Data-driven central air conditioner cooling station energy efficiency optimization control method and system
By installing microbial sensors in central air conditioning chiller stations and analyzing social media data, combined with blockchain and quantum computing, the operating strategies of air conditioning equipment are optimized, solving the problem of limited energy efficiency improvement potential in existing technologies and achieving improvements in energy efficiency and user satisfaction.
Patent Information
- Application Number
- CN202511699887.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-19
- Publication Date
- 2026-01-09
AI Technical Summary
Existing central air conditioning chiller plant control systems rely on empirical rules, which cannot adapt to complex and ever-changing operating conditions, and ignore the impact of user experience and microbial environment on energy efficiency, resulting in limited room for energy efficiency improvement.
By employing a data-driven approach, microbial sensors are installed in air conditioning chiller plants to monitor bacterial concentrations. Combined with social media data analysis of user feedback, a multidimensional dataset is constructed. Blockchain and quantum computing are then used to optimize the operation strategies of air conditioning equipment, thereby achieving energy efficiency optimization.
It improves the energy efficiency optimization efficiency of central air conditioning chiller stations, can respond to user needs in real time, prevent microbial contamination, provide an intuitive energy efficiency improvement path, and quickly converge to the global optimal solution.
Smart Images

Figure CN121297202A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of central air conditioning, and more particularly to a data-driven method and system for optimizing the energy efficiency of central air conditioning chiller plants. Background Technology
[0002] With the rapid development of society and economy and the continuous improvement of people's living standards, central air conditioning systems are increasingly widely used in various buildings, covering many places such as commercial office buildings, hotels, hospitals, and large shopping malls. Currently, the control systems of central air conditioning refrigeration stations have the following problems: Central air conditioning chiller plant control largely relies on experience-based rules. Operators adjust equipment operating parameters, such as temperature setpoints and fan speeds, based on past experience and judgment of system operating status. While this method is simple and easy to implement, experience-based rules are often difficult to adapt to various complex and changing operating conditions, making it impossible to achieve optimal operation of the chiller plant and resulting in limited room for energy efficiency improvement. Secondly, existing energy efficiency optimization control methods rely solely on internal operating parameters of the chiller plant, such as temperature, pressure, and flow rate, while neglecting other important factors affecting the chiller plant's energy efficiency, such as user experience and feedback on the air conditioning system, and the indoor microbial environment. Although these factors do not directly participate in the physical operation of the chiller plant, they have a significant impact on the overall performance of the system and user satisfaction. Control methods that lack multi-source data fusion cannot comprehensively and accurately assess the operating status of the chiller plant, causing inconvenience to actual use.
[0003] Therefore, it is necessary to provide a new data-driven method and system for optimizing the energy efficiency of central air conditioning chiller plants to solve the above-mentioned technical problems. Summary of the Invention
[0004] To address the aforementioned technical problems, this invention provides a data-driven method and system for optimizing the energy efficiency of central air conditioning chiller plants.
[0005] The data-driven energy efficiency optimization control method for central air conditioning chiller stations provided by this invention includes the following steps: Step 1: Data Acquisition: First, microbial sensors are installed in key areas of the central air conditioning chiller to monitor the bacterial concentration in real time. At the same time, parameters of the traditional central air conditioning chiller are collected. The microbial concentration data and the traditional central air conditioning chiller parameters are spatiotemporally aligned to construct a multidimensional dataset and preprocessed. Step 2, Data Import: After completing the data collection inside the central air conditioning chiller station, use the API interface to crawl posts related to the air conditioning system on social media, filter advertisements and non-local posts, calculate the sentiment ratio of negative posts and associate them with specific building areas; Step 3, Model Building: Based on the internal data of the chiller plant collected in Step 1 and the social media data introduced in Step 2, the chiller units, cooling towers and water pumps are modeled as independent intelligent agents. The objective function of each intelligent agent is set as local energy efficiency optimization. Power resources are dynamically allocated according to the current energy efficiency of the air conditioning equipment. The more efficient ones get more power. At this time, the air conditioning equipment shares the operating status through blockchain, and smart contracts coordinate the actions of the air conditioning equipment to build a competition-cooperation model. The shared operating status includes the real-time cooling capacity of the cooling tower, and the smart contract coordinates the actions of the air conditioning equipment, including the chiller unit adjusting the outlet water temperature according to the cooling tower status. Step 4, Model Training: Using the competition-cooperation model constructed in Step 3, the on / off states of the air conditioning equipment are encoded as qubits, and the optimal strategy combination is solved using the D-Wave quantum annealing machine or the QAOA variable quantum algorithm. Step 5: Smart control strategy formulation: Each air conditioning unit is used as a blockchain node to store local data and smart contracts. When the proportion of negative emotions on social media rises, that is, when the concentration of microorganisms exceeds the standard, the smart contract automatically triggers control commands to adjust the temperature setpoint or start the Pareto front maintenance. The cooling tower fan speed and water pump frequency are regarded as ant paths, and the energy efficiency optimal solution corresponds to the shortest path. Step Six: Implementation and Continuous Optimization: Apply the intelligent control strategy developed in Step Five to the actual operation of the central air conditioning chiller station, continuously collect new operating data and user feedback data, and use the new operating data and user feedback data to update and optimize the competition-cooperation model.
[0006] Preferably, in step one, the fungi include bacteria and fungi.
[0007] Preferably, in step one, the key areas include the air filter inlet, the condensate pan, and the inner wall of the duct.
[0008] Preferably, in step one, the parameters of a traditional central air conditioning chiller include the temperature, humidity, and air velocity of the central air conditioning chiller.
[0009] Preferably, in step one, the microbial sensor includes a sensor based on ATP biofluorescence detection or PCR technology.
[0010] Preferably, in step one, the time alignment includes aggregating the average values of traditional central air conditioning chiller parameters within a 5-10 minute time window based on the sampling time of the microbial sensor, and associating the microbial data with the traditional parameters to the same physical location through a region identifier.
[0011] Preferably, in step two, the sentiment ratio of the negative posts is the ratio of the number of negative posts to the total number of posts, and the specific building area includes the geographical location tag of the building.
[0012] Preferably, in steps three, four, and five, the air conditioning equipment includes a chiller, a cooling tower, a water pump, and a sensor.
[0013] Preferably, in step five, the local data includes temperature and energy consumption.
[0014] The data-driven energy efficiency optimization and control system for central air conditioning chiller includes a data acquisition module, which is used to collect microbial concentration data and traditional central air conditioning chiller parameters in key areas of the central air conditioning chiller. The microbial concentration data and traditional central air conditioning chiller parameters are spatiotemporally aligned to construct a multidimensional dataset and preprocessed. The data import module is used to crawl and filter posts related to air conditioning systems from social media data after preprocessing, calculate the sentiment ratio of negative posts, and associate the ratio with specific building areas. The model building module is used to model chillers, cooling towers and water pumps as independent intelligent agents based on the data provided by the data acquisition module and the data import module. Local energy efficiency optimization is set as the objective function for each intelligent agent. Power resources are dynamically allocated according to the current energy efficiency of the air conditioning equipment. The more efficient air conditioning equipment receives more power. Blockchain technology is used to realize the sharing of the operating status between air conditioning equipment. The actions of air conditioning equipment are coordinated through smart contracts to build a competition-cooperation model. The model training module is used to encode the on / off states of air conditioning equipment into qubits using a competition-cooperation model, and to solve for the optimal strategy combination using the D-Wave quantum annealing machine or the QAOA variable quantum algorithm. The intelligent control strategy formulation module is used to store local data and smart contracts with each air conditioning unit as a blockchain node. It monitors the proportion of negative sentiment on social media and the concentration of microorganisms in real time. When the proportion of negative sentiment on social media rises or the concentration of microorganisms exceeds the standard, the smart contract is triggered. The smart contract automatically triggers control commands to adjust the temperature setpoint or start the Pareto front maintenance. The cooling tower fan speed and water pump frequency are regarded as ant paths. The energy efficiency optimal solution corresponds to the shortest path, realizing the intelligent control of the cooling plant. The implementation and continuous optimization module is used to apply the intelligent control strategies formulated by the intelligent control strategy formulation module to the actual operation of the central air conditioning chiller, and to continuously collect new operating data and user feedback data to update and optimize the competition-cooperation model.
[0015] Compared with related technologies, the data-driven central air conditioning chiller plant energy efficiency optimization control method and system provided by the present invention have the following beneficial effects: Existing methods rely solely on internal parameters of the chiller plant, neglecting the implicit impact of user experience and the microbial environment on energy efficiency. This invention analyzes real-time user complaints and evaluations of the air conditioning system, quantifies user satisfaction, and correlates it with specific building areas. This provides user-oriented optimization targets for control strategies. The system automatically adjusts temperature setpoints when negative emotions rise and monitors microbial concentrations in key areas of the chiller plant in real time to prevent equipment corrosion or air quality problems caused by excessive microbial levels. Simultaneously, it aligns microbial data with operating parameters in time and space to uncover their potential correlation with energy efficiency. This allows the system to comprehensively assess the operating status of the chiller plant, thus solving, to some extent, the inconveniences caused by relying on empirical rules or solely on internal parameters of the chiller plant. This invention treats the cooling tower fan speed and water pump frequency as ant paths, with the energy-efficient solution corresponding to the shortest path. By simulating the pheromone transmission mechanism in an ant colony, the energy-efficient solution, i.e. the corresponding shortest path, is found. This improves computational efficiency and provides maintenance personnel with an intuitive path to improve energy efficiency. Traditional optimization methods, such as genetic algorithms, are prone to getting stuck in local optima when dealing with high-dimensional, nonlinear chiller control problems. This invention encodes the on / off states of air conditioning equipment into qubits and uses the D-Wave quantum annealing machine or QAOA algorithm to solve for the optimal strategy combination. The parallelism and entanglement characteristics of quantum computing enable it to efficiently explore the solution space and quickly converge to the global optimal solution, that is, the minimum energy consumption state of the equipment combination, which significantly improves optimization efficiency and is especially suitable for real-time decision-making in large-scale chiller systems. Attached Figure Description
[0016] Figure 1 A flowchart illustrating the data-driven energy efficiency optimization control method for central air conditioning chiller plants provided by this invention; Figure 2 This is a system block diagram of a data-driven energy efficiency optimization control system for central air conditioning chiller plants. Detailed Implementation
[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments.
[0018] Please refer to the following: Figures 1 to 2 ,in, Figure 1 A flowchart illustrating the data-driven energy efficiency optimization control method for central air conditioning chiller plants provided by this invention; Figure 2 This is a system block diagram of a data-driven energy efficiency optimization control system for central air conditioning chiller plants. Example 1
[0019] For reference Figure 1 As shown, the data-driven energy efficiency optimization control method for central air conditioning chiller plants includes the following operational steps: Step 1: Data Acquisition. First, install microbial sensors based on ATP biofluorescence detection or PCR technology at the air filter inlet, condensate pan, and inner wall of the duct in the central air conditioning chiller to monitor the concentration of bacteria and fungi in real time. If the sensors are equipped, the sampling frequency is once every 10 minutes. At the same time, collect the temperature, humidity, air velocity, pressure, power consumption, etc. of the traditional central air conditioning chiller. Ensure that the time is synchronized with the chiller control system through the NTP protocol, and ensure that the sensor location corresponds to the microbial sensor area, such as the same duct section or cooling tower area. Spatiotemporally align the microbial concentration data with the parameters of the traditional central air conditioning chiller to construct a multidimensional dataset, and preprocess the multidimensional dataset. It should be noted that the NTP protocol, also known as the Network Time Protocol, is a protocol used to synchronize the clocks of devices in a computer network. Its main goal is to ensure that all devices in the network have a consistent time reference so that they can work together to perform timestamp recording, data synchronization, and various computing tasks. NTP uses a layered structure to ensure time synchronization, so that all devices in the network can obtain accurate time information. It should be noted that time alignment includes aggregating the average values of traditional central air conditioning chiller parameters within a 5-10 minute time window based on the sampling time of the microbial sensor, and associating the microbial data with the traditional parameters to the same physical location through area identifiers, such as: air supply zone A, cooling tower 1; It should be noted that the purpose of the preprocessing operation is to remove outliers, such as extreme values caused by sensor failure, and to use linear interpolation or prediction methods based on historical patterns to interpolate and complete the missing data. It is worth noting that for linear interpolation: based on the values of the known data points adjacent to the missing data points, assuming that the data changes uniformly between adjacent points, the missing values are calculated and filled according to a linear proportional relationship; Historical pattern-based forecasting: Utilizing patterns in historical data under similar times or conditions, such as periodicity or trends, to predict missing values through statistical models or algorithms, such as moving averages and time series analysis. It is worth noting that microbial sensors based on PCR technology, such as real-time quantitative PCR sensors, work by adding fluorescent labels to the PCR reaction system and quantitatively analyzing the amount of initial template DNA by monitoring the intensity of the fluorescence signal in real time, which is used to monitor the concentration of bacteria and fungi. Step 2: Data Import: After completing the internal data collection of the central air conditioning chiller station, retrieve posts related to the air conditioning system from social media via API, filter out advertisements and non-local posts, and calculate the sentiment ratio of negative posts. And associate it with specific building areas; It should be noted that advertising and promotional posts are filtered by identifying content keywords or publisher types; non-local posts are filtered by geolocation tags or IP addresses. Before calculating the sentiment ratio of negative posts, natural language processing tools such as VADER and TextBlob are used to calculate the sentiment tendency of posts to determine whether they are positive, negative, or neutral. Then, the proportion of negative posts is counted and classified by building area, such as office building A and shopping mall B. The proportion of negative sentiment on social media is mapped to specific areas of cold sites by associating them with building IDs or cold site partition identifiers. Step 3, Model Building: Based on the internal data of the chiller plant collected in Step 1 and the social media data introduced in Step 2, the chiller units, cooling towers and water pumps are modeled as independent intelligent agents. The objective function of each intelligent agent is set as local energy efficiency optimization. Power resources are dynamically allocated according to the current energy efficiency of the air conditioning equipment. The more efficient ones get more power. At this time, the air conditioning equipment shares the operating status through blockchain, and smart contracts coordinate the actions of the air conditioning equipment to build a competition-cooperation model. Among them, the shared operating status includes the real-time cooling capacity of the cooling tower, and the smart contract coordinates the actions of the air conditioning equipment, including the chiller unit adjusting the outlet water temperature according to the cooling tower status; It should be noted that the specific steps for model building are as follows: S1. Define agent attributes: This includes defining the chiller agent, cooling tower agent, and water pump agent; For chiller unit intelligent agents: State variables: current operating power (kW), energy efficiency ratio (COP), outlet water temperature (°C), load rate (%).
[0020] Objective function: Minimize energy consumption per unit cooling capacity, i.e. Energy consumption is measured in kW / cdotph, and cooling capacity is measured in tons. For cooling tower agents: State variables: real-time cooling capacity (kW), inlet water temperature (°C), fan speed (rpm).
[0021] Objective function: Maximize cooling efficiency, i.e. Cooling capacity is measured in kW; For the pump agent: State variables: current frequency (Hz), flow rate (m³ / h), energy consumption (kW).
[0022] Objective function: Minimize energy consumption per unit flow, i.e. The unit of flow rate is m³; S2. Constructing independent intelligent agents: Treat each device, including chiller, cooling tower and water pump, as an independent intelligent agent, assign it state variables and objective function, collect state variables such as power, temperature and flow rate in real time through sensors, calculate the energy consumption per unit cooling capacity every 5 minutes, and update the objective function value periodically. S3. Dynamic allocation of power resources: Obtain real-time state variables from the intelligent agent, such as COP value and cooling efficiency, and conduct energy efficiency assessment of chiller units, cooling towers and water pumps. Chiller units: Calculate the COP value, i.e. ; Cooling tower: Calculate cooling efficiency, i.e. ; Water pump: Calculate energy consumption per unit flow rate, i.e. ; At this point, the units are sorted from highest to lowest energy efficiency value to generate a priority list, such as chillers with COP=5.0 being ranked first. S4. Dynamic Power Quota Allocation: Set the upper limit of available power for the chiller plant. For example, if the upper limit is 1000kW, the power quota will be allocated as follows: the most efficient plant will be allocated 40% of the budget, i.e., 400kW; the second most efficient plant will be allocated 30%, i.e., 300kW; and the least efficient plant will be allocated 30%, i.e., 300kW. The energy efficiency ranking will be recalculated and the allocation will be adjusted every 10 minutes to ensure real-time optimization. S5. Blockchain State Sharing: Each agent encrypts its state variables, such as the real-time cooling capacity of the cooling tower, and uploads them to the blockchain network Hyperledger Fabric. The blockchain's chaincode ensures that the data is immutable. Other agents can query the shared state, and the power allocation results are recorded on the blockchain. S6. Define smart contract rules: Set trigger conditions and action logic. Set the trigger conditions as follows: the cooling tower cooling capacity decreases by 10%, i.e., the current value is less than the historical average value × 90%, and the chiller unit outlet water temperature deviates from the set value by ±2℃. Set the action logic as follows: if the cooling tower inlet water temperature increases by 2℃, the chiller unit load decreases by 10%; if it decreases by 2℃, the load increases by 10%. Record the adjustment action to the blockchain. S7. Deploy smart contracts: Choose a blockchain platform: Ethereum or Hyperledger Fabric. Write the contract logic using Solidity or Go / Java and put it on the chain. During the on-chain process, Ethereum, i.e., the public chain, is deployed to the Rinkeby testnet through the Remix IDE tool, while Hyperledger is deployed to the consortium chain node through the chaincode container of Fabric. S8. Decentralized execution: Contract actions are executed by consensus among blockchain nodes to avoid single points of failure. Each agent stores the contract address and triggering conditions, and checks them periodically. If multiple agents have conflicting triggering conditions, they are adjudicated according to priority, i.e., energy efficiency ranking, and a competition-cooperation model is established.
[0023] Step 4, Model Training: Using the competition-cooperation model constructed in Step 3, the on / off states of the air conditioning equipment are encoded as qubits, and the optimal strategy combination is solved using the D-Wave quantum annealing machine or the QAOA variable quantum algorithm. It is worth noting that, during use, the on / off state of the air conditioning equipment can be encoded as qubits, with 0 representing off and 1 representing on. A quantum state is constructed to represent the combined states of all equipment. For example, 3 equipment corresponds to 3 qubits, resulting in a total of 8 states. The optimal strategy combination is solved using the D-Wave quantum annealing machine or the QAOA quantum approximation optimization algorithm. The objective function is defined as minimizing the total energy consumption of the system. The solution space is explored through quantum entanglement and superposition to find the global optimal solution. The quantum solution is then converted into a binary representation, with 00 representing equipment 1 off, equipment 2 on, and equipment 3 off. The system also verifies whether the equipment load exceeds the rated value. It should be noted that the specific steps for model training are as follows: Step 1: Encoding the on / off state of air conditioning equipment with qubits: Encode the on / off state of each air conditioning unit into qubits, where 0 represents the device being off and 1 represents the device being on. Construct a quantum superposition state to represent all possible combinations of device states:
[0024] Where n is the number of devices. This represents the i-th device state. For complex amplitudes, Indicates the combined state of all devices; Step 2, Definition of Objective Function: Define the objective function as the total energy consumption of the chiller system, denoted as... ;
[0025] in, Let m be the power of device k in state s, and m be the total number of devices. Step 3: Solving for the optimal strategy combination using quantum algorithms: For one of them, such as the D-Wave quantum annealing machine, the objective function is transformed into QUBO form:
[0026] in, , represents the on / off state of the k-th device, where 0 indicates off and 1 indicates on; Indicates the cost or power consumption of enabling device (k); : This is the penalty coefficient, used to force the constraint to be met. C: The set of all device pairs that require constraints to be imposed. , represents the rated load limit of device i in state j; This represents the value of the i-th qubit in the j-th iteration or state. This indicates the switching state of device i under a certain combination state j; m is the total number of devices; At this point, encode the QUBO problem into a format acceptable to D-Wave, such as .json or .iqp, submit the task via the D-Wave API, and set the parameters: Chain Strength: Controls the coupling strength of the qubits, set to 1.2 times the problem energy scale.
[0027] Annealing Time: Adjust according to the complexity of the problem, such as from 20μs to 200μs; After the parameters are set, D-Wave returns the optimal solution. For example, 101 is the optimal solution, which means that device 1 is off, device 2 is on, and device 3 is off. At this time, verify whether the solution meets the constraints, such as device 2 load rate = 80% ≤ 100%. Step 4: Data Conversion and Verification: Using the quantum solution D-Wave's return value of 101, and based on real-time equipment parameters (e.g., chiller unit current load rate = 80%), check if the activated equipment exceeds its rated value. If equipment 2's rated load is 100kW and the current load is 80kW, the solution is valid. If it exceeds the limit (e.g., load = 110kW), mark the solution as infeasible, retrain the model, and calculate the total system energy consumption corresponding to the optimal solution. The optimization effect is evaluated by comparing it with historical data, such as a 15% reduction in energy consumption; Step 5: Iterative optimization: Incorporate newly collected operational data, such as temperature adjustments caused by user complaints and the proportion of social media sentiment, into the training set, reconstruct the competition-cooperation model, and adjust the QUBO penalty coefficient λ based on the validation results to improve the solution accuracy.
[0028] Step 5: Smart control strategy formulation: Each air conditioning unit is used as a blockchain node to store local data and smart contracts. When the proportion of negative emotions on social media rises, that is, when the concentration of microorganisms exceeds the standard, the smart contract automatically triggers control commands to adjust the temperature setpoint or start the Pareto front maintenance. The cooling tower fan speed and water pump frequency are regarded as ant paths, and the energy efficiency optimal solution corresponds to the shortest path. It should be noted that during blockchain node deployment, each air conditioning unit acts as a blockchain node, storing local data such as operation logs and smart contracts. Nodes communicate via a P2P network. When the microbial concentration exceeds a threshold, such as a preset threshold of 50 CFU / m³, the smart contract automatically performs maintenance actions. When the proportion of negative sentiment on social media increases by 10%, the temperature setpoint is adjusted, such as reducing the supply air temperature by 1°C. The ant colony algorithm is optimized, treating the cooling tower fan speed and water pump frequency as adjustable parameters in the ant colony algorithm. The algorithm finds the energy-efficient solution corresponding to the shortest path, defining the path cost as the total system energy consumption. Ants select parameter combinations based on pheromone concentration, iteratively updating the optimal path. The smart contract generates control commands, such as adjusting the water pump frequency to 40Hz, which are then sent to the equipment through the chiller control system, such as the BACnet protocol.
[0029] Step Six: Implementation and Continuous Optimization: Apply the intelligent control strategy developed in Step Five to the actual operation of the central air conditioning chiller station, continuously collect new operating data and user feedback data, and use the new operating data and user feedback data to update and optimize the competition-cooperation model.
[0030] Specifically, existing methods rely solely on internal parameters of the chiller plant, neglecting the implicit impact of user experience and the microbial environment on energy efficiency. This invention analyzes real-time user complaints and evaluations of the air conditioning system, quantifies user satisfaction, and correlates it with specific building areas. This provides user-oriented optimization targets for control strategies, automatically adjusts temperature setpoints when negative emotions rise, and monitors the concentration of bacteria in key areas of the chiller plant in real time to prevent equipment corrosion or air quality problems caused by excessive microorganisms. At the same time, it aligns microbial data with operating parameters in time and space to explore their potential correlation with energy efficiency, enabling the system to comprehensively evaluate the operating status of the chiller plant. To a certain extent, this solves the inconvenience caused by relying on empirical rules or solely on internal parameters of the chiller plant. Furthermore, this invention treats the cooling tower fan speed and water pump frequency as ant paths, with the energy-efficient solution corresponding to the shortest path. By simulating the pheromone transmission mechanism in an ant colony, the energy-efficient solution, i.e. the corresponding shortest path, is found, which improves computational efficiency and provides maintenance personnel with an intuitive path for energy efficiency improvement. Furthermore, traditional optimization methods, such as genetic algorithms, are prone to getting trapped in local optima when dealing with high-dimensional, nonlinear chiller control problems. This invention encodes the on / off states of air conditioning equipment into qubits and uses the D-Wave quantum annealing machine or QAOA algorithm to solve for the optimal strategy combination. The parallelism and entanglement characteristics of quantum computing enable it to efficiently explore the solution space and quickly converge to the global optimal solution, that is, the minimum energy consumption state of the equipment combination, which significantly improves optimization efficiency and is especially suitable for real-time decision-making in large-scale chiller systems.
[0031] Example 2 For reference Figure 2 As shown, the data-driven energy efficiency optimization control system for central air conditioning chiller includes a data acquisition module, which is used to collect microbial concentration data and traditional central air conditioning chiller parameters in key areas of the central air conditioning chiller. The microbial concentration data and traditional central air conditioning chiller parameters are spatiotemporally aligned to construct a multidimensional dataset and preprocessed. The data import module is used to crawl and filter posts related to air conditioning systems from social media data after preprocessing, calculate the sentiment ratio of negative posts, and associate the ratio with specific building areas. The model building module is used to model chillers, cooling towers and water pumps as independent intelligent agents based on the data provided by the data acquisition module and the data import module. Local energy efficiency optimization is set as the objective function for each intelligent agent. Power resources are dynamically allocated according to the current energy efficiency of the air conditioning equipment. The more efficient air conditioning equipment receives more power. Blockchain technology is used to realize the sharing of the operating status between air conditioning equipment. The actions of air conditioning equipment are coordinated through smart contracts to build a competition-cooperation model. The model training module is used to encode the on / off states of air conditioning equipment into qubits using a competition-cooperation model, and to solve for the optimal strategy combination using the D-Wave quantum annealing machine or the QAOA variable quantum algorithm. The intelligent control strategy formulation module is used to store local data and smart contracts with each air conditioning unit as a blockchain node. It monitors the proportion of negative sentiment on social media and the concentration of microorganisms in real time. When the proportion of negative sentiment on social media rises or the concentration of microorganisms exceeds the standard, the smart contract is triggered. The smart contract automatically triggers control commands to adjust the temperature setpoint or start the Pareto front maintenance. The cooling tower fan speed and water pump frequency are regarded as ant paths. The energy efficiency optimal solution corresponds to the shortest path, realizing the intelligent control of the cooling plant. The implementation and continuous optimization module is used to apply the intelligent control strategies formulated by the intelligent control strategy formulation module to the actual operation of the central air conditioning chiller, and to continuously collect new operating data and user feedback data to update and optimize the competition-cooperation model.
[0032] Specifically, by capturing negative posts related to air conditioning systems on social media, such as stuffy indoor air and unpleasant odors, calculating the emotional ratio and linking it to specific building areas, the user's subjective experience is quantified into optimizable goals. When the emotional ratio rises, cooling or purification actions are triggered, achieving triple optimization of energy efficiency, comfort, and health.
[0033] The circuits and controls involved in this invention are all existing technologies and will not be described in detail here.
[0034] The above are merely embodiments of the present invention and do not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.
Claims
1. A data-driven central air conditioning cooling station energy efficiency optimization control method, characterized in that, The method comprises the following steps: Step one, data collection: first, install a microorganism sensor in the key area of the central air conditioning cold station to monitor the concentration of bacteria in real time, collect traditional central air conditioning cold station parameters, and time-space align the microorganism concentration data with the traditional central air conditioning cold station parameters to build a multi-dimensional data set and perform preprocessing; Step two, data introduction: after completing the data collection inside the central air conditioning cold station, the posts related to the air conditioning system in the social media are grabbed through the API interface, the advertisements and non-local posts are filtered, the proportion of negative posts is calculated and associated with the specific building area; Step three, model building: based on the internal data of the cold station collected in step one and the social media data introduced in step two, the chiller, cooling tower and water pump are modeled as independent agents, the objective function of each agent is set as local energy efficiency optimization, the power resources are dynamically allocated according to the current energy efficiency of the air conditioning equipment, and the more efficient one obtains more power, at this time the air conditioning equipment shares the running state through the blockchain, the smart contract coordinates the actions of the air conditioning equipment, and a competition-cooperation model is built; The shared running state includes the real-time cooling capacity of the cooling tower, and the smart contract coordinating the actions of the air conditioning equipment includes adjusting the outlet water temperature of the chiller according to the cooling tower state; Step four, model training: using the competition-cooperation model built in step three, the on-off state of the air conditioning equipment is coded as a quantum bit, and the D-Wave quantum annealing machine or QAOA variational quantum algorithm is used to solve the optimal strategy combination; Step five, intelligent control strategy formulation: taking each air conditioning equipment as a blockchain node, storing local data and smart contracts, when the proportion of negative emotions in social media rises, i.e. the microorganism concentration exceeds the standard, the smart contract automatically triggers the control instruction to adjust the temperature setting value or start the maintenance Pareto frontier, and the cooling tower fan speed and water pump frequency are regarded as ant path, and the most energy efficient solution corresponds to the shortest path; Step six, implementation and continuous optimization: the intelligent control strategy formulated in step five is applied to the actual operation of the central air conditioning cold station, new operation data and user feedback data are continuously collected, and the competition-cooperation model is updated and optimized using the new operation data and user feedback data. 2.The data-driven central air conditioning cold station energy efficiency optimization control method of claim 1, wherein, In the step one, the bacteria include bacteria and fungi. 3.The data-driven central air conditioning cold station energy efficiency optimization control method of claim 2, wherein, In the step one, the key areas include the air filter inlet, the condensate water tray and the inner wall of the air duct.
4. The data-driven central air conditioning cold station energy efficiency optimization control method of claim 3, wherein, In the step one, the traditional central air conditioning cold station parameters include the temperature, humidity and air flow rate of the central air conditioning cold station.
5. The data-driven central air conditioning cold station energy efficiency optimization control method of claim 4, wherein, In the step one, the microorganism sensor includes a sensor based on ATP bioluminescence detection or PCR technology. 6.The data-driven central air conditioning cold station energy efficiency optimization control method of claim 5, wherein, In the step one, the time alignment includes taking the sampling time of the microorganism sensor as the benchmark, taking the average value of the traditional central air conditioning cold station parameters within 5-10 minutes for time window aggregation, and associating the microorganism data and the traditional parameters to the same physical location through the area identifier.
7. The data-driven central air conditioning cold station energy efficiency optimization control method of claim 6, wherein, In the step two, the proportion of negative emotions of the negative posts is the ratio of the number of negative posts to the total number of posts, and the specific building area includes the geographical location label of the building. 8.The data-driven central air conditioning cold station energy efficiency optimization control method of claim 7, wherein, In the third step, the fourth step and the fifth step, the air conditioning equipment includes a water chiller, a cooling tower, a water pump and a sensor. 9.The data-driven central air conditioning cold station energy efficiency optimization control method of claim 8, wherein, In the fifth step, the local data includes temperature and energy consumption.
10. The data-driven central air conditioning cooling station energy efficiency optimization control system of any one of claims 1-9, wherein, The data acquisition module is configured to collect microorganism concentration data and traditional central air conditioning station parameters in key areas of the central air conditioning cold station, align the microorganism concentration data with the traditional central air conditioning station parameters in time and space, construct a multi-dimensional data set, and perform preprocessing. The data introduction module is configured to, after preprocessing, capture and filter posts related to the air conditioning system in the social media data, calculate the proportion of negative posts, and associate the proportion with a specific building area. The model construction module is configured to model the water chiller, the cooling tower and the water pump as independent agents based on the data provided by the data acquisition module and the data introduction module, set local energy efficiency optimization as a target function for each agent, dynamically allocate power resources according to the current energy efficiency of the air conditioning equipment, and allocate more power to the air conditioning equipment with higher efficiency. The operation status of the air conditioning equipment is shared by using the blockchain technology, the actions of the air conditioning equipment are coordinated by the smart contract, and a competition-cooperation model is constructed. The model training module is configured to use the competition-cooperation model to encode the on-off state of the air conditioning equipment as a quantum bit, and use a D-Wave quantum annealing machine or a QAOA variational quantum algorithm to solve the optimal strategy combination. The intelligent control strategy formulation module is configured to use each air conditioning equipment as a blockchain node to store local data and a smart contract, monitor the negative sentiment proportion of social media and the microorganism concentration in real time, trigger the smart contract when the negative sentiment proportion of social media rises or the microorganism concentration exceeds the standard, automatically trigger the control instruction of the smart contract, adjust the temperature setting value or start the maintenance Pareto frontier, and regard the cooling tower fan speed and the water pump frequency as an ant path, so that the energy efficiency optimal solution corresponds to the shortest path, and the intelligent control of the cold station is realized. The implementation and continuous optimization module is configured to apply the intelligent control strategy formulated by the intelligent control strategy formulation module to the actual operation of the central air conditioning cold station, continuously collect new operation data and user feedback data, and update and optimize the competition-cooperation model.
