Intelligent building energy management method, system, equipment and medium

Abstract

The invention relates to the technical field of energy management, in particular to an intelligent building energy management method, system and device and a medium. According to the invention, the collaborative multi-agent reinforcement learning technology is adopted to realize comprehensive energy management of electric, heat and cold loads and energy storage and distributed energy systems in the smart building: firstly, through a multi-objective optimization strategy, economy, energy efficiency, environmental influence and user comfort are balanced, and comprehensive optimization of energy utilization is realized; secondly, the system can adapt to demand fluctuation and uncertainty of energy output, and stability and reliability of energy supply are ensured; furthermore, a real-time monitoring and self-adaptive learning mechanism enables the system to quickly respond to external environment change, so that the flexibility and the accuracy of decision making are improved; the expandability and modular design of the scheme are easy to adapt to smart buildings of different scales, and an efficient, intelligent and environment-friendly solution is provided for building energy management.

Description

Technical Field

[0001] The present disclosure relates to the field of energy management technology, and in particular to a smart building energy management method, system, device and medium. Background Art

[0002] Building energy consumption accounts for the largest part of global energy consumption, and building energy consumption mainly comes from heating, ventilation and air conditioning (HVAC) systems, hot water equipment, elevators and lighting systems. Energy management systems are cutting-edge methods for optimizing building energy consumption and efficiency. Formulating effective control strategies for these devices can significantly save energy. Due to the wide variety of building energy-consuming equipment, there are many types of energy consumption control objects that need to be considered in smart buildings, and the control objects are often complex, so it is not easy to formulate effective control strategies.

[0003] The existing technology mainly has the following problems:

[0004] (1) Rule-based feedback control, which formulates control plans in advance, does not take into account the many uncertain factors of building comprehensive energy on the source side and the energy consumption side, and the actual application effect is not good.

[0005] (2) Predictive control based on uncertainty information or model-based predictive control requires an accurate model and the modeling process is costly. Summary of the invention

[0006] In order to solve the above problems, the present invention provides a smart building energy management method, system, device and medium.

[0007] In a first aspect, a smart building energy management method is provided, the method comprising:

[0008] Identify smart building energy management equipment and information;

[0009] Based on management equipment and information, a mathematical model of building system energy management is constructed; the energy management mathematical model is modeled using Markov decision process; a training scenario is constructed for the energy management mathematical model, and an intelligent agent for energy management is obtained through training;

[0010] The real-time data of management equipment and information is input into the intelligent agent to obtain the energy scheduling plan for each device.

[0011] Further, management equipment, including: electrical load, heat load, cooling load, energy storage system and distributed energy system; information, including: external environment information;

[0012] Electrical load, including electrical equipment in the building;

[0013] Thermal load, including heating and hot water supply systems in buildings;

[0014] Cooling load, including air conditioning and refrigeration equipment in the building;

[0015] Energy storage systems, including in-building electrical energy storage, thermal energy storage, and electric vehicles;

[0016] Distributed energy systems, including photovoltaic power generation systems and combined heat and power systems;

[0017] External environmental information, including real-time electricity prices, gas prices, weather forecasts, and external temperature and humidity information.

[0018] Furthermore, the objective function of the energy management mathematical model includes:

[0019] The objective function of the energy management mathematical model is obtained by weighting the following indicators:

[0020] The economic goal is to minimize the energy cost of building operation, the energy efficiency goal is to maximize the efficiency of energy utilization, the environmental goal is to minimize the environmental impact of carbon emissions, and the comfort goal is to consider the comfort of users.

[0021] Furthermore, the economic goal of minimizing the energy cost of building operation is as follows:

[0022]

[0023] Where: T is the scheduling period; t is the scheduling time, which is a discrete value; β e (t) and β g (t) are the electricity purchase price and gas purchase price corresponding to the dispatch time t; E in (t) and G in (t) are the electricity purchase capacity and gas purchase capacity corresponding to the dispatch time t respectively;

[0024] The energy efficiency target of maximizing energy utilization efficiency is as follows:

[0025]

[0026] Where: D e (t) is the power load corresponding to the dispatching time t; D h (t) is the heat load corresponding to the dispatching time t; ω g is the calorific value of natural gas;

[0027] The environmental goal of minimizing the impact of carbon emissions is as follows:

[0028]

[0029] In the formula, and are the carbon emission coefficients of electricity and natural gas in the external system, respectively;

[0030] Considering the comfort target of user comfort, the formula is as follows:

[0031]

[0032] In the formula, S e , S h are user discomfort caused by unit electrical load and thermal load deviation respectively; ΔD e (t),ΔD h (t) are the offsets of the electrical load and thermal load compared to the optimal comfort load at time t;

[0033] The objective function of the energy management mathematical model is as follows:

[0034] F=ω1*F1-ω2*F2+ω3*F3+ω4*F4

[0035] ω1+ω2+ω3+ω4=1

[0036] 0≤ω1,ω2,ω3,ω4≤1

[0037] Where ω1, ω2, ω3, and ω4 are the weight coefficients of each sub-goal respectively.

[0038] Furthermore, the constraints of the energy management mathematical model include:

[0039] Energy balance constraints, CHP equipment operation constraints, electric energy storage equipment operation constraints, thermal energy storage equipment operation constraints, electric vehicle constraints and electric heat balance constraints.

[0040] Furthermore, the energy balance constraint is given by the following formula:

[0041]

[0042] Where: D e (t) is the power load corresponding to the dispatch time t; μ e To consider the power supply efficiency of line loss; and is the PEV battery charging and discharging amount at scheduling time t, and their efficiencies are and E pv (t) is the photovoltaic power generation power at the scheduling time t, μ pv Power supply efficiency for photovoltaic systems; D is the power generation of CHP system at the dispatching time t; h (t) is the heat load corresponding to the dispatching time t; is the heating load of the CHP system corresponding to the scheduling time t; π(t) represents the proportion of gas consumed by CHP equipment, and the remaining gas is supplied to other gas equipment, μB Indicates the heating efficiency of other gas equipment;

[0043] The operating constraints of CHP equipment are as follows:

[0044]

[0045] In the formula, and represent the CHP power generation and heating efficiency respectively;

[0046] Operating capacity limitations of CHP:

[0047]

[0048] The operating constraints of electric energy storage equipment are as follows:

[0049]

[0050] Where: Indicates the minimum discharge power of the electric energy storage device; Indicates the maximum discharge power of the electric energy storage device; Indicates the discharge power of the electric energy storage device during period t; Indicates the minimum charging power of the electric energy storage device; Indicates the maximum charging power of the electric energy storage device; represents the charging power of the electric energy storage device during period t; A binary variable representing the discharge state of the electrical energy storage device; A binary variable representing the state of charge of the electrical energy storage device; Indicates the charging efficiency of the electric energy storage device; Indicates the discharge efficiency of the electrical energy storage device; E ES (t) represents the energy value of the electric energy storage device during time period t; E ES (t-1) represents the energy value of the electric energy storage device during the period t-1; Indicates the minimum energy value of the electric energy storage device within the dispatch period; Indicates the maximum energy value of the electric energy storage device within the dispatch period;

[0051] The operating constraints of thermal energy storage equipment are as follows:

[0052]

[0053] Where: Indicates the minimum heat release power of the thermal energy storage device; Indicates the maximum heat release power of the thermal energy storage device; It represents the heat release power of the thermal energy storage device during period t; Indicates the minimum heat storage power of the thermal energy storage device; Indicates the maximum heat storage power of the thermal energy storage device; It represents the thermal storage power of the thermal energy storage device during period t; A binary variable representing the heat storage state of the thermal energy storage device; A binary variable representing the heat release state of the thermal energy storage device; Indicates the heat storage efficiency of thermal energy storage equipment; Indicates the heat release efficiency of thermal energy storage equipment; E HS (t) represents the energy value of the thermal energy storage device during time period t; E HS (t-1) represents the energy value of the thermal energy storage device during the period t-1; Indicates the minimum energy value of the thermal energy storage device within the scheduling period; Indicates the maximum energy value of the thermal energy storage device within the scheduling period;

[0054] Electric vehicle constraints, the formula is as follows:

[0055]

[0056]

[0057] E PEV (t)≤Cap PEV

[0058]

[0059] Where: t dep Indicates the PEV departure time; t arr Indicates the PEV arrival time; E PEV (t arr ) indicates the charge state of the PEV when it arrives at the home; and Respectively represent the maximum charging and discharging power of PEV; E PEV (t) represents the PEV battery capacity at the dispatch time t; Cap PEV Indicates the PEV battery capacity; represents the desired state of charge of the PEV battery determined by the user at the time of departure; It is the minimum state of charge of the PEV battery;

[0060] The electric and thermal balance constraint is as follows:

[0061]

[0062] Where: and They represent the non-dispatchable electric load and non-dispatchable thermal load within the commissioning time t respectively; and They represent the dispatchable electric load and dispatchable thermal load within the commissioning time t respectively;

[0063]

[0064]

[0065] Where: and Represents an adjustable electrical load and thermal load of the breaking type, which is controlled by switching ON / OFF and remains ON for a specified time after the switch is turned ON; and Respectively represent the binary states of the on-off adjustable electric load and thermal load, where a value of 1 indicates that the device is on, and a value of 0 indicates that the device is off; b i and f i Indicates the start and end time of the operation of the breaking type adjustable load; and Respectively represent the operating time required for the breaking adjustable electrical load and thermal load; and They represent flexible adjustable electrical load and thermal load respectively, and their power consumption can be adjusted up or down. and Indicates the upper and lower limits of the flexible adjustable electric load; and Indicates the adjustable upper and lower limits of the flexible adjustable thermal electric load.

[0066] Furthermore, the energy management mathematical model is modeled using a Markov decision process, including:

[0067] Establish electric energy storage equipment agent, thermal energy storage equipment agent, electric vehicle equipment agent, CHP equipment agent, disconnection type adjustable load agent and flexible adjustable load agent, and set the action space, state space and reward space for each agent;

[0068] The energy storage device agent is used to control the charging or discharging state of the energy storage device. Its action space includes: charging, discharging, and no action. The actions are divided into the following ten sets:

[0069]

[0070] In the formula, the action value Represents the ratio of charging or discharging power to rated power, which controls the charging or discharging rate;

[0071] The reward design includes the difference between the electricity price in each state and the average of the electricity price as the basis for calculating the reward amount. The formula is as follows:

[0072]

[0073] In the formula, β e (t) represents the electricity price at dispatch time t; represents the average daily electricity price;

[0074] Thermal energy storage device agent, used to control the heating or heat release state of thermal energy storage devices; its action space includes: heating, heat release, no action, and the actions are divided into the following ten sets:

[0075]

[0076] In the formula, the action value Represents the ratio of heating or heat release power to rated power, which controls the heating or heat release rate;

[0077] The reward design includes: the average value of the gas price difference and the electric power price in each state is used as the basis for calculating the reward amount. The formula is as follows:

[0078]

[0079] In the formula, β g (t) represents the gas price at dispatch time t; represents the average daily gas price;

[0080] The electric vehicle device agent is used to control the charging or discharging state of the electric vehicle device; its action space includes: charging, discharging, and no action; the actions are divided into the following ten sets:

[0081]

[0082] In the formula, the action value Represents the ratio of charging or discharging power to rated power, which controls the charging or discharging rate;

[0083] The reward design includes: the difference between the electricity price in each state and the average of the electricity price are used as the basis for calculating the reward amount. The formula is as follows:

[0084]

[0085] In the formula, β e (t) represents the electricity price at dispatch time t; represents the average daily electricity price;

[0086] The CHP equipment agent is used to control the gas consumption ratio π coefficient to control the contribution of CHP in power production. Its action space formula is as follows:

[0087] a CHP =0,0.01,0.02,...,1

[0088] The reward function formula of the CHP agent is as follows:

[0089] R CHP =-a CHP *G in (t)*(β e (t)-β g (t))

[0090] The breaking type adjustable load agent is used to control the on or off state of the breaking type load. The formula is as follows:

[0091]

[0092] The agent determines the working time of the machine by choosing actions, and the formula is as follows:

[0093]

[0094] Flexible adjustable load agent is used for energy consumption management of adjustable load. The action of flexible adjustable load agent is the energy consumed in each time step. The formula is as follows:

[0095]

[0096] Each adjustable device uses an agent agent and constructs its own Q function for each agent, which is updated during training by the following formula:

[0097]

[0098] In the formula, and They represent the state and action of agent numbered i at time step t, Indicates that the agent numbered i is in state S t and perform action A t The expected return when R t+1 represents the reward obtained at event step t+1; γ is the discount factor, which indicates the importance of future rewards; θ is the learning rate, which affects the speed at which new information covers old information; It represents the maximum value of Q value among all possible actions in the state of the next time step t+1;

[0099] Input real-time data into the intelligent agent to obtain energy scheduling for each device, including:

[0100] Real-time monitoring of the status of each device and subsystem through a sensor network, including changes in demand for electricity, heat, and cooling loads, the charging and discharging status of energy storage devices, photovoltaic power generation, and external environmental information;

[0101] Real-time data is input into the intelligent agent, which selects the optimal action based on the current state and performs energy scheduling for each device.

[0102] In a second aspect, a smart building energy management system includes: a device unit, a modeling unit, and a scheduling unit;

[0103] Equipment unit, used to determine smart building energy management equipment and information;

[0104] A modeling unit is used to construct a mathematical model of energy management of a building system based on management equipment and information; the mathematical model of energy management is modeled using a Markov decision process; a training scenario is constructed for the mathematical model of energy management, and an intelligent agent for energy management is obtained through training;

[0105] The scheduling unit is used to input the real-time data of management equipment and information into the intelligent agent to obtain the energy scheduling plan for each device.

[0106] In a third aspect, an electronic device includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0107] a memory storing a computer program;

[0108] The processor is used to implement the above-mentioned smart building energy management method when executing the computer program stored in the memory.

[0109] In a fourth aspect, a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned smart building energy management method is implemented.

[0110] The present disclosure has at least the following beneficial effects:

[0111] The present invention realizes the comprehensive energy management of electricity, heat, cooling loads, energy storage and distributed energy systems in smart buildings by adopting collaborative multi-agent reinforcement learning technology: first, it balances economy, energy efficiency, environmental impact and user comfort through multi-objective optimization strategy, and realizes comprehensive optimization of energy utilization; second, the system can adapt to demand fluctuations and uncertainty of energy production, ensuring the stability and reliability of energy supply; third, real-time monitoring and adaptive learning mechanisms enable the system to respond quickly to changes in the external environment, improving the flexibility and accuracy of decision-making; the scalability and modular design of the solution can easily adapt to smart buildings of different sizes, providing an efficient, intelligent and environmentally friendly solution for building energy management.

[0112] The present disclosure adopts comprehensive energy management. The system not only considers the management of electrical load, but also integrates the thermal load, cooling load and energy storage system to achieve coordinated optimization management of multiple energy sources.

[0113] The present disclosure adopts multi-objective optimization and achieves trade-offs and optimization among objectives by defining a multi-objective function including economy, energy efficiency, environmental impact and user comfort.

[0114] The present disclosure adopts autonomous management of independent intelligent agents. Each device or subsystem acts as an independent intelligent agent and is responsible for its own energy scheduling and management decisions, thereby improving the flexibility and response speed of the system.

[0115] Other features and advantages of the present disclosure will be described in the following description, and partly become apparent from the description, or be understood by implementing the present disclosure. The purpose and other advantages of the present disclosure can be realized and obtained by the structures indicated in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0116] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0117] Figure 1 This is a flowchart of the management method of the embodiment of the present disclosure;

[0118] Figure 2 This is a schematic diagram of the structure of the management device according to an embodiment of the present disclosure;

[0119] Figure 3 It is a schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0120] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0121] In a first aspect, a smart building energy management method is provided, the method comprising:

[0122] S101, determine smart building energy management equipment and information;

[0123] S102, constructing a mathematical model of energy management of a building system based on management equipment and information; the mathematical model of energy management is modeled using a Markov decision process; constructing a training scenario for the mathematical model of energy management, and training to obtain an intelligent agent for energy management;

[0124] S103, inputting the real-time data of management equipment and information into the intelligent agent to obtain an energy scheduling plan for each device.

[0125] The specific implementation is as follows:

[0126] For the target building, smart building, extract the following management equipment information:

[0127] Electrical load: includes electrical equipment in the building, such as lighting, air conditioning, office equipment, etc.

[0128] Heat load: includes heating, hot water supply and other systems within the building.

[0129] Cooling load: includes air conditioning and refrigeration equipment in the building.

[0130] Energy storage system: including in-building electrical energy storage equipment (ES), thermal energy storage equipment (HS), and electric vehicles (EV).

[0131] Distributed energy systems, including photovoltaic systems (PV) and combined heat and power systems (CHP).

[0132] External environmental information: including real-time electricity prices, gas prices, weather forecasts, external temperature and humidity, etc.

[0133] Constructing a mathematical model of building system energy management specifically includes defining an energy management objective function and determining the constraint equations of the management equipment in step S1. The energy management objective function includes: an economic objective of minimizing the energy cost of building operation, an energy efficiency objective of maximizing energy utilization efficiency, an environmental objective of minimizing the environmental impact of carbon emissions, and a comfort objective of considering user comfort; the energy management objective function is a weighted combination of the above four objective functions.

[0134] The mathematical model of target building energy management is modeled as a Markov decision process, which specifically includes treating each device or subsystem (such as electrical load, energy storage, distributed energy) as an independent intelligent agent. Each intelligent agent is responsible for its own energy scheduling and management decisions, and setting an action space and action reward function for each agent.

[0135] The electricity, heat and cooling demand, photovoltaic power generation fluctuations, electric vehicle arrival / departure time and charge state at arrival are taken as uncertain variables; multiple scenarios are constructed and divided into training and testing categories. The proposed method is performed on the training scenario to train the Q function of the agent.

[0136] Decision-making process: Real-time monitoring of the status of each device and subsystem through a sensor network, including changes in demand for electricity, heat, and cooling loads, the charging and discharging status of energy storage equipment, photovoltaic power generation, etc.; external environmental information (such as electricity prices, gas prices, and weather).

[0137] State input: Real-time data is input into the reinforcement learning model. The agent selects the optimal action based on the current state and schedules energy for each device.

[0138] Action execution: Execute various scheduling operations according to the agent's decision.

[0139] Feedback update: Record the actual execution results, feed the results back to the reinforcement learning model, and continuously optimize the Q function and decision-making strategy of the agent.

[0140] The objective function is defined as follows:

[0141] Economic goals of minimizing energy costs in building operations:

[0142]

[0143] Where:

[0144] T is the scheduling period;

[0145] t is the scheduling time, which is a discrete value;

[0146] β e (t) and β g (t) are the electricity purchase price and gas purchase price corresponding to the dispatch time t respectively;

[0147] E in (t) and G in (t) are the electricity purchase capacity and gas purchase capacity corresponding to the dispatch time t.

[0148] Energy efficiency goals to maximize energy efficiency:

[0149]

[0150] Where:

[0151] D e (t) is the power load corresponding to the dispatch time t;

[0152] D h (t) is the heat load corresponding to the dispatching time t;

[0153] ω g is the calorific value of natural gas;

[0154] Environmental goals to minimize carbon emissions impact:

[0155]

[0156] Where: and are the carbon emission coefficients of electricity and natural gas in the external system, respectively;

[0157] Comfort goals that take user comfort into account:

[0158]

[0159] Where: S e , S h are user discomfort caused by deviation of unit electrical load and thermal load, respectively;

[0160] ΔD e (t),ΔD h (t) are the offsets of the electrical load and thermal load compared to the optimal comfort load at time t;

[0161] The energy management optimization objective function is the weighted sum of the above sub-objective functions, and the mathematical expression is as follows:

[0162] F=ω1*F1-ω2*F2+ω3*F3+ω4*F4

[0163] ω1+ω2+ω3+ω4=1

[0164] 0≤ω1,ω2,ω3,ω4≤1

[0165] Among them, ω1, ω2, ω3, and ω4 are the weight coefficients of each sub-goal respectively.

[0166] The constraints are defined as follows:

[0167] A1: Energy balance constraint:

[0168]

[0169] Where:

[0170] D e (t) is the power load corresponding to the dispatch time t;

[0171] μ e To consider the power supply efficiency of line loss;

[0172] and is the PEV battery charging and discharging amount at scheduling time t, and their efficiencies are and

[0173] E pv (t) is the photovoltaic power generation power at the scheduling time t, μ pvPower supply efficiency for photovoltaic systems;

[0174] is the power generation of CHP system at the dispatching time t;

[0175] D h (t) is the heat load corresponding to the dispatching time t;

[0176] is the heating load of the CHP system corresponding to the scheduling time t;

[0177] π(t) represents the proportion of gas consumed by CHP equipment, and the remaining gas is supplied to other gas equipment, μ B Indicates the heating efficiency of other gas equipment.

[0178] A2: CHP equipment operation constraints:

[0179]

[0180] Where: and represent the CHP power generation and heating efficiency respectively;

[0181] Operating capacity limitations of CHP:

[0182]

[0183] A3: Operation constraints of electric energy storage equipment:

[0184]

[0185] Where:

[0186] Indicates the minimum discharge power of the electric energy storage device;

[0187] Indicates the maximum discharge power of the electric energy storage device;

[0188] Indicates the discharge power of the electric energy storage device during period t;

[0189] Indicates the minimum charging power of the electric energy storage device;

[0190] Indicates the maximum charging power of the electric energy storage device;

[0191] represents the charging power of the electric energy storage device during period t;

[0192] A binary variable representing the discharge state of the electrical energy storage device;

[0193] A binary variable representing the state of charge of the electrical energy storage device;

[0194] Indicates the charging efficiency of the electric energy storage device;

[0195] Indicates the discharge efficiency of the electrical energy storage device;

[0196] E ES (t) represents the energy value of the electric energy storage device during time period t; E ES (t-1) represents the energy value of the electric energy storage device during the period t-1; Indicates the minimum energy value of the electric energy storage device within the dispatch period; Indicates the maximum energy value of the electric energy storage device within the dispatch period. A4: Thermal energy storage device operation constraints:

[0197]

[0198] Where:

[0199] Indicates the minimum heat release power of the thermal energy storage device;

[0200] Indicates the maximum heat release power of the thermal energy storage device;

[0201] It represents the heat release power of the thermal energy storage device during period t;

[0202] Indicates the minimum heat storage power of the thermal energy storage device;

[0203] Indicates the maximum heat storage power of the thermal energy storage device;

[0204] It represents the thermal storage power of the thermal energy storage device during period t;

[0205] A binary variable representing the heat storage state of the thermal energy storage device; A binary variable representing the heat release state of the thermal energy storage device;

[0206] Indicates the heat storage efficiency of thermal energy storage equipment;

[0207] Indicates the heat release efficiency of thermal energy storage equipment;

[0208] E HS (t) represents the energy value of the thermal energy storage device during time period t; E HS (t-1) represents the energy value of the thermal energy storage device during the period t-1; Indicates the minimum energy value of the thermal energy storage device within the scheduling period; Indicates the maximum energy value of the thermal energy storage device within the dispatch period. A5: Electric vehicle constraints:

[0209]

[0210] E PEV (t)≤Cap PEV

[0211]

[0212] Where:

[0213] t dep Indicates the PEV departure time;

[0214] t arr Indicates the PEV arrival time;

[0215] E PEV (t arr ) indicates the charge state of the PEV when it arrives at the home;

[0216] and Respectively represent the maximum charging and discharging power of PEV; E PEV (t) represents the PEV battery capacity at the dispatch time t;

[0217] Cap PEV Indicates the PEV battery capacity;

[0218] represents the desired state of charge of the PEV battery determined by the user at the time of departure;

[0219] It is the minimum state of charge of the PEV battery;

[0220] A6: Electric and thermal balance constraints:

[0221]

[0222] Where: The constraint formulas for responsive and non-responsive heat loads are the same as above;

[0223] and They represent the non-dispatchable electric load and non-dispatchable thermal load within the commissioning time t respectively; and They represent the dispatchable electric load and dispatchable thermal load within the commissioning time t respectively;

[0224]

[0225] Where:

[0226] and Indicates an interrupting type adjustable electrical load and thermal load, which is controlled by a switch ON / OFF and must remain ON for a specified time after the switch is turned ON.

[0227] and They represent the binary states of the disconnecting adjustable electric load and thermal load respectively, with a value of 1 indicating that the device is on and a value of 0 indicating that the device is off;

[0228] b i and f i Indicates the start and end time of the operation of the breaking type adjustable load;

[0229] and Respectively represent the operating time required for the breaking adjustable electrical load and thermal load;

[0230] and They represent flexible adjustable electrical load and thermal load respectively, and their power consumption can be adjusted up or down. and Indicates the upper and lower limits of the flexible adjustable electric load; and Indicates the upper and lower limits of the flexible adjustable thermal power load;

[0231] Markov process transformation defines the action space, state space and reward space for each agent, including:

[0232] B1: Electric energy storage device agent: controls the charging or discharging status of the electric energy storage device;

[0233] The action space includes: charging, discharging, and no action. The actions are divided into the following ten sets:

[0234]

[0235] Action Value Represents the ratio of charging or discharging power to rated power, which controls the charging or discharging rate;

[0236] Reward design: The difference between the electricity price in each state and the average of the electricity price are used as the basis for calculating the reward amount:

[0237]

[0238] Where:

[0239] β e (t) represents the electricity price at dispatch time t;

[0240] represents the average daily electricity price;

[0241] B2: Thermal energy storage device agent: controls the heating or heat release state of the thermal energy storage device;

[0242] The action space includes: heating, releasing heat, and no action. The actions are divided into the following ten sets:

[0243]

[0244] Action Value Represents the ratio of heating or heat release power to rated power, which controls the heating or heat release rate;

[0245] Reward design: The average value of the gas price difference and the electricity price in each state is used as the basis for calculating the reward amount:

[0246]

[0247] Where:

[0248] β g (t) represents the gas price at dispatch time t;

[0249] represents the average daily gas price;

[0250] B3: Electric vehicle equipment agent: controls the charging or discharging status of electric vehicle equipment;

[0251] The action space includes: charging, discharging, and no action. The actions are divided into the following ten sets:

[0252]

[0253] Action Value Represents the ratio of charging or discharging power to rated power, which controls the charging or discharging rate;

[0254] Reward design: The difference between the electricity price in each state and the average of the electricity price are used as the basis for calculating the reward amount:

[0255]

[0256] Where:

[0257] β e (t) represents the electricity price at dispatch time t;

[0258] represents the average daily electricity price;

[0259] B4: CHP equipment agent: Control the gas consumption ratio π coefficient to control the contribution of CHP in power production. Its action space is designed as:

[0260] a CHP =0,0.01,0.02,...,1

[0261] The reward function of the CHP agent is defined as:

[0262] R CHP =-a CHP *G in (t)*(β e (t)-β g (t))

[0263] B5: On / off adjustable load agent: Control the on or off state of the on / off load:

[0264]

[0265] The agent determines the working time of the machine by choosing actions:

[0266]

[0267] B6: Flexible adjustable load agent: Since the energy consumption of the flexible adjustable load can be controlled in each time step, considering the working conditions of the flexible adjustable load, the action of the flexible adjustable load agent is the energy consumed in each time step:

[0268]

[0269] Each adjustable device uses an agent agent and constructs its own Q function for each agent, which is updated during training by the following formula:

[0270]

[0271] Where: and They represent the state and action of agent numbered i at time step t, Indicates that the agent numbered i is in state S t and perform action A t The expected return. t+1 represents the reward obtained at event step t+1; γ is the discount factor, which indicates the importance of future rewards; θ is the learning rate, which affects the speed at which new information covers old information; max At+1 Q(S t+1 ,A t+1 ) represents the maximum Q value of all possible actions in the state of the next time step t+1.

[0272] like Figure 2As shown, a smart building energy management system includes: a device unit 201, a modeling unit 202 and a scheduling unit 203;

[0273] The equipment unit 201 is used to determine the smart building energy management equipment and information;

[0274] The modeling unit 202 is used to construct a mathematical model of energy management of the building system based on the management equipment and information; the mathematical model of energy management is modeled by using a Markov decision process; a training scenario is constructed for the mathematical model of energy management, and an intelligent agent for energy management is obtained by training;

[0275] The scheduling unit 203 is used to input the real-time data of management equipment and information into the intelligent agent to obtain the energy scheduling plan for each device.

[0276] like Figure 3 As shown, the present disclosure provides an electronic device, including a processor 301, a communication interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communication interface 302 and the memory 303 communicate with each other through the communication bus 304;

[0277] Memory 303, storing computer programs;

[0278] The processor 301 is configured to implement the above method when executing the computer program stored in the memory 303 .

[0279] The present disclosure provides a computer-readable storage medium storing a computer program, wherein the computer program implements the above method when executed by a processor.

[0280] The computer-readable storage medium may be included in the device / apparatus described in the above embodiment; or it may exist independently without being assembled into the device / apparatus. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0281] According to an embodiment of the present disclosure, a computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, apparatus, or device.

[0282] Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure.

Claims

1. A smart building energy management method, characterized in that: The method comprises: Identify smart building energy management equipment and information; Based on management equipment and information, a mathematical model of building system energy management is constructed; the energy management mathematical model is modeled using Markov decision process; a training scenario is constructed for the energy management mathematical model, and an intelligent agent for energy management is obtained through training; The real-time data of management equipment and information is input into the intelligent agent to obtain the energy scheduling plan for each device.

2. The smart building energy management method according to claim 1, characterized in that: Management equipment, including: electrical load, thermal load, cooling load, energy storage system and distributed energy system; information, including: external environment information; Electrical load, including electrical equipment in the building; Thermal load, including heating and hot water supply systems in buildings; Cooling load, including air conditioning and refrigeration equipment in the building; Energy storage systems, including in-building electrical energy storage, thermal energy storage, and electric vehicles; Distributed energy systems, including photovoltaic power generation systems and combined heat and power systems; External environmental information, including real-time electricity prices, gas prices, weather forecasts, and external temperature and humidity information.

3. The smart building energy management method according to claim 1, characterized in that: The objective function of the energy management mathematical model includes: The objective function of the energy management mathematical model is obtained by weighting the following indicators: The economic goal is to minimize the energy cost of building operation, the energy efficiency goal is to maximize the efficiency of energy utilization, the environmental goal is to minimize the environmental impact of carbon emissions, and the comfort goal is to consider the comfort of users.

4. The smart building energy management method according to claim 3, characterized in that: The economic goal of minimizing the energy cost of building operation is as follows: Where: T is the scheduling period; t is the scheduling time, which is a discrete value; β e (t) and β g (t) are the electricity purchase price and gas purchase price corresponding to the dispatch time t; E in (t) and G in (t) are the electricity purchase capacity and gas purchase capacity corresponding to the dispatch time t respectively; The energy efficiency target of maximizing energy utilization efficiency is as follows: Where: D e (t) is the power load corresponding to the dispatching time t; D h (t) is the heat load corresponding to the dispatching time t; ω g is the calorific value of natural gas; The environmental goal of minimizing the impact of carbon emissions is as follows: In the formula, and are the carbon emission coefficients of electricity and natural gas in the external system, respectively; Considering the comfort target of user comfort, the formula is as follows: In the formula, S e , S h are user discomfort caused by unit electrical load and thermal load deviation respectively; ΔD e (t),ΔD h (t) are the offsets of the electrical load and thermal load compared to the optimal comfort load at time t; The objective function of the energy management mathematical model is as follows: F=ω1*F1-ω2*F2+ω3*F3+ω4*F4 ω1+ω2+ω3+ω4=1 0≤ω1,ω2,ω3,ω4≤1 Where ω1, ω2, ω3, and ω4 are the weight coefficients of each sub-goal respectively.

5. The smart building energy management method according to claim 1, characterized in that: The constraints of the energy management mathematical model include: Energy balance constraints, CHP equipment operation constraints, electric energy storage equipment operation constraints, thermal energy storage equipment operation constraints, electric vehicle constraints and electric heat balance constraints.

6. A smart building energy management method according to claim 5, characterized in that: Energy balance constraint, the formula is as follows: Where: D e (t) is the power load corresponding to the dispatching time t; μ e To consider the power supply efficiency of line loss; and is the PEV battery charging and discharging amount at scheduling time t, and their efficiencies are and E pv (t) is the photovoltaic power generation power at the scheduling time t, μ pv Power supply efficiency for photovoltaic systems; D is the power generation of CHP system at the dispatching time t; h (t) is the heat load corresponding to the scheduling time t; is the heating load of the CHP system corresponding to the scheduling time t; π(t) represents the proportion of gas consumed by CHP equipment, and the remaining gas is supplied to other gas equipment, μ B Indicates the heating efficiency of other gas equipment; The operating constraints of CHP equipment are as follows: In the formula, and represent the CHP power generation and heating efficiency respectively; Operating capacity limitations of CHP: The operating constraints of electric energy storage equipment are as follows: Where: Indicates the minimum discharge power of the electric energy storage device; Indicates the maximum discharge power of the electric energy storage device; Indicates the discharge power of the electric energy storage device during period t; Indicates the minimum charging power of the electric energy storage device; Indicates the maximum charging power of the electric energy storage device; represents the charging power of the electric energy storage device during period t; A binary variable representing the discharge state of the electrical energy storage device; A binary variable representing the state of charge of the electrical energy storage device; Indicates the charging efficiency of the electric energy storage device; Indicates the discharge efficiency of the electrical energy storage device; E ES (t) represents the energy value of the electric energy storage device during time period t; E ES (t-1) represents the energy value of the electric energy storage device during the period t-1; Indicates the minimum energy value of the electric energy storage device within the dispatch period; Indicates the maximum energy value of the electric energy storage device within the dispatch period; The operating constraints of thermal energy storage equipment are as follows: Where: Indicates the minimum heat release power of the thermal energy storage device; Indicates the maximum heat release power of the thermal energy storage device; It represents the heat release power of the thermal energy storage device during period t; Indicates the minimum heat storage power of the thermal energy storage device; Indicates the maximum heat storage power of the thermal energy storage device; It represents the thermal storage power of the thermal energy storage device during period t; A binary variable representing the heat storage state of the thermal energy storage device; A binary variable representing the heat release state of the thermal energy storage device; Indicates the heat storage efficiency of thermal energy storage equipment; Indicates the heat release efficiency of thermal energy storage equipment; E HS (t) represents the energy value of the thermal energy storage device during time period t; E HS (t-1) represents the energy value of the thermal energy storage device during the period t-1; Indicates the minimum energy value of the thermal energy storage device within the scheduling period; Indicates the maximum energy value of the thermal energy storage device within the scheduling period; Electric vehicle constraints, the formula is as follows: Where: t dep Indicates the PEV departure time; t arr Indicates the PEV arrival time; E PEV (t arr ) indicates the charge state of the PEV when it arrives at the home; and Respectively represent the maximum charging and discharging power of PEV; E PEV (t) represents the PEV battery capacity at the dispatch time t; Cap PEV Indicates the PEV battery capacity; represents the desired state of charge of the PEV battery determined by the user at the time of departure; It is the minimum state of charge of the PEV battery; The electric and thermal balance constraint is as follows: Where: and They represent the non-dispatchable electric load and non-dispatchable thermal load within the commissioning time t respectively; and They represent the dispatchable electric load and dispatchable thermal load within the commissioning time t respectively; Where: and Represents an adjustable electrical load and thermal load of the breaking type, which is controlled by switching ON / OFF and remains ON for a specified time after the switch is turned ON; and Respectively represent the binary states of the on-off adjustable electric load and thermal load, where a value of 1 indicates that the device is on, and a value of 0 indicates that the device is off; b i and f i Indicates the start and end time of the operation of the breaking type adjustable load; and Respectively represent the operating time required for the breaking adjustable electrical load and thermal load; and They represent flexible adjustable electrical load and thermal load respectively, and their power consumption can be adjusted up or down. and Indicates the upper and lower limits of the flexible adjustable electric load; and Indicates the adjustable upper and lower limits of the flexible adjustable thermal electric load.

7. The smart building energy management method according to claim 1, characterized in that: The energy management mathematical model is modeled using the Markov decision process, including: Establish electric energy storage equipment agents, thermal energy storage equipment agents, electric vehicle equipment agents, CHP equipment agents, disconnection-type adjustable load agents, and flexible adjustable load agents, and set action space, state space, and reward space for each agent; The energy storage device agent is used to control the charging or discharging state of the energy storage device. Its action space includes: charging, discharging, and no action. The actions are divided into the following ten sets: In the formula, the action value Represents the ratio of charging or discharging power to rated power, which controls the charging or discharging rate; The reward design includes the difference between the electricity price in each state and the average of the electricity price as the basis for calculating the reward amount. The formula is as follows: In the formula, β e (t) represents the electricity price at dispatch time t; represents the average daily electricity price; Thermal energy storage device agent, used to control the heating or heat release state of thermal energy storage devices; its action space includes: heating, heat release, no action, and the actions are divided into the following ten sets: In the formula, the action value Represents the ratio of heating or heat release power to rated power, which controls the heating or heat release rate; The reward design includes: the average value of the gas price difference and the electric power price in each state is used as the basis for calculating the reward amount. The formula is as follows: In the formula, β g (t) represents the gas price at dispatch time t; represents the average daily gas price; The electric vehicle device agent is used to control the charging or discharging state of the electric vehicle device; its action space includes: charging, discharging, and no action; the actions are divided into the following ten sets: In the formula, the action value Represents the ratio of charging or discharging power to rated power, which controls the charging or discharging rate; The reward design includes: the difference between the electricity price in each state and the average of the electricity price are used as the basis for calculating the reward amount. The formula is as follows: In the formula, β e (t) represents the electricity price at dispatch time t; represents the average daily electricity price; The CHP equipment agent is used to control the gas consumption ratio π coefficient to control the contribution of CHP in power production. Its action space formula is as follows: a CHP ＝0,0.01,0.02,...,1 The reward function formula of the CHP agent is as follows: R CHP =-a CHP *G in (t)*(β e (t)-β g (t)) The breaking type adjustable load agent is used to control the on or off state of the breaking type load. The formula is as follows: The agent determines the working time of the machine by choosing actions, and the formula is as follows: Flexible adjustable load agent is used for energy consumption management of adjustable load. The action of flexible adjustable load agent is the energy consumed in each time step. The formula is as follows: Each adjustable device uses an agent agent and constructs its own Q function for each agent, which is updated during training by the following formula: In the formula, and They represent the state and action of agent numbered i at time step t, Indicates that the agent numbered i is in state S t and perform action A t The expected return when t+1 represents the reward obtained at event step t+1; γ is the discount factor, which indicates the importance of future rewards; θ is the learning rate, which affects the speed at which new information covers old information; It represents the maximum value of Q value among all possible actions in the state of the next time step t+1; Input real-time data into the intelligent agent to obtain energy scheduling for each device, including: Real-time monitoring of the status of each device and subsystem through a sensor network, including changes in demand for electricity, heat, and cooling loads, the charging and discharging status of energy storage devices, photovoltaic power generation, and external environmental information; Real-time data is input into the intelligent agent, which selects the optimal action based on the current state and performs energy scheduling for each device.

8. A smart building energy management system, characterized in that: include: Equipment unit, modeling unit and scheduling unit; Equipment unit, used to determine smart building energy management equipment and information; A modeling unit is used to construct a mathematical model of energy management of a building system based on management equipment and information; the mathematical model of energy management is modeled using a Markov decision process; Construct training scenarios for the energy management mathematical model and train an intelligent agent for energy management; The scheduling unit is used to input the real-time data of management equipment and information into the intelligent agent to obtain the energy scheduling plan for each device.

9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; a memory storing a computer program; The processor is used to implement a smart building energy management method as described in any one of claims 1 to 7 when executing a computer program stored in a memory.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it implements a smart building energy management method as described in any one of claims 1-7.

Images