP model parameters, and finally, output the final optimization parameters, including: Extract the total exergy cost, energy efficiency ratio, load rate, chilled water flow rate, and current operating power of the equipment from the configuration parameter table, and extract the ambient temperature and humidity from the multimodal data; A GP model was constructed, and historical environmental temperature and humidity, equipment operating power, and chilled water flow rate were extracted from historical operating data to train the GP model. The ambient temperature and humidity, the current operating power of the equipment, and the chilled water flow rate are input into the GP model. The predicted mean and predicted variance are calculated through the posterior distribution. The predicted mean and predicted variance are formed into a predicted distribution through the Gaussian regression process. The predicted ambient temperature and humidity, equipment operating power, and chilled water flow rate for future times are output. Define the Wasserstein uncertainty set, use the steady-state thermal model to calculate the thermal equilibrium, set the thermal equilibrium threshold based on statistical analysis, if the deviation between the left and right sides of the thermal equilibrium is greater than the thermal equilibrium threshold, adjust the GP model parameters by maximizing the marginal likelihood, recalculate the predicted mean and predicted variance through the posterior distribution, update the predicted distribution, and stop when the deviation between the left and right sides of the thermal equilibrium is less than or equal to the thermal equilibrium threshold. Based on the predicted ambient temperature and humidity at future times, the PMV score is calculated using the PMV model. Define the objective function and set the heat balance formula as the constraint condition, transform it into a quadratic bivariate optimization problem, and define the Hamiltonian formula. The Hamiltonian is obtained by discretizing the equipment operating power, chilled water flow rate and PMV score for predicting future time into qubit states and substituting them into the Hamiltonian formula. The Hamiltonian is saved as a matrix, the number of samplings and annealing time are set, and the quantum annealing algorithm is used to predict the combination of equipment operating power and chilled water flow rate in the future through parallel search of quantum superposition states. The energy distribution is obtained by the number of samplings, the energy mean is calculated by weighted average, and solutions with less than 50% energy are screened out. Based on discretization calculation, the results are mapped back to continuous variables to generate candidate parameter groups. Based on the verification of thermal equilibrium and PMV, the parameter groups that pass the verification are selected from the candidate parameter groups, the objective function value is calculated, and the parameter with the smallest objective value is selected as the initial running optimization parameter. The initial operation optimization parameters are used to obtain real-time chilled water flow rate, equipment operating power and PMV score, and the deviation rate is calculated. If a deviation rate threshold is set based on the statistical significance test, and the deviation rate is greater than the deviation rate threshold, the real-time collected chilled water supply and return water temperature difference, power, and PMV score will be input into the GP model to update the model parameters. Optimization will stop when the deviation rate is less than or equal to the deviation rate threshold, and the operating optimization parameters, including the predicted chilled water flow rate, equipment operating power, and PMV score, will be output. Adjust and optimize equipment frequency based on operational optimization parameters, perform real-time monitoring, and set alarm mechanisms; Store the collected and analyzed multimodal data.

2. The energy-saving optimization method for water-cooled air conditioning in manufacturing plants as described in claim 1, characterized in that: The adjustment and optimization of equipment frequency based on operational optimization parameters includes: The operating optimization parameters will be used to adjust and optimize the water pump frequency, cooling tower fan speed, and chiller unit frequency converter frequency.

3. The energy-saving optimization method for water-cooled air conditioning in manufacturing plants as described in claim 2, characterized in that: The real-time monitoring and alarm setting mechanism includes: Every c minutes, the energy efficiency ratio, PMV score, and total power are recalculated. Thresholds for energy efficiency ratio, PMV score, and total efficiency are set. If the energy efficiency ratio is greater than the preset threshold, the chiller frequency coefficient is adjusted. If the PMV score is less than the preset threshold, the cooling tower fan speed coefficient is adjusted. If the total power exceeds the total power threshold, an alarm is triggered, and staff are notified to conduct an inspection.

4. The energy-saving optimization method for water-cooled air conditioning in manufacturing plants as described in claim 3, characterized in that: The multimodal data is first preprocessed, including size standardization, pixel normalization, mean filtering and noise reduction of infrared images, CLAHE enhancement of image features, and edge detection to extract thermal features of the images, as well as ambient temperature, ambient humidity, current operating power of the equipment, chilled water supply and return water temperatures, and chilled water flow rate.

5. The energy-saving optimization method for water-cooled air conditioning in manufacturing plants as described in claim 4, characterized in that: The multimodal data collected and analyzed includes: The collected multimodal data, along with the operational optimization parameters and adjustment frequency data generated from the analysis, are used to generate a complete evaluation report through data visualization and then uploaded to the central database.

6. An energy-saving optimization system for water-cooled air conditioning in a manufacturing plant, based on the energy-saving optimization method for water-cooled air conditioning in a manufacturing plant according to any one of claims 1 to 5, characterized in that: include, The collection and configuration module is used to collect multimodal data, build an initial gene library, screen it using the Pareto front, construct the pyrokinetic coupling weight matrix through FCM, form clustered gene libraries through spectral decomposition and K-means, update the clustered gene libraries using a genetic algorithm, and output a configuration parameter table. The prediction and optimization module is used to make predictions using the GP model. It sets the objective function and thermal balance constraints, transforms the problem into a quadratic binary optimization problem, uses the quantum annealing algorithm for parallel search, and outputs the initial running optimization parameters through discretization mapping and filtering. It then uses these parameters to calculate the deviation rate, updates the GP model parameters, and outputs the running optimization parameters. The monitoring module is adjusted to optimize equipment frequency based on operational optimization parameters, perform real-time monitoring, and set alarm mechanisms. The storage and analysis module is used to store the collected and analyzed multimodal data.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the energy-saving optimization method for water-cooled air conditioning in manufacturing plants as described in any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the energy-saving optimization method for water-cooled air conditioning in manufacturing plants as described in any one of claims 1 to 5.

Images