An optimization method and terminal for power grid demand response considering power consumption comfort

Abstract

The present invention discloses an optimization method and terminal for power grid demand response taking into account the comfort of electricity consumption. According to network constraints and the RES and rigid load of the power system, the ideal load curve CDL of the power grid node is calculated; according to the historical electricity consumption data of the user, the weekly load curve and the adjustment range of the user are generated; the CDL, the weekly load curve and the adjustment range are sent to the home energy management system HEMS, and the HEMS establishes and solves the optimal decision model in combination with the user's comfort factor to obtain an optimized electricity management plan; the present invention considers the user's electricity comfort and subjective electricity preference, establishes an optimal decision model, and automatically solves the best electricity plan for the user according to the optimal decision model. The plan uses the preference coefficient to balance the contradiction between following the CDL and maintaining the comfort of electricity consumption, so that the user can participate in the demand response while maintaining the comfort of electricity consumption, thereby being able to more accurately and effectively adjust the user's electricity consumption plan and improve the effectiveness of the demand response.

Description

Technical Field

[0001] The present invention relates to the technical field of power consumption optimization, and in particular to a method and a terminal for optimizing power grid demand response taking power consumption comfort into consideration. Background Art

[0002] The new power system for a carbon-neutral future has a stronger ability to absorb renewable energy sources (RES). However, due to the volatility and intermittency of RES, the flexible regulation capability of the source side is greatly weakened, which brings great power balance challenges to the power system. At the same time, with the increase in the number of flexible loads, the demand side has great potential for flexible regulation (such as air conditioners, water heaters, and electric vehicles). Demand response (DR) provides more and more flexibility by managing these flexible loads, which is of great significance to effectively reduce the regulation burden of the new power system.

[0003] According to whether the electricity price is directly used as a guiding signal, DR is mainly divided into price-based DR (PDR) and incentive-based DR (IDR). "S. Mohajeryami, I. Moghaddam, M. Doostan, B. Vatani, and P. Schwarz, "A novel economic model for price-based demand response," Electric Power Systems Research, vol. 135, pp. 1-9, Jun. 2016." theoretically analyzes the economic principles of peak-valley time-of-use electricity prices, real-time electricity prices, and peak electricity prices to promote peak-valley filling of power grids. In response to the shortcomings of time-of-use electricity prices, "J. Yang, J. Zhao, F. Wen, and Z. Dong, "A model of customizing electricity retail prices based on load profile clustering analysis," IEEE Transactions on Smart Grid, vol. 10, no. 3, pp. 3374-3386, May. 2019." proposed a personalized price customization method based on data mining. IDR can be divided into interruptible load control, direct load control, and emergency demand response. Users are compensated according to the change in their power consumption curve compared with the customer load baseline (Customer Baseline Load, CBL), which is the predicted load curve of the user without a demand response event. Many studies have explored the estimation and reconstruction of CBL. CDL describes what kind of load curve is friendly to the power system and uses it as a target curve to guide users to participate in demand response autonomously. CDL is calculated through system-wide global optimization to achieve specific demand response purposes (for example, promoting renewable energy consumption). Therefore, the higher the similarity of the power consumption curve with CDL, the greater the contribution made by the user. CDL-based DR is a goal-oriented mechanism with the advantages of avoiding overshoot and rebound, which can promote the large-scale promotion and normalized implementation of DR.

[0004] However, the formulation of CDL lacks consideration of the underlying network, resulting in inaccurate guidance to users. In other words, it is difficult for the current CDL to tap into the flexibility of users under network constraints. In addition, during the implementation of demand response based on CDL, the comfort of power consumption and the subjective preferences of users are ignored, reducing the effectiveness of demand response. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide an optimization method and terminal for power grid demand response taking into account power consumption comfort, which can more accurately and effectively adjust user power consumption plans and improve the effectiveness of demand response.

[0006] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0007] A method for optimizing power grid demand response taking into account power consumption comfort comprises the following steps:

[0008] S1. Calculate the ideal load curve CDL of the grid node according to the network constraints, RES and rigid load of the power system;

[0009] S2. Generate the user's weekly load curve and adjustment range based on the user's historical electricity consumption data;

[0010] S3. Send the CDL, the weekly load curve and the adjustment range to the home energy management system HEMS, which establishes and solves the optimal decision model in combination with the user's comfort factor to obtain an optimized power management plan.

[0011] In order to solve the above technical problems, another technical solution adopted by the present invention is:

[0012] An optimization terminal for grid demand response taking into account electricity comfort comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0013] S1. Calculate the ideal load curve CDL of the grid node according to the network constraints, RES and rigid load of the power system;

[0014] S2. Generate the user's weekly load curve and adjustment range based on the user's historical electricity consumption data;

[0015] S3. Send the CDL, the weekly load curve and the adjustment range to the home energy management system HEMS, which establishes and solves the optimal decision model in combination with the user's comfort factor to obtain an optimized power management plan.

[0016] The beneficial effects of the present invention are as follows: an optimization method and terminal for grid demand response taking into account electricity comfort of the present invention calculates the CDL of each node according to network constraints and uncontrollable parts of the power system (RES and rigid loads). As the ideal load curve of each node, the CDL will be published to the corresponding users to guide the users to reshape and adjust electricity consumption; typical weekly load curves and adjustment ranges are extracted to characterize the user's adjustability; the home energy management system considers the user's electricity comfort and subjective electricity preference, establishes an optimal decision model, and automatically solves the best electricity plan for the user according to the optimal decision model. The plan uses the preference coefficient to balance the contradiction between following the CDL and maintaining electricity comfort, so that users can participate in demand response while maintaining electricity comfort, thereby being able to more accurately and effectively adjust the user's electricity plan and improve the effectiveness of demand response. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a method for optimizing grid demand response taking into account electricity comfort according to an embodiment of the present invention;

[0018] Figure 2 This is a structural diagram of an optimization terminal for grid demand response taking into account electricity comfort according to an embodiment of the present invention;

[0019] Figure 3 A load guideline diagram of an optimization method for grid demand response taking into account electricity comfort according to an embodiment of the present invention;

[0020] Figure 4 A demand response architecture diagram based on load guidelines of an optimization method for grid demand response taking into account electricity comfort according to an embodiment of the present invention;

[0021] Figure 5 A schematic diagram of an improved IEEE 9 bus system for an optimization method for grid demand response taking into account electricity comfort according to an embodiment of the present invention;

[0022] Figure 6 A schematic diagram of load power of a demand response user 1 of an optimization method for grid demand response taking into account electricity comfort according to an embodiment of the present invention;

[0023] Figure 7 A schematic diagram of load power of a demand response user 2 of an optimization method for grid demand response taking into account electricity comfort according to an embodiment of the present invention;

[0024] Figure 8 A schematic diagram of load power of a demand response user 3 of an optimization method for grid demand response taking into account electricity comfort according to an embodiment of the present invention;

[0025] Fig. 9A schematic diagram of power comparison between two situations of participating in demand response and not participating in demand response in an optimization method for power grid demand response taking into account power consumption comfort according to an embodiment of the present invention;

[0026] Fig.10 A schematic diagram of changes in RES power abandonment with λ and DR percentages in an optimization method for grid demand response taking into account power comfort in an embodiment of the present invention;

[0027] Description of labels:

[0028] 1. An optimization terminal for grid demand response taking into account electricity comfort; 2. A processor; 3. A memory. DETAILED DESCRIPTION

[0029] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in combination with the implementation modes and the accompanying drawings.

[0030] Please refer to Figure 1 , an optimization method for power grid demand response taking into account power consumption comfort, comprising the steps of:

[0031] S1. Calculate the ideal load curve CDL of the grid node according to the network constraints, RES and rigid load of the power system;

[0032] S2. Generate the user's weekly load curve and adjustment range based on the user's historical electricity consumption data;

[0033] S3. Send the CDL, the weekly load curve and the adjustment range to the home energy management system HEMS, which establishes and solves the optimal decision model in combination with the user's comfort factor to obtain an optimized power management plan.

[0034] From the above description, it can be seen that the beneficial effects of the present invention are: an optimization method and terminal for power grid demand response taking into account electricity comfort of the present invention calculates the CDL of each node according to network constraints and uncontrollable parts of the power system (RES and rigid loads). As the ideal load curve of each node, CDL will be published to the corresponding users to guide users to reshape and adjust electricity consumption; typical weekly load curves and adjustment ranges are extracted to characterize the user's adjustability; the home energy management system considers the user's electricity comfort and subjective electricity preference, and establishes an optimal decision model, and automatically solves the best electricity plan for the user according to the optimal decision model. The plan uses the preference coefficient to balance the contradiction between following CDL and maintaining electricity comfort, so that users can participate in demand response while maintaining electricity comfort, thereby being able to more accurately and effectively adjust the user's electricity plan and improve the effectiveness of demand response.

[0035] Furthermore, the calculation of CDL is as follows:

[0036] Under the constraints (2)-(4), the power output of the conventional generator set is determined with the goal of minimizing the total operating cost (1):

[0037]

[0038]

[0039]

[0040]

[0041] Among them, a j ,b j ,c j is the cost parameter of conventional thermal power unit j; ζ is the time slot set indexed by t, and the cardinality of the set is T; j represents the index of the conventional thermal power unit user; i represents the index of the RES user, m represents the index of the rigid load user, n represents the index of the DR user, and J, I, M and N are the sets corresponding to j, i, m and n; represents the power output of conventional thermal power unit j; represents the maximum power output of RES i; Indicates the power value of the rigid load m; is the total load of demand response user n; represents the upper limit of the power of conventional thermal power unit j, represents the lower power limit of conventional thermal power unit j; represents the maximum upward slope of conventional thermal power unit j, represents the maximum downward slope of conventional thermal power unit j, ζ -1 ={t|2,3,…,T};

[0042] The CDL calculation model is as follows:

[0043]

[0044]

[0045]

[0046] Among them, CDL n is the normalized CDL of demand response user n, which is a T-dimensional vector; CDL avg represents the average CDL, which is obtained by the following formula (8);

[0047]

[0048] represents the optimal power output of the conventional generator set obtained by solving equations (1)-(4); Γ is a line set indexed by l, with a cardinality of L; P l line,max is the upper limit of the transmission capacity of line l, which is obtained by the following formula (9);

[0049] P line =X×P Δ ; (9)

[0050] N D is the cardinality of the user set; P line is the line power, P line is an L×T dimensional matrix; X is the power transfer distribution factor, X is an L×K dimensional matrix; P Δ is the node injection power, which is a K × T dimensional matrix, obtained from (10);

[0051]

[0052] κ is the set of nodes indexed by k, with cardinality K, J k represents the set of conventional thermal power users in node k, I k represents the set of RES users in node k, M k represents the set of rigid load users in node k, N k represents the set of DR users in node k.

[0053] It can be seen from the above description that without considering network constraints, all users from different nodes share the same CDL, which ensures fairness when CDL is used as a criterion for user contribution evaluation. However, the CDL of each node may be different due to the influence of network constraints, so the difference between each CDL should be minimized to ensure the fairness of the evaluation. The modeling of the CDL of each node in the present invention does not rely on artificial assumptions. The CDL can be directly obtained based on conventional units, new energy, rigid loads and network topology. The calculation process is clear, and there is no need to use the contribution ratio of the generator set to the node load (this information cannot be obtained, and artificial assumptions are subjective). The calculation results are reasonable and fair, and their effectiveness can be demonstrated. There is no need for multiple iterations to solve, which greatly improves the solution efficiency.

[0054] Furthermore, the weekly load curve is calculated as follows:

[0055] Obtain the user's historical electricity consumption data;

[0056] Divide the total time period T into N Υ The time period is represented as s, and the set Υ is obtained, s∈Υ, so the historical load of user n in each time period of a day can be obtained:

[0057]

[0058] in is a time set of s periods, whose cardinality is S; q∈{1,2,…,7} represents the qth day of a week;

[0059] The typical load for the qth day of the week is calculated as:

[0060]

[0061] Among them, Q q is the set of all dates belonging to the qth day of the week, Q represents the set Q q The number of elements.

[0062] It can be seen from the above description that the present invention adopts the concept of typical weekly load to describe the typical load curve of the user, which improves the typicality and representativeness of the curve.

[0063] Furthermore, the adjustment range is obtained specifically as follows:

[0064] Compare the weekly load with the extreme values ​​of historical power consumption of users participating in DR to obtain the regulation range of DR users:

[0065]

[0066] in, and are the adjustment limits of user n at time t on the qth day of a week; and are the maximum and minimum historical electricity consumption of user n on the qth day of a week at time t,

[0067] From the above description, it can be seen that the regulation range of DR users is obtained by comparing the typical weekly load with the extreme values ​​of historical electricity consumption of users participating in DR.

[0068] Furthermore, the optimal decision model is established:

[0069]

[0070]

[0071]

[0072] in, is the power consumption of user n at time t; n represents the comfort coefficient of user n, λ n ∈[0,1]; is the typical weekly load of user n at time t. The value depends on the day of the week when the user participates in demand response and makes decisions. When the real-time time is the qth day of the week,

[0073] From the above description, it can be seen that users will adjust their electricity consumption to improve the similarity between their load curve and CDL. For the system, the higher the similarity, the greater the absorption of RES. However, for the user itself, excessive adjustment will lead to a loss of electricity comfort. The conflict between promoting the regulation of renewable energy and maintaining the comfort of electricity consumption should be balanced. The present invention introduces the user's comfort factor λ in the optimal decision model. n , a response strategy can be flexibly formulated for users: if the user has a high requirement for comfort, the coefficient can be set larger; if the user does not have a high requirement for comfort, the coefficient can be set smaller, so that the user can participate in demand response as much as possible and make greater contributions.

[0074] Please refer to Figure 2 , an optimization terminal for grid demand response taking into account electricity comfort, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0075] S1. Calculate the ideal load curve CDL of the grid node according to the network constraints, RES and rigid load of the power system;

[0076] S2. Generate the user's weekly load curve and adjustment range based on the user's historical electricity consumption data;

[0077] S3. Send the CDL, the weekly load curve and the adjustment range to the home energy management system HEMS, which establishes and solves the optimal decision model in combination with the user's comfort factor to obtain an optimized power management plan.

[0078] From the above description, it can be seen that the beneficial effects of the present invention are: an optimization method and terminal for power grid demand response taking into account electricity comfort of the present invention calculates the CDL of each node according to network constraints and uncontrollable parts of the power system (RES and rigid loads). As the ideal load curve of each node, CDL will be published to the corresponding users to guide users to reshape and adjust electricity consumption; typical weekly load curves and adjustment ranges are extracted to characterize the user's adjustability; the home energy management system considers the user's electricity comfort and subjective electricity preference, and establishes an optimal decision model, and automatically solves the best electricity plan for the user according to the optimal decision model. The plan uses the preference coefficient to balance the contradiction between following CDL and maintaining electricity comfort, so that users can participate in demand response while maintaining electricity comfort, thereby being able to more accurately and effectively adjust the user's electricity plan and improve the effectiveness of demand response.

[0079] Furthermore, the calculation of CDL is as follows:

[0080] Under the constraints (2)-(4), the power output of the conventional generator set is determined with the goal of minimizing the total operating cost (1):

[0081]

[0082]

[0083]

[0084]

[0085] Among them, a j ,b j ,c j is the cost parameter of conventional thermal power unit j; ζ is the time slot set indexed by t, and the cardinality of the set is T; j represents the index of the conventional thermal power unit user; i represents the index of the RES user, m represents the index of the rigid load user, n represents the index of the DR user, and J, I, M and N are the sets corresponding to j, i, m and n; represents the power output of conventional thermal power unit j; represents the maximum power output of RES i; Indicates the power value of the rigid load m; is the total load of demand response user n; represents the upper limit of the power of conventional thermal power unit j, represents the lower power limit of conventional thermal power unit j; represents the maximum upward slope of conventional thermal power unit j, represents the maximum downward slope of conventional thermal power unit j, ζ -1 ={t|2,3,…,T};

[0086] The CDL calculation model is as follows:

[0087]

[0088]

[0089]

[0090] Among them, CDL n is the normalized CDL of demand response user n, which is a T-dimensional vector; CDL avg represents the average CDL, which is obtained by the following formula (8);

[0091]

[0092] represents the optimal power output of the conventional generator set obtained by solving equations (1)-(4); Γ is a line set indexed by l, with a cardinality of L; P l line,max is the upper limit of the transmission capacity of line l, which is obtained by the following formula (9);

[0093] P line =X×P Δ ; (9)

[0094] N D is the cardinality of the user set; P line is the line power, P line is an L×T dimensional matrix; X is the power transfer distribution factor, X is an L×K dimensional matrix; P Δ is the node injection power, which is a K × T dimensional matrix, obtained from (10);

[0095]

[0096] κ is the set of nodes indexed by k, with cardinality K, J k represents the set of conventional thermal power users in node k, I k represents the set of RES users in node k, M k represents the set of rigid load users in node k, N k represents the set of DR users in node k.

[0097] It can be seen from the above description that without considering network constraints, all users from different nodes share the same CDL, which ensures fairness when CDL is used as a criterion for user contribution evaluation. However, the CDL of each node may be different due to the influence of network constraints, so the difference between each CDL should be minimized to ensure the fairness of the evaluation. The modeling of the CDL of each node in the present invention does not rely on artificial assumptions. The CDL can be directly obtained based on conventional units, new energy, rigid loads and network topology. The calculation process is clear, and there is no need to use the contribution ratio of the generator set to the node load (this information cannot be obtained, and artificial assumptions are subjective). The calculation results are reasonable and fair, and their effectiveness can be demonstrated. There is no need for multiple iterations to solve, which greatly improves the solution efficiency.

[0098] Furthermore, the weekly load curve is calculated as follows:

[0099] Obtain the user's historical electricity consumption data;

[0100] Divide the total time period T into N γ The time period is represented as s, and the set γ is obtained, s∈γ, so the historical load of user n in each time period of a day can be obtained:

[0101]

[0102] in is a time set of s periods, whose cardinality is S; q∈{1,2,…,7} represents the qth day of a week;

[0103] The typical load for the qth day of the week is calculated as:

[0104]

[0105] Among them, Q q is the set of all dates belonging to the qth day of the week, Q represents the set Q q The number of elements.

[0106] It can be seen from the above description that the present invention adopts the concept of typical weekly load to describe the typical load curve of the user, which improves the typicality and representativeness of the curve.

[0107] Furthermore, the adjustment range is obtained specifically as follows:

[0108] Compare the weekly load with the extreme values ​​of historical power consumption of users participating in DR to obtain the regulation range of DR users:

[0109]

[0110] in, and are the adjustment limits of user n at time t on the qth day of a week; and are the maximum and minimum historical electricity consumption of user n on the qth day of a week at time t,

[0111] From the above description, it can be seen that the regulation range of DR users is obtained by comparing the typical weekly load with the extreme values ​​of historical electricity consumption of users participating in DR.

[0112] Furthermore, the optimal decision model is established:

[0113]

[0114]

[0115]

[0116] in, is the power consumption of user n at time t; n represents the comfort coefficient of user n, λ n ∈[0,1]; is the typical weekly load of user n at time t. The value depends on the day of the week when the user participates in demand response and makes decisions. When the real-time time is the qth day of the week,

[0117] From the above description, it can be seen that users will adjust their electricity consumption to improve the similarity between their load curve and CDL. For the system, the higher the similarity, the greater the absorption of RES. However, for the user itself, excessive adjustment will lead to a loss of electricity comfort. The conflict between promoting the regulation of renewable energy and maintaining the comfort of electricity consumption should be balanced. The present invention introduces the user's comfort factor λ in the optimal decision model. n , a response strategy can be flexibly formulated for users: if the user has a high requirement for comfort, the coefficient can be set larger; if the user does not have a high requirement for comfort, the coefficient can be set smaller, so that the user can participate in demand response as much as possible and make greater contributions.

[0118] The present invention provides a method and terminal for optimizing power grid demand response taking into account power consumption comfort, which is suitable for optimizing power system demand response, making demand response more effective in adjusting participants' power consumption plans, thereby ensuring the effectiveness of demand response.

[0119] Please refer to Figure 1 as well as Figures 3 to 10 , Embodiment 1 of the present invention is:

[0120] A method for optimizing power grid demand response taking into account power consumption comfort comprises the following steps:

[0121] S1. Calculate the ideal load curve CDL of the grid node according to the network constraints, RES and rigid load of the power system.

[0122] In this embodiment, the CDL modeling is:

[0123] Power fluctuations in the power system originate from uncontrollable devices (e.g., rigid loads and RES). On the other hand, controllable parts (e.g., conventional generators and flexible loads) help offset fluctuations. CDL is the ideal load curve shape for each node user, which can alleviate fluctuations caused by the unadjustable part. It should be noted that when calculating CDL, we assume that conventional units output according to the optimal economic operation and that renewable energy is at maximum output. In actual demand response, there is always a deviation between the load curve and CDL, so conventional generators need to be adjusted or even renewable energy output needs to be reduced.

[0124] Under the constraints (2)-(4), the power output of the conventional generator set is determined with the goal of minimizing the total operating cost (1):

[0125]

[0126]

[0127]

[0128]

[0129] Among them, a j ,b j ,c j is the cost parameter of conventional thermal power unit j; ζ is the time slot set indexed by t, and the cardinality of the set is T; j represents the index of the conventional thermal power unit user; i represents the index of the RES user, m represents the index of the rigid load user, n represents the index of the DR user, and J, I, M and N are the sets corresponding to j, i, m and n; represents the power output of conventional thermal power unit j; represents the maximum power output of RES i; Indicates the power value of the rigid load m; is the total load of demand response user n; represents the upper limit of the power of conventional thermal power unit j, represents the lower power limit of conventional thermal power unit j; represents the maximum upward slope of conventional thermal power unit j, represents the maximum downward slope of conventional thermal power unit j, ζ -1 ={t|2,3,…,T}.

[0130] It should be noted that (2) represents the balance constraint of the sum of the powers of all time slots, and the power balance satisfied by each time slot will be further described in the following CDL formula.

[0131] Without considering network constraints, all users from different nodes share the same CDL, which ensures fairness when using CDL as a criterion for user contribution evaluation. However, the CDL of each node may be different due to the influence of network constraints. The schematic diagram of the CDL formula considering network constraints is shown in Figure 3 shown.

[0132] The difference between each CDL should be minimized to ensure the fairness of the evaluation. Therefore, the CDL calculation model is as follows:

[0133]

[0134]

[0135]

[0136] Among them, CDL n is the normalized CDL of demand response user n, which is a T-dimensional vector; CDL avg represents the average CDL, which is obtained by the following formula (8);

[0137]

[0138] represents the optimal power output of the conventional generator set obtained by solving equations (1)-(4); Γ is a line set indexed by l, with a cardinality of L; P l line,max is the upper limit of the transmission capacity of line l, which is obtained by the following formula (9);

[0139] P line =X×P Δ ; (9)

[0140] N D is the cardinality of the user set; P line is the line power, P line is an L×T dimensional matrix; X is the power transfer distribution factor, X is an L×K dimensional matrix; P Δ is the node injection power, which is a K × T dimensional matrix, obtained from (10);

[0141]

[0142] κ is the set of nodes indexed by k, with cardinality K, J k represents the set of conventional thermal power users in node k, I krepresents the set of RES users in node k, M k represents the set of rigid load users in node k, N k represents the set of DR users in node k.

[0143] S2. Generate the user's weekly load curve and adjustment range based on the user's historical electricity consumption data.

[0144] In this embodiment, we characterize the user's regulation capability through weekly load curve and regulation range:

[0145] The concept of typical weekly load is used to describe the typical load curve of the user, which improves the typicality and representativeness of the curve. The data used is the historical power consumption of the user without DR. First, the total time period T is divided into N Υ The time period is represented as s, and the set Υ is obtained, s∈Υ, so the historical load of user n in each time period of a day can be obtained:

[0146]

[0147] in is a time set of s periods, whose cardinality is S; q∈{1,2,…,7} represents the qth day of a week;

[0148] The typical load for the qth day of the week is calculated as:

[0149]

[0150] Among them, Q q is the set of all dates belonging to the qth day of the week, Q represents the set Q q It should be noted that the typical weekly load curve may be different to take into account the seasonal effects of electricity consumption. Therefore, it is necessary to recalculate the typical weekly load every month or two months based on the data in the corresponding time range.

[0151] The regulation range of DR users is derived by comparing the typical weekly load with the extreme values ​​of the historical electricity consumption of DR participating users:

[0152]

[0153] in, and are the adjustment limits of user n at time t on the qth day of a week;

[0154] and are the maximum and minimum historical electricity consumption of user n on the qth day of a week at time t,

[0155]

[0156] S3. Send the CDL, the weekly load curve and the adjustment range to the home energy management system HEMS, which establishes and solves the optimal decision model in combination with the user's comfort factor to obtain an optimized power management plan.

[0157] In this embodiment, after CDL is published to each node, the user will adjust the electricity consumption to improve the similarity between its load curve and CDL. For the system, the higher the similarity, the greater the absorption of RES. However, for the user itself, excessive adjustment will lead to a loss of electricity comfort. The conflict between promoting the regulation of renewable energy and maintaining the comfort of electricity consumption should be balanced. To this end. Establish a user optimal decision model and use the preference coefficient to express the user's preference for electricity comfort:

[0158]

[0159]

[0160]

[0161] in, is the power consumption of user n at time t; n represents the comfort coefficient of user n, λ n ∈[0,1]; is the typical weekly load of user n at time t. The value depends on the day of the week when the user participates in demand response and makes decisions. When the real-time time is the qth day of the week,

[0162] Coefficient λ n The higher the value, the higher the requirement for power consumption comfort. n When it reaches 1, user n exits DR and follows a typical consumption pattern, making no contribution to system regulation. n When it reaches 0, the user n will spare no effort to reduce the deviation between the load curve and CDL without considering comfort. The user can set the coefficient value in advance according to his or her subjective preference. Then, HEMS automatically provides the best consumption plan based on the decision model.

[0163] The proposed optimal decision model is a convex quadratic optimization problem. Therefore, its solution can be proved to be unique and can be solved by the interior point method or optimizer such as GUROBI.

[0164] like Figure 4The framework of CDL-based DR is shown in the figure. First, the load service agency (LSE) calculates the CDL of each node according to the network constraints and the uncontrollable parts of the power system (RES and rigid loads). As the ideal load curve for each node, CDL will be published to the corresponding users to guide them to reshape and adjust their electricity consumption. Then, typical weekly load curves and adjustment ranges are extracted to characterize the user's adjustability. Finally, the home energy management system (HEMS) automatically solves the optimal electricity consumption plan for the user based on the optimal decision model. The plan uses the preference coefficient to balance the contradiction between following CDL and maintaining electricity comfort.

[0165] Figure 5 The modified IEEE 9-bus system shown is used to test the effectiveness of the proposed model. The test system consists of 3 conventional generators, 1 photovoltaic (PV) generator, 2 wind turbines, and 3 node loads, including DR users and non-DR users. The demand response users of each node load are aggregated and treated as a whole to participate in demand response, and then make the best decision in the test. The demand response percentage is defined as the ratio of the total power consumption of demand response users to the total energy consumption of all users. The RES maximum output data and user power consumption data are from PJM and Open Energy Data Initiative, respectively. We use the GUROBI optimizer under the MATLAB platform to implement the simulation test on a 3.2GHz laptop with 16GB RAM. Section B discusses the decision results of user power consumption from the user level. Due to space limitations, a typical weekly load on the first day of the week is taken as an example. Section C analyzes the effect of user participation in DR from the system level.

[0166] Figure 6-8 Describes the power consumption of demand response users, with a demand response percentage of 45% and preference coefficients ranging from 0 to 1. Figure 6-8 As shown in Figure 1, under network constraints, the CDL of each user is different. According to the basic network topology, user 1 belongs to node 5 where the wind turbine is located, so the profile of CDL is similar to the profile of wind power output (e.g. Figure 6 ), thus accommodating more wind power and reducing the probability of congestion on adjacent lines. Figure 7 and Figure 8The CDL of users 2 and 3 are more similar to the superposition of PV and wind power generation, respectively, because users 2 and 3 are located between PV and wind turbines. In addition, the simulation results clearly show that as λ increases, the users' power consumption deviates from their CDL and approaches their typical weekly load, thereby improving their power consumption comfort. Therefore, users can achieve DR goals by flexibly adjusting λ while maintaining their own comfort in power consumption. It is worth noting that due to limited adjustability, even if λ reaches 0, the user's power consumption cannot be consistent with their CDL.

[0167] Fig. 9 The net load and renewable energy curtailment with or without DR are shown. When there is no DR, λ is set to 1. When DR is involved, λ is set to 0.7 in the test. Net load is defined as the total load minus RES output, which reflects the system power fluctuation that needs to be balanced by traditional power units. The results show that the net load with DR increases during valleys and decreases during peaks compared to without DR. Therefore, the proposed DR scheme effectively reduces the peak-to-valley difference of net load and improves the flexibility of system regulation. It can also be seen that the RES output without DR is lower than the maximum output of 5h-7h and 21h-24h, so the renewable energy curtailment power at the corresponding time. After implementing demand response, RES output increases and approaches its maximum output compared to the case without demand response, thereby avoiding RES curtailment from 5h-6h and 22h-24h, and significantly reducing RES curtailment from 7h and 21h.

[0168] Fig.10 Describes the change of renewable energy curtailment power with λ and DR percentage. Fig.10 As shown in the figure, the power of renewable energy abandonment decreases as λ decreases, which means that users contribute more to RES absorption by sacrificing their electricity comfort. When λ is reduced to a lower level (such as 0.5), the power of renewable energy abandonment is reduced to 0, indicating that RES is fully absorbed. Therefore, there is no need to further reduce λ, which will result in more loss of electricity comfort for users but will not benefit the absorption of renewable energy. In addition, it can be seen that when λ is large (for example, 0.6-0.7), RES abandonment decreases as the DR percentage increases. Therefore, increasing the number of demand response users can also promote the absorption of renewable energy, although they are more willing to maintain their original electricity usage habits for their own electricity comfort.

[0169] Please refer to Figure 2 , Embodiment 2 of the present invention is:

[0170] An optimization terminal 1 for grid demand response taking into account electricity comfort, comprising a processor 2, a memory 3, and a computer program stored in the memory 3 and executable on the processor 2, wherein the processor 2 implements the steps of an optimization method for grid demand response taking into account electricity comfort in the above embodiment 1 when executing the computer program.

[0171] In summary, the present invention provides an optimization method and terminal for grid demand response taking into account electricity comfort, calculates the CDL of each node according to network constraints and the uncontrollable parts of the power system (RES and rigid loads), and as the ideal load curve of each node, the CDL will be published to the corresponding users to guide the users to reshape and adjust their electricity consumption; extracts typical weekly load curves and adjustment ranges to characterize the user's adjustability; the home energy management system considers the user's electricity comfort and subjective electricity preference, establishes an optimal decision model, and automatically solves the best electricity plan for the user according to the optimal decision model. The plan uses the preference coefficient to balance the contradiction between following the CDL and maintaining electricity comfort, so that users can participate in demand response while maintaining electricity comfort, thereby being able to more accurately and effectively adjust the user's electricity plan and improve the effectiveness of demand response.

[0172] CDL is issued to guide users' power consumption management, taking into account users' regulation capabilities and electricity comfort, thereby improving the effectiveness of demand response implementation.

[0173] Users of each node use the CDL of their own node as a guidance signal, rather than the CDL of the entire system, which can free up their flexibility within network constraints.

[0174] Users can participate in DR to promote the system's consumption of RES while maintaining their electricity comfort level.

[0175] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An optimization method for grid demand response considering electricity comfort, It is characterized in that Includes steps: S1. Calculate the ideal load curve CDL of the grid node according to the network constraints, RES and rigid load of the power system; The calculation of CDL is as follows: Under the constraints (2)-(4), the power output of the conventional generator set is determined with the goal of minimizing the total operating cost (1): in, b j ,c j is the cost parameter of conventional thermal power unit j; ζ is the time slot set indexed by t, and the cardinality of the time slot set ζ is T; j represents the index of the conventional thermal power unit user; i represents the index of the RES user, m represents the index of the rigid load user, and n represents the index of the demand response user. J, I, M and N are the sets corresponding to j, i, m and n; represents the power output of conventional thermal power unit j; represents the maximum power output of RES i; Indicates the power value of the rigid load m; is the total load of demand response user n; represents the upper limit of the power of conventional thermal power unit j, represents the lower power limit of conventional thermal power unit j; represents the maximum upward slope of conventional thermal power unit j, represents the maximum downward slope of conventional thermal power unit j, ζ -1 ={t|2,3,…,T}; The CDL calculation model is as follows: Among them, CDL n is the normalized CDL of demand response user n, which is a T-dimensional vector; CDL avg represents the average CDL, which is obtained by the following formula (8); represents the optimal power output of the conventional generator set obtained by solving equations (1)-(4); Γ is a line set indexed by l, with a cardinality of L; P l line,max is the upper limit of the transmission capacity of line l, which is obtained by the following formula (9); P line =X×P Δ ; (9) N D is the cardinality of the user set; P line is the line power, P line is an L×T dimensional matrix; X is the power transfer distribution factor, X is an L×K dimensional matrix; P Δ is the node injection power, which is a K × T dimensional matrix, obtained from (10); κ is the set of nodes indexed by k, with cardinality K, J k represents the set of conventional thermal power users in node k, I k represents the set of RES users in node k, M k represents the set of rigid load users in node k, N k represents the set of demand response users in node k; S2. Generate the user's weekly load curve and adjustment range based on the user's historical electricity consumption data; The weekly load curve is calculated as: Obtain the user's historical electricity consumption data; Divide the total time period T into N Υ The time period is represented as s, and the set Υ is obtained, s∈Υ, so the historical load of user n in each time period of a day is obtained: in is a time set of s periods, with cardinality S; q∈{1,2,…,7} represents the qth day of a week; The typical load for the qth day of the week is calculated as: Among them, Q q is the set of all dates belonging to the qth day of the week, Q represents the set Q q The number of elements of ; S3. Send the CDL, the weekly load curve and the adjustment range to the home energy management system HEMS, which establishes and solves the optimal decision model in combination with the user's comfort factor to obtain an optimized power management plan.

2. According to claim 1, a method for optimizing grid demand response taking into account electricity comfort, It is characterized in that The acquisition of the adjustment range is specifically as follows: Compare the weekly load with the extreme values ​​of historical electricity consumption of users participating in DR to obtain the regulation range of demand response users: in, and are the adjustment limits of user n at time t on the qth day of a week; and are the maximum and minimum historical electricity consumption of user n on the qth day of a week at time t, 3. The method for optimizing grid demand response taking into account electricity comfort according to claim 1, It is characterized in that Establishing the optimal decision model: in, is the power consumption of demand response user n at time t; n represents the comfort factor of demand response user n, λ n ∈[0,1]; is the typical weekly load of the extracted demand response user n at time t. The value depends on the day of the week when the user participates in the demand response and makes a decision. When the real-time time is the qth day of the week, 4. An optimization terminal for grid demand response taking into account electricity comfort, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the steps in the method for optimizing grid demand response taking into account electricity comfort as described in any one of claims 1 to 3 above are implemented.

Images