Zero-carbon smart energy system optimization scheduling method, device and electronic equipment

Abstract

The present invention discloses a method, device, and electronic device for optimizing and scheduling a zero-carbon smart energy system. The method comprises: obtaining historical data of the zero-carbon smart energy system, randomly selecting N scenarios based on the uncertain variables and their upper and lower limits in the zero-carbon smart energy system; wherein N is a non-zero natural number, and the upper and lower limits of the uncertain variables are determined by the historical data; obtaining an uncertainty set of the N scenarios based on the uncertainty of each of the uncertain variables in the N scenarios during a target scheduling period; establishing a scheduling model for the zero-carbon smart energy system based on the uncertainty set of the N scenarios; solving the scheduling model based on a scenario tree method to obtain an operating strategy for the zero-carbon smart energy system during the target scheduling period; and controlling the equipment in the zero-carbon smart energy system to operate according to the operating strategy. The above method can effectively balance computational complexity and feasibility in the random optimization of the zero-carbon smart energy system, thereby improving system operating efficiency.

Description

Technical Field

[0001] The present invention relates to the field of energy system technology, and in particular to a zero-carbon smart energy system optimization scheduling method, device and electronic equipment. Background Art

[0002] The zero-carbon smart energy system is a system that achieves complete energy consumption and utilization through flexible conversion between multiple renewable energy media such as hydrogen, electricity, cold, and heat, as well as interactive coordination between hydrogen storage and traditional temperature storage equipment. It can achieve zero carbon emissions based on renewable new energy.

[0003] However, since the multi-energy demands of zero-carbon smart energy systems often have dynamic time-varying and high uncertainty, the multi-stage randomness on both the supply and demand sides must be considered when optimizing the operation of zero-carbon smart energy systems. However, the existing scenario tree random optimization method often needs to generate a large number of scenarios to describe the uncertainty of the zero-carbon smart energy system to ensure that the obtained solution is feasible and accurate, which results in the computational cost of solving the feasible solution being very high.

[0004] Therefore, how to balance computational complexity and feasibility in the stochastic optimization of zero-carbon smart energy systems is an urgent problem that needs to be solved. Summary of the Invention

[0005] The purpose of the present invention is to at least solve the technical problems existing in the prior art and to provide a zero-carbon smart energy system optimization scheduling method, device and electronic equipment.

[0006] In a first aspect, the present invention provides a zero-carbon smart energy system optimization scheduling method, comprising:

[0007] Acquiring historical data of the zero-carbon smart energy system, wherein the historical data includes historical energy demand data and historical meteorological data;

[0008] Randomly select N scenarios based on the uncertain variables and their upper and lower limits in the zero-carbon smart energy system; where N is a non-zero natural number, and the upper and lower limits of the uncertain variables are determined by the historical meteorological data and the historical energy demand data;

[0009] Based on the uncertainty of each of the uncertain variables in the N scenarios during the target scheduling period, an uncertainty set of the N scenarios is obtained;

[0010] Establishing a scheduling model for the zero-carbon smart energy system based on the uncertainty set of N scenarios;

[0011] Solving the scheduling model based on a scenario tree method to obtain an operation strategy of the zero-carbon smart energy system during a target scheduling period;

[0012] Control the equipment in the zero-carbon smart energy system to operate according to the operation strategy.

[0013] As a further improvement, the uncertain variables and their upper and lower limits in the zero-carbon smart energy system are represented by a convex polyhedron set Ω, specifically:

[0014] Ω=Ω1×Ω2×...Ω t ...×Ω T

[0015]

[0016] Where × is the Cartesian product, K is the number of uncertain variables, T is the total target scheduling period, and Ω t represents the uncertainty set at time period t; and ξ k,t represent the upper and lower limits of the uncertain variables respectively.

[0017] As a further improvement, based on the uncertainty of each of the uncertain variables in the N scenarios during the target scheduling period, an uncertainty set of N scenarios is obtained, specifically:

[0018] The uncertainty sets of N scenarios are determined by random linear combinations of vertices in the convex polyhedron set Ω. The uncertainty sets of each scenario are expressed as:

[0019] Ωs=(ξ s,1,t ,ξ s,2,t ,...,ξ s,k,t ,...,ξ s,K,t ) T

[0020]

[0021] Among them, s represents the scene, ξ s,k,t represents the uncertainty variable in the t-th period in scene s, α represents the selection factor, and the selection factor is a random variable that conforms to the truncated normal distribution.

[0022] As a further improvement, the method further includes:

[0023] Acquire real-time data of the zero-carbon smart energy system, including perception data and behavior data of system resident personnel, indoor and outdoor environment data, and equipment operation data.

[0024] The real-time data is standardized to obtain a standardized data set.

[0025] As a further improvement, after controlling the operating status of the equipment in the zero-carbon smart energy system according to the optimal operation scheduling strategy set, the method further includes:

[0026] Establishing a Markov decision process based on the current real-time data, equipment operation strategy, and energy consumption data of the zero-carbon smart energy system; wherein the real-time data is used to construct a state space, the equipment operation strategy is used to construct a behavior space, and the energy consumption data of the equipment is used to determine the reward;

[0027] The Markov decision process is solved using a reinforcement learning method to obtain a revised operation strategy for the zero-carbon smart energy system, and the devices in the zero-carbon smart energy system are controlled in real time to operate according to the revised operation strategy.

[0028] As a further improvement, the perception data of the current system resident personnel of the zero-carbon smart energy system is set as a prerequisite for the equipment operation strategy to reduce the behavior space.

[0029] In a second aspect, the present invention further provides a zero-carbon smart energy system optimization and scheduling device, comprising:

[0030] A data acquisition unit, configured to acquire historical data of the zero-carbon smart energy system, wherein the historical data includes historical energy demand data and historical meteorological data;

[0031] A scenario selection unit, configured to randomly select N scenarios based on the uncertain variables and their upper and lower limits in the zero-carbon smart energy system; wherein N is a non-zero natural number, and the upper and lower limits of the uncertain variables are determined by the historical meteorological data and the historical energy demand data;

[0032] a strategy determination unit, configured to obtain an uncertainty set for the N scenarios based on the uncertainty of each of the uncertain variables in the N scenarios during a target scheduling period; establish a scheduling model for the zero-carbon smart energy system based on the uncertainty set for the N scenarios; and solve the scheduling model based on a scenario tree method to obtain an operation strategy for the zero-carbon smart energy system during the target scheduling period;

[0033] An operation control unit is used to control the equipment in the zero-carbon smart energy system to operate according to the operation strategy.

[0034] As a further improvement, the data acquisition unit further includes:

[0035] Acquire real-time data of the zero-carbon smart energy system, including perception data and behavior data of system resident personnel, indoor and outdoor environment data, and equipment operation data.

[0036] The real-time data is standardized to obtain a standardized data set.

[0037] As a further improvement, the device further comprises:

[0038] a real-time correction unit, configured to establish a Markov decision process based on the current real-time data, device operation strategy, and energy consumption data of the zero-carbon smart energy system; wherein the real-time data is used to construct a state space, the device operation strategy is used to construct a behavior space, and the energy consumption data of the device is used to determine a reward;

[0039] The Markov decision process is solved using a reinforcement learning method to obtain a revised operation strategy for the zero-carbon smart energy system, and the devices in the zero-carbon smart energy system are controlled in real time to operate according to the revised operation strategy.

[0040] In a third aspect, the present invention provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the method described in the first aspect through the computer program.

[0041] Compared with the existing technology, the zero-carbon smart energy system optimization scheduling method provided by the present invention has at least the following beneficial effects:

[0042] This method determines the range of uncertain variables in the zero-carbon smart energy system based on historical weather and energy demand conditions, which can improve the effectiveness of the selected scenarios under specific scenario scales. By further solving the optimal strategy for the operation of the zero-carbon smart energy system under the selected scenario, it takes into account the feasibility and computational complexity of the system operation strategy. Compared with the traditional scenario tree method, the performance can be effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the implementation. Obviously, the drawings described below are only some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 This is a flowchart of a method for optimizing and scheduling a zero-carbon smart energy system according to an embodiment of the present invention;

[0045] Figure 2 It is a structural diagram of a zero-carbon smart energy system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] This section will describe in detail the specific embodiments of the present invention. The preferred embodiments of the present invention are shown in the accompanying drawings. The purpose of the accompanying drawings is to supplement the description of the text part of the specification with graphics, so that people can intuitively and vividly understand each technical feature and the overall technical solution of the present invention, but it should not be understood as a limitation on the scope of protection of the present invention.

[0047] like Figure 1As shown, in the first aspect, an embodiment of the present invention provides a zero-carbon smart energy system optimization scheduling method, comprising the following steps.

[0048] S1: Acquire historical data of the zero-carbon smart energy system, where the historical data includes historical energy demand data and historical meteorological data.

[0049] S2: Randomly select N scenarios based on the uncertain variables and their upper and lower limits in the zero-carbon smart energy system; wherein N is a non-zero natural number, and the upper and lower limits of the uncertain variables are determined by the historical meteorological data and the historical energy demand data.

[0050] It should be noted that users can select uncertain variables in the zero-carbon smart energy system according to actual application needs, and further determine the range of each uncertain variable based on the historical demand data and meteorological data of the zero-carbon smart energy system; in addition, the number of scenarios N is a user-preset value.

[0051] S3: Based on the uncertainty of each of the uncertain variables in the N scenarios during the target scheduling period, an uncertainty set of the N scenarios is obtained; and a scheduling model of the zero-carbon smart energy system is established according to the uncertainty set of the N scenarios.

[0052] Specifically, within the N selected scenarios, the uncertainty of the uncertain variables in each scenario during the target scheduling period can be described by a linear combination of these variables within upper and lower bounds. In the scheduling model established based on the uncertainty set of N scenarios, the objective can be to minimize the average cost of the selected scenarios, which can be investment costs, operating costs, etc.

[0053] S4: Solve the scheduling model based on a scenario tree method to obtain an operation strategy of the zero-carbon smart energy system during a target scheduling period.

[0054] S5: Control the devices in the zero-carbon smart energy system to operate according to the operation strategy.

[0055] The above-mentioned embodiment of the present invention determines the upper and lower limits of uncertain variables in the zero-carbon smart energy system by combining historical weather and energy demand conditions, thereby improving the effectiveness of the selected scenario, and further solving the optimal strategy for the operation of the zero-carbon smart energy system under the selected scenario, thereby achieving the effect of taking into account the feasibility of the system operation strategy and the computational complexity.

[0056] The implementation process of the technical solution of the present invention will be described below through several specific embodiments.

[0057] Please refer to Figure 2In an embodiment of the present invention, the equipment in the zero-carbon smart energy system includes a hydrogen fuel cell cogeneration unit, a geothermal energy utilization unit, a water electrolysis hydrogen production unit, a photovoltaic power generation device, an electric boiler, an electric refrigeration unit, a heat exchanger, a hot water tank, a cold water tank, a hydrogen storage tank, a pressure regulating valve group, a shut-off valve, an AC / DC device, and a DC / DC device, etc.

[0058] Among them, the hydrogen fuel cell cogeneration unit includes a fuel cell. The fuel cell equipment operates throughout the year and is used for heating in winter. The waste heat in summer and transition season is used for heat storage to maintain the thermal balance of the ground source heat pump throughout the year; the geothermal energy utilization unit includes a ground source heat pump and a shallow geothermal well. The ground source heat pump is used to absorb heat from the soil for heating in winter and store cold energy for use in summer; in summer, it is used to transfer heat from the building to the earth to cool the building, and at the same time store heat in the earth for use in winter.

[0059] In one embodiment, real-time data of the zero-carbon smart energy system needs to be obtained, and the real-time data includes perception data and behavior data of system residents, indoor and outdoor environment data, and equipment operation data.

[0060] Specifically, equipment operation data can be obtained through the sensor network, which includes water temperature sensors, hydrogen pressure sensors, current sensors, voltage sensors, water metering devices, electricity metering devices and gas metering devices; indoor and outdoor environmental data are obtained through temperature and humidity sensors; and the perception data of system resident personnel can be obtained through behavior perception devices, demand collection devices and comfort collection devices.

[0061] Considering that the internal data from the device and the external data from the environment are neither completely independent nor interrelated, the various data structures are complex and diverse. In addition to traditional structured data, they also contain a large amount of semi-structured and unstructured data, and the sampling frequency and life cycle of these data are also different. In this regard, this embodiment abstracts the physical data and normalizes the information data to standardize the real-time data, ensuring that a standardized and unified data set is obtained, providing effective data support for system optimization and scheduling.

[0062] In one embodiment, the uncertain variables and their upper and lower limits in the zero-carbon smart energy system can be represented by a convex polyhedron set Ω, specifically expressed as:

[0063] Ω=Ω1×Ω2×...Ω t ...×Ω T

[0064]

[0065] Where × is the Cartesian product, K is the number of uncertain variables, T is the total target scheduling period, and Ω trepresents the uncertainty set at time period t, specifically the exponential operation of each uncertain variable; and ξ k,t represent the upper and lower limits of the uncertain variables respectively.

[0066] Based on the characteristics of convex polyhedron sets, linear combinations of their vertices can be used to describe any uncertainty situation, but due to the existence of 2 KT Vertices, the computational complexity is too high, and the uncertain variables are continuous in the convex polyhedron set. Therefore, in order to balance the computational complexity and feasibility, the embodiment of the present invention can select a limited number of scenes based on the random linear combination of the vertices in the convex polyhedron set to reduce the computational complexity. The number of scenes N can be set according to actual application requirements.

[0067] After obtaining the uncertainty set of the target scenario, a scheduling model of the zero-carbon smart energy system is constructed, and the scheduling model is solved based on the scenario tree method.

[0068] Specifically, the uncertainty set of each scenario is expressed as:

[0069] Ωs=(ξ s,1,t ,ξ s,2,t ,...,ξ s,k,t ,...,ξ s,K,t ) T

[0070]

[0071] Among them, s represents the scene, ξ s,k,t represents the uncertainty variable in the t-th period in scene s, α represents the selection factor, and the selection factor is a random variable that conforms to the truncated normal distribution, and its mean can be 0.5.

[0072] In this embodiment, the uncertain variables can be set as solar radiation r, electricity demand e load , cold demand q demand and heat demand demand , that is, K = 4; then the uncertainty set of scene s is expressed as:

[0073]

[0074] In one embodiment, the zero-carbon smart energy system scheduling model constructed based on the uncertainty set of the selected scenarios is a mixed integer linear programming model, whose constraints include system equipment constraints, energy balance constraints, etc. Considering the uncertainties on both the supply and demand sides of the zero-carbon smart energy system, the goal can be set to minimize the average cost of all selected scenarios within the target scheduling period, and the stochastic optimization problem can be solved using solvers such as CPLEX to obtain the operation strategy of the zero-carbon smart energy system in the target scheduling period.

[0075] Specifically, the average cost of the selected scenario may be system investment cost, operation cost or carbon emission, etc., which is not limited in the present invention.

[0076] In one embodiment, after obtaining the operation strategy of the zero-carbon smart energy system during the target scheduling period, the equipment control module in the operation control unit can be used to control the equipment in the zero-carbon smart energy system to operate according to the operation strategy, and at the same time, the grid-connected control module in the operation control unit can be used to control the zero-carbon smart energy system to be grid-connected or islanded.

[0077] Specifically, the grid-connected control module is used to ensure that the hydrogen fuel cell generator set and photovoltaic power generation device are safely and stably connected to the grid at a certain power and voltage, and specifically includes: a generator speed control system, a generator excitation system, a reverse power limiting device and a mains fault protection device.

[0078] Among them, the generator speed control system is used to adjust the frequency and active power of the fuel cell generator set; the generator excitation system is used to adjust the voltage and reactive power of the fuel cell generator set; the reverse power limiting device is set on the line connecting the zero-carbon smart energy system and the mains power, and is used to issue an early warning and reduce the generator output when the grid power used by the zero-carbon smart energy system is less than the set value, and to issue a tripping command and grid decoupling operation when the grid power used by the zero-carbon smart energy system is less than the threshold value; the mains fault protection device is used to detect sudden changes in the mains phase angle and the mains frequency change rate.

[0079] In one embodiment, the operation of a zero-carbon smart energy system is directly related to human behavioral needs. Therefore, fully leveraging human-machine hybrid information is crucial for improving system performance. However, human behavioral needs are difficult to accurately describe using mechanistic models, and the temporal and spatial coupling of load demand with multiple energy media, as well as the complex system structure and multi-energy conversion relationships, make it difficult for existing technologies to achieve efficient system optimization decisions using mechanistic-based methods.

[0080] In this regard, an embodiment of the present invention provides a technical solution for real-time correction of the operation strategy of a zero-carbon smart energy system based on dynamic data such as equipment operation status, personnel perception information and environmental information.

[0081] In this embodiment, a Markov decision process can be established based on the current real-time data, equipment operation strategy and energy consumption data of the zero-carbon smart energy system, and then the reinforcement learning method is used to solve the Markov decision process to obtain a corrected operation strategy of the zero-carbon smart energy system, and the equipment in the zero-carbon smart energy system can be controlled in real time to operate according to the corrected operation strategy.

[0082] Specifically, the perception data and behavior data of the current resident personnel in the zero-carbon smart energy system, indoor and outdoor environment data, and equipment operation data can be used as the states in the Markov decision process to construct the system's state space S; then the behavior space A is constructed based on the decisions that can be executed by the equipment in the system; and considering the coupling relationship between personnel needs and the energy consumption of system equipment, the real-time energy consumption of the system equipment is used as the reward R of the decision-making behavior at the corresponding moment, and the value function Q is determined by accumulating the reward values ​​of multiple time periods, and the Q value is updated by time difference.

[0083] It should be noted that, in this embodiment, the perception data of the current system resident personnel can be set as a prerequisite for the device operation strategy to be followed, so as to reduce the behavior space A and thus reduce the amount of calculation.

[0084] Furthermore, the training value network is used to estimate the Q value, and a target value network is separately set to generate a target Q value to handle the TD (Temporal-Difference) error in the temporal difference algorithm.

[0085] Specifically, this embodiment can update the Q value through the following time difference method:

[0086] Q i (s,a)←Q i (s,a)+α[r+γQ i-1 (s',a')-Q i (s,a)]

[0087] 0≤γ≤1

[0088] Among them, i represents the number of iterative updates, α represents the learning rate, r represents the immediate return, and γ represents the attenuation coefficient of the return value.

[0089] Furthermore, an experience replay pool is designed to store the transfer samples (s t ,a t ,r t ,s t-1 ) as data support, where t represents the tth moment; using the large amount of data in the experience replay pool, combined with reinforcement learning methods such as Qlearning and DeepQ-Network, the intelligence level is continuously improved through offline training to obtain the revised operation strategy of the zero-carbon smart energy system.

[0090] The above-described embodiment of the present invention analyzes the stochastic characteristics of both the supply and demand sides of a zero-carbon smart energy system to construct a convex polyhedron set that considers the coupling of multiple energy supply and demand, as well as the elasticity of personnel demand. To improve the effectiveness of scenario selection, a selection factor that considers historical meteorological and demand data is used to randomly generate linear combinations of the vertices in the convex polyhedron set. This selects a finite number of scenarios, and then, through a stochastic optimization method that minimizes the average cost of all selected scenarios, derives the operating strategies for each device in the zero-carbon smart energy system. This method effectively balances strategy feasibility and computational complexity, effectively improving the economic efficiency of the system. It also avoids the drawbacks of traditional scenario tree optimization methods, such as model error and high dimensionality, significantly reducing the scale of the stochastic optimization problem and improving the efficiency of the algorithm.

[0091] Furthermore, the embodiment of the present invention also considers the impact of personnel on system operation and provides a technical solution for correcting equipment operation strategies based on the dynamic update of real-time system information, making system operation optimization safer and more efficient.

[0092] In a second aspect, another embodiment of the present invention further provides a zero-carbon smart energy system optimization and scheduling device, including a data acquisition unit, a scenario selection unit, a strategy determination unit and an operation control unit.

[0093] The data acquisition unit is used to acquire historical data of the zero-carbon smart energy system, where the historical data includes historical energy demand data and historical meteorological data.

[0094] Specifically, the data acquisition unit is also used to obtain real-time data of the zero-carbon smart energy system, which includes perception data and behavior data of system residents, indoor and outdoor environment data, and equipment operation data, and standardizes the real-time data to obtain a standardized data set.

[0095] The scenario selection unit is used to randomly select N scenarios based on the uncertain variables and their upper and lower limits in the zero-carbon smart energy system; wherein N is a non-zero natural number, and the upper and lower limits of the uncertain variables are determined by the historical meteorological data and the historical energy demand data.

[0096] The strategy determination unit is used to obtain the uncertainty sets of N scenarios based on the uncertainty of each of the uncertain variables in the N scenarios during the target scheduling period; establish a scheduling model of the zero-carbon smart energy system according to the uncertainty sets of the N scenarios; and solve the scheduling model based on the scenario tree method to obtain the operation strategy of the zero-carbon smart energy system during the target scheduling period.

[0097] The operation control unit is used to control the equipment in the zero-carbon smart energy system to operate according to the operation strategy.

[0098] In another embodiment, the device also includes a real-time correction unit for establishing a Markov decision process based on the current real-time data, equipment operation strategy and energy consumption data of the zero-carbon smart energy system; wherein the real-time data is used to construct the state space, the equipment operation strategy is used to construct the behavior space, and the energy consumption data of the equipment is used to determine the reward; the Markov decision process is solved by using a reinforcement learning method to obtain a corrected operation strategy of the zero-carbon smart energy system, and the equipment in the zero-carbon smart energy system is controlled in real time to operate according to the corrected operation strategy.

[0099] Since the information interaction, execution process and other contents between the units in the above-mentioned device are based on the same concept as the embodiment of the method of the present invention, the specific contents can be found in the description of the embodiment of the method of the present invention and will not be repeated here.

[0100] In a third aspect, the present invention provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the zero-carbon smart energy system optimization scheduling described in the first aspect through the computer program.

[0101] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing related hardware through a computer program. The program can be stored in a computer-monitorable storage medium, and when executed, the program can include the processes in the above-described method embodiments. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0102] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A zero-carbon smart energy system optimization scheduling method, characterized in that: include: Acquiring historical data of the zero-carbon smart energy system, wherein the historical data includes historical energy demand data and historical meteorological data; Acquiring real-time data of the zero-carbon smart energy system, including perception data and behavior data of system resident personnel, indoor and outdoor environment data, and equipment operation data; Performing standardization processing on the real-time data to obtain a standardized data set; Randomly select N scenarios based on the uncertain variables and their upper and lower limits in the zero-carbon smart energy system; where N is a non-zero natural number, and the upper and lower limits of the uncertain variables are determined by the historical meteorological data and the historical energy demand data; Based on the uncertainty of each of the uncertain variables in the N scenarios during the target scheduling period, an uncertainty set of the N scenarios is obtained; Establishing a scheduling model for the zero-carbon smart energy system based on the uncertainty set of N scenarios; Solving the scheduling model based on a scenario tree method to obtain an operation strategy of the zero-carbon smart energy system during a target scheduling period; Controlling the equipment in the zero-carbon smart energy system to operate according to the operation strategy; Establishing a Markov decision process based on the current real-time data, equipment operation strategy, and energy consumption data of the zero-carbon smart energy system; wherein the real-time data is used to construct a state space, the equipment operation strategy is used to construct a behavior space, and the energy consumption data of the equipment is used to determine the reward; Solving the Markov decision process using a reinforcement learning method to obtain a revised operating strategy for the zero-carbon smart energy system, and controlling the devices in the zero-carbon smart energy system to operate according to the revised operating strategy in real time; The perception data of the current system resident personnel of the zero-carbon smart energy system is set as a prerequisite for the device operation strategy to reduce the behavior space.

2. The zero-carbon smart energy system optimization scheduling method according to claim 1 is characterized in that: The uncertain variables and their upper and lower limits in the zero-carbon smart energy system are represented by the convex polyhedron set Ω, specifically: Ω=Ω1×Ω2×...Ω t ...×Ω T Where × is the Cartesian product, K is the number of uncertain variables, T is the total target scheduling period, and Ω t represents the uncertainty set at time period t; and ξ k,t represent the upper and lower limits of the uncertain variables respectively.

3. The zero-carbon smart energy system optimization scheduling method according to claim 2 is characterized in that: Based on the uncertainty of each of the uncertain variables in the N scenarios during the target scheduling period, an uncertainty set of the N scenarios is obtained, specifically: The uncertainty sets of N scenarios are determined by random linear combinations of vertices in the convex polyhedron set Ω. The uncertainty sets of each scenario are expressed as: Ωs=(ξ s,1,t ,x s,2,t ,...,x s,k,t ,...,x s,K,t ) T Among them, s represents the scene, ξ s,k,t represents the uncertainty variable in the t-th period in scene s, α represents the selection factor, and the selection factor is a random variable that conforms to the truncated normal distribution.

4. A zero-carbon smart energy system optimization and scheduling device, characterized in that: include: a data acquisition unit configured to acquire historical data of the zero-carbon smart energy system, the historical data including historical energy demand data and historical meteorological data; acquire real-time data of the zero-carbon smart energy system, the real-time data including perception data and behavior data of system resident personnel, indoor and outdoor environmental data, and equipment operation data; and perform standardization processing on the real-time data to obtain a standardized data set; A scenario selection unit, configured to randomly select N scenarios based on the uncertain variables and their upper and lower limits in the zero-carbon smart energy system; wherein N is a non-zero natural number, and the upper and lower limits of the uncertain variables are determined by the historical meteorological data and the historical energy demand data; a strategy determination unit, configured to obtain an uncertainty set for the N scenarios based on the uncertainty of each of the uncertain variables in the N scenarios during a target scheduling period; establish a scheduling model for the zero-carbon smart energy system based on the uncertainty set for the N scenarios; and solve the scheduling model based on a scenario tree method to obtain an operation strategy for the zero-carbon smart energy system during the target scheduling period; An operation control unit, configured to control the equipment in the zero-carbon smart energy system to operate according to the operation strategy; A real-time correction unit is used to establish a Markov decision process based on the current real-time data, equipment operation strategy and energy consumption data of the zero-carbon smart energy system; wherein, the real-time data is used to construct the state space, the equipment operation strategy is used to construct the behavior space, and the energy consumption data of the equipment is used to determine the reward; the Markov decision process is solved by using the reinforcement learning method to obtain the corrected operation strategy of the zero-carbon smart energy system, and the equipment in the zero-carbon smart energy system is controlled in real time to operate according to the corrected operation strategy; the perception data of the current system resident personnel of the zero-carbon smart energy system is set as a prerequisite for the equipment operation strategy to reduce the behavior space.

5. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 3 through the computer program.

Images