Hierarchical evaluation method and hierarchical evaluation system for schedulable demand response potential of high-energy-load industrial user

Abstract

The invention discloses a high-energy-load industrial user schedulable demand response potential hierarchical evaluation method and system, and relates to the technical field of high-energy-load industrial user schedulable demand response potential hierarchical evaluation, and the method comprises the steps: inputting a user historical annual load curve and a response parameter of a flexible resource; obtaining typical load curves of K users; constructing a theoretical demand response potential evaluation model of the high-energy-load industrial user, and calculating the theoretical demand response potential of the user in the demand response time period based on the typical load curve of the user; constructing an interactive response model of three flexible resources of translation, transferability and load reduction of the high-energy-load industrial user; and calculating the schedulable demand response potential of the user in the demand response time period. According to the method, the DTW K-means + + algorithm is adopted, similarity matching is carried out on the sequences through the dynamic programming technology, the similarity between samples can be measured more effectively, and support is provided for the key technology needed by high-energy-load industrial users participating in demand response.

Description

Technical Field

[0001] The present invention relates to the technical field of hierarchical evaluation of dispatchable demand response potential of high-energy-consuming industrial users, and in particular to a method and system for hierarchical evaluation of dispatchable demand response potential of high-energy-consuming industrial users. Background Art

[0002] As the proportion of new energy continues to increase, coping with the volatility and uncertainty of new energy power and ensuring the balance of power supply and demand have become the biggest challenges facing the dispatching and operation of new power systems. Demand response, as an effective means of optimizing the dispatching of power systems, can improve the stability and economy of power systems by encouraging users to reduce power consumption during peak hours or increase power consumption during off-peak hours. The industrial sector, especially high-energy-load industrial users (such as electrolytic aluminum, steel, cement and other industries), has great potential for demand response due to its high energy consumption and heavy load, and has become an important target of demand-side management.

[0003] There have been many studies on the evaluation of demand response potential of high-energy-load industrial users, but the existing studies have the following deficiencies: the identification of typical power consumption patterns of users often uses traditional clustering methods to cluster the characteristic indicators contained in the load data. Although this method greatly reduces the amount of calculation, it ignores the large amount of information contained in the time series; in addition, the current evaluation methods are often single-dimensional and fail to conduct a hierarchical analysis of the demand response potential of high-energy-load industrial users from multiple levels and multiple angles, making it difficult to fully and accurately grasp the user's dispatchable potential. In this context, there is an urgent need for a demand response potential hierarchical evaluation method that can comprehensively consider the characteristics of high-energy-load industrial users and provide multi-level evaluation results, so as to provide technical support for power system optimization scheduling, load management, and user participation in demand response. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is: how to solve the problem that current evaluation methods often use traditional clustering methods to cluster characteristic indicators contained in load data in a single dimension.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a hierarchical evaluation method for the dispatchable demand response potential of high-energy-load industrial users, comprising: determining the demand response time period specified by the day-ahead invitation in which the high-energy-load industrial users are to participate, and inputting the user's historical annual load curve and the response parameters of the flexibility resources; performing cluster analysis based on the DTW K-means++ algorithm after preprocessing the historical annual load curves of the high-energy-load industrial users, and using the elbow rule to determine the optimal number of clusters to obtain K typical load curves of the users; constructing a theoretical demand response potential evaluation model for high-energy-load industrial users, and calculating the theoretical demand response potential of the users within the demand response time period based on the users' typical load curves; constructing an interactive response model for three types of flexibility resources of high-energy-load industrial users, namely, shiftable, transferable, and reducible loads; constructing a dispatchable demand response potential evaluation model for high-energy-load industrial users considering the technical constraints of the flexibility resources and the output constraints of the users, and calculating the dispatchable demand response potential of the users within the demand response time period.

[0007] As a preferred scheme of the hierarchical evaluation method of the dispatchable demand response potential of high-load industrial users described in the present invention, wherein: the demand response time period includes the time period for determining the valley filling type day-ahead invitation and the peak shaving type day-ahead invitation demand response according to the system peak-shaving demand; the response parameters of the flexibility resources include the proportion of three types of flexibility resources, namely, shiftable, transferable, and reducible loads, the acceptable load transfer time period, the load shifting duration, the maximum number of load transfers, and the maximum load output change per unit time.

[0008] As a preferred solution of the hierarchical evaluation method for the dispatchable demand response potential of high-load industrial users described in the present invention, the preprocessing includes smoothing, denoising and normalizing the annual load curve by using the mean smoothing method; the mean smoothing method includes inputting the annual load curve of 24 sampling points of the user within 365 days, the number of smoothing times N, the smoothing threshold th, calculating the absolute value of the difference between two consecutive sampling points, recorded as dl, judging whether the difference dl is less than the smoothing threshold th, if dl is less than th, it is replaced by the average value of the two, if dl is greater than or equal to th, the sampling point value remains unchanged, according to the number of smoothing times N, the cyclic execution judges whether the difference dl is less than the smoothing threshold th, and outputs the annual load curve after data smoothing; the cluster analysis includes introducing the DTW distance instead of the Euclidean distance to measure the similarity between samples, and using the DTW K-means++ algorithm to perform cluster analysis on the annual load curve; the DTW distance is expressed as:

[0009]

[0010] Among them, d DTW (L,D) is the DTW distance between time series L and D, Y(w i) is the path distance of the i-th point in the regular path, i is the i-th point in the regular path, p is the total number of points in the regular path, w i is the point coordinate of the i-th point in the regularized path; the elbow rule is used to determine the optimal number of clusters, including using the K-means++ algorithm to calculate y initial cluster centers, calculating the DTW distances of the remaining samples to the cluster centers, assigning each sample to the category closest to the cluster center, calculating the average value of the data in each category, obtaining a new cluster center, and repeating the operation until the cluster center remains unchanged; drawing an elbow diagram, selecting the inflection point as the number of cluster centers, repeating the steps of determining the cluster centers, and determining the optimal number of clusters k.

[0011] As a preferred solution of the hierarchical evaluation method for the dispatchable demand response potential of high-energy-consuming industrial users of the present invention, the theoretical demand response potential evaluation model of the high-energy-consuming industrial users is expressed as:

[0012]

[0013] Among them, P shave The theoretical peak shaving potential of high-power industrial users, P fill Theoretical valley-filling potential for high-energy-consuming industrial users, N shave is the total number of hours in the peak shaving period, N fill is the total number of hours in the valley filling period, P t is the load value of the user at time t, P avg is the average load of the user, t is the time t, Ω shave is the peak shaving period, Ω fill It is the valley filling period.

[0014] As a preferred solution of the hierarchical evaluation method of the dispatchable demand response potential of high-energy-load industrial users of the present invention, the interactive response model includes an interactive response model of high-energy-load industrial users' load shifting, an interactive response model of high-energy-load industrial users' load transfer, and an interactive response model of high-energy-load industrial users' load reduction; the interactive response model of high-energy-load industrial users' load shifting includes calling compensation costs and constraints, which are expressed as:

[0015]

[0016] S shift =[t shift- ,t shift+ -T shift +1]U{t s}

[0017]

[0018] Among them, Cshift is the total interactive benefit of the translatable load, c shift is the unit capacity compensation price of the movable load at time t, is the moving-in response state of the translatable load in period t, is the displacement response state of the translatable load in period t, P t shift,in is the load translation of the user in period t, [t shift- ,t shift+ ] is the acceptable load transfer time interval, t s is the load removal start time, T shift is the load translation duration, T is the load translation duration, S shift is the set of starting time periods for translation, and τ1 is the starting time of translation.

[0019] As a preferred solution of the hierarchical evaluation method of the dispatchable demand response potential of high-energy-load industrial users of the present invention, the interactive response model of the transferable load of the high-energy-load industrial users includes the call compensation fee and the constraint, which is expressed as:

[0020]

[0021] Among them, C trans is the total interactive benefit of transferable load, c trans is the unit capacity compensation price of the transferable load in period t, The shift-in response state of the transferable load in period t, is the shift-out response state of the transferable load in period t, P t trans,in is the load transfer amount of the user in period t, P t trans,out is the load removal of the user in period t, P t trans,max is the maximum load that can be transferred in period t, is the minimum removal duration of the transferable load, is the transfer time interval where the load is acceptable. is the load transfer-out period interval that is acceptable, and τ2 is the transfer start time; the interactive response model of the load reduction of the high-load industrial user includes the call compensation fee and constraints, which is expressed as:

[0022]

[0023] Among them, C cut is the total interactive benefit of load reduction, c cut is the unit capacity compensation price of the load that can be reduced in period t, is the response state of the load that can be reduced in period t, P t cut is the load reduction of the user in period t, is the maximum number of load reductions allowed in a dispatch cycle, is the minimum load reduction duration, is the maximum load reduction duration that can be reduced, is the load reduction period, P t cut,max is the maximum load reduction, and τ3 is the start time of reduction.

[0024] As a preferred scheme of the hierarchical evaluation method for the dispatchable demand response potential of high-energy-load industrial users described in the present invention, the dispatchable demand response potential evaluation model includes the comprehensive optimization of the load variance of high-energy-load industrial users and the electricity cost after participating in the demand response, and takes the independent optimal solutions of the two sub-goals as their respective benchmark values, and constructs a comprehensive objective function after weighted processing.

[0025] Another object of the present invention is to provide a system for a hierarchical evaluation method of the dispatchable demand response potential of high-energy-load industrial users, which can solve the problem of hierarchical evaluation of the dispatchable demand response potential of high-energy-load industrial users by constructing a hierarchical evaluation system for the dispatchable demand response potential of high-energy-load industrial users.

[0026] In order to solve the above technical problems, the present invention provides the following technical solutions: a hierarchical evaluation system for the dispatchable demand response potential of high-load industrial users, including a data acquisition module, a cluster analysis module, a theoretical demand response potential module, an interactive response model module and a dispatchable demand response potential evaluation module; the data acquisition module is used to determine the demand response time period specified by the day-ahead invitation to which the high-load industrial users want to participate, and input the user's historical annual load curve and the response parameters of the flexible resources; the cluster analysis module is used to perform cluster analysis based on the DTWK-means++ algorithm after preprocessing according to the historical annual load curve of the high-load industrial users, and use the elbow rule to determine the most The optimal number of clusters is obtained to obtain the typical load curves of K users; the theoretical demand response potential module is used to construct a theoretical demand response potential evaluation model for high-energy-load industrial users, and based on the typical load curves of users, the theoretical demand response potential of users within the demand response time period is calculated; the interactive response model module is used to construct an interactive response model of three types of flexibility resources for high-energy-load industrial users, namely, shiftable, transferable and reducible loads; the dispatchable demand response potential evaluation module is used to construct a dispatchable demand response potential evaluation model for high-energy-load industrial users considering the technical constraints of flexibility resources and the output constraints of users, and calculate the dispatchable demand response potential of users within the demand response time period.

[0027] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned hierarchical evaluation method for the dispatchable demand response potential of high-energy-consuming industrial users are implemented.

[0028] A computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the above-mentioned hierarchical evaluation method for the dispatchable demand response potential of high-energy-consuming industrial users.

[0029] The beneficial effects of the present invention are as follows: the hierarchical evaluation method for the dispatchable demand response potential of high-energy-load industrial users provided by the present invention first constructs a theoretical demand response potential evaluation model, and then constructs the dispatchable demand response potential of high-energy-load industrial users considering technology, output and economy, forming a set of hierarchical evaluation methods for demand response potential for high-energy-load industrial users, which can provide support for the key technologies required for high-energy-load industrial users to participate in demand response.

[0030] Existing load curve clustering research mostly uses traditional clustering algorithms, which can only reflect the numerical differences between the sampling points corresponding to two samples and ignore the large amount of information contained in the time series. The present invention adopts the DTW K-means++ algorithm and uses dynamic programming technology to match the sequences by similarity, which can more effectively measure the similarity between samples. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0032] Figure 1 An overall flow chart of a hierarchical evaluation method for the dispatchable demand response potential of high-energy-consuming industrial users provided in the first embodiment of the present invention.

[0033] Figure 2 A flow chart of a hierarchical evaluation method for the dispatchable demand response potential of high-energy-consuming industrial users provided in the first embodiment of the present invention.

[0034] Figure 3 A schematic diagram of improving the theoretical demand response potential of high-energy-consuming industrial users according to the hierarchical evaluation method for the dispatchable demand response potential of high-energy-consuming industrial users provided in the first embodiment of the present invention.

[0035] Figure 4 A structural diagram of a hierarchical assessment system for dispatchable demand response potential of high-energy-consuming industrial users provided in the second embodiment of the present invention.

[0036] Figure 5 The third embodiment of the present invention provides a method for evaluating the dispatchable demand response potential of high-energy-consuming industrial users, and the distribution of cement plant flexibility resources before evaluation.

[0037] Figure 6 The elbow diagram of the hierarchical evaluation method for dispatchable demand response potential of high-energy-consuming industrial users provided in the third embodiment of the present invention determines k.

[0038] Figure 7 A clustering result diagram of a hierarchical evaluation method for dispatchable demand response potential of high-energy-consuming industrial users provided in the third embodiment of the present invention.

[0039] Figure 8 This is the cement plant dispatchable demand response potential assessment result of the high-energy-consuming industrial user dispatchable demand response potential hierarchical assessment method provided in the third embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0041] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0042] Example 1, reference Figure 1 to Figure 3 , which is the first embodiment of the present invention, and provides a hierarchical evaluation method for the dispatchable demand response potential of high-load industrial users, including: determining the demand response time period specified by the day-ahead invitation that the high-load industrial users want to participate in, and inputting the user's historical annual load curve and the response parameters of the flexibility resources; according to the historical annual load curve of the high-load industrial users, cluster analysis is performed based on the DTW K-means++ algorithm after preprocessing, and the elbow rule is used to determine the optimal number of clusters to obtain K typical load curves of the users; a theoretical demand response potential evaluation model for high-load industrial users is constructed, and based on the user's typical load curve, the theoretical demand response potential of the user in the demand response time period is calculated; an interactive response model of three types of flexibility resources of high-load industrial users, namely, shiftable, transferable, and reducible loads, is constructed; a dispatchable demand response potential evaluation model for high-load industrial users is constructed considering the technical constraints of the flexibility resources and the output constraints of the users, and the dispatchable demand response potential of the users in the demand response time period is calculated.

[0043] like Figure 2 As shown, the specific evaluation steps include:

[0044] S1. Determine the demand response time period specified in the day-ahead invitation that the high-energy-consuming industrial user wants to participate in, and input the user's historical annual load curve and the response parameters of the flexibility resources.

[0045] According to the system peak-shaving demand, determine the time period for high-load industrial users to participate in valley-filling day-ahead invitations and peak-shaving day-ahead invitations, and input the user's historical annual load curve and the response parameters of the flexibility resources, including the proportion of the three types of flexibility resources, namely, shiftable, transferable, and reducible loads, the acceptable load transfer time period, the load shifting duration, the maximum number of load transfers, the maximum load output change per unit time, etc.

[0046] S2. According to the historical annual load curves of high-energy-consuming industrial users, cluster analysis is performed based on the DTW K-means++ algorithm after preprocessing, and the elbow rule is used to determine the optimal number of clusters to obtain the typical load curves of K users.

[0047] S2.1. The preprocessing process of the load curve is to use the mean smoothing method to smooth, reduce noise and normalize the annual load curve. The smoothing algorithm process is:

[0048] (1) Input the annual load curve of the user at 24 sampling points within 365 days, the number of smoothing times N, and the smoothing threshold th.

[0049] (2) Calculate the absolute value of the difference between two consecutive sampling points, denoted as dl.

[0050] (3) Determine whether the difference dl is less than the smoothing threshold th. If so, replace it with the average of the two, otherwise the sampling point value remains unchanged.

[0051] (4) According to the number of smoothing times N, the third step is executed repeatedly.

[0052] (5) Annual load curve after output data smoothing.

[0053] S2.2. DTW distance is introduced to replace Euclidean distance to measure the similarity between samples, and DTW K-means++ algorithm is used to perform cluster analysis on the annual load curve.

[0054] For any two time series L = {l1,l2,…,l m} and D={d1,d2,…,d n}, construct auxiliary matrix A∈R m ×n , matrix element a i,j Representative i and d j The moving point is initially at a 1,1 The position of the element, for element a ijThe moving point of the position can only be a when it moves to the element next time. i+1,j 、a i+1,j+1 、a i,j+1 The position of an element in the element, repeatedly moving the moving point to a m,n Until now, the coordinates of all points on the moving path of the moving point form a regular path W = {w1, w2, ... w i ,…,w p Under the condition of satisfying the regular path constraint, according to dynamic time warping, the calculation formula of DTW distance is:

[0055]

[0056] Among them, d DTW (L,D) is the DTW distance between time series L and D, Y(w i ) is the path distance of the i-th point in the regular path.

[0057] S2.3. Use the DTW K-means++ algorithm to perform cluster analysis and use the elbow rule to determine the optimal number of clusters. The specific process is as follows:

[0058] (1) Select y cluster centers, specifically randomly select a sample point as the first cluster center, and calculate the DTW distance D(x) of each remaining sample point from the selected cluster center. i ), for each sample point, the probability of being selected as the next cluster center is:

[0059]

[0060] Select the sample with the largest P as the next cluster center, and repeat the above process until y cluster centers are selected.

[0061] (2) Calculate the DTW distance from the remaining samples to the cluster center.

[0062] (3) Assign each sample to the category closest to the cluster center.

[0063] (4) Calculate the average value of the data in each category and obtain the new cluster center.

[0064] (5) Repeat steps 2 to 4 until the cluster center remains unchanged.

[0065] (6) Draw the elbow diagram, select the inflection point as the number of cluster centers, repeat steps 2 to 5, and determine the optimal number of clusters k.

[0066] S3. Construct a theoretical demand response potential evaluation model for high-energy-consuming industrial users, and calculate the theoretical demand response potential of users within the demand response time period based on their typical load curves.

[0067] Based on the typical load curve of the user, the theoretical demand response potential of the user within the demand response time period is calculated. The schematic diagram of the theoretical demand response potential of high-energy-consuming industrial users is shown in Figure 3 As shown in the figure, a theoretical demand response potential assessment model for high-energy-consuming industrial users is constructed:

[0068]

[0069] Among them, P shave and P fill are the theoretical peak shaving and valley filling potentials of high-energy-consuming industrial users, N shave and N fill are the total hours in the peak shaving and valley filling periods, respectively. t is the load value of the user at time t, P avg is the average load of users.

[0070] S4. Construct an interactive response model for three types of flexibility resources for high-energy-load industrial users: load translation, transferability, and load reduction.

[0071] The interactive response model includes an interactive response model for high-energy-load industrial users to translate loads, an interactive response model for high-energy-load industrial users to transfer loads, and an interactive response model for high-energy-load industrial users to reduce loads.

[0072] S4.1. Establish an interactive response model for high-energy industrial users to shift loads, including invocation compensation costs and related constraints:

[0073]

[0074] S shift =[t shift- ,t shift+ -T shift +1]U{t s}(5)

[0075]

[0076] Among them, C shift is the total interactive benefit of the translatable load, c shift is the unit capacity compensation price of the shiftable load in period t, and are the moving-in and moving-out response states of the movable load in period t, P t shift,in is the load translation of the user in period t, [t shift- ,t shift+ ] is the acceptable load transfer time interval, ts is the load transfer start time, T shift is the load translation duration.

[0077] S4.2. Establish an interactive response model for high-energy-consuming industrial users to transfer loads, including call compensation costs and related constraints:

[0078]

[0079] Among them, C trans is the total interactive benefit of transferable load, c trans is the unit capacity compensation price of the transferable load in period t, and are the shift-in and shift-out response states of the transferable load in period t, respectively; P t trans,in and P t trans,out are the load inflow and outflow of the user in period t, P t trans,max is the maximum load that can be transferred in period t, is the minimum removal duration of the transferable load, and They are the load-acceptable transfer-in time period and load-out time period respectively.

[0080] S4.3. Establish an interactive response model for high-energy-consuming industrial users to reduce load, including invoking compensation costs and related constraints:

[0081]

[0082] Among them, C cut is the total interactive benefit of load reduction, c cut is the unit capacity compensation price of the load that can be reduced in period t, is the response state of the load that can be reduced in period t, P t cut is the load reduction of the user in period t, is the maximum number of load reductions allowed in a dispatch cycle, and are the minimum load reduction duration and the maximum load reduction duration, respectively. Indicates the time period during which load can be reduced.

[0083] S5. Considering the technical constraints of flexible resources and the output constraints of users, a dispatchable demand response potential evaluation model for high-energy-consuming industrial users is constructed to calculate the dispatchable demand response potential of users during the demand response period.

[0084] Considering the comprehensive optimality of the load variance of high-energy-consuming industrial users and the electricity cost after participating in demand response, the independent optimal solutions of the two sub-goals are taken as their respective benchmark values, and the comprehensive objective function is constructed after weighted processing. The objective function and constraints are as follows:

[0085] Objective function:

[0086]

[0087] F2=C BUY -C DR (20)

[0088]

[0089] C DR =C shift +C trans +C cut (twenty two)

[0090] Among them, F is the total objective function, F1 and F2 are the load variance of high-energy-consuming industrial users and the electricity cost after participating in demand response, respectively. * and are the independent optimal solutions for each sub-goal, ω1 and ω2 are the weights of each sub-goal, C BUY The cost of electricity purchased by users, C DR is the compensation income of users participating in demand response, L(t) is the load value of users at time t, e p 、e f 、e v They are the peak, flat and valley electricity prices corresponding to time-of-use electricity prices, and Δt is the duration of adjacent moments.

[0091] Constraints:

[0092] (1) Technical constraints:

[0093] Constraints (5)-(7), Constraints (9)-(12), Constraints (14)-(17).

[0094] (2) Output constraints:

[0095]

[0096] Where, k is the energy consumption value (MWh / t), E N Minimum production requirement.

[0097] Output high-energy-consuming industrial user theory - hierarchical evaluation results of dispatchable demand response potential.

[0098] Example 2, reference Figure 4, which is the second embodiment of the present invention, is different from the previous embodiment in that it provides a hierarchical evaluation system for the dispatchable demand response potential of high-energy-load industrial users, including: a data acquisition module 100, a cluster analysis module 200, a theoretical demand response potential module 300, an interactive response model module 400 and a dispatchable demand response potential evaluation module 500.

[0099] The data collection module 100 is used to determine the demand response time period specified by the day-ahead invitation that the high-energy-consuming industrial user wants to participate in, and input the user's historical annual load curve and the response parameters of the flexibility resources.

[0100] The cluster analysis module 200 is used to perform cluster analysis based on the DTWK-means++ algorithm after preprocessing according to the historical annual load curves of high-energy-consuming industrial users, and to determine the optimal number of clusters using the elbow rule to obtain typical load curves of K users.

[0101] The theoretical demand response potential module 300 is used to construct a theoretical demand response potential evaluation model for high-energy-consuming industrial users, and calculate the theoretical demand response potential of users within a demand response time period based on the typical load curve of the users.

[0102] The interactive response model module 400 is used to construct an interactive response model of three types of flexibility resources for high-energy-load industrial users: load translation, transferability, and load reduction.

[0103] The dispatchable demand response potential assessment module 500 is used to construct a dispatchable demand response potential assessment model for high-energy-load industrial users by considering the technical constraints of flexible resources and the production constraints of users, and calculate the dispatchable demand response potential of users in the demand response time period.

[0104] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0105] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0106] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0107] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0108] Example 3, reference Figures 5 to 8 , which is the third embodiment of the present invention, and which is different from the first two embodiments in that it is used to verify and illustrate the technical effects adopted in the present invention, so as to verify the real effects of this method.

[0109] A typical cement plant was selected as the research object. The daily load curve data recorded by the power data acquisition system has a granularity of 1h / point, with a total of 24 sampling points per day. After data preprocessing, 365 valid sample curves of the cement plant were obtained in one year. The flexibility resource response parameters of the cement plant are shown in Table 1. The flexibility resource distribution is shown in Figure 5As shown. According to the time-of-use electricity price, a day is divided into 6 time periods for analysis. The peak period is 9:00-11:00 and 17:00-21:00, and the electricity price is 0.82￥ / kWh; the normal period is 7:00-8:00 and 12:00-16:00, and the electricity price is 0.55￥ / kWh; the valley period is 0:00-6:00 and 22:00-23:00, and the electricity price is 0.27￥ / kWh.

[0110] Table 1 Flexibility resource response parameters of cement plants

[0111]

[0112] When using the DTW K-means++ algorithm for clustering, it is necessary to determine the number of clusters k, which is determined using the elbow method, such as Figure 6 shown.

[0113] It can be seen from the figure that the elbow is k = 2, the number of clusters is selected as 2, and the clustering results are as follows Figure 7 shown.

[0114] Based on the theoretical demand response potential evaluation model of high-energy-consuming industrial users, the dispatchable demand response potential evaluation model of high-energy-consuming industrial users is constructed by considering the technical constraints of flexible resources and the output constraints of users. The demand response potential hierarchical evaluation is performed on the load curves under two typical electricity consumption modes of cement plants. The hierarchical evaluation results of the dispatchable demand response potential of cement plants obtained from clustering result 1 are shown in Table 2 and Figure 8 shown.

[0115] Table 2 Results of theoretical demand response potential assessment

[0116]

[0117] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A hierarchical evaluation method for the dispatchable demand response potential of high-energy-consuming industrial users, characterized by: include, Determine the demand response time period specified in the day-ahead invitation that the high-energy-consuming industrial user wants to participate in, and input the user's historical annual load curve and the response parameters of the flexibility resource; According to the historical annual load curves of high-energy-consuming industrial users, cluster analysis is performed based on the DTW K-means++ algorithm after preprocessing, and the optimal number of clusters is determined using the elbow rule to obtain the typical load curves of K users; Construct a theoretical demand response potential evaluation model for high-energy-consuming industrial users, and calculate the theoretical demand response potential of users within the demand response period based on their typical load curves; Construct an interactive response model for three types of flexibility resources for high-energy-consuming industrial users: load shifting, transferability, and load reduction; Considering the technical constraints of flexible resources and the output constraints of users, a dispatchable demand response potential evaluation model for high-energy-consuming industrial users is constructed to calculate the dispatchable demand response potential of users during the demand response period.

2. The hierarchical evaluation method for dispatchable demand response potential of high-energy-consuming industrial users according to claim 1 is characterized in that: The demand response time period includes the time period for determining valley filling type day-ahead invitation and peak shaving type day-ahead invitation demand response according to the system peak shaving demand; The response parameters of the flexibility resources include the proportion of the three types of flexibility resources: shiftable, transferable, and reducible loads, the acceptable load transfer time period, the load shifting duration, the maximum number of load transfers, and the maximum load output change per unit time.

3. The hierarchical evaluation method for the dispatchable demand response potential of high-energy-consuming industrial users according to claim 2 is characterized in that: The preprocessing includes smoothing, denoising and normalizing the annual load curve using a mean smoothing method; The mean smoothing method includes inputting the annual load curve of 24 sampling points within 365 days of the user, the number of smoothing times N, the smoothing threshold th, calculating the absolute value of the difference between two consecutive sampling points, recorded as dl, judging whether the difference dl is less than the smoothing threshold th, if dl is less than th, replacing it with the average value of the two, if dl is greater than or equal to th, the sampling point value remains unchanged, and according to the number of smoothing times N, cyclically executing to judge whether the difference dl is less than the smoothing threshold th, and outputting the annual load curve after data smoothing; The cluster analysis includes introducing DTW distance instead of Euclidean distance to measure the similarity between samples, and using DTW K-means++ algorithm to perform cluster analysis on the annual load curve; The DTW distance is expressed as, Among them, d DTW (L,D) is the DTW distance between time series L and D, Y(w i ) is the path distance of the i-th point in the regular path, i is the i-th point in the regular path, p is the total number of points in the regular path, w i is the point coordinate of the i-th point in the regularized path; The method of using the elbow rule to determine the optimal number of clusters includes using the K-means++ algorithm to calculate y initial cluster centers, calculating the DTW distances of the remaining samples to the cluster centers, assigning each sample to the category closest to the cluster center, calculating the average value of the data in each category, obtaining a new cluster center, and repeating the operation until the cluster center remains unchanged; Draw the elbow diagram, select the inflection point as the number of cluster centers, repeat the steps of determining cluster centers, and determine the optimal number of clusters k.

4. The hierarchical evaluation method for dispatchable demand response potential of high-energy-consuming industrial users according to claim 3 is characterized in that: The theoretical demand response potential evaluation model of high-energy-consuming industrial users is expressed as: Among them, P shave The theoretical peak shaving potential of high-power industrial users, P fill Theoretical valley-filling potential for high-energy-consuming industrial users, N shave is the total number of hours in the peak shaving period, N fill is the total number of hours in the valley filling period, P t is the load value of the user at time t, P avg is the average load of the user, t is the time t, Ω shave is the peak shaving period, Ω fill It is the valley filling period.

5. The hierarchical evaluation method for dispatchable demand response potential of high-energy-consuming industrial users according to claim 4 is characterized in that: The interactive response model includes an interactive response model for high-energy-load industrial users to translate loads, an interactive response model for high-energy-load industrial users to transfer loads, and an interactive response model for high-energy-load industrial users to reduce loads; The interactive response model of the high-energy industrial user's shiftable load includes calling compensation costs and constraints, which is expressed as: S shift =[t shift- ,t shift+ -T shift +1]U{t s } Among them, C shift is the total interactive benefit of the translatable load, c shift is the unit capacity compensation price of the movable load at time t, is the moving-in response state of the translatable load in period t, is the displacement response state of the translatable load in period t, P t shift,in is the load translation of the user in period t, [t shift- ,t shift+ ] is the acceptable load transfer time interval, t s is the load removal start time, T shift is the load translation duration, T is the load translation duration, S shift is the set of starting time periods for translation, and τ1 is the starting time of translation.

6. The hierarchical evaluation method for dispatchable demand response potential of high-energy-consuming industrial users according to claim 5 is characterized in that: The interactive response model of the high-energy-consuming industrial user's transferable load includes the call compensation cost and constraints, which is expressed as: Among them, C trans is the total interactive benefit of transferable load, c trans is the unit capacity compensation price of the transferable load in period t, The shift-in response state of the transferable load in period t, is the removal response state of the transferable load in period t, P t trans,in is the load transfer amount of the user in period t, P t trans,out is the load removal of the user in period t, P t trans,max is the maximum load that can be transferred in period t, is the minimum removal duration of the transferable load, is the transfer time interval where the load is acceptable. is the transfer-out period interval that the load can accept, and τ2 is the transfer start time; The interactive response model of high-energy-consuming industrial users that can reduce load includes calling compensation costs and constraints, which is expressed as: Among them, C cut is the total interactive benefit of load reduction, c cut is the unit capacity compensation price of the load that can be reduced in period t, is the response state of the load that can be reduced in period t, P t cut is the load reduction of the user in period t, is the maximum number of load reductions allowed in a dispatch cycle, is the minimum load reduction duration, is the maximum load reduction duration that can be reduced, is the load reduction period, P t cut,max is the maximum load reduction, and τ3 is the start time of reduction.

7. The hierarchical evaluation method for dispatchable demand response potential of high-energy-consuming industrial users according to claim 6 is characterized in that: The dispatchable demand response potential assessment model includes a comprehensive optimization of the load variance of high-energy-consuming industrial users and the electricity cost after participating in demand response, taking the independent optimal solutions of the two sub-goals as their respective benchmark values, and constructing a comprehensive objective function after weighted processing.

8. A system using the hierarchical evaluation method for dispatchable demand response potential of high-energy-consuming industrial users as described in any one of claims 1 to 7, characterized in that: It includes a data collection module (100), a cluster analysis module (200), a theoretical demand response potential module (300), an interactive response model module (400) and a dispatchable demand response potential evaluation module (500); The data collection module (100) is used to determine the demand response time period specified by the day-ahead invitation that the high-energy-consuming industrial user wants to participate in, and input the user's historical annual load curve and the response parameters of the flexibility resources; The cluster analysis module (200) is used to perform cluster analysis based on the DTW K-means++ algorithm after preprocessing the historical annual load curves of high-energy-consuming industrial users, and to determine the optimal number of clusters using the elbow rule to obtain typical load curves of K users; The theoretical demand response potential module (300) is used to construct a theoretical demand response potential evaluation model for high-energy-consuming industrial users, and to calculate the theoretical demand response potential of the user within a demand response time period based on the user's typical load curve; The interactive response model module (400) is used to construct an interactive response model of three types of flexibility resources for high-energy-load industrial users: load translation, transferability, and load reduction; The dispatchable demand response potential assessment module (500) is used to construct a dispatchable demand response potential assessment model for high-energy-load industrial users by taking into account the technical constraints of flexible resources and the output constraints of users, and calculate the dispatchable demand response potential of users in the demand response time period.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the hierarchical evaluation method for the dispatchable demand response potential of high-energy-consuming industrial users described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the hierarchical evaluation method for the dispatchable demand response potential of high-energy-consuming industrial users described in any one of claims 1 to 7 are implemented.

Images