Distributed Optimal Scheduling Method for Building Clusters Considering Building Characteristics and Electricity Trading

Abstract

The present invention proposes a distributed optimization scheduling method for building groups that takes into account building characteristics and electricity trading. The method includes: fine-grained modeling of various types of buildings, and establishing multi-objective operation optimization models for different buildings with economic and comfort requirements as goals, taking into account building envelope structures, functional characteristics, personnel behavior and distributed energy generation; at the overall coordination and sharing level among different buildings, a new energy trading platform and trading strategy are formulated according to building grade classification and distributed energy generation; the transaction electricity price after the transaction between buildings is fed back to each building for a new round of self-optimization, and through repeated iterations of building self-optimization and trading platform quotations, energy coordination and sharing among various types of buildings is finally completed.

Description

Technical Field

[0001] The present invention belongs to the field of integrated energy technology for power systems, and specifically relates to a distributed optimization scheduling method for building groups that takes building characteristics and electric energy trading into consideration. Background Art

[0002] Unlike the real-time matching of electricity supply and consumption, indoor temperature changes in enclosed buildings exhibit a thermal delay. This characteristic allows buildings to be considered thermal virtual energy storage devices. This thermal storage feature allows for time-shifting of thermal energy demand, significantly contributing to peak load shifting and reducing system operating costs. Due to the volatility of the output of new energy equipment, directly connecting distributed energy resources to the main grid will directly affect the accuracy of their power generation forecast data and the stable operation of the main grid system. By integrating distributed energy resources with buildings, buildings can adjust their operations based on the distributed energy generation situation, significantly minimizing the impact of these fluctuations on the main grid system.

[0003] Taking into account the virtual energy storage characteristics of buildings, further refining the building energy consumption model and constructing a suitable energy trading platform are of great significance for fully tapping the potential of buildings in peak shaving and valley filling and renewable energy consumption in the context of vigorously developing new energy and multi-energy complementary energy Internet. Summary of the Invention

[0004] The purpose of the present invention is to address the fact that current research on electricity trading and sharing between microgrids cannot reflect the operating characteristics of each building within the microgrid, and compared with microgrids with a larger market base and a larger trading volume, most buildings do not have the ability to independently participate in market competition between microgrids due to their small trading volume. Therefore, this method takes buildings as the main body and provides a distributed optimization scheduling method for building groups that takes into account building characteristics and electricity trading. Through this method, the building's own optimization process is realized and a point-to-point energy trading platform and strategy are constructed based on this, thereby promoting the complementarity and interaction of resources between buildings, achieving a local balance between supply and demand power, and improving the system's ability to absorb new energy.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A distributed optimization scheduling method for building clusters considering building characteristics and power trading.

[0007] (1) Classify buildings according to their functions and establish corresponding thermodynamic models;

[0008] (2) Establish a corresponding building operation optimization model based on the functional characteristics and personnel characteristics of the building, with economy and comfort as the goals;

[0009] (3) Construct a corresponding energy trading platform and trading strategy based on the optimization results of each building's own operation, the characteristics of the power market formed between buildings, and the distributed energy generation situation;

[0010] (4) The results of inter-building transactions are returned to the building's own operation optimization model for iteration to achieve optimal optimization.

[0011] The classification of buildings in step (1) is to divide buildings into residential buildings, commercial buildings and special buildings according to their functions. The thermodynamic model is:

[0012]

[0013] Where: H SUN Indicates the heat transferred by solar thermal radiation, I SUN F is the thermal radiation power of the sun, which means the amount of heat received per square meter per unit time when the light is irradiated vertically; win is the total area of ​​the building's exterior windows; SC is the exterior window shading coefficient, the value of which is related to whether the exterior window has a sunshade and the glass material itself; H rand Indicates the heating power of indoor heat sources, mainly referring to the heating power of human body and electrical equipment, N peo Refers to the total number of people in the room at that moment, Q peo Refers to per capita heat dissipation; P equi Refers to the total power value of all indoor equipment, ε e is the heat dissipation ratio of the equipment; H HVAC Indicates the cooling / heating power of the air conditioning system. In this method, summer cooling is taken as an example, so it is expressed as cooling power in this formula; K wall The heat transfer coefficient between the building's exterior wall and the outdoors is the amount of heat transferred per second for every 1 degree difference between indoor and outdoor temperatures during steady-state heat transfer. K win F is the heat transfer coefficient of the building's exterior windows, which has a similar meaning to the heat transfer coefficient of the building's exterior walls; wall and F win are the exterior wall area and exterior window area of ​​the building; T room Indicates the indoor temperature, T out Indicates outdoor temperature, T room.t Indicates the indoor temperature of the current period, T room.t+1 Indicates the indoor temperature of the next period; ρ air is the indoor air density, C air is the specific heat capacity of air, V room is the indoor air volume.

[0014] In step (2), after considering the differences in building thermodynamic models and the different occupant characteristics caused by building classification, each building operation optimization model is constructed with economy and comfort as the goals, and a unified optimization goal and standard is established among different buildings. The specific objective function is as follows:

[0015]

[0016] In the above formula, C Trade.t represents the economic cost of system operation, which means the electricity transaction cost between the building and the distribution network and other buildings; C Ma.t C represents the maintenance cost of each device. This method mainly considers the maintenance cost of the HVAC system and photovoltaic power generation system; Tem.t The penalty cost for affecting the user's temperature comfort; C Net.b and C Net.s They represent the electricity selling and purchasing prices of the distribution network in the current period, P Net.b and P Net.s They represent the amount of electricity purchased and sold by the building at the distribution network during the current time period. There can only be one state of electricity purchase and electricity sale in the same time period; δ HVAC , δ pv Respectively represent the use and maintenance cost per unit power of the HVAC system and photovoltaic power generation system per unit time period; P HVAC 、P pv denote the power of the HVAC system and the photovoltaic power generation system respectively; γ is the temperature penalty factor, which can be regarded as the user’s sensitivity to temperature comfort, T set The indoor optimal temperature is set. The greater the deviation from the set temperature, the greater the temperature penalty cost.

[0017] The transaction strategy described in step (3) is as follows: in the transaction process, each building mainly provides three types of information: transaction quotation, transaction electricity and photovoltaic power generation. The transaction market will compare and judge based on the quotation information of each building. When the highest quotation of the electricity buyer is higher than the lowest quotation of the electricity seller, it means that the transaction conditions are met. If the buildings have the same quotation at this time, they are distinguished according to the building level, and the building with the higher priority level is traded first; the priority level of the buildings is mainly based on the following two factors: according to the function of the building, special buildings with important loads are classified as the highest level, and residential buildings are classified as the lowest level; for both the electricity buyer and the electricity seller, if the proportion of the building's to-be-traded electricity to the total to-be-traded electricity is higher, the building's electricity purchase and sales are more stable, and the priority level is higher. If the proportion is the same, in order to promote the consumption of distributed energy, the building with a larger distributed energy generation capacity is recognized as a higher level; after both parties meet the transaction conditions, the transaction price is the average of the two quotations, and the transaction electricity is the party with less to-be-traded electricity. After the transaction is completed, each party updates its own to-be-traded information and proceeds to the next round of transactions.

[0018] The step (4) returns the result of the inter-building transaction to the building's own operation optimization model for iteration to achieve the best optimization. The specific process is as follows:

[0019] After the transaction is completed, each building will receive the final electricity transaction cost and operating cost of its own system.

[0020] C Trade.t =C Net.b P Net.b -C Net.s P Net.s

[0021] Where C Net.b and C Net.s They represent the electricity selling and purchasing prices in the current trading market, P Net.b and P Net.s They represent the amount of electricity purchased and sold by the building at the distribution network during the current period. The transaction price of electricity after the transaction is iterated back into the building operation optimization for further optimization and subsequent market transactions. If the number of iterations reaches the set maximum value K or the difference in the total operating cost of the system after two iterations is less than 5%, the iteration is considered complete. At this time, the building has reached the optimal solution considering the building's virtual energy storage characteristics and its own operation optimization after P2P transactions.

[0022] The beneficial effects obtained by the present invention are:

[0023] This method proposes a distributed optimization scheduling model for building clusters that considers building characteristics and electricity trading. It establishes a building operation optimization model with economy and comfort as its goals. Through a new continuous auction trading mechanism that considers market relations and transaction risks, it incentivizes and guides each building to achieve building cluster energy sharing under various uncertain conditions. The new continuous auction trading mechanism proposed in this method can complete market bidding based on market relations and building operation conditions. Each building's quotation is updated in real time based on the overall transaction situation, and has good stability and dynamic characteristics. Through the update of transaction prices, each building is guided to complete iterative optimization, improving the economic efficiency of the building while further exploring the ability of each building to regulate load; while achieving energy sharing between buildings, it also improves the energy sharing capabilities of the building cluster and the distributed energy absorption capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a structural diagram of the building group system.

[0025] Figure 2 It is a structural diagram of the building thermodynamic model.

[0026] Figure 3 It is a temperature change diagram for optimized operation of special buildings.

[0027] Figure 4It is the P2P transaction process.

[0028] Figure 5 This is the trading situation at 9 o'clock in the trading center.

[0029] Figure 6 It is the trading situation at 15:00 in the trading center.

[0030] Figure 7 It is a building iterative optimization flow chart. DETAILED DESCRIPTION

[0031] To more clearly understand the above-mentioned objectives, features, and advantages of the present invention, the following non-limiting detailed description of the technical solution of the present invention is provided in conjunction with the accompanying drawings and specific embodiments. A distributed optimization scheduling method for a building cluster that considers building characteristics and power trading includes the following:

[0032] 1. Classify buildings according to their functions and establish corresponding thermodynamic models.

[0033] Constructing a building cluster energy management framework:

[0034] like Figure 1 As shown in the figure, consider a building cluster consisting of multiple buildings, each interconnected via power and communication networks. This method categorizes buildings into three main categories: commercial, residential, and special buildings. Each building has a distributed energy (DES) site, including photovoltaic (PV) and wind power (WP). In this method, DES primarily considers PV. Each building has a terminal energy management system (EMS) that integrates building operations and DES generation to optimize its own operations. After optimizing its own operations, the building submits information such as transaction power, transaction quotes, and DES generation to a peer-to-peer (P2P) trading market. Ultimately, the building completes power transactions with other buildings in the P2P market. Any remaining surplus or deficit after the transaction is completed is traded with the distribution network.

[0035] Building thermodynamic model establishment:

[0036]

[0037] The following conclusions can be drawn from the investigation and analysis of various types of buildings in cities today. Modern buildings in cities are generally integrated buildings with large glass curtain walls. Any one or more floors in the building share the same temperature state. Therefore, the thermodynamic model adopts the classic model that considers solar radiation heat gain, external window radiation heat dissipation, external wall radiation heat dissipation and various indoor heat sources, such as Figure 2 As shown in the figure, the biggest feature of this thermodynamic model is that it regards the entire building as a whole, which well reflects the heat gain and heat dissipation of the entire system. Different types of buildings have different heat storage characteristics due to the different materials and areas of their exterior walls, interior walls, floor slabs, roofs and skylights. Therefore, the thermodynamic model mainly considers the structural area F of each type of building. wall 、F win and heat transfer coefficient K wall , K win difference.

[0038] 2. Based on the functional characteristics and personnel characteristics of the building, establish a corresponding building operation optimization model with economy and comfort as the goals.

[0039] By leveraging the time-delayed nature of building heat conduction, known as heat storage, and based on the building's heat balance equation, a quantitative mathematical relationship between indoor temperature, building heat load, cooling power, and outside temperature is established from the perspective of energy conservation. This allows for the construction of a virtual energy storage system for heat within the building. This system is then integrated into the building's operational optimization model, incorporating temperature comfort and its penalty function into the optimization objective. This allows for optimal management of the charging and discharging of the building's virtual energy storage, which is reflected in changes in the building's indoor temperature. By incorporating a virtual energy storage system into the building, the building's operating costs can be reduced to a certain extent.

[0040] Taking into account the differences in building thermodynamic models and the different occupant characteristics caused by building classification, each building operation optimization model is constructed with economy and comfort as the goals, and unified optimization goals and standards are established across different buildings. The specific objective function is as follows:

[0041]

[0042] In the above formula, C Trade.t represents the economic cost of system operation, which means the electricity transaction cost between the building and the distribution network and other buildings; C Ma.t C represents the maintenance cost of each device. This method mainly considers the maintenance cost of the HVAC system and photovoltaic power generation system; Tem.t The penalty cost for affecting the user's temperature comfort; C Net.b and C Net.s They represent the electricity selling and purchasing prices of the distribution network in the current period, P Net.b and P Net.s They represent the amount of electricity purchased and sold by the building at the distribution network during the current time period. There can only be one state of electricity purchase and electricity sale in the same time period; δ HVAC , δ pv Respectively represent the use and maintenance cost per unit power of the HVAC system and photovoltaic power generation system per unit time period; PHVAC 、P pv denote the power of the HVAC system and the photovoltaic power generation system respectively; γ is the temperature penalty factor, which can be regarded as the user’s sensitivity to temperature comfort, T set The indoor optimal temperature is set. The greater the deviation from the set temperature, the greater the temperature penalty cost.

[0043] Here we show the results of the operation optimization of a special building. Figure 3 As shown, special buildings have the following main differences in operation optimization compared with the other two types of buildings: special buildings have higher electrical loads and heat loads, more people, and the impact of personnel uncertainty is more serious than the other two types of buildings; special buildings have smaller glass curtain wall areas, and the building's virtual energy storage characteristics are better. At the same time, the heat transferred by solar thermal radiation is less, and the impact of light intensity uncertainty is also smaller; special buildings have higher temperature range requirements during operation, requiring the building to remain within the temperature requirement range within 24 hours. In this method, the optimal temperature for people in special buildings is set to 24 degrees, and the building is allowed to fluctuate by 2.5 degrees during operation. Figure 3 It can be seen that the operation of special buildings meets the above temperature conditions and maintains a fluctuating state. The factors affecting temperature fluctuations are mainly the electricity price of the distribution network and the temperature penalty factor. The characteristics of buildings as virtual energy storage systems are quite different from those of traditional power storage systems. Compared with the high charge and discharge frequency of power storage systems, the virtual energy storage system of buildings has a lower charge and discharge frequency because of its faster energy dissipation and greater influence from external factors such as light and outdoor temperature. Figure 3 The figure shows a total of seven changes within 24 hours. The essence of optimizing building virtual energy storage is to utilize electricity prices at different times, preemptively cooling when prices are low and reducing cooling power when prices are high, thereby saving operating costs.

[0044] 3. Build a corresponding energy trading platform and trading strategy based on the optimization results of each building’s own operation, the characteristics of the power market composed of buildings, and the distributed energy power generation situation.

[0045] P2P trading platform uses a distributed trading platform such as Figure 1 As shown, each building is an independent entity that provides relevant quotation information to the P2P transaction center and continuously modifies its transaction quotation based on market information to finally reach a transaction. The process of P2P transaction is as follows Figure 4 shown.

[0046] In this transaction process, each building primarily provides three types of information: transaction bids, transaction power, and photovoltaic power generation. The market compares and determines the bids from each building. When the highest bid from the buyer exceeds the lowest bid from the seller, the transaction is concluded. If two buildings offer the same bid, they are prioritized based on building tiers, with the building with the higher priority receiving the transaction. Building priority is determined based on two factors: Buildings with critical loads are prioritized based on their functional characteristics, with residential buildings receiving the lowest. For both buyers and sellers, the higher the proportion of a building's pending transaction power to the total pending transaction power, the more stable its electricity purchases and sales, and the higher its priority. If the proportions are the same, the building with the greater distributed energy generation capacity is given a higher tier to promote the integration of distributed energy resources. Once both parties meet the transaction conditions, the transaction price is the average of their bids, with the transaction power amount being the building with the lower pending transaction power. After the transaction is completed, each party updates its pending transaction information and proceeds to the next round of transactions.

[0047] When most of the building transactions are completed, there are still buildings that have not completed their target transaction volume or the transaction rounds have reached the maximum number and the trading platform is closed. In order to complete the transaction targets, the buildings that have not completed the transaction targets will settle the remaining electricity to be traded with the distribution network.

[0048] Figure 5 and Figure 6 For specific transaction situations, Figure 5 This is the trading situation at 9 o'clock in the trading center. Figure 6 The transaction status at 15:00 in the trading center is shown in the figure. The black line represents the electricity seller and the blue line represents the electricity buyer. When the highest electricity price of the electricity buyer is higher than the lowest electricity price of the electricity seller, the two lines intersect, indicating that the transaction is successful. Figure 5 and Figure 6 The difference in the transaction quotation curves between the purchaser and the seller can be analyzed to show that at 9 o'clock in the trading center, it is a buyer's market. At this time, the total amount of electricity sold is greater than the amount of electricity purchased. Therefore, after multiple rounds of transaction quotations, the purchaser can purchase electricity at a lower price, but for the seller, the final electricity selling price is also better than the clearing price of the same distribution network. At 15 o'clock in the trading center, it is a seller's market. At this time, the total amount of electricity purchased is greater than the amount of electricity sold. After multiple rounds of transaction quotations, the seller can sell electricity at a higher price. Figure 6 The phenomenon of prices rising first and then falling on the electricity seller side is mainly due to the combined effects of the pessimism coefficient and the risk coefficient. As the transaction progresses, the influence of the pessimism coefficient will become greater than that of the risk coefficient. Therefore, in order to ensure the final transaction, the electricity seller will experience a gradual price decrease. The above analysis verifies that the trading platform and trading strategy proposed in this method can operate effectively.

[0049] 4. Return the results of inter-building transactions to the building’s own operation optimization model for iteration to achieve optimal optimization.

[0050] After the transaction is completed, each building will get the final electricity transaction cost and operation cost of its own system, as shown in the following formula. In the initial building operation optimization, the system operation economic cost is the transaction cost between the building and the distribution network, P Net.b is the electricity price of the distribution network, P Net.s The electricity purchase price of the distribution network. There is only one state of electricity purchase and sale in the same period.

[0051] C Trade.t =C Net.b P Net.b -C Net.s P Net.s (2)

[0052] The above method will have a large deviation from the system cost after the transaction is completed. Therefore, in this method, the electricity transaction price after the transaction is iterated back to the building operation optimization for further optimization and subsequent market transactions. The specific process is as follows: Figure 7 shown.

[0053] Taking into account the impact of calculation time, if the number of iterations reaches the set maximum value K or the difference in the total operating cost of the system after two iterations is less than 5%, the iteration is considered complete. At this time, the building reaches the optimal solution considering the building's virtual energy storage characteristics and its own operation optimization after P2P transactions.

Claims

1. A distributed optimization scheduling method for building groups considering building characteristics and power trading, characterized in that: The steps include: (1) Classify buildings according to their functions and establish corresponding thermodynamic models; (2) Establish a corresponding building operation optimization model based on the functional characteristics and personnel characteristics of the building, with economy and comfort as the goals; (3) Construct a corresponding energy trading platform and trading strategy based on the optimization results of each building's own operation, the characteristics of the power market formed between buildings, and the distributed energy generation situation; (4) Return the results of inter-building transactions to the building's own operation optimization model for iteration to achieve optimal optimization; The classification of buildings in step (1) is to divide buildings into residential buildings, commercial buildings and special buildings according to their functions. The thermodynamic model is: Where: H SUN Indicates the heat transferred by solar thermal radiation, I SUN F is the thermal radiation power of the sun, which means the amount of heat received per square meter per unit time when the light is irradiated vertically; win is the total area of ​​the building's exterior windows; SC is the exterior window shading coefficient, the value of which is related to whether the exterior window has a sunshade and the glass material itself; H rand Indicates the heating power of indoor heat sources, mainly referring to the heating power of human body and electrical equipment, N peo Refers to the total number of people in the room at that moment, Q peo Refers to per capita heat dissipation; P equi Refers to the total power value of all indoor equipment, ε e is the heat dissipation ratio of the equipment; H HVAC Indicates the cooling / heating power of the air conditioning system. In this method, summer cooling is taken as an example, so it is expressed as cooling power in this formula; K wall The heat transfer coefficient between the building's exterior wall and the outdoors is the amount of heat transferred per second for every 1 degree difference between indoor and outdoor temperatures during steady-state heat transfer. K win is the heat transfer coefficient of the building's exterior windows, which has a similar meaning to the heat transfer coefficient of the building's exterior walls; F wall and F win They are the exterior wall area and exterior window area of ​​the building; T room Indicates the indoor temperature, T out Indicates outdoor temperature, T room.t Indicates the indoor temperature of the current period, T room.t+1 Indicates the indoor temperature of the next period; ρ air is the indoor air density, C air is the specific heat capacity of air, V room is the indoor air volume; In step (2), after considering the differences in building thermodynamic models and the different occupant characteristics caused by building classification, each building operation optimization model is constructed with economy and comfort as the goals, and a unified optimization goal and standard is established among different buildings. The specific objective function is as follows: In the above formula, C Trade.t represents the economic cost of system operation, which means the electricity transaction cost between the building and the distribution network and other buildings; C Ma.t C represents the maintenance cost of each device. This method mainly considers the maintenance cost of the HVAC system and photovoltaic power generation system; Tem.t The penalty cost for affecting the user's temperature comfort; C Net.b and C Net.s They represent the electricity sales and purchase prices of the current trading market, and in the initial optimization, they are the electricity purchase and sale prices of the distribution network. Net.b and P Net.s Respectively represent the electricity purchase and sales of the building in the current time period. There can only be one state of electricity purchase and sales in the same time period; δ HVAC , δ pv Respectively represent the use and maintenance cost per unit power of the HVAC system and photovoltaic power generation system per unit time period; P HVAC 、P pv denote the power of the HVAC system and the photovoltaic power generation system respectively; γ is the temperature penalty factor, which can be regarded as the user’s sensitivity to temperature comfort, T set The indoor optimal temperature is set. The greater the deviation from the set temperature, the greater the temperature penalty cost.

2. The distributed optimization scheduling method for building groups considering building characteristics and power trading as claimed in claim 1 is characterized in that: The transaction strategy described in step (3) is as follows: in the transaction process, each building mainly provides three types of information: transaction quotation, transaction electricity and photovoltaic power generation. The transaction market will compare and judge based on the quotation information of each building. When the highest quotation of the electricity buyer is higher than the lowest quotation of the electricity seller, it means that the transaction conditions are met. If the buildings have the same quotation at this time, they are distinguished according to the building level, and the building with the higher priority level is traded first; the priority level of the buildings is mainly based on the following two factors: according to the function of the building, special buildings with important loads are classified as the highest level, and residential buildings are classified as the lowest level; for both the electricity buyer and the electricity seller, if the proportion of the building's to-be-traded electricity to the total to-be-traded electricity is higher, the building's electricity purchase and sales are more stable, and the priority level is higher. If the proportion is the same, in order to promote the consumption of distributed energy, the building with a larger distributed energy generation capacity is recognized as a higher level; after both parties meet the transaction conditions, the transaction price is the average of the two quotations, and the transaction electricity is the party with less to-be-traded electricity. After the transaction is completed, each party updates its own to-be-traded information and proceeds to the next round of transactions.

3. The distributed optimization scheduling method for building groups considering building characteristics and power trading as claimed in claim 1 is characterized in that: The step (4) returns the result of the inter-building transaction to the building's own operation optimization model for iteration to achieve the best optimization. The specific process is as follows: After the transaction is completed, each building will receive the final electricity transaction cost and operating cost of its own system. C Trade.t =C Net.b P Net.b -C Net.s P Net.s Where C Net.b and C Net.s They represent the electricity selling and purchasing prices in the current trading market, P Net.b and P Net.s They represent the amount of electricity purchased and sold by the building at the distribution network during the current period. The transaction price of electricity after the transaction is iterated back into the building operation optimization for further optimization and subsequent market transactions. If the number of iterations reaches the set maximum value K or the difference in the total operating cost of the system after two iterations is less than 5%, the iteration is considered complete. At this time, the building has reached the optimal solution considering the building's virtual energy storage characteristics and its own operation optimization after P2P transactions.

Images