562 results about "Heat load" patented technology

The invention provides a liquid cooling fusion energy consumption adjustment optimization method and system based on dynamic adjustment, and the method comprises the steps: S1, obtaining the thermal load data of calculation equipment, the state data of a liquid cooling system, the state data of an air cooling system, and the constraint data of the system, and constructing a cooling fusion real-time state map; s2, performing calibration operation on the cooling prediction model based on the cooling fusion real-time state atlas, after calibration is completed, extracting cooling key dynamic feature vectors, obtaining short-time cooling prediction data based on the cooling prediction model and the cooling fusion real-time state atlas, and constructing a cooling conflict objective function; s3, obtaining a cooling random sample set, performing initialization operation on a PSO algorithm based on the cooling random sample set, and generating an optimal control parameter combination; and S4, the cooling system is controlled in real time based on the optimal control parameter combination, the step S1 to the step S4 are executed in a circulating mode, and by means of the method, optimal energy consumption control can be found in real time under the condition that the cooling capacity of the cooling system is kept.

The invention discloses a multi-parameter MPC two-phase liquid cooling energy efficiency optimization distribution control system and method, and relates to the technical field of liquid cooling energy efficiency optimization. The multi-parameter MPC two-phase liquid cooling energy efficiency optimization distribution control system and method comprises the following steps: S1, collecting and preprocessing two-phase liquid cooling control data; s2, the thermal load level of each area is evaluated periodically, a discrete time linear state space model is constructed, and the thermal load change rate is extracted; s3, the cold distribution amount of each area is quantified periodically, the pump speed and the valve opening target value are calculated, and an adjustment control instruction is generated; and S4, high heat risk and low heat stability areas are recognized, the cooling capacity compensation requirement is evaluated, a cooling capacity enhancement and recovery distribution value is output, and a corresponding control instruction is generated. The problems that an existing two-phase liquid cooling system generally only depends on temperature feedback for adjustment, so that cooling capacity distribution response lags behind, a short-time overheating phenomenon is prone to occurring in a high-heat area, and cooling capacity waste exists in a low-heat area at the same time are solved.

The invention relates to the technical field of data processing, and provides a multi-mode water-cooling heat dissipation system and method.The thermal load urgency degree is generated by obtaining the temperature change rate and the thermal power density of a component in real time, the running state of the component is monitored, and a user scene identifier is output; comparing the temperature difference value of the cooling liquid at the inlet and outlet of the radiator with the cooling liquid flow to estimate the instantaneous heat dissipating capacity; predicting a temperature change track based on the thermal load urgency degree and the user scene identifier, calibrating the temperature change track in combination with the instantaneous heat dissipation amount, determining a candidate mode set and an optimization weight through the user scene identifier, and selecting a target heat dissipation mode; temperature distribution of different areas of the radiator is obtained according to the target heat dissipation mode so as to distribute the target rotating speed of the fan, respond to the local high heat load, control the multi-way valve to dynamically adjust the topological structure of the cooling liquid flowing path and distribute the cooling liquid flow, and deep cooperation of the fan and the cooling liquid is achieved.

The invention discloses a micro-channel liquid cooling method, device, equipment and medium for heat dissipation of an integrated circuit chip, and the method comprises the steps: collecting temperature data of different depths of the surface and the interior of the chip in real time through a temperature sensor array, an embedded temperature probe and an infrared detector, and generating a three-dimensional temperature field matrix; carrying out region division according to the microchannel physical layout of the chip, mapping a three-dimensional temperature field matrix, and generating a region temperature field matrix; calculating a thermal load change rate of the regional temperature field matrix, and generating a main flow regulation parameter by adopting a PI D algorithm based on the change rate; calculating branch flow adjusting parameters by combining the temperature field matrixes of the adjacent areas and a pre-constructed topological association table, and fusing the main flow adjusting parameters and the branch flow adjusting parameters to generate a flow adjusting quantity instruction. According to the invention, dynamic and accurate regulation and control of chip heat dissipation are realized, and the heat dissipation efficiency is effectively improved.

The invention relates to the technical field of equipment parameter optimization, and discloses a liquid cooling system parameter optimization method and system for a data center. The method comprises the steps of performing wavelet noise reduction processing on thermal load data of equipment, and constructing a time sequence heat generation model; the thermodynamic model and the time sequence heat generation model are fused to generate a comprehensive heat model, an objective function and constraint conditions are set, and a parameter combination is output; the parameter optimization model generates a standard parameter combination based on the parameter combination, the standard parameter combination is substituted into the parameter optimization model for evaluation, and an optimal parameter combination is output; the operation parameters of the liquid cooling system are adjusted in real time, the actual system operation data of the liquid cooling system are collected, the actual system operation data are compared with the expected equipment thermal load data of the data center, the parameter deviation is obtained, and the operation parameters are dynamically adjusted based on the parameter deviation. According to the invention, the efficiency and accuracy of parameter optimization of the liquid cooling system are improved.

The invention relates to the technical field of intelligent building heat load prediction and heat balance regulation and control, and discloses an intelligent building heat balance dynamic regulation and control method and system based on load prediction, and the method comprises the steps: constructing a time-frequency causal double-flow analysis network, mutual enhancement of the time-frequency characteristics and the causal relationship is realized; designing a hierarchical causal discovery algorithm, and mining a multilevel causal structure; developing a causal enhanced time-frequency representation learning method, and fusing time-frequency and causal information; establishing an intervention decision framework based on anti-fact analysis, and evaluating an intervention effect; an abnormal mode self-evolution recognition system is realized, and new abnormal modes are continuously learned; an interpretable abnormity diagnosis mechanism is developed, and a clear diagnosis report is provided; according to the method, the thermal load prediction accuracy is improved, the abnormal early warning capability is enhanced, the intervention efficiency is improved, the system interpretability is enhanced, and the optimization of the energy utilization efficiency is realized.

The invention relates to the technical field of immersion liquid cooling data center cooling, and particularly provides a single-phase immersion liquid cooling server circulation cooling method and system, and the method comprises the steps: comparing the real-time temperature of a cooling liquid with a preset temperature threshold value, transmitting a first program instruction by a control system, and carrying out an immersion liquid cooling cabinet internal circulation mode; or a second program instruction is sent, and a heat exchange circulation mode is carried out; after the heat exchange circulation mode is started, a main liquid outlet pipeline is opened; the cooling liquid flows back to the liquid outlet of the immersed liquid cooling cabinet through the backflow branch pipeline, the real-time temperature of the cooling liquid and the preset temperature threshold value are continuously judged, and switching between the internal circulation mode and the heat exchange circulation mode of the immersed liquid cooling cabinet is achieved according to the comparison result. The system comprises an immersed liquid cooling cabinet, a branch pipeline valve, a main pipeline valve, a main liquid outlet pipeline, a pump, a heat exchanger and a branch pipeline. According to the invention, the topology reconstruction capability of the cooling liquid flow is realized, and local circulation and global circulation are dynamically switched according to the thermal load.

The invention discloses an air pressure-boiling point self-adaptive regulation and control system and method of an energy storage immersed phase change liquid cooling system, and relates to the technical field of energy storage systems. The method comprises the following steps: calculating a thermal load of a system in real time, performing multi-objective optimization grading, and dynamically determining an optimal boiling point set value under a current working condition; resolving and obtaining a target air pressure set value by using the air pressure-boiling point relation mapping library; an improved composite control algorithm integrating feedforward control, fuzzy adaptive PID, integral saturation resistance and change rate limitation is adopted to drive an air pressure regulation execution mechanism, and regulation and control of the air pressure in the system are achieved. The system and the method have the advantages of high control precision, high response speed, good stability, high adaptability and the like.

The invention relates to the technical field of computer aided design, in particular to an energy-saving building design method based on digital simulation, which comprises the following steps: analyzing air velocity, temperature gradient, density and pressure distribution based on indoor aerodynamic parameters, identifying air vortex structure characteristics, and generating an energy-saving building environment data set. According to the method, aerodynamic characteristics in a building are accurately recognized, the vortex structure analysis capacity is improved, the air flowing path is optimized, aerodynamic parameters are combined, thermal resistance and heat conduction materials are matched, the building envelope layer is adjusted, heat flow transmission is guided, heat loss is reduced, the heat energy utilization rate is increased, and the spectral characteristics of the materials are adjusted according to illumination redundancy; the lighting efficiency is optimized, the heat load is reduced, outer wall heat transfer is optimized in combination with airflow characteristics, the ventilation rate is dynamically regulated and controlled, airflow short circuit is avoided, heat flow, illumination and ventilation parameters are integrated, precise optimization of building energy consumption is achieved, and the energy efficiency regulation and control level is improved.

The invention discloses a comprehensive energy low-carbon optimization scheduling method and system based on a master-slave game. The method comprises the following steps: constructing a hydrogen-containing comprehensive energy system model, and inputting data of an external energy supply side, energy conversion and energy storage equipment and a load side to obtain initialized system operation parameters; building a user side model based on the model, generating an initial electricity / heat load demand, calculating a user alternative energy consumption rate and dynamically adjusting an electricity / heat energy conversion ratio in combination with a price difference of real-time electricity and heat prices and user energy consumption behavior characteristics, and forming a new electricity / heat load distribution scheme; the integrated energy operator generates an electricity / heat price signal and formulates a dual excitation strategy, the user aggregator adjusts the user energy load and response strategy according to the signal and the strategy and feeds back the strategy to the integrated energy operator, and the user aggregator and the integrated energy operator adjust bidirectional feedback iteration to solve Stackelberg balance through the electricity / heat price signal and the load to obtain an optimal scheduling scheme. According to the invention, economic low-carbon operation and benefit win-win between subjects can be realized.

The invention relates to the technical field of data management, in particular to a pet soft food production data management method and system.The method comprises the following steps of analyzing heat energy and rhythm synchronization based on damp and hot section operation data, judging batch data continuity, extracting fluctuation data to recognize quality inspection deviation, and matching packaging and discharging to analyze response consistency. And abnormal batches are extracted to locate an alarm concentration area. According to the method, by monitoring the heat energy flow and the conveying period, synchronization of the heat load and the putting rhythm is achieved, data consistency and collaboration are improved, data interruption and overlapping are avoided in combination with the batch time difference, fluctuation of indexes such as PH is periodically recognized, the quality inspection gathering time period is determined, packaging and discharging response offset is analyzed, and the equipment collaboration efficiency is quantified; energy records and alarm logs are crossly screened, abnormal batches are accurately positioned, a multi-parameter time sequence coupling analysis chain is constructed, cross-section quality response and traceability are enhanced, and centralized troubleshooting and early warning of the abnormal batches are realized.

The invention relates to the technical field of liquid cooling heat dissipation, in particular to a liquid cooling distribution device, liquid cooling heat dissipation equipment and a liquid cooling control method. The liquid cooling distribution device comprises a main liquid supply collecting pipe, a flow adjusting assembly, a multi-channel liquid distribution assembly, a plurality of cooling medium conveying pipelines and a central control unit, a proportioning valve of the flow adjusting assembly is connected into the main liquid supply collecting pipe in parallel, and a liquid distributor of the multi-channel liquid distribution assembly is correspondingly connected with the proportioning valve. A parallel branch of the liquid separator is connected with the liquid outlet node through the reversing valve, and the two ends of the cooling medium conveying pipeline are connected with the liquid outlet node and target heat dissipation equipment respectively; the central control unit achieves accurate distribution and dynamic adjustment of the flow of the cooling liquid by controlling the proportion adjusting valve and the reversing valve, the problem of uneven flow distribution caused by pipeline resistance difference is solved, the flow can be adjusted according to real-time heat loads, and system energy consumption is optimized.

The invention relates to the technical field of task scheduling, in particular to a task scheduling optimization system for a smelting process, which comprises a furnace temperature identification module, a hot load analysis module, a load screening module, a sorting regulation and control module and a task joint control module. According to the method, a stable operation section is screened based on thermocouple time sequence temperature difference fluctuation characteristics, task identification is ensured to be in tight fit with a real metallurgical process state, furnace profile thermal parameters and raw material attributes are introduced to dynamically estimate task thermal loads, and the heat adaptation relation between tasks and equipment is determined; safety load interval limitation is set in the screening task and furnace body combination process, the safety and matching precision of resource allocation are improved, meanwhile, cooling period limitation is embedded in task sorting to recognize delay tasks, scheduling continuity and task connection reasonability are enhanced, the capacity of a scheduling strategy for responding to actual operation constraints is improved, and the service life of the scheduling strategy is prolonged. The overall optimization of accurate task screening, reasonable load distribution and ordered scheduling sequence is realized.

The invention provides an intelligent cooling capacity distribution method and system for a data center refrigeration system, and relates to the technical field of data center refrigeration. Through the thermal model based on data driving, airflow distribution and temperature field change are accurately simulated, thermal load prediction errors are effectively reduced, and high-precision data support is provided for cold distribution, so that the energy utilization efficiency and the refrigeration efficiency are improved, and energy efficiency improvement and reliable operation of a data center refrigeration system are achieved. And meanwhile, the interaction among heat transfer, power consumption and a control strategy is comprehensively considered, the overall efficiency loss caused by local optimization is avoided, and the comprehensive energy efficiency of the refrigerating system is improved. Besides, the cooling capacity distribution optimization model is constructed based on the power-cooling capacity characteristic curve, so that the optimal balance between the cooling power and the cooling index is realized, and the PUE is effectively reduced.

The invention provides an LED lamp adaptive speed regulation heat dissipation method and system based on real-time junction temperature monitoring. The method comprises the following steps: S100, acquiring a PWM signal or an analog signal of an LED driving command signal in real time; s200, on the basis of the change of the driving instruction signal, feedforward prediction is carried out by using a dynamic thermal inertia mapping model, and a feedforward fan control quantity is generated; s300, LED junction temperature is monitored in real time through a forward voltage method, and feedback fan control quantity is generated through a feedback controller according to the deviation between the real-time junction temperature and the target junction temperature; and S400, synthesizing the feedforward control quantity and the feedback control quantity to generate a fan PWM control signal, outputting the fan PWM control signal to a fan driving circuit, and adjusting the rotating speed of the cooling fan. According to the method, driving signals and junction temperature information are processed in parallel, and feedforward and feedback control is combined, so that self-adaptive accurate regulation and control are realized, thermal load change is quickly responded, system lag is effectively inhibited, and LED heat dissipation performance is optimized.

Thermal management systems include a vapor cycle and a liquid cycle sharing a common working fluid. The vapor cycle includes, along a vapor cycle flow path, a compressor and a condenser. The liquid cycle includes, along a liquid cycle flow path, a fluid driver, a load, a regulator valve, and a phase change material heat exchanger. A cold sink is thermally coupled to a heat load. A separator is configured to separate liquid and vapor portions of the working fluid and direct the liquid into the liquid cycle and the vapor into the vapor cycle. The separator is part of both the vapor cycle and the liquid cycle. The regulator valve controls a temperature of the working fluid within the liquid cycle at a location upstream of the phase change material heat exchanger to control a mode of operation of the phase change material heat exchanger.

The invention relates to the technical field of heat pump system control, in particular to a heat pump system control method and system based on compressor rotating speed feedback. The method comprises the steps that a refrigerant side parameter cluster, an air side parameter cluster and a heat load side parameter cluster of the heat pump system in a preset sliding window are obtained, and a first refrigerating capacity, a second refrigerating capacity and a third refrigerating capacity are determined through the different parameter clusters so as to determine fault source positioning indexes corresponding to the different parameter clusters; comparing the fault source positioning index with a preset fault threshold value, and determining a fault sensor cluster from a refrigerant side parameter cluster, an air side parameter cluster and a heat load side parameter cluster; and controlling the rotating speed of a compressor of the heat pump system according to a determination result of the fault sensor cluster. According to the technical scheme, the rotating speed of the compressor of the heat pump system can be effectively controlled.

The invention relates to a centralized heating balanced regulation and control method, device and equipment and a medium. The method comprises the steps that firstly, the heat supply area of a user and the heat use state of a neighbor are collected, the basic flow of the jet device is calculated according to an industrial standard formula, and final flow containing resistance characteristic data is obtained through secondary correction by combining a neighbor heat use state correction coefficient and an environment temperature influence factor. Extracting flow and inlet and outlet pressure difference from jet device parameters, adjusting a valve to enable the resistance proportion to reach a preset interval, calculating parameters such as unit area heat consumption by combining real-time flow and supply and return water temperature difference, generating regulation and control basic data through coupling analysis, adjusting the jet device when the regulation and control basic data reach the standard, and generating a dynamic scheme by combining a thermal load prediction model. And finally, capturing the flow pulse signal according to a fixed period to calculate a difference value, generating a valve control signal according to a threshold value, acquiring the water supply and return temperature after execution, and coupling with a dynamic scheme to generate a regulation feedback signal. By adopting the method, accurate regulation and control can be realized through multi-dimensional analysis, and the flow distribution accuracy and the heat supply balance are improved.

The invention discloses a copious cooling rectification real-time monitoring and correction system based on digital twinning, and the system comprises a data processing module which is used for collecting and preprocessing data of a sensor; the graph structure construction module is used for constructing a twin graph structure and generating a topological adjacency matrix; the feature extraction module is used for executing graph embedding operation and extracting virtual twin features; the feature decoupling module is used for carrying out dynamic decoupling on the virtual twin tensor and constructing a feature sequence; the twinning correction module is used for inputting the state mapping vector into a twinning correction network and correcting a virtual twinning tensor; the variational inference module is used for inferring an output confidence interval and generating a reliability mask; and the parameter optimization module is used for adjusting the reflux ratio, the tower top temperature and the reboiling heat load in the credible area to realize real-time monitoring and correction. According to the invention, the intelligent monitoring capability and the self-adaptive control level of the cryogenic distillation process are obviously improved, and the system robustness, the energy-saving property and the industrial applicability are higher.

The invention discloses a dynamic building thermal load simulation method and device, heat supply equipment and a medium, and relates to the technical field of load prediction.The method comprises the steps that when the environment temperature changes, the first building unit area thermal load before change is determined, and the second building unit area thermal load after change is determined; determining a first heat change of the building external envelope structure and a second heat change of the building internal envelope structure caused by the environment temperature change; according to the difference value of the unit area thermal load of the first building and the unit area thermal load of the second building, the delay time is determined in combination with the heat change of the building envelope; and based on the delay time, the outdoor temperature before the delay time and the current indoor temperature are utilized to correct the unit area thermal load of the second building. By adopting the method, the calculation complexity of dynamic building thermal load simulation can be simplified, and the calculation of the delay time is more practical and operable.

The invention provides a spindle constant-temperature AI closed-loop heat engine system, and relates to the technical field of numerical control machine tool control and heat management. The invention aims to solve the problems of response lag, uncontrollable thermal stress, single heating strategy and the like of the traditional spindle heat engine method. The system comprises a multi-dimensional thermal state sensing module used for constructing a multi-node discrete temperature field representing the overall thermal distribution of the main shaft; the real-time heat balance feedback module is used for judging the cold / hot starting state of the main shaft according to the initial temperature difference and executing a differentiated feedback control strategy; and an AI prediction forward compensation module that predicts a future cutting thermal load by pre-reading a machining instruction and generates a feedforward thermal compensation instruction. According to the method, the thermal stress is controlled by introducing the radial thermal gradient index as the safety constraint, negative precooling or positive preheating is achieved in combination with the AI predictive capacity, a feedback and feedforward composite control mode is formed, and the dynamic thermal stability and the machining precision of the main shaft under the complex working condition are improved.

The invention relates to an energy-saving control strategy for a medical building air-conditioning ventilation system based on climate change characteristics and load prediction. The energy-saving control strategy comprises the following steps of: acquiring all original data by utilizing various sensors, actuators and the like of a sensing layer; calculating, optimizing and deciding on an intelligent control layer (including building comprehensive cooling and heating load prediction, clean operation department area cooling and heating demand calculation, dynamic water supply temperature optimization, a particle swarm optimization (PSO) algorithm for domain optimization, constraint check and other algorithm modules) on a central management server; and an optimization result is issued to the execution layer and the executed equipment layer, so that energy-saving control is finally realized. The method overcomes the defects that the existing hospital building load prediction technology mostly adopts a macroscopic and rough prediction model, and the global optimization particle swarm optimization algorithm is slow in convergence and long in solving time when coping with a large medical building adopting multiple types of cold and heat source equipment. The device is suitable for large-scale hospital projects in hot-summer and warm-winter areas.

The invention discloses a thermal load prediction method and system based on multi-stage fusion optimization, and the method employs a gray wolf optimization algorithm to replace a conventional empirical method to optimize key parameters of variational mode decomposition, and automatically determines an optimal mode number and a penalty factor through multi-objective optimization. The obtained parameter combination is introduced into a variational mode decomposition algorithm, and then the variational mode decomposition algorithm is utilized to decompose thermal load historical data into a plurality of stable mode components, so that the complexity and noise interference of the data are reduced, and the decomposition quality of variational mode decomposition is remarkably improved; then, the hyper-parameters of the long-short-term memory network are optimized by combining a sparrow search algorithm, so that an optimal hyper-parameter combination is found, and finally, a thermal load data set is predicted through a long-short-term memory model under the obtained optimal hyper-parameter combination.

The invention relates to the technical field of ground source heat pump operation control, in particular to an optimal adjustment method of a ground source heat pump system, which comprises the following steps: S1, collecting underground soil temperature and system heat load of the ground source heat pump system; s2, calculating a soil heat balance index based on the underground soil temperature; s3, according to the soil heat balance index and the system heat load, the optimized operation mode of the ground source heat pump system is judged; and S4, according to the optimized operation mode judged in the S3, the operation parameters of the ground source heat pump system are adjusted, and a corresponding execution mechanism is driven. According to the method, the operation mode is judged by introducing the soil heat balance index and combining the system heat load, meanwhile, the flow of the circulating water pump, the frequency of the compressor and the opening degree of the heat exchange valve are subjected to interval adjustment, and unification of energy efficiency optimization and long-term balance of the soil heat environment is achieved.

The invention relates to a multi-energy-load scene generation method and device based on a diffusion model. Multi-energy load data and condition information are input into a time-space and cross-dimension decomposition reconstruction network of a classifier-free condition diffusion architecture; projecting the multi-energy load data into time sequence embedding and cross-dimension feature embedding; coding the condition signal and the diffusion step number together to obtain a condition embedding vector; the reconstruction network comprises a time channel and a feature channel, a global context relationship is captured by an encoder, and condition information is fused based on a condition embedding vector through a decoder; wherein an STAR module is arranged behind the decoder of the feature channel; and integrating the output of the time channel and the output of the feature channel, and then carrying out layer-by-layer denoising to generate a multi-energy-load scene. Compared with the prior art, the method has the advantages that autocorrelation, cross correlation and dislocation correlation among cold, hot and electric loads of the building group and renewable energy output are considered, and high-fidelity scene generation under condition control is realized.

The invention discloses a building energy consumption management and control method, belongs to the technical field of building energy-saving control, and solves the problems of response lag and energy waste caused by independent operation of a lighting system and an HVAC system. According to the technical scheme, a building parameter database is pre-stored; environment data and equipment power consumption are collected in real time; calculating a solar radiation thermal gain coefficient and a thermal load influence quantity of illumination on the HVAC; when the dynamic coupling factor exceeds a first threshold value or the variation of the illumination intensity exceeds a second threshold value within a preset time, triggering collaborative optimization; a lighting target power consumption value (the compensated lighting power consumption is not lower than a preset proportion of the basic lighting demand value) is synchronously generated after response, and the HVAC set temperature is corrected based on the thermal gain coefficient; and finally a control instruction is issued. The method is mainly used for real-time energy consumption collaborative optimization of large-scale public buildings.

The invention relates to the technical field of automatic control, in particular to a PE thermal insulation pipe production equipment control method and system.The method comprises the following steps that based on raw material particle size distribution characteristics, heating temperature control parameters are adjusted, the viscosity state is recognized in combination with screw torque and rotating speed, and the feeding rate of each section is adjusted according to the viscosity difference; forming uniformity is calculated and foaming agent injection parameters are adjusted in combination with a foam structure, and flow velocity parameters of a cooling area are adjusted in combination with thermal load evaluation. According to the invention, by introducing raw material particle size analysis parameters and combining torque and viscosity change characteristics, the heating temperature control and extrusion viscosity adjustment process is optimized, the adaptability of equipment to material physical property change is enhanced, the foaming agent injection process is fed back by using foam structure distribution, the foaming uniformity and injection accuracy are improved, and the foaming efficiency is improved. And by adopting linkage analysis of heat release intensity and cooling rate, dynamic regulation and control of the cooling circulation rate are realized, and the consistency and stability of cooling shaping are improved.

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 discloses an efficient circulating water heating system based on a heat pump and a control method thereof.The system comprises a multi-stage heat pump system, a circulating water heating loop and an intelligent control unit, and the multi-stage heat pump system is used for increasing heat energy of a low-temperature heat source stage by stage to heat circulating water; the intelligent control unit is integrated with a data acquisition module, a load prediction module, a system modeling module, a parameter optimization module, an instruction issuing module and a performance evaluation module, and can realize thermal load dynamic prediction and thermal performance modeling and optimization control of the heat pump system. The control method is based on deep learning and model prediction control, combines real-time data and an external scheduling instruction, calculates an optimal operation parameter to improve the system energy efficiency ratio or reduce the operation cost, and continuously optimizes prediction and modeling precision through an adaptive learning mechanism. The system has the capacity of intelligent prediction, accurate control, efficient heat transfer and self-optimization, is suitable for occasions such as industries and buildings with high requirements for hot water supply energy efficiency, and remarkably improves the operation efficiency and stability of the system.

The invention relates to a heat supply prediction method and system based on spatial-temporal feature fusion deep learning, and the method comprises the following steps: S1, carrying out the collection and fusion of multi-source heterogeneous data, and constructing an integrated data set; s2, preprocessing the data; s3, constructing a graph structure model of the heat supply system, and constructing a weighted undirected graph; s4, constructing and executing forward calculation of the space-time double-flow deep network; s5, designing a composite loss function including mean square error loss and physical constraint loss, and performing joint optimization training on the space-time double-flow deep network; and S6, performing multi-step heat supply load prediction by using the trained model, outputting a heat supply load curve of each heat exchange station in a specified time period in the future, and integrating a prediction result with a heat supply scheduling system. The method has the advantages that the prediction precision is improved compared with that of a traditional machine learning model by capturing the spatial-temporal characteristics at the same time, and the advantages are more remarkable in extreme weather.