851 results about "Central air conditioning" patented technology

The invention relates to an ice crystal evaporator and an ice crystal water cooling device producing from the same, belonging to the technical field of refrigeration. The ice crystal evaporator is characterized in that a liquid inlet channel is formed by an evaporator tube, a liquid inlet and a liquid outlet in a tubular casing; and the outer channel of the evaporator tube of the tubular casing is divided into more than one S-shaped channel forming a water return channel through a water inlet of a water divider and a water outlet of a water collector. The ice crystal water cooling device is characterized in that the return air inlet of the ice crystal evaporator is communicated with a compressor, a condenser, a liquid feeding valve, a dry filter, a liquid feeding electromagnetic valve anda thermostatic expansion valve in sequence, and then is communicated with the liquid inlet of the ice crystal evaporator; a cooling water inlet / outlet of the condenser is connected with a cooling tower in series; and a water inlet / outlet of the ice crystal evaporator is communicated with a water circulating pump and an ice crystal energy storage tank in series. The invention has the advantages ofhigh cooling efficiency, no icing and low cooling temperature and is suitable for production of energy storage central air-conditioning, marine fishing, seafood processing, fruit, vegetable or meat processing, and the like; the cooled medium is municipal water.

The invention relates to an intelligent regulation and control method for a fan coil of a central air conditioner, relates to an internet of things technology, and aims to solve the problems that multi-source environment parameter data is large in noise, working condition modeling is inaccurate, and equipment regulation response and energy consumption optimization are not coordinated. A regulation and control system integrating multi-parameter distributed sensing, data preprocessing, feature extraction and cross-domain knowledge migration is provided. A high-quality structured data set is formed through multi-dimensional real-time acquisition of room temperature and humidity, fan current, valve opening, human traffic, external weather and building information and in combination with denoising, normalization, alignment and missing completion. And a multi-target neural network and digital twinborn simulation are utilized to realize load trend prediction, equipment abnormity early warning and virtual deduction of an adjustment strategy, an optimal adjustment time sequence window and an activation threshold are output, and dynamic adaptive control is realized in combination with reinforcement learning.

The invention relates to the field of heating ventilation air conditioner automatic control, and discloses an efficient central air conditioner cold station optimization control system and method based on physical AI. The system comprises a data acquisition module, an energy efficiency modeling module, a load prediction module, a rolling optimization module, an execution control module and a feedback correction module. The data acquisition module forms a time sequence data set; the energy efficiency modeling module adopts transfer learning to obtain a system energy efficiency model of a target domain cold station; the load prediction module outputs a predicted cold load sequence; the rolling optimization module is used for solving an optimal control sequence under the constraints of cooling capacity balance, an equipment operation boundary and a temperature difference threshold by taking a predicted cooling load sequence, a system energy efficiency model and a current equipment operation state as input under a model prediction control framework; the execution control module issues a first control action to control the water supply temperature, the water pump and fan frequency and unit start and stop; and the feedback correction module calculates a residual error and is used for closed-loop correction of the next period.

The invention discloses a fault early warning method and system for the equipment health state of a central air-conditioning system, and belongs to the technical field of central air-conditioning system operation monitoring. Multi-source operation state data in the central air-conditioning system are obtained, a state feature matrix is constructed, potential target feature points are extracted, and an equipment health state evolution path is constructed; performing multi-stage time sequence segmentation on the evolution path, identifying similar evolution trend fragments based on a density peak clustering algorithm, and classifying typical degradation modes; according to the evolution trend and the health scoring result, constructing a dynamic early warning threshold value; when the deviation trend between the monitored operation state characteristics and the early warning threshold value exceeds a set range, structured fault early warning information is output and synchronously fed back to a central air conditioner control system, and early intervention and maintenance scheduling are achieved; the method has high trend recognition capacity and prediction perspectiveness, and the accuracy of fault early warning and the intelligent operation level of the central air conditioning system are improved.

The invention discloses a refrigeration host parallel system prediction and energy efficiency control method and system, relates to the technical field of refrigeration system energy efficiency optimization control, and is used for solving the problem that energy efficiency optimization and load dynamic response are difficult to cooperate when a plurality of refrigeration hosts are in parallel operation under variable working conditions. According to the method, a working condition-load-energy efficiency mapping relation is established by constructing a refrigeration host three-dimensional energy efficiency mapping table, day-ahead and intra-day load prediction is executed in parallel based on historical data and environmental parameters, and a comprehensive load value is obtained through prediction error inverse proportion dynamic weighted fusion. And enumerating an optimal host operation combination and issuing a control instruction by taking the maximization of the comprehensive revenue function as a target, and updating prediction model parameters on line by utilizing a residual error of an actual load and a prediction value. The method is mainly used for energy efficiency optimization and self-adaptive control of a central air conditioning system, a ground source heat pump system or an industrial refrigerating unit group, the overall energy efficiency of the system can be effectively improved, and frequent start and stop are reduced.

The invention relates to an energy efficiency optimization method for a dynamic load of a central air conditioner, which comprises the following steps of: establishing a response relationship among a cold and heat source, a transmission and distribution system and end equipment, identifying a nonlinear decoupling characteristic of a cold and heat response behavior by monitoring operation deflection of each level of equipment under a low-load working condition, and defining and updating a cooling capacity response deviation threshold in real time; on the basis of the cooling capacity response deviation threshold, delay compression control is conducted on response actions of the refrigerant side and the air medium side, and by adjusting the air side response time sequence in advance, dynamic thermodynamic equilibrium is formed between refrigerant transfer delay and air medium response in advance; by establishing the response relationship among the cold and heat source, the transmission and distribution system and the end equipment and introducing a cold capacity response deviation threshold dynamic updating mechanism, nonlinear coupling mismatch in a cold and heat transmission chain can be accurately identified, so that the system is more flexible to operate under low-load and variable-load conditions, frequent start and stop or redundant operation of the cold and heat source is avoided, and the system reliability is improved. And therefore, the overall COP value of the system is remarkably increased, and unit cooling capacity consumption is reduced.

The invention discloses an optimal control method and system for a central air conditioner. The method comprises the steps that a starting instruction is responded; selecting an operation parameter combination from a preset database and sending a test request to the central air-conditioning system; energy efficiency detection results are received, wherein the energy efficiency detection results comprise energy consumption, equipment efficiency and user thermal comfort indexes; determining a next round of test combination based on the detection result and the mapping relation table; and selecting the combination with the highest energy efficiency from the tested combinations as a final result and configuring the final result to the system. Energy efficiency self-adaptive optimization is achieved by establishing the preset database and the mapping relation table in combination with a multi-round test mechanism, the test is intelligently guided according to the divided energy efficiency value interval, blind traversal is avoided, and the search efficiency is improved. A termination test strategy is set to effectively control a period and reduce interference, a load prediction model is introduced to make optimization more targeted, energy waste caused by lagging response is avoided, key parameters are collaboratively optimized, comfort is improved, and energy consumption is reduced.

The invention provides an energy-saving central air conditioner, which belongs to the technical field of central air conditioners, and comprises an air conditioner body and a pipeline arranged at the lower end of the air conditioner body, and further comprises an air inlet adjusting assembly, an induction dust collecting assembly and a cleaning and collecting assembly, air is purified through electrostatic adsorption of the adsorption plate, dust accumulation and local blockage conditions are accurately detected through cooperation of an elastic part, a magnetic coating and a Hall sensor, the air inlet quality of the air conditioner and system operation are prevented from being affected by feedback signals, tiny impurities can be effectively captured by the adsorption plate, and the air cleanliness is improved; a wind power and knocking combined mode is adopted for dust removal, a filter screen plate and an adsorption plate are cleaned in all directions, and inlet air cleanliness is guaranteed; when the ash discharging port is not used for removing ash, the magnetic blocking plate blocks the ash discharging port to prevent leakage, and the pressing rod limits the position of the connecting plate to ensure the monitoring precision of the sensing rod, reduce part shaking, prevent dust from falling off in advance and ensure the adsorption effect; the baffle is connected with related parts to control the opening degree of the pipeline, dust is prevented from entering the wall during dust removal, and dust collection is guided.

The invention discloses an air conditioning system control method and system based on reinforcement learning and a twinborn model strategy. The method comprises the steps that S1, a data-driven twinborn model of a central air conditioning system is constructed; s2, performing offline pre-training based on a virtual environment formed by a twin model; s3, performing online integration on a plurality of pre-trained reinforcement learning algorithms and the twin model to form a reinforcement learning strategy integration framework; and S4, the final control action is issued to execution equipment of the central air conditioning system. According to the invention, by constructing the serial data-driven twinborn model, the running state, energy consumption and cooling and heating load supply-demand relationship of the central air-conditioning system are accurately simulated, a high-fidelity virtual environment is provided for various reinforcement learning algorithms, offline pre-training is realized, an initial control strategy is generated, and risks caused by blind exploration in a real system are avoided. Continuous self-adaptive optimization is achieved, and the stability and the energy efficiency level of intelligent control of the air conditioning system are remarkably improved.

The invention relates to the technical field of energy management, in particular to a dynamic energy management method and system for a central air-conditioning system based on a large model, and the method specifically comprises the steps: outputting a short-term load demand prediction curve in a rolling manner through a time sequence depth prediction model; carrying out inversion calculation on equivalent heat capacity and equivalent heat resistance of the building material through a physical information neural network, and quantifying thermal inertia parameters of the building material in real time; evaluating a phase change delay time coefficient of the building material; obtaining a photovoltaic power generation power prediction curve of a power grid system and a power grid time-of-use electricity price signal, and determining an optimal energy charging and discharging time window of the building material; and adopting a multi-objective decision algorithm to decide and generate an optimal operation strategy of the central air-conditioning system. The problem that in the prior art, an energy management system matched with the thermal dynamic characteristics of a building is lacked is solved.

The invention discloses an intelligent regulation and control method and system for the environmental comfort degree of a central air conditioner terminal, and relates to the technical field of air conditioner regulation and control. The intelligent regulation and control method comprises the following steps that a pressure fluctuation track generated by aerodynamic disturbance is collected in the operation process of a draught fan, and the pressure fluctuation track is converted into energy spectrum distribution in a time sequence; establishing dynamic energy representation of the fan frequency; based on dynamic energy representation of fan frequency, a sound vibration field is projected in a spatial range, excitation response of inherent frequency is synchronously recorded through a room structure sensing array, and instantaneous energy coupling fingerprints of a room structure are formed. Through dynamic energy characterization and structural response coupling analysis, real-time identification and quantitative control of the standing wave noise risk are realized, a phase drift mechanism is introduced at a risk critical point, a fan disturbance phase is actively adjusted, standing wave superposition is effectively blocked, and acoustic safety and operation stability of a tail end environment are improved.

The invention discloses a central air conditioner pollution degree autonomous evaluation robot based on multi-modal perception, and relates to the technical field of intelligent detection, the robot comprises a voxel space division module, a data acquisition module, a pollution characteristic mapping modeling module, a physical pollution detection module, a chemical pollution detection module and a biological pollution detection module; through three-dimensional point cloud data scanning and real-time data acquisition, the air pipe wall surface roughness, the temperature delay time, the gas fluctuation, the heat retention degree and the microbial metabolism gas concentration are monitored in real time, and a three-dimensional pollution characteristic mapping model is generated according to the pollution characteristic data set; by calculating a physical pollution index PPI, a chemical pollution index CPI and a biological pollution index BPI and comparing and analyzing with corresponding threshold values, whether dust retention, gas concentration and microorganism quantity in an air pipe exceed standards or not is judged in real time, cleaning, filtering and sterilizing strategies are automatically triggered, and the air quality and operation safety of a central air conditioning system are effectively guaranteed.

The invention discloses an air purification and temperature and humidity coordination control method and system for a central air conditioner, and the method comprises the steps: learning past and future environment change trends based on a Bi-LSTM bidirectional neural network, adding a one-dimensional CNN convolution layer before Bi-LSTM, and extracting a local space mode of sensor data; introducing an attention mechanism to dynamically weight a pollutant concentration peak value and a temperature and humidity sudden drop point in the input sequence; inputting the multi-modal data into the hybrid prediction model for prediction, and outputting a prediction result; establishing a dynamic optimization model by taking minimization of energy consumption, maximization of air cleanliness and maintenance of temperature and humidity stability as targets through an NSGA-II multi-objective optimization algorithm; and inputting the prediction result, the multi-modal data and a threshold value set by a user into a dynamic optimization model, and outputting an air purification adjustment parameter and a temperature and humidity adjustment parameter. The efficiency and accuracy of temperature and humidity control and air purification control are improved.

The invention relates to the technical field of central air conditioner control, in particular to an energy-saving optimization control method and system for a central air conditioner. The method comprises the following steps: constructing a Q value table, obtaining current moment parameter data, selecting a corresponding action according to a state section corresponding to the current moment parameter data and the Q value table, controlling the working state of the central air conditioner, obtaining a reward value and a new state according to the selected action, updating a Q value, and using the updated Q value for next moment control. According to the scheme, the working state of the central air conditioner can be intelligently adjusted according to the indoor environment and user requirements, more accurate environment control is achieved, and the user experience is improved.

The invention relates to the technical field of water treatment, and discloses a central air-conditioning condensate water disinfection and purification device which comprises a treatment tank main body, an ozone generator and an ozone generator real-time regulation and control system. According to the system, a demand coefficient is dynamically obtained based on residual ozone concentration and ultraviolet absorbance through a pollution load evaluation module; the ozone mass transfer efficiency evaluation module calculates a mass transfer efficiency coefficient based on the tail gas and inlet gas ozone concentration and the gas flow; the matching evaluation module combines the demand coefficient, the mass transfer efficiency coefficient and the water temperature influence factor to generate a matching factor; the generator health monitoring module evaluates a health degree coefficient based on parameters such as air inlet pressure; and the power comprehensive decision-making module fuses the health degree coefficient and the matching factor, and outputs a target power set value through a multi-parameter regulation and control model. The intelligent sensing and dynamic optimization of ozone feeding are realized, the problems of traditional control stiffness, low mass transfer efficiency and insufficient equipment guarantee are solved, and the energy utilization rate and the operation reliability are improved while the disinfection effect is ensured.

The invention relates to the technical field of intelligent city management, in particular to an intelligent city carbon emission panoramic management and planning system, which is characterized in that a multi-modal behavior sensing unit acquires multi-body differentiated use behavior data of a central air conditioning system, a shared parking lot and a heat supply pipe network in real time through a heterogeneous sensing network deployed in a shared facility; the dynamic emission traceability unit receives the data, the data and the total energy consumption of the shared facility are subjected to space-time coupling calculation through a carbon flow tracing algorithm engine, a multi-subject responsibility carbon footprint is generated, and the responsibility weight distribution unit dynamically calculates the contribution degree of each subject based on the carbon footprint by adopting a Sharpley value-Bayesian game fusion model; and the responsibility allocation proportion is output to the block chain incentive execution unit, the block chain incentive execution unit automatically triggers a differentiated reward and punishment mechanism through an intelligent contract according to the responsibility allocation proportion, and for behavior abnormal subjects, facility use permission can be degraded, carbon emission management accuracy and fairness can be improved, and urban low-carbon development can be assisted.

The invention relates to a combined cooling and heating collaborative optimization method for a central air conditioner, which comprises the following steps of: when cold and heat load fluctuation is detected, according to the response speed difference of a cooling and heating system, asymmetrically controlling the time sequence and pre-starting or hot standby of equipment, so that the time sequence of a cold and heat source is coordinated; potential physical interference nodes or cross paths are identified in the operation process of the cooling and heating system, a hydraulic or energy transmission path is temporarily passivated, and the reverse coupling influence between the cooling and heating system is reduced in combination with hot side delayed water feeding and cold side path adjustment; according to the method, the asymmetric control time sequence and the cold and hot response priority queue are introduced, the control windows can be dynamically distributed according to the cold and hot source response speed difference, the system control conflict problem caused by simultaneous adjustment of the cold and hot sources is avoided, the cold and hot system rush adjustment phenomenon is effectively relieved, and response adjustment is more orderly. By means of the pre-starting device and the hot standby mechanism, the device can be awakened in advance or maintained in the heat balance pre-starting state before load fluctuation.

The utility model relates to the field of air conditioner air outlet devices, in particular to a central air conditioner air outlet noise reduction device which comprises an air receiving barrel, a bucket-shaped air expanding cover, an air outlet cover, a noise reduction frame, an air direction guiding frame and a protection frame. A hopper-shaped air expanding cover is arranged at the lower end of the air receiving barrel, an air outlet cover is arranged at the lower end of the hopper-shaped air expanding cover, an embedded groove is formed in the end, away from the hopper-shaped air expanding cover, of the air outlet cover, and a noise reduction frame, an air direction guiding frame and a protection frame are sequentially arranged in the embedded groove. According to the air conditioner air outlet device, noise generated by airflow resonance is effectively reduced through the air expansion flow guide design of the hopper-shaped air expansion cover and the sound absorption cotton plate filling of the sound attenuation and noise reduction frame, airflow can be evenly and softly blown to a room through the design of the air direction guide frame, and the problems that airflow in a traditional air conditioner air outlet device is too strong and uneven in distribution are solved.

The invention relates to a central air-conditioning system precooling stage optimization operation method based on adaptive updating, and is suitable for the field of central air-conditioning system optimization control. The method comprises the steps that the pre-cooling time of all areas of a building on the previous day and system operation parameters of the pre-cooling stage on the previous day are obtained; acquiring initial environment parameters of the current day and the previous day; based on the system operation parameters of the previous day, the pre-cooling time deviation between the areas of the previous day and the initial environment parameter deviation between the current day and the previous day, the system operation parameters of the current day are determined in combination with the deviation weight coefficients of the current day; executing the pre-cooling stage of the current day based on the system operation parameters of the current day, and obtaining the pre-cooling time of each area of the building on the current day; and on the basis of the pre-cooling time and the initial environment parameters of each area of the building on the current day, with the purpose of minimizing the total pre-cooling time of the system, a Q-learning algorithm is adopted to update the weight coefficient of each deviation, and the weight coefficient of each deviation on the next day is obtained.

The application discloses a parallel cooling tower group hydraulic balance adjusting device and a parallel cooling tower group, and belongs to the technical field of centralized central air conditioner hydraulic balance adjusting. The device comprises a shell, a water inlet channel for cooling circulating water, at least two water outlet channels for discharging the cooling circulating water, and a static pressure shunt cavity connecting the water inlet channel and the water outlet channel; a shunt plate is fixed in the static pressure shunt cavity and abuts against the top wall surface and the bottom wall surface of the static pressure shunt cavity, so that the static pressure shunt cavity is divided into at least two shunt cavities with equal volumes; wherein the water inlet channel comprises a first end and a second end, the first end extends into the static pressure shunt cavity, and is arranged to be inclined towards the top wall surface of the static pressure shunt cavity. The application can ensure equal distribution of water flow on both sides of the cooling tower and minimize the total resistance of the cooling water system.

The invention provides a dynamic optimization control method for a building chilled water storage central air-conditioning system, which comprises the following steps of: predicting a historical cold load of the building air-conditioning system through a pre-trained cold load prediction model, and calculating a cold load prediction error rate and uncertainty distribution; performing power consumption prediction by using the power consumption prediction model; sampling the cold load prediction error rate in the next 24 hours by using a Monte Carlo method; and in combination with the initial predicted cooling load and the dynamic time-of-use electricity price, with the goal of minimizing the daily electricity consumption cost, an optimal control signal of the hourly cold charging or discharging capacity of the chilled water storage device in the next 24 hours is obtained through a particle swarm algorithm. According to the method, quantitative modeling is carried out on random distribution of the error rate of cold load prediction, and the control signal of the cold charging or discharging capacity of the chilled water storage device is optimized through the Monte Carlo method and the particle swarm optimization, so that the optimal control robustness under the condition that the cold load prediction is uncertain is improved; and the operation cost of the chilled water storage central air-conditioning system under the dynamic electricity price is saved.

The invention provides an energy efficiency monitoring method and system for real-time operation of a central air conditioner host system, and the method comprises the steps: collecting operation data of an air conditioner system, and fusing the operation data according to a preset data weight to obtain fused data; and performing time alignment and space mapping on the fused data to obtain standardized data, and performing dimension feature extraction on the standardized data to obtain dimension feature data. And performing spatial feature extraction on the dimension feature data through a preset convolutional neural network to obtain spatial feature data. And performing time sequence feature extraction on the spatial feature data through a preset long-short-term memory network to obtain an initial load prediction value. And correcting the initial load prediction value according to a preset time sequence change regularization item to obtain a corrected load prediction value. And the accuracy of predicting the load of the air conditioning system is improved.

The invention relates to the technical field of environment control, and discloses a time-sharing optimization control method for a cold storage central air conditioner based on a large model, which comprises the following steps of: firstly, acquiring a system cold demand quantity and an environment working condition predicted value at a future moment through a pre-trained deep learning model, and estimating a current cold storage charging and discharging state in combination with a thermal front degradation correction algorithm; then, a non-linear flow resistance model associated with the porosity and the cold storage state is established, and a hydraulic high-energy-consumption interval in which fluid resistance is highly sensitive to flow change is recognized; and on the basis, deducing the energy efficiency trend of the water chilling unit by utilizing the environment temperature to determine load distribution, actively inhibiting cold storage output in a high flow resistance interval through energy efficiency gating logic, and finally generating a time sequence control strategy containing flow and load instructions to realize optimal scheduling of system-level energy efficiency.

The invention relates to the technical field of automatic control, in particular to a pipeline blockage risk identification and early warning method for multiple water chilling units of a central air conditioner, which comprises the following steps: collecting a cold water pressure difference signal, processing temperature difference data, extracting offset characteristics, analyzing a temperature track, identifying sudden change and matching flow response, and judging response abnormity. According to the method, the blockage head is identified through the differential pressure change rate jump, the response sensitivity is enhanced, the temperature difference continuous offset and the differential pressure event are combined, the identification accuracy and time efficiency are improved, the temperature sudden change and the flow response rhythm are superposed, a physical chain verification mechanism is constructed, and the high-frequency abnormal pipe section is identified. The reliability of analysis logic is enhanced, response interval and adjustment period consistency analysis is introduced, the adaptive capacity under complex working conditions is improved, pipe sections are collected according to abnormal frequencies, a multi-feature structured result is output, the data support of maintenance decision is enhanced, and the perspectiveness of blockage recognition is improved through multi-source feature fusion and time sequence rule judgment.

This application provides a central air conditioning system and its control method, relating to the field of air conditioning technology. The central air conditioning system includes: a cooling tower; a chiller unit; a first temperature sensor; a humidity sensor; and a controller configured to: acquire the current operating condition of the cooling tower and multiple control strategies under the current operating condition; determine the design operating power of the cooling tower corresponding to each control strategy under the current operating condition based on the current operating condition, the multiple control strategies, and a first energy consumption model; determine a target control strategy based on the design operating power of the cooling tower corresponding to each control strategy, the target control strategy being the control strategy with the lowest corresponding design operating power among all control strategies; determine a target operating frequency of the cooling tower based on the design operating power corresponding to the target control strategy and a second energy consumption model; and control the cooling tower to operate according to the target control strategy and the target operating frequency.

The invention relates to the technical field of central air conditioner control, and discloses a central air conditioner energy-saving disinfection and killing control method and system based on Internet of Things monitoring, and the method comprises the steps that temperature and humidity time sequence data monitored by the Internet of Things of a central air conditioner and a current and voltage parameter matrix of a high-voltage electrostatic disinfection and killing module are obtained; performing multi-modal feature fusion calculation to generate a multi-dimensional state matrix; establishing an energy consumption-disinfection associated nodal topology network, simulating an energy transfer path and outputting a synergistic effect probability value set; performing space-time correlation mapping operation to generate a sterilization-energy consumption coupling prediction curve; calculating an energy-saving efficiency factor and generating a standardized coupling coefficient thermodynamic diagram; and extracting the operation quantitative index, dynamically comparing the operation quantitative index with a preset threshold value, and generating a regulation and control scheme when the index breaks through the threshold value. According to the method, collaborative optimization control over energy conservation and disinfection of the central air conditioning system is achieved, the energy utilization efficiency is improved, the disinfection effect is guaranteed, and high adaptability and expandability are achieved.

The invention relates to a central air conditioner partition control method and device, an air conditioner and a computer readable medium. The method comprises the steps that under the condition that air supply adjustment on a target area is triggered this time according to a first control period, current environment data of the target area are collected, and the first control period is the time interval determined when air supply adjustment is conducted on the target area last time and conducted this time; the target area is any one of a plurality of areas governed by the central air conditioner; determining the current environmental disturbance intensity of the target area according to the environmental data; and air is supplied to the target area according to a target control strategy matched with the environment disturbance intensity, a second control period is determined according to the environment disturbance intensity, and air supply adjustment on the target area next time is triggered after the second control period. The control period of each area is dynamically adjusted according to the actual environment disturbance of each area, and the technical problem that a fixed control period cannot adapt to the real dynamic environment of each area is solved.

The invention relates to the technical field of deep learning, in particular to a federal learning collaborative modeling method for energy efficiency optimization of a central air-conditioning system, which comprises the following steps: acquiring data of a system operation state, calculating to obtain real-time flow data, and calculating a real-time load value by combining the real-time flow data with water supply and return temperatures; inputting the environment data into the reference generation model to obtain a load mode; inputting system operation data into the energy efficiency model to generate a dynamic remodeling coefficient; through a parameter transformation network, taking the dynamic remodeling coefficient as a transformation parameter of the network, and generating a load prediction curve; based on the load prediction curve, solving is carried out through an optimal control algorithm, a cooperative control instruction is generated, and the operation frequency is adjusted through the control instruction; using the dynamic remodeling coefficient as a label, training an energy efficiency model and generating an update parameter; and the server aggregates the parameters to generate a global optimization model, then distributes the global optimization model, and iteratively updates the energy efficiency model. According to the method, system energy efficiency optimization is realized through federated learning collaborative modeling.