Big data industrial park multi-type building group electric energy sharing method and device

Abstract

The invention provides a big data industrial park multi-type building group electric energy sharing method and device, and belongs to the technical field of building group electric energy sharing optimization. The method comprises the steps that a big data industrial park electric energy sharing framework is built, and three typical building group optimization operation models of a data center building group, an office building group and a residential building group in a big data industrial park are built by considering the space-time adjustable characteristics of different loads; constructing a multi-building-group asymmetric Nash negotiation model by considering the contribution degree of each building group during electric energy sharing, and equivalently converting the multi-building-group asymmetric Nash negotiation model into a first-stage interaction power decision optimization problem and a second-stage transaction price decision optimization problem; a two-stage decision optimization problem is solved in a distributed manner by adopting an enhanced adaptive algorithm, so that the convergence speed of the model is accelerated. The method can assist the big data industrial park to realize efficient utilization of electric energy and cost saving, and has a wide application prospect.

Description

Technical Field

[0001] The present invention relates to a method and a device for sharing electric energy among multiple types of building groups in a big data industrial park, and belongs to the technical field of optimizing electric energy sharing among building groups. Background Art

[0002] With the rapid growth of data traffic under the digital economy, the scale of the big data industry continues to expand, and entities in various industries at home and abroad have built data centers to cope with the large-scale demand for computing power in the whole society. Data centers are the strategic core of my country's new infrastructure construction during the "14th Five-Year Plan" period. In order to achieve intensive land resource utilization and resource sharing and mutual use, the big data industrial park has become an important gathering place for big data companies in various industries and their data center buildings. As the energy demand side, the big data industrial park is usually divided into different functional areas such as commercial and residential areas to meet the production and living needs of the personnel of the park enterprises. However, the high electricity cost has become a major constraint on the development of the big data industrial park.

[0003] In the context of building a new power system, the building clusters in different functional areas of the big data industrial park are usually equipped with renewable energy such as distributed photovoltaics, making them prosumers with both energy production and consumption attributes. Users of building clusters in each functional area actively adjust their computing power and power load based on electricity price signals, formulate reasonable self-optimizing equipment operation strategies and electricity purchase and sales plans, thereby promoting local energy consumption and saving electricity costs. However, different functional areas have different load characteristics, and there is potential for source-load complementarity between them. Therefore, promoting the sharing and mutual assistance of electricity among multiple types of building clusters in different functional areas of the big data industrial park in the market environment can achieve efficient utilization of source-load resources of different entities.

[0004] Cooperative games take into account the fairness between different stakeholders and give participants the freedom to choose other partners. Shapley values ​​are often used to motivate participants to form cooperative alliances and fairly distribute cooperative benefits. However, as the number of cooperative participants increases, the Shapley value method will produce the problem of "combinatorial explosion", which will cause a surge in the scale of calculation and a long calculation time. In order to overcome this defect, domestic and foreign scholars usually use the Nash negotiation method to determine the transaction power and reasonably distribute the cooperative surplus. Since different building groups in the park are managed and operated by different stakeholders, the alternating direction multiplier method (ADMM) as a distributed solution algorithm can protect user privacy. However, the standard ADMM does not always guarantee the convergence of problem solving in practical applications, and the penalty coefficient remains unchanged during the iteration process. There is no dynamic adjustment, which may cause the results to be difficult to converge or even oscillate, affecting the accuracy of the final results. Summary of the invention

[0005] The purpose of the present invention is to take multi-type building groups in a big data industrial park as the research object, focus on the cooperative optimization operation strategy of the building groups, provide a method and device for sharing electricity among multi-type building groups in a big data industrial park, establish an asymmetric Nash negotiation problem and decompose the problem into a two-stage sequential optimization problem, and use an enhanced adaptive PCB-ADMM method to solve it, thereby improving the economic benefits of multi-type building groups and enhancing the fairness of profit distribution.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a method for sharing electric energy among multiple types of buildings in a big data industrial park, comprising:

[0008] Construct a power sharing system for multiple types of buildings in the big data industrial park, and establish a self-optimizing response optimization strategy for each building group;

[0009] Based on the self-optimizing response optimization strategy of each building group, a multi-type building group power sharing optimization operation strategy based on asymmetric Nash negotiation is constructed;

[0010] The multi-building group power sharing optimization operation strategy based on asymmetric Nash negotiation is decomposed into a two-stage sequential optimization decision problem, wherein the first stage is the transaction power decision problem of multi-type building groups in the big data industrial park, and the second stage is the transaction price decision problem of multi-type building groups in the big data industrial park;

[0011] The two-stage sequential optimization decision-making problem is solved to obtain the optimal power sharing strategy for multiple types of buildings in the big data industrial park.

[0012] Preferably, the multi-type building group power sharing system of the big data industrial park is constructed, and the system includes: a data center building group, an office building group and a residential building group, each building group contains multiple buildings and is managed by a unique operator, and each building is equipped with distributed photovoltaic renewable energy, basic load and different types of adjustable resources;

[0013] The adjustable resources of the data center building are energy storage systems, servers and auxiliary equipment that can handle time-space migratable workloads;

[0014] The adjustable resources of the office building include energy storage system and central air conditioning load;

[0015] The adjustable resources of the residential building are electric vehicle load and transferable load.

[0016] There is an interaction of electricity and electricity price information between each building group, and they can independently choose whether to participate in electricity transactions among multiple types of building groups in the market through self-optimization response optimization strategies; when a transaction is reached between different building groups, the big data industrial park as a whole transmits electricity and purchases and sells electricity with the external power grid, so it needs to pay a transmission fee to the power grid company, which is borne jointly by the building groups engaged in electricity transactions.

[0017] Preferably, the self-optimizing response optimization strategy of each building group includes:

[0018] The objective function of the self-optimizing response optimization strategy for the data center building group is:

[0019]

[0020] Among them, C n is the operating cost of data center building cluster n, It is a collection of data center buildings. as well as They represent the cost of purchasing and selling electricity from the data center building group n to the superior power grid, the cost of electricity transaction with other buildings, the cost of grid access fee, and the cost of energy storage operation degradation. and They represent the power purchased and sold by the data center building group n to the upper power grid at time t, and They represent the electricity price that users pay to the upper grid at time t, Δt and denote the time step and set respectively, represents the set of all building groups, and Represents data center building group n and other building groups respectively Between transaction power and transaction price, π WC Indicates the price of Internet access fee. and They represent the energy storage charging power and discharging power of building j in the data center building group n at time t, are the energy storage purchase cost, number of battery charge and discharge cycles in the life cycle, available discharge depth, and rated capacity of building j in data center building group n, respectively. n represents the set of buildings in building group n;

[0021] Based on the objective function of the self-optimizing response optimization strategy of the data center building group, the autonomous selection of whether to participate in the power transaction of multiple types of building groups in the market is realized, and the implementation method is:

[0022] If individual rationality constraints are met: Then the data center building group n participates in the power transaction, otherwise, it does not participate in the power transaction, where, for the breaking point of the negotiations;

[0023] The objective function of the office building cluster self-optimization response optimization strategy is:

[0024]

[0025] Among them, C n represents the operating cost of office building cluster n, represents the collection of office buildings, and They represent the cost of office building group n purchasing and selling electricity to the superior power grid, the cost of electricity transaction with other buildings, the cost of grid access fee, the cost of energy storage operation degradation, and the discomfort cost of all users caused by indoor temperature changes. is the indoor temperature and set temperature of building j in office building group n at time t; is the temperature discomfort coefficient; and The calculation method of is consistent with the calculation method in the self-optimization response optimization strategy of the data center building group;

[0026] Based on the objective function of the office building group self-optimization response optimization strategy, the autonomous selection of whether to participate in the power transaction of multiple types of building groups in the market is realized, and the implementation method is:

[0027] If individual rationality constraints are met: Then the office building group n participates in the power transaction, otherwise, it does not participate in the power transaction, where, for the breaking point of the negotiations;

[0028] The objective function of the self-optimizing response optimization strategy for residential building groups is:

[0029]

[0030] Among them, C n represents the operating cost of residential building cluster n, and They represent the cost of electricity purchase and sale from residential building n to the upper grid, the cost of electricity transaction with other buildings, the cost of grid access fee, and the battery degradation cost of all electric vehicle users. and They represent the number of users in building j in residential building group n at time t. EV charging and discharging power, are the EV purchase cost, battery charge and discharge cycle times, available discharge depth, and rated capacity of user i in building j in residential building group n, respectively. and The calculation method of is consistent with the calculation method in the self-optimization response optimization strategy of the data center building group;

[0031] Based on the objective function of the self-optimizing response optimization strategy of the residential building group, the autonomous selection of whether to participate in the power transaction of multiple types of building groups in the market is realized, and the implementation method is:

[0032] If individual rationality constraints are met: Then the residential building group n participates in the electricity trading, otherwise, it does not participate in the electricity trading, where, The breaking point of the negotiations.

[0033] Preferably, the data center building group self-optimizing response optimization strategy also needs to meet the time-space transferable workload operation constraints, energy storage operation constraints, single data center building group line constraints and data center building group power balance constraints;

[0034] The office building group self-optimization response optimization strategy must also meet the central air-conditioning operation constraints, the line constraints of a single office building group, and the power balance constraints of the office building group;

[0035] The residential building cluster self-optimizing response optimization strategy also needs to meet electric vehicle operation constraints, transferable load operation constraints, single residential building cluster line constraints and residential building cluster power balance constraints.

[0036] Preferably, the self-optimizing response optimization strategy based on each building group is used to construct a multi-type building group power sharing optimization operation strategy based on asymmetric Nash negotiation, including:

[0037] Taking the maximum total benefit as the goal of cooperation, the objective function of the optimal operation strategy of multi-type building group power sharing based on asymmetric Nash negotiation is obtained as follows:

[0038]

[0039] And the constraints must be met: and

[0040] In the formula, Represents all building groups, represents the target value of the optimal operation strategy of building group n before Nash negotiation, that is, the negotiation breakdown point, φ n represents the participation contribution of building group n, They represent the electricity purchase price and electricity sales price of the user to the upper grid at time t, respectively, WC Indicates the price of Internet access fee. The electricity price negotiated for building groups n and m at time t.

[0041] Preferably, the participation contribution of the building group n is expressed as follows:

[0042]

[0043] in, represents the interaction power between building groups n and m.

[0044] Preferably, the transaction power decision problem of the multi-type building group in the big data industrial park is optimized with the goal of minimizing the total operating cost of the multi-type building group cooperation alliance, and the objective function is:

[0045] The objective function of the transaction price decision problem of multiple types of building groups in the big data industrial park is:

[0046]

[0047] in, is the interaction power between building groups n and m obtained in the first stage, Based on The calculated participation contribution of building group n is It is the objective function value obtained by the first stage optimization.

[0048] Preferably, the method of solving the two-stage sequential optimization decision problem using the enhanced adaptive PCB-ADMM algorithm includes:

[0049] 1) Set the maximum number of iterations And initialize the number of iterations k = 1, initialize the original residual and the dual residual And the corresponding convergence threshold and Initialize the penalty coefficient and Lagrange multipliers Set the correction step size α P1 , calculate the negotiation breakdown power value and set the initial interaction power is 0;

[0050] 2) Update the penalty coefficient as follows:

[0051]

[0052] 3) Update the predicted values ​​of interaction power and Lagrange multipliers by block coordinate descent and

[0053] 4) According to and Correction to obtain the k+1th iteration and

[0054] 5) Based on Calculate the original residual of the k+1th iteration and the dual residual And determine whether the convergence condition is met If satisfied or Then the iteration process stops, otherwise set k=k+1 and return to 2).

[0055] In a second aspect, the present invention provides a device for sharing power among multiple types of buildings in a big data industrial park, which is used to implement the above-mentioned method for sharing power among multiple types of buildings in a big data industrial park, and the device comprises:

[0056] The self-optimization module is used to build a power sharing system for multiple types of building groups in the big data industrial park and establish a self-optimization response optimization strategy for each building group;

[0057] A sharing optimization module, used for constructing a multi-type building group power sharing optimization operation strategy based on asymmetric Nash negotiation based on the self-optimizing response optimization strategy of each building group;

[0058] A decision-making problem module is used to decompose the multi-building group power sharing optimization operation strategy based on asymmetric Nash negotiation into a two-stage sequential optimization decision problem, wherein the first stage is the transaction power decision problem of multi-type building groups in the big data industrial park, and the second stage is the transaction price decision problem of multi-type building groups in the big data industrial park;

[0059] The result output module is used to solve the two-stage sequential optimization decision-making problem and obtain the optimal power sharing strategy for multiple types of buildings in the big data industrial park.

[0060] In a third aspect, the present invention provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, which, when executed by a computing device, enable the computing device to execute any one of the above-mentioned methods for sharing power among multiple types of buildings in a big data industrial park.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] The present invention aims at the problem of power sharing of multiple types of building groups in the big data industrial park, considers the contribution of each building group when sharing power, constructs an asymmetric Nash negotiation model for multiple building groups, and converts it into a one-stage interactive power decision optimization problem and a two-stage transaction price decision optimization problem. In order to protect the privacy of each building group, an enhanced adaptive PCB-ADMM solution method is proposed, which improves the problem convergence performance and has technical feasibility. In terms of market implementation, with the expansion of the scale of the big data industry and the increase in electricity costs, the invention can help the big data industrial park achieve efficient use of electricity and cost savings, and has broad application prospects. In terms of economic benefit prediction, by promoting mutual assistance in power sharing, the invention will bring significant electricity bill savings to the industrial park, promote the research and development and application of related technologies, and drive industrial upgrading and economic development. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 A schematic diagram of a flow chart of a method for sharing electric energy among multiple types of buildings in a big data industrial park provided by the present invention;

[0064] Figure 2 A schematic diagram of a power sharing framework for multiple types of buildings in a big data industrial park provided by the present invention;

[0065] Figure 3 A schematic diagram of a two-stage sequential optimization of the power sharing problem of multiple types of buildings provided by the present invention;

[0066] Figure 4 A flow chart of an enhanced adaptive PCB-ADMM algorithm provided by the present invention;

[0067] Figure 5 The electricity transaction price between different building groups provided in one embodiment of the present invention;

[0068] Figure 6 The interactive power between different building groups provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0069] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0070] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions according to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.

[0071] It should be emphasized that the term “include / comprises” when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.

[0072] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0073] It should be emphasized here that the step marks mentioned below are not intended to limit the order of the steps, but it should be understood that the steps can be executed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be executed simultaneously.

[0074] This embodiment provides a method for sharing power among multiple types of buildings in a big data industrial park. Figure 1 As shown, the specific steps are as follows:

[0075] Step 1: Build a power sharing system for multiple types of buildings in the big data industrial park, and establish a self-optimizing response optimization strategy for multiple types of buildings.

[0076] like Figure 2 The big data industrial park multi-type building group power sharing system architecture shown in the figure, according to different functions, the big data industrial park mainly includes data center building group (DCBG), office building group (OBG) and residential building group ( group, RBG, each building group contains multiple buildings and is managed by a single operator. Each building is equipped with renewable energy such as distributed photovoltaics, basic loads, and different types of adjustable resources. Among them, the adjustable resources of the data center building are mainly energy storage systems, servers and auxiliary equipment that can handle time-space migratable workloads; the main adjustable resources of office buildings include energy storage systems and central air-conditioning loads; the adjustable resources of residential buildings are mainly electric vehicle loads and transferable loads. There is an interaction of electricity and electricity price information between each building group, and through internal optimization operation strategies, it can independently choose whether to participate in the electricity trading of multiple types of building groups in the market to achieve mutual complementarity of electricity energy. When a transaction is reached between different building groups, the big data industrial park as a whole can transmit and purchase electricity with the external power grid through the distribution lines laid by the power grid company, so it needs to pay a certain amount of network access fees to the power grid company, and this part of the cost is borne by the building groups that conduct power trading.

[0077] Under this framework, DCBG, OBG and RBG optimize the time and space adjustable resources in their respective building groups with the goal of economy.

[0078] The objective function of DCBG is:

[0079]

[0080] Among them, C n is the operating cost of data center building cluster n, It is a collection of data center buildings. as well as They represent the cost of electricity purchase and sale from the data center building group n to the superior power grid, the cost of electricity transaction with other building groups, the cost of grid access fee, and the cost of energy storage operation degradation, which are expressed as:

[0081]

[0082] in, and They represent the power purchased and sold by the data center building group n to the upper power grid at time t respectively; and They represent the electricity price that users pay to the upper grid at time t; Δt and denote the time step and set respectively; Represents the set of all building groups; and Represents data center building group n and other building groups respectively The transaction power and transaction price between π WC Indicates the price of Internet access fee; and They represent the energy storage charging power and discharging power of building j in the data center building group n at time t respectively; are the energy storage purchase cost, number of battery charge and discharge cycles in the life cycle, available discharge depth, and rated capacity of building j in data center building group n, respectively. n Represents the set of buildings in building group n.

[0083] The DCBG self-optimizing response optimization strategy also includes time-space transferable workload operation constraints, energy storage operation constraints, single data center building line constraints, and DCBG power balance constraints, as follows:

[0084] 1) Time-space transferable workload operation constraints:

[0085] Batch workloads (BWs) run constraints:

[0086]

[0087] Where: and They represent the predicted and actual batch load quantities of building j in data center building cluster n at time τ; TD n,j is the maximum number of delay periods of building j in data center building cluster n; The number of workloads that each server can handle per unit time, i.e., the service rate (assuming that the servers in building j in data center building cluster n are homogeneous); is the number of active servers in building j of data center building cluster n processing batch load at time t.

[0088] Interactive workloads (IWs) running constraints:

[0089]

[0090] Where: represents the predicted value of the number of interactive workloads arriving at the front-end server of data center building cluster n at time t, It represents the maximum response time for processing interactive workloads that meets user needs; and They represent the number of interactive workloads actually processed by building j in data center building cluster n at time t and the number of servers required, respectively.

[0091] Therefore, given that the workload to be processed includes BWs and IWs, the total power consumption of each building j in the data center building cluster n is

[0092] Where: and It represents the rated power of building j in the data center building cluster n when the building is idle and the CPU utilization reaches the maximum state; represents the power usage effectiveness (PUE) of building j in data center building cluster n; and represents the number of workloads to be processed and the number of active servers in building j in data center building cluster n at time t; Indicates the maximum number of active servers.

[0093] 2) Energy storage operation constraints:

[0094] Energy storage operation constraints include energy balance constraints, state of charge constraints (SOC) and charge and discharge power constraints:

[0095]

[0096] Where: represents the rated capacity of energy storage in building j of data center building n, represents the actual capacity of the energy storage of building j in data center building n at time t, represents the actual capacity of the energy storage of building j in data center building n at time t, Indicates the energy storage of building j in data center building n in the Actual capacity at the moment; and are the charging efficiency and discharging efficiency of energy storage respectively; is the charge state of the energy storage of building j in the data center building group n at time t; and for The upper and lower limits of and are the maximum charging and discharging powers of energy storage in building j in data center building cluster n, respectively; and are the state variables of energy storage charging and discharging, and When energy storage is charged, and Time storage energy discharge.

[0097] 3) Line constraints of a single data center building group:

[0098]

[0099] Where: is the predicted value of photovoltaic power output of building j in data center building group n at time t; is the maximum transmission limit between building j in data center building cluster n and the power grid line.

[0100] 4) Power balance constraints:

[0101]

[0102] Where: It represents the maximum power of electricity purchased and sold by the data center building group n due to line capacity limitation. and are the electricity purchase and sales of data center building group n at time t, and represents the power purchase and sale state variable of data center building group n at time t, and When it means buying electricity, and When it is on, it means electricity is sold.

[0103] In this embodiment, based on the objective function, the autonomous selection of whether to participate in the power trading of multiple types of building groups in the market is realized, and the implementation method is:

[0104] If individual rationality constraints are met: Then the data center building group n participates in the power transaction, otherwise, it does not participate in the power transaction, where, The breaking point of the negotiations.

[0105] The objective function of OBG is:

[0106]

[0107] Among them, C n represents the operating cost of office building cluster n, represents the collection of office buildings, as well as They represent the cost of office building group n purchasing and selling electricity to the superior power grid, the cost of electricity transaction with other buildings, and the cost of electricity transmission through the grid.

[0108] The cost of electricity consumption, the cost of energy storage operation degradation, and the discomfort cost caused by indoor temperature changes for all users are and is the indoor temperature and set temperature of building j in office building group n at time t; is the temperature discomfort coefficient.

[0109] OBG's self-optimizing response optimization strategy also includes central air-conditioning operation constraints, single office building line constraints, and OBG power balance constraints. The meaning of some variables and the energy storage operation constraint model are consistent with DCBG, as follows:

[0110] 1) Central air conditioning operation constraints:

[0111] Assuming that the parameters of all rooms in the office building are the same, the heat transfer effect between the walls inside the building is ignored, and the total area of ​​the exterior walls and windows of the entire building is used to approximate the exterior wall area and window area of ​​all rooms. The operating constraints of the central air conditioner are:

[0112]

[0113] In the formula, and represents the indoor temperature and wall temperature of building j in office building group n at time t, Q t is the solar radiation, T t out is the outdoor temperature; and They represent the heat capacity of the interior and wall of building j in office building group n respectively; γ is the solar radiation efficiency; S n,j and W n,j denote the exterior wall area and window area of ​​building j in office building cluster n respectively; and They represent the heat transfer efficiency between “interior-wall” and “wall-exterior” of building j in office building cluster n respectively; and They respectively represent the electric cooling energy efficiency coefficient of the central air-conditioning in building j in office building group n and the electric cooling power at time t.

[0114] After the above discretization processing, we can get:

[0115]

[0116] Taking into account user comfort, the indoor temperature of the building needs to meet the comfortable temperature range, that is:

[0117]

[0118] Where: and are the upper and lower bounds of the comfortable temperature of building j in office building cluster n, respectively.

[0119] 2) Line constraints for a single office building group:

[0120]

[0121] Where: represents the basic power load of building j in office building group n at time t; It represents the maximum transmission limit between building j in office building group n and the power grid line.

[0122] 3) Power balance constraints for office buildings:

[0123]

[0124] The remaining power purchase and sales constraints refer to the remaining three constraints of power balance constraints in DCBG.

[0125] In this embodiment, based on the objective function, the autonomous selection of whether to participate in the electricity trading of multiple types of building groups in the market is realized, and the implementation method is:

[0126] If individual rationality constraints are met: Then the office building group n participates in the power transaction, otherwise, it does not participate in the power transaction, where, The breaking point of the negotiations.

[0127] The objective function of RBG is:

[0128]

[0129] Among them, C n is the operating cost of residential building group n, represents a collection of residential buildings. as well as They represent the cost of electricity purchase and sale from the residential building group n to the superior power grid, the cost of electricity transaction with other buildings, the cost of grid access fee, and the battery operation degradation cost of all electric vehicle users. and They represent the number of users in building j in residential building group n at time t. EV charging and discharging power, are the EV purchase cost, number of battery charge and discharge cycles during the life cycle, available discharge depth, and rated capacity of user i in building j in residential building cluster n.

[0130] The self-optimizing response optimization strategy for residential buildings also includes electric vehicle operation constraints, transferable load operation constraints, single residential building line constraints, and RBG power balance constraints.

[0131] 1) Electric vehicle operation constraints:

[0132] Assume that the dispatch period of each electric vehicle user is

[0133]

[0134] Where: and denote the rated capacity of electric vehicles of user i in building j in residential building cluster n and the actual capacity at time t respectively; and are the charging efficiency and discharging efficiency of the electric vehicle of user i in building j in residential building cluster n, respectively; is the charge state of the electric vehicle of user i in building j in residential building group n at time t; and for The lower and upper limits of and are the maximum charging and discharging powers of the electric vehicle of user i in building j in residential building cluster n, respectively; and are the state variables of electric vehicle charging and discharging, and When charging electric vehicles, and When the electric vehicle discharges; and are the start and end times of the dispatchable electric vehicle of user i in building j in residential building group n, and The actual state of charge and the set state of charge of the electric vehicle of user i in building j in residential building group n at the end time.

[0135] 2) Transferable load operation constraints:

[0136] Transferable load refers to the amount of time that can be spent within an acceptable period of time. For residential building users, the transferable load usually refers to loads such as washing machines, dryers, and dishwashers. The model can be expressed as:

[0137]

[0138] Where: Indicates whether the transferable load is in working condition. Indicates the working status. Indicates non-working status; Indicates the total time required to complete the work task; and They represent the rated power and actual power of the transferable load respectively.

[0139] 3) Line constraints for a single residential building group:

[0140]

[0141] Where: Represents the transmission limit between building j in residential building group n and the power grid line.

[0142] 4) RBG power balance constraint:

[0143]

[0144] For other power purchase and sales constraints, refer to the power balance constraints of DCBG.

[0145] In this embodiment, based on the objective function, the autonomous selection of whether to participate in the electricity trading of multiple types of building groups in the market is realized, and the implementation method is:

[0146] If individual rationality constraints are met: Then the residential building group n participates in the electricity trading, otherwise, it does not participate in the electricity trading, where, The breaking point of the negotiations.

[0147] Step 2: Based on the multi-type building group power sharing system of the big data industrial park established in step 1, construct a multi-building group power sharing optimization operation strategy based on asymmetric Nash negotiation.

[0148] Construct an optimal operation strategy for multi-building power sharing based on asymmetric Nash negotiation, considering the participation contribution of multiple types of building groups φ n , the remaining benefits of cooperation can be distributed more fairly in the transaction process. Each building group takes the maximum total benefit as the goal of participating in cooperation through power sharing, that is:

[0149]

[0150] In the formula Represents all building groups, It indicates the target value of the optimized operation strategy of building group n before Nash negotiation, that is, the negotiation breaking point,

[0151] According to the interaction power between building groups n and m The participation contribution of each building group n is φ n It can be defined as:

[0152]

[0153] In addition to satisfying the above model constraints, each building group considers any building group In addition to being willing to participate in cooperation, individual rational constraints must also be met:

[0154] In order to ensure the stability of multiple types of building groups participating in Nash negotiations through power sharing, it is necessary to ensure that each building group has corresponding economic incentives. Specifically, the electricity price for each period negotiated through bargaining It cannot be higher than the difference between the price of electricity purchased from the power grid and the price of the transmission fee, nor lower than the sum of the price of electricity sold to the power grid and the price of the transmission fee, that is:

[0155] Step 3: Decompose the multi-building group power sharing optimization operation strategy based on asymmetric Nash negotiation proposed in step 2 into a two-stage sequential optimization decision problem. The first stage is the transaction power decision problem of multi-type building groups in the big data industrial park, and the second stage is the transaction price decision problem of multi-type building groups in the big data industrial park. The two-stage decision and optimization process is as follows: Figure 3 As shown, including:

[0156] The first stage of transaction power decision: the optimization is carried out with the goal of minimizing the total operating cost of the multi-type building group cooperation alliance. The objective function is: Given that the transaction costs of electricity purchase and sale between different types of building groups need to be balanced, that is, Then the objective function can be changed to:

[0157] The second stage transaction price decision: given the first stage interaction power After that, each building group determines the transaction price of electricity purchase and sale between itself and other building groups through bargaining. Since the asymmetric Nash negotiation problem contains the product term of energy cost, it is a non-convex nonlinear optimization problem and cannot be solved directly. Therefore, it is convexified into a summation problem through the logarithmic function, and the maximum value problem is rewritten as a minimum value problem. The objective function of the transaction price decision in the second stage is:

[0158]

[0159] In the formula Based on The calculated contribution is It is the objective function value obtained by the first stage optimization.

[0160] Step 4: Use the enhanced adaptive PCB-ADMM (prediction-correction-based alternating direction multiplier method) algorithm to solve the two-stage sequential optimization decision problem in step 3 and obtain the optimal power sharing strategy for multiple types of buildings in the big data industrial park. The implementation process is as follows: Figure 4 As shown, taking the first stage problem P1 as an example, it specifically includes:

[0161] 41) Set the maximum number of iterations And initialize the number of iterations k = 1, initialize the original residual and the dual residual And the corresponding convergence threshold and Initialize the penalty coefficient and Lagrange multipliers Set the correction step size α P1 , calculate the negotiation breakdown power value and set the initial interaction power is 0;

[0162] 42) Update the penalty coefficient

[0163] 43) Prediction step, that is, updating the predicted values ​​of the interaction power and Lagrange multipliers by block coordinate descent (BCD) method and

[0164]

[0165] Both represent the Lagrangian function of problem P1,

[0166] 44) Correction steps, according to and Correction to obtain the k+1th iteration

[0167]

[0168]

[0169] 45) Pass Calculate the original residual of the k+1th iteration and the dual residual And determine whether the convergence condition is met If satisfied or Then the iteration process stops, otherwise set k=k+1 and return to 42).

[0170] Based on the above-mentioned method for sharing power among multiple types of buildings in a big data industrial park, in one embodiment, a 24-hour time window is used and the resolution is 1 hour. The electricity price data adopts the two-part industrial and commercial electricity price in Shanghai in summer. The photovoltaic grid-connected electricity price is 0.3 yuan / kWh, the energy storage operation and maintenance price is 0.1 yuan / kWh, and the grid access fee paid by multiple types of building groups to the power grid company during power exchange is 0.025 yuan / kWh. In the enhanced adaptive PCB-ADMM algorithm for solving the two-stage sequential problem, the maximum number of iterations is set to 150 times, the initial values ​​of the primal residual and the dual residual are both set to 1, and the relevant convergence thresholds are all set to 10 -3 The initial penalty coefficient is 10 -3 , the initial Lagrange multiplier is 0, and the correction step size is 0.95.

[0171] Based on the above initial conditions, Figure 5 The electricity trading prices between different building groups are given. It can be seen that each building group only trades energy when there is photovoltaic power output, and the final negotiated transaction price is between the photovoltaic grid-connected electricity price and the price of purchasing electricity from the power grid, which verifies the effectiveness of the proposed strategy. Figure 6The interaction power between different building groups is given, where A←B is a positive value indicating that A purchases electricity from B, and a negative value indicating that A sells electricity to B. DCBG1 belongs to a power-deficient data center building group, and needs to purchase electricity from DCBG2, OBG, and RBG at the same time during the period of 8:00-14:00; DCBG2 belongs to a power-rich data center building group, which can not only sell electricity to DCBG1, but also sell it to OBG and RBG during 13:00-16:00 and 7:00-16:00 respectively. In addition, in certain time periods, there is a situation where a building group buys electricity by bargaining with other building groups and then resells it to other building groups. For example, at 14:00, RBG purchases electricity from DCBG2. In addition to its own use, it also sells it to DCBG1 and OBG. This is because the price agreed upon by RBG and DCBG2 is relatively low, and there is room for resale. The results of the above embodiment verify the effectiveness of the method and system for sharing electricity among multiple types of building groups in a big data industrial park proposed in the present invention.

[0172] Based on the above invention concept, the embodiment of the present invention further provides a device for sharing power of multiple types of buildings in a big data industrial park, which is used to implement the above method for sharing power of multiple types of buildings in a big data industrial park. The device includes:

[0173] The self-optimization module is used to build a power sharing system for multiple types of building groups in the big data industrial park and establish a self-optimization response optimization strategy for each building group;

[0174] A sharing optimization module, used for constructing a multi-type building group power sharing optimization operation strategy based on asymmetric Nash negotiation based on the self-optimizing response optimization strategy of each building group;

[0175] A decision-making problem module is used to decompose the multi-building group power sharing optimization operation strategy based on asymmetric Nash negotiation into a two-stage sequential optimization decision problem, wherein the first stage is the transaction power decision problem of multi-type building groups in the big data industrial park, and the second stage is the transaction price decision problem of multi-type building groups in the big data industrial park;

[0176] The result output module is used to solve the two-stage sequential optimization decision-making problem and obtain the optimal power sharing strategy for multiple types of buildings in the big data industrial park.

[0177] It is worth pointing out that the device embodiment corresponds to the above-mentioned method embodiment, and the implementation methods of the above-mentioned method embodiments are all applicable to the device embodiment and can achieve the same or similar technical effects, so they will not be repeated here.

[0178] Based on the above-mentioned inventive concept, an embodiment of the present invention also provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, and when the instructions are executed by a computing device, the computing device executes any one of the above-mentioned methods for sharing electricity among multiple types of buildings in a big data industrial park.

[0179] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0180] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0181] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0182] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0183] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A method for sharing electric energy among multiple types of buildings in a big data industrial park, characterized in that: include: Construct a power sharing system for multiple types of buildings in the big data industrial park, and establish a self-optimizing response optimization strategy for each building group; Based on the self-optimizing response optimization strategy of each building group, a multi-type building group power sharing optimization operation strategy based on asymmetric Nash negotiation is constructed; The multi-building group power sharing optimization operation strategy based on asymmetric Nash negotiation is decomposed into a two-stage sequential optimization decision problem, wherein the first stage is the transaction power decision problem of multi-type building groups in the big data industrial park, and the second stage is the transaction price decision problem of multi-type building groups in the big data industrial park; The two-stage sequential optimization decision-making problem is solved to obtain the optimal power sharing strategy for multiple types of buildings in the big data industrial park.

2. The method for sharing electric energy among multiple types of buildings in a big data industrial park according to claim 1 is characterized in that: The multi-type building group power sharing system of the big data industrial park is constructed, and the system includes: a data center building group, an office building group and a residential building group. Each building group contains multiple buildings and is managed by a unique operator. Each building is equipped with distributed photovoltaic renewable energy, basic load and different types of adjustable resources; The adjustable resources of the data center building are energy storage systems, servers and auxiliary equipment that can handle time-space migratable workloads; The adjustable resources of the office building include energy storage system and central air conditioning load; The adjustable resources of the residential building are electric vehicle load and transferable load. There is an interaction of electricity and electricity price information between each building group, and they can independently choose whether to participate in electricity transactions among multiple types of building groups in the market through self-optimization response optimization strategies; when a transaction is reached between different building groups, the big data industrial park as a whole transmits electricity and purchases and sells electricity with the external power grid, so it needs to pay a transmission fee to the power grid company, which is borne jointly by the building groups engaged in electricity transactions.

3. The method for sharing electric energy among multiple types of buildings in a big data industrial park according to claim 2 is characterized in that: The self-optimizing response optimization strategy of each building group includes: The objective function of the self-optimizing response optimization strategy for the data center building group is: Among them, C n is the operating cost of data center building cluster n, It is a collection of data center buildings. as well as They represent the cost of purchasing and selling electricity from the data center building group n to the superior power grid, the cost of electricity transaction with other buildings, the cost of grid access fee, and the cost of energy storage operation degradation. and They represent the power purchased and sold by the data center building group n to the upper power grid at time t, and They represent the electricity price that users pay to the upper grid at time t, Δt and denote the time step and set respectively, represents the set of all building groups, and Represents data center building group n and other building groups respectively Between transaction power and transaction price, π WC Indicates the price of Internet access fee. and They represent the energy storage charging power and discharging power of building j in the data center building group n at time t, are the energy storage purchase cost, number of battery charge and discharge cycles in the life cycle, available discharge depth, and rated capacity of building j in data center building group n, respectively. n represents the set of buildings in building group n; Based on the objective function of the self-optimizing response optimization strategy of the data center building group, the autonomous selection of whether to participate in the power transaction of multiple types of building groups in the market is realized, and the implementation method is: If individual rationality constraints are met: Then the data center building group n participates in the power transaction, otherwise, it does not participate in the power transaction, where, for the breaking point of the negotiations; The objective function of the office building cluster self-optimization response optimization strategy is: Among them, C n represents the operating cost of office building cluster n, represents the collection of office buildings, and They represent the cost of office building group n purchasing and selling electricity to the superior power grid, the cost of electricity transaction with other buildings, the cost of grid access fee, the cost of energy storage operation degradation, and the discomfort cost of all users caused by indoor temperature changes. and is the indoor temperature and set temperature of building j in office building group n at time t; is the temperature discomfort coefficient; and The calculation method of is consistent with the calculation method in the self-optimization response optimization strategy of the data center building group; Based on the objective function of the office building group self-optimization response optimization strategy, the autonomous selection of whether to participate in the power transaction of multiple types of building groups in the market is realized, and the implementation method is: If individual rationality constraints are met: Then the office building group n participates in the power transaction, otherwise, it does not participate in the power transaction, where, for the breaking point of the negotiations; The objective function of the self-optimizing response optimization strategy for residential building groups is: Among them, C n represents the operating cost of residential building cluster n, and They represent the cost of electricity purchase and sale from residential building n to the upper grid, the cost of electricity transaction with other buildings, the cost of grid access fee, and the battery degradation cost of all electric vehicle users. and They represent the number of users in building j in residential building group n at time t. EV charging and discharging power, are the EV purchase cost, battery charge and discharge cycle times, available discharge depth, and rated capacity of user i in building j in residential building group n, respectively. and The calculation method of is consistent with the calculation method in the self-optimization response optimization strategy of the data center building group; Based on the objective function of the self-optimizing response optimization strategy of the residential building group, the autonomous selection of whether to participate in the power transaction of multiple types of building groups in the market is realized, and the implementation method is: If individual rationality constraints are met: Then the residential building group n participates in the electricity trading, otherwise, it does not participate in the electricity trading, where, The breaking point of the negotiations.

4. The method for sharing electric energy among multiple types of buildings in a big data industrial park according to claim 3 is characterized in that: The data center building group self-optimization response optimization strategy also needs to meet the time and space transferable workload operation constraints, energy storage operation constraints, single data center building group line constraints and data center building group power balance constraints; The office building group self-optimization response optimization strategy must also meet the central air-conditioning operation constraints, the line constraints of a single office building group, and the power balance constraints of the office building group; The residential building cluster self-optimizing response optimization strategy also needs to meet electric vehicle operation constraints, transferable load operation constraints, single residential building cluster line constraints and residential building cluster power balance constraints.

5. The method for sharing electric energy among multiple types of buildings in a big data industrial park according to claim 3 is characterized in that: The self-optimizing response optimization strategy based on each building group is used to construct a multi-type building group power sharing optimization operation strategy based on asymmetric Nash negotiation, including: Taking the maximum total benefit as the goal of cooperation, the objective function of the optimal operation strategy of multi-type building group power sharing based on asymmetric Nash negotiation is obtained as follows: And the constraints must be met: In the formula, Represents all building groups, represents the target value of the optimal operation strategy of building group n before Nash negotiation, that is, the negotiation breakdown point, φ n represents the participation contribution of building group n, They represent the electricity purchase price and electricity sales price of the user to the upper grid at time t, respectively, WC Indicates the price of Internet access fee. The electricity price negotiated for building groups n and m at time t.

6. The method for sharing electric energy among multiple types of buildings in a big data industrial park according to claim 5 is characterized in that: The participation contribution of the building group n is expressed as follows: in, represents the interaction power between building groups n and m.

7. The method for sharing electric energy among multiple types of buildings in a big data industrial park according to claim 6 is characterized in that: The transaction power decision problem of multi-type building groups in the big data industrial park is optimized with the goal of minimizing the total operating cost of the multi-type building group cooperation alliance. The objective function is: The objective function of the transaction price decision problem of multiple types of building groups in the big data industrial park is: in, is the interaction power between building groups n and m obtained in the first stage, Based on The calculated participation contribution of building group n is It is the objective function value obtained by the first stage optimization.

8. The method for sharing electric energy among multiple types of buildings in a big data industrial park according to claim 7 is characterized in that: The method of solving the two-stage sequential optimization decision problem by using the enhanced adaptive PCB-ADMM algorithm includes: 1) Set the maximum number of iterations And initialize the number of iterations k = 1, initialize the original residual and the dual residual And the corresponding convergence threshold and Initialize the penalty coefficient and Lagrange multipliers Set the correction step size α P1 , calculate the negotiation breakdown power value and set the initial interaction power is 0; 2) Update the penalty coefficient as follows: 3) Update the predicted values ​​of interaction power and Lagrange multipliers by block coordinate descent and 4) According to and Correction to obtain the k+1th iteration and 5) Based on Calculate the original residual of the k+1th iteration and the dual residual And determine whether the convergence condition is met If satisfied or Then the iteration process stops, otherwise set k=k+1 and return to 2).

9. A power sharing device for multiple types of buildings in a big data industrial park, characterized in that: The device is used to implement the method for sharing power among multiple types of buildings in a big data industrial park as described in any one of claims 1 to 8, and comprises: The self-optimization module is used to build a power sharing system for multiple types of building groups in the big data industrial park and establish a self-optimization response optimization strategy for each building group; A sharing optimization module, used for constructing a multi-type building group power sharing optimization operation strategy based on asymmetric Nash negotiation based on the self-optimizing response optimization strategy of each building group; A decision-making problem module is used to decompose the multi-building group power sharing optimization operation strategy based on asymmetric Nash negotiation into a two-stage sequential optimization decision problem, wherein the first stage is the transaction power decision problem of multi-type building groups in the big data industrial park, and the second stage is the transaction price decision problem of multi-type building groups in the big data industrial park; The result output module is used to solve the two-stage sequential optimization decision-making problem and obtain the optimal power sharing strategy for multiple types of buildings in the big data industrial park.

10. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include instructions, which, when executed by a computing device, enable the computing device to execute any one of the methods for sharing power among multiple types of buildings in a big data industrial park according to claims 1 to 8.

Images