600 results about "Energy cost" patented technology

The invention discloses a multi-level energy management system based on multi-dimensional data, and relates to the technical field of energy intelligent management, and the system comprises a multi-source data collection module which collects power utilization, environment and equipment state data in real time; the data fusion processing module is used for processing abnormal values through an algorithm and fusing multi-scale data features; the energy state evaluation module is used for realizing equipment state evaluation and early warning by using a fusion algorithm and a prediction model; the multi-level energy scheduling module adopts an optimization algorithm to balance the energy cost, the production efficiency and the carbon emission, and dynamically adjusts the strategy; the energy performance analysis module is used for developing an analysis tool and an evaluation model; and the decision support module is used for configuring an expert knowledge base and developing a fault diagnosis system and a knowledge graph. Through multi-dimensional data acquisition and multi-level management, the energy data error is greatly reduced, the comprehensive energy cost and carbon emission are remarkably reduced, the cost of participating in enterprise operation is reduced, and the accuracy, safety and sustainability of energy management are improved.

The invention discloses an acquisition and processing system and method for dry data of a transformer, and relates to the technical field of data processing. Comprising the following steps: step 1, multi-source data real-time acquisition and edge preprocessing; 2, constructing a drying end point prediction model; 3, constructing a composite objective function, and solving the composite objective function by adopting a multi-objective evolutionary algorithm; 4, performing real-time correction and feedback; according to the method, key parameters in the drying treatment process are monitored in real time, an accurate drying end point prediction model is automatically constructed, heating and air exhaust curves are dynamically optimized according to the real-time working condition and the production strategy, and traditional manual blind judgment and static presetting are replaced; and meanwhile, the operation energy consumption is remarkably reduced, the optimal balance of the energy consumption and the production efficiency is realized through a multi-objective evolutionary algorithm and an online Kalman filtering correction closed loop mechanism, the drying quality and the equipment safety are improved, the energy cost is greatly saved, and the dual requirements of modern intelligent manufacturing for flexibility, energy conservation and consumption reduction are met.

The invention discloses an airport energy balance management method, system and device and a medium, and relates to the field of energy management. The method comprises the following steps: constructing an energy topological graph according to multi-source data; predicting energy load demands of the airport in different time periods according to the energy topological graph, determining the supply amount of renewable energy in different time periods according to the predicted illumination intensity and wind power, and determining a time-sharing energy supply and demand balance probability according to the energy load demands and the supply amount; determining a preliminary energy management plan according to the cost parameter, the time-sharing energy supply and demand balance probability and the equipment operation constraint, wherein the preliminary energy management plan comprises an energy storage system charging and discharging time sequence, a standby generator starting and stopping plan and a power grid electricity purchasing proportion; and executing the preliminary energy management plan through the digital twin of the airport to obtain first energy execution data, and adjusting the preliminary energy management plan through the difference between the first energy execution data and the second energy execution data at the current moment. By implementing the technical scheme provided by the invention, the energy cost is reduced.

The invention discloses a low-carbon building monitoring system and method based on digital twinning, and relates to the technical field of digital twinning, the system comprises a data acquisition unit, a computing power cooperation unit, a regulation and control execution unit and a user interaction unit, the data acquisition unit carries out regional targeted timing monitoring, builds a unified semantic standard and a dynamic knowledge graph, and sends the unified semantic standard and dynamic knowledge graph to the user interaction unit; the comprehensiveness and efficiency of data association analysis are improved; a three-level computing power collaboration system is constructed through the computing power collaboration unit, building monitoring information is integrated, complex simulation of a digital twin module is carried out, an energy transmission path and equipment fault risks are accurately tracked, prediction optimization and long-term regulation and control management are carried out, and the construction efficiency is improved. Therefore, a full-life-cycle integrated management framework in a macroscopic geographical environment, a mesoscopic energy link and microscopic equipment of a building cluster is built, cross-domain collaborative analysis is realized, the real-time requirement is met, multi-target optimization design for balancing energy cost, energy efficiency and comfort is realized, and carbon neutralization targets in the building industry are assisted to fall into the ground.

The invention discloses an energy storage equipment power balance dynamic load automatic regulation and control method and system, and the method comprises the steps: building a charging power optimization model based on an electricity price policy divided in peak and valley time periods, combining historical electricity demand data, and obtaining the optimal charging power in each time period; the method comprises the following steps: constructing a discharge power dynamic adjustment model through operation state parameters acquired by a real-time environment parameter acquisition device in combination with real-time electricity demand data, and acquiring optimal discharge power in each time period; parameters of the two models are optimized, and the load state based on the optimal charging and discharging power is balanced; based on the historical regulation and control data, the overload condition under the abnormal factors is predicted; and on the basis of a power imbalance condition, preferentially ensuring power supply of a key load according to a power distribution priority and executing energy storage equipment state self-checking. The method has the advantages that the charging power and the discharging power are optimized in combination with the electricity price and the real-time requirement, so that the load state is balanced, the energy cost is reduced, and the equipment overload risk is avoided.

The invention provides a heat storage heat pump system control method based on a physical information neural network, and belongs to the technical field of heat storage pump system intelligent control. Aiming at the problems that in the prior art, an algorithm is difficult to adapt to dynamic energy consumption requirements, engineering application of a model is difficult due to building space heterogeneity, high-order RC model prediction credibility is weak, engineering feasibility is poor and the like, a solution combining a physical information sequence to sequence neural network technology and a finite-state machine control strategy is provided. On the model level, a 2R2C resistance-capacitance RC model of building temperature change is established, and then a PI-Seq2seq prediction model is proposed based on the physical model. On the control flow optimization level, on the basis of an industrial and commercial time-of-use electricity price policy, an FSM control model is designed, a system state set is defined, parameters and a transfer function are input, and a control rule is constructed in combination with the working period of a building heat pump and the characteristics of a heat storage tank. And finally, energy consumption cost optimization and indoor temperature stabilization under the peak-valley electricity price are realized.

The invention is suitable for the technical field of mobile edge computing, and provides a hybrid energy WPT-MEC system-based task unloading and resource allocation method, which comprises the following steps of: establishing a system model; an optimization problem with the goal of minimizing the long-term average power grid energy consumption is drawn up and converted into a Markov decision process; implementing an action space reduction strategy, and executing data normalization processing; constructing an enhanced depth deterministic policy gradient (EDDPG) algorithm; and generating and outputting an optimal strategy according to a real-time system state by using the trained network. According to the method, the wireless power transmission (WPT) technology, the green energy harvesting technology and the mobile edge computing technology are combined, so that the energy use duration and the computing capacity of the wireless equipment are improved. The online decision-making mechanism based on the EDDPG algorithm can dynamically optimize task unloading and resource allocation strategies, adapts to a dynamic environment, reduces energy cost and environmental pollution, and provides reliable technical support for large-scale deployment of the Internet of Things.

The invention relates to the technical field of high-performance computing resource management, and discloses a computing resource allocation method for a distributed supercomputing center. The method comprises the following steps: on the basis of obtaining real-time computing task and supercomputing center resource data and uniformly quantifying, integrally predicting resource requirements of future tasks; constructing a mixed integer linear programming model with the minimization of the total operation cost as a single target, wherein the total operation cost is the sum of the energy cost, the carbon emission cost, the data transmission cost and the SLA default penalty cost; solving the model by taking the time-varying electricity price, the green energy ratio, the resource capacity and the network parameters of each center as constraint conditions to generate an optimal resource allocation scheme; and then, by dynamically monitoring the resource state and the task progress, the model is triggered to resolve when the resource utilization rate is detected to be unbalanced or default risks, so that self-adaptive adjustment is realized. According to the invention, global collaborative resource allocation across super computing centers is realized, and operation economy, environmental sustainability and service reliability are considered.

The invention discloses an intelligent agent spraying decoking system for energy-saving decoking of a hearth, and relates to the technical field of decoking of industrial boilers, the intelligent agent spraying decoking system comprises a sensing module used for collecting coke scale characteristic data, and the data comprises the temperature, thickness, component distribution and adhesion strength of coke scale; the module is combined with a temperature sensor, a component analyzer and image detection equipment for cooperative work, and multi-dimensional scorch scale characteristic information is obtained in real time. According to the intelligent agent spraying decoking system for hearth energy saving and decoking, through cooperative use of the sensing module, the control module, the agent storage module, the agent spraying module and the feedback module, intelligent and accurate decoking is achieved, the diversity of coke scales can be effectively dealt with, the heat efficiency of a boiler is improved, the energy cost is reduced, and the safety problem caused by the coke scales can be prevented; the problems that a traditional decoking method is low in efficiency, poor in applicability and insufficient in adaptability to coke scale types are solved.

A federated learning framework including household agents configured to continuously learn model parameters for managing charging periods and discharging periods of household batteries, and microgrid agents to maximize use of local energy based on a pricing policy, including accessing power from other microgrids when there is insufficient local energy to cover local demand. and selling surplus energy to the other microgrids when power generation by the microgrid surpasses the local demand. Each household machine learning agent is configured to control household energy demand from and supply to a microgrid which they are connected in order to minimize household energy cost while adapting to changes in the energy price that is determined based on the pricing policy of the microgrid agent that encourages reduction of carbon emission. A federated learning engine combines the model parameters from the household machine learning agents to update a global household machine learning agent.

The present disclosure discloses a system (106) and a method (400) for energy management in a network (108). The method may include initialization by defining input and output parameters (baseline energy cost, optimized energy cost). The method may include performing real-time data collection, systematically gathering real-time metrics and calculating the baseline energy cost. The method may include conducting AI / ML-based prediction for traffic and load forecasting and coverage demand analysis, using historical and real-time data to predict future patterns, and allowing for resource optimization and proactive adjustments. The method may include dynamically adjusting network parameters and calculating and optimizing real-time energy cost (EC), continuously monitoring the network's current EC based on dynamically adjusted parameters. Compare optimized EC with baseline EC, adjusting parameters as needed.

The invention relates to a powder metallurgy high-performance copper-based alloy friction plate for a torque limiter of a wind power coupler and a preparation method of the friction plate. The friction plate is composed of, by mass, 9%-14% of carbon, 1%-3% of tin, 2%-8% of friction components, 12%-18% of iron, 0%-3% of nickel and the balance copper. The preparation method comprises the steps of burdening, mixing, forming, sintering and machining. The prepared high-performance copper-based alloy friction plate is particularly suitable for a torque limiter of a wind power generation coupler. The alloy disclosed by the invention is reasonable in component design, simple and controllable in preparation process, easy to realize batch, large-scale and automatic production, high in production efficiency, energy-saving and moderate in cost; the obtained product is excellent in performance, and can greatly improve the torque control precision, save the maintenance cost and improve the operation safety and reliability of wind power equipment when being applied to a wind power coupling torque limiter.

The invention relates to an industrial and commercial park shared storage and charging system intelligent scheduling method based on reinforcement learning, and the method comprises the steps: carrying out the modeling of various loads in an industrial and commercial park, building a model of an optical storage and charging system, analyzing the charging and discharging characteristics of the optical storage and charging system, and calculating the degradation cost of a battery. By designing a reasonable reward function, the reinforcement learning algorithm is guided to make optimal charging and discharging decisions at different moments so as to maximize the overall benefits of the industrial and commercial park. And the reinforcement learning agent continuously selects actions according to the current state, executes a corresponding charging and discharging strategy, and optimizes the strategy by feeding back a reward signal. Through multiple iterations, the optimal scheduling mode is found step by step, the balance between power supply and demand is ensured, energy waste is reduced, and the service life of a battery is prolonged. According to the invention, the use efficiency of energy in industrial and commercial parks is improved, the energy cost can be greatly reduced, the double improvement of economy and sustainability is realized, and the safety and stability of a 750KV and above power transmission system can be improved through seamless connection with a power grid through dynamic load adjustment.

The invention discloses a photovoltaic power generation and rod pumped well group off-peak power utilization coupling control method, and relates to the technical field of oilfield development and new energy utilization. The method comprises the following steps of: acquiring data such as photovoltaic power, irradiance, temperature, wind speed, an indicator diagram of a pumping unit, electrical parameters, a working fluid level and stroke frequency, and constructing a multi-modal data set comprising external meteorological characteristics, a shaft operation time sequence and a well group space relationship; a time sequence model combining reversible normalization and cross-variable attention is adopted to predict photovoltaic output, and a multi-mode liquid supply state recognition model is utilized to obtain the liquid supply sufficiency level of each well. On the basis, a peak shifting power consumption optimization model which takes minimum energy consumption and light abandoning quantity as targets and meets the requirements of yield, power grid capacity, start-stop times and liquid supply constraint is constructed, reinforcement learning is introduced to generate well group start-stop and stroke frequency strategies, and dynamic matching of photovoltaic output and pumping unit load is realized according to real-time data rolling updating, so that the optimal power consumption is realized. The photovoltaic utilization rate is improved; and the energy cost is reduced.

The invention relates to a photovoltaic low-carbon park smart energy management system and method based on digital twinning. Accurate mapping and simulation analysis of a park energy system are realized by constructing a digital twinborn model, dynamic optimization scheduling of energy equipment is realized by adopting a multi-agent cooperative control algorithm, an energy optimization strategy is formulated by taking low carbon as a target, the health state of the equipment is predicted by utilizing machine learning, and an intelligent operation and maintenance plan is generated. And decision support is provided through energy big data analysis. The system integrates multi-source data, and efficient utilization and low-carbon operation of park energy are realized through the steps of digital twin modeling, intelligent agent collaboration, low-carbon scheduling, equipment prediction, big data analysis and the like. Practical application shows that the system can improve the photovoltaic efficiency by 8%, reduce the energy cost by 18%, reduce the carbon emission intensity by 30%, and significantly improve the park energy management level and the low-carbon degree.

The present disclosure discloses a system (106) and a method (400) for energy cost (EC) optimization based on one or more beamforming actions in a network (108) The method may include initialization by defining input and output parameters (baseline energy cost, optimized energy cost). The method may include performing real-time data collection, systematically gathering real-time metrics and calculating the baseline energy cost. The method may include conducting AI / ML-based prediction for traffic and load forecasting and coverage demand analysis, using historical and real-time data to predict future patterns, and allowing for resource optimization and proactive adjustments. The method may include dynamically adjusting network parameters and calculating and optimizing real-time energy cost (EC), continuously monitoring the network's current EC based on dynamically adjusted parameters. Compare optimized EC with baseline EC, adjusting parameters as needed.

The invention discloses a side plate type heat exchanger performance testing system and method, and relates to the technical field of heat exchanger testing, and the system comprises a computer which is used for constructing a digital twinborn model, generating multiple groups of dynamic working condition data, inputting the multiple groups of dynamic working condition data into the digital twinborn model to obtain dynamic performance data, screening according to the sensitivity coefficient of each piece of dynamic performance data to obtain sensitive working condition data; the test equipment executes the sensitive working condition data and collects actual measurement state data of the side plate type heat exchanger; and the computer judges whether the performance of the side plate heat exchanger meets the design requirement according to the actually measured state data. Physical modeling is carried out on the heat exchanger in a digital twinning mode, heat exchange simulation is carried out on the heat exchanger, invalid test working conditions are deleted from a parameter set, only sensitive working conditions are reserved, and time, manpower and energy cost needed by testing is greatly saved.

The invention belongs to the technical field of detection equipment, and discloses full-automatic concrete fracture resistance and compression resistance detection equipment aiming at the problem that the position of a sample is changed in fracture resistance detection due to the fact that no mechanism is used for clamping the concrete sample when a workpiece is detected by the existing fracture resistance equipment. The rack is provided with a supporting rod, the top of the supporting rod is provided with a top plate, the rack is provided with an anti-bending detection mechanism and a compression resistance detection mechanism, by adopting the structural design, when an anti-bending experiment is carried out, a concrete sample can be clamped, so that the stability of the anti-bending experiment is ensured, and the sample is prevented from splashing; on the premise that other power is not added, the test sample is clamped while fracture resistance detection is achieved, and the energy cost is saved.

A renewable power system with a twin-configurable architecture is described. The system includes a renewable energy source (RES), an energy storage system (ESS), and at least one controllable load (CL) (e.g., AI training / datacenter). The system can serve as a baseload or semi-baseload plant for CL(s) and / or as a peaker or semi-peaker plant for an electric grid, or vice-versa, and optionally in parallel, can also provide ancillary services to the electric grid and / or to the CL(s). In certain embodiments, e.g. solar PV RES(es), the system can have capacity factors of at least about 60% and up to 100%, higher asset utilization, better economics for the RES-ESS, improved system performance, and lower energy costs as compared with known systems without a CL(s). By making load a variable, and integral part of the system, sophisticated resource allocation strategies, including AI algorithms, can be developed not previously possible with known systems lacking a CL(s).

The invention discloses a virtual power plant source load interaction optimization scheduling model based on low-carbon response and a solving algorithm, and belongs to the technical field of power system optimization scheduling. A low-carbon scheduling framework containing a distributed power supply, energy storage, a flexible load and a carbon transaction mechanism is constructed, the carbon emission intensity of each link is quantified to form a carbon flow scheduling signal, and a dynamic carbon emission factor and energy cost are coupled. A multi-objective optimization model is established, a complex function is processed by piecewise linearization, and a hybrid algorithm of an improved genetic algorithm and a commercial solver is designed to improve the solving efficiency. The prediction error is dynamically corrected through a'prediction-optimization-feedback 'closed loop, and the strategy is adjusted. According to the scheme, low-carbon and economic collaborative optimization is realized, renewable energy consumption and system stability are enhanced, user satisfaction and real-time scheduling are considered, and a solution is provided for low-carbon intelligent operation of the power distribution network.

The invention provides a carbon powder production line energy consumption optimization method and system, and is applied to the technical field of energy consumption optimization in the production and manufacturing field, and the method comprises the steps: obtaining a production order, decomposing the production order into subtasks and time-of-use electricity price information, dividing a time slot, calculating the energy cost, constructing a mathematical optimization model according to the energy cost, and solving the mathematical optimization model. And a group of optimal decision variable values are found, so that the total energy cost is minimum, and all production constraints are met. Through the steps, the method integrates the production task, the equipment capability, the dynamic energy price and the production constraint into an optimization framework, and finds the production scheduling scheme with the lowest energy cost through the solving model, so that the technical problem of reducing the energy consumption cost while meeting the production requirement is solved, and the method has the advantages of reducing the energy consumption and improving the production efficiency. And the production plan is optimized.

A system and method for an AI enhanced smart grid framework with optimized incentive capabilities. The present invention optimizes energy efficiency by generating personalized retrofit plans, optimal energy dispatch schedules, and customized financial incentives. Machine learning algorithms analyze homeowner profiles, energy consumption patterns, and market conditions to recommend tailored energy-saving solutions. The system integrates with funding institutions and aggregators to streamline retrofit financing and carbon credit management. By leveraging AI and ML techniques, this innovative approach aims to accelerate the adoption of energy-efficient technologies, reduce homeowner energy costs, and contribute to a more sustainable energy future.

The invention provides an optimal control method and system for power system carbon emission, and relates to the technical field of carbon emission optimal control, and the method comprises the steps: collecting a real-time load, a power grid carbon emission factor and time-of-use electricity price, employing a clustering algorithm to automatically divide a power utilization period, carrying out the modeling of the dynamic change of the carbon emission factor at different time periods, and carrying out the optimal control of the power system carbon emission. An energy storage charging and discharging path is optimized according to the principle of peak period discharging, smoothing and valley period charging, and energy loss and the health state of an energy storage unit in the charging and discharging process are included in carbon emission accounting; and meanwhile, a multi-target collaborative optimization model is constructed to realize tradeoff optimization of the total carbon emission and the power consumption cost, so that the time sequence characteristics of the actual production load and the power grid carbon emission factor are accurately matched, the carbon emission collection precision and the scheduling optimization level are improved, and the carbon emission and the energy consumption cost in the production cycle are effectively reduced.

A system and method for managing and distributing an electrical power system is provided. In one embodiment, the system and method comprise at least one powered device that receives electrical power from an electrical power system utilizing multiple energy sources. Structure and devices are provided within the system to command, regulate, monitor, and transmit power to a device or devices in an efficient, reliable manner based on resource availability, energy cost, and environmental factors. The system selects and manages the resources to optimize the energy output to the device(s) to achieve the specific requirements in resiliency, cost, and the environmental impact of the application.

The invention provides a method and device for determining steelmaking-continuous casting production multi-process collaborative demand response, and relates to the technical field of manufacturing, and the method comprises the steps that target parameters in the steelmaking-continuous casting production process are obtained; based on a production plan, each work starting moment, each work ending moment, each unit energy consumption and time-of-use electricity prices of different time periods included in the target parameters, by taking the minimum energy cost and the minimum waiting duration as targets and based on an MILP method, establishing a heat coordinated optimization material flow mathematical model in the steelmaking-continuous casting production process and solving the heat coordinated optimization material flow mathematical model; and a coordinated optimization demand response strategy is obtained. According to the method, the coordinated optimization demand response strategy in the steelmaking-continuous casting production process is quickly obtained by solving the heat coordinated optimization material flow mathematical model, so that the determination efficiency of the heat distribution strategy is improved, the internal flexibility adjustment potential of iron and steel enterprises can be mined by the obtained coordinated optimization demand response strategy, and the economic benefit of the enterprises is improved. And friendly interaction between iron and steel enterprises and a power grid is promoted.

The invention belongs to the field of heat supply, and discloses a heat supply system and power grid interaction method, device and equipment and a storage medium, and the method comprises the steps: obtaining the real-time electricity price, the dynamic threshold electricity price and the operation data of a phase change heat storage system, inputting the obtained data into a pre-established Markov decision process model, the Markov decision process model comprises a heat storage process sub-model and a heat supply process sub-model; according to the real-time electricity price, the dynamic threshold electricity price and the thermal load, determining that the moment attribute of the current moment is a heat storage moment or a heat supply moment, and based on the determined moment attribute, selecting a target sub-model from two sub-models of a Markov decision process model; and solving the target sub-model according to a DQN algorithm, and controlling the operation of the phase change heat storage and supply system according to a solving result. Through the method, the building energy demand can be accurately met, the user energy cost is reduced, the real-time supply and demand balance of a power system is promoted, and the power supply pressure at the power peak is relieved.

The present invention provides a method, device, system and electronic equipment for managing photovoltaic storage and charging energy, which are applied to the technical field of photovoltaic storage and charging energy management. The method dynamically adjusts the photovoltaic output, charging power and energy storage output at the next moment according to the electricity price period by obtaining the current electricity purchase price, electricity selling price and load power. The energy storage is discharged at the maximum during high-price periods, charged at the maximum during low-price periods, and the power is balanced during medium-price periods to ensure operation within the preset load limit, and the confirmed parameters are sent to the device side to achieve efficient energy management. The present invention comprehensively considers the real-time dynamic changes in the purchase price and the selling price of electricity, and adjusts the photovoltaic output, charging efficiency and energy storage output in combination with the electricity price period, thereby optimizing energy use, improving energy utilization and reducing overall energy costs.

The invention relates to the technical field of energy and power engineering, and discloses an efficient thermoelectric decoupling system based on phase change energy storage, which comprises an energy storage unit module, a regulation and control algorithm module and a scheduling optimization module. The energy storage unit module adopts a nanotechnology novel phase change material, so that the energy storage density and stability are improved; the regulation and control algorithm module realizes accurate control of charging and discharging through a multi-stage regulation and control strategy; and the scheduling optimization module predicts energy demand and production by using a machine learning algorithm, and dynamically adjusts an energy distribution strategy. The problems that a traditional heat storage material is low in energy storage density and poor in stability and lacks an intelligent scheduling mechanism can be solved, the energy utilization efficiency and operation reliability of a thermoelectric system are improved, and the energy cost is reduced.

The invention provides a passenger compartment air conditioner control method based on model predictive control. The method comprises the steps that a passenger compartment model is constructed, and an original data set containing environment parameters, air conditioner control parameters and passenger compartment response parameters is obtained through experimental design; preprocessing the data; a neural network model is trained through the preprocessed data and serves as a prediction model to learn the nonlinear mapping relation among the air conditioner operation parameters, the environment parameters and the passenger compartment states; calling the prediction model to execute MPC rolling prediction, and generating a future state variable sequence; constructing a target function containing a state tracking error and control energy cost, obtaining an optimal control sequence through optimization solution, and acting the first control quantity on an actuator; and dynamically adjusting a prediction model or a control parameter according to real-time feedback. The technical problems that physical modeling is difficult due to high nonlinearity of a passenger compartment air conditioning system, and a traditional control strategy cannot give consideration to high-precision prediction and multi-working-condition adaptability can be solved.

The invention discloses a micro-grid-heating ventilation air-conditioning coordinated optimization method based on deep reinforcement learning, and relates to the field of building energy system management and scheduling, and the method comprises the following steps: S1, constructing a refined model of a heating ventilation air-conditioning system and a micro-grid system, which considers the comfort level of people and the building energy consumption cost; s2, aiming at the refined model constructed in the step S1, adopting a double-layer optimization control strategy to construct a micro-grid energy management system framework integrated with the heating ventilation air conditioning system; s3, based on the step S2, constructing a multi-energy coupling model of the micro-grid-heating ventilation air conditioning system, converting a time sequence optimization problem into a Markov decision process MDP, and performing mathematical representation; and S4, training the intelligent agent by adopting an improved priority experience playback depth deterministic policy gradient algorithm PER-DDPG. According to the method, a collaborative scheduling optimization framework of the micro-grid and the heating ventilation air conditioning system is constructed, and the energy cost of a building system can be effectively reduced under the condition that the comfort level of personnel is effectively guaranteed.