Method and device for predicting production load rate of enterprise

Abstract

The application discloses a kind of enterprise production load rate prediction method and device, it is related to electric power big data technical field.The method comprises: obtaining the electricity consumption of target enterprise in the period to be predicted;According to the electricity consumption and the production load rate prediction model generated in advance, the production load rate prediction value of the period to be predicted is determined;Wherein, the production load rate prediction model is obtained according to the clustering analysis of the multiple historical electricity data of target enterprise and the corresponding multiple historical production load rate data.This application can obtain the quantitative value of enterprise production load rate according to power data, compared with prior art, which can only judge the enterprise stop limit production according to the electricity data, which has been significantly improved;This application only needs the yield data corresponding to the characteristic electricity of enterprise to establish the relationship between electricity and yield;This application can be applied to various industrial enterprises, has wide applicability, and the technical cost is lower, easy to popularize and apply.

Description

Technical Field

[0001] This application relates to the field of power big data technology, specifically to a method and apparatus for predicting enterprise production load rate. Background Technology

[0002] Electricity big data is fundamental data for socio-economic operations. By analyzing electricity big data, we can understand the operational patterns of enterprises and further determine their production and operational status based on real-time electricity consumption data. In recent years, electricity big data has played an increasingly important role in areas such as resumption of work and production analysis, air pollution prevention and control monitoring, and orderly electricity use.

[0003] A company's production load rate refers to the ratio of its actual output to its maximum output; it is a direct parameter characterizing the company's production status. Electricity consumption is an important indicator of a company's production load rate. Currently, existing technologies characterize a company's production load rate by comparing current electricity consumption with maximum electricity consumption. However, due to the complexity of actual production conditions, the electricity consumption per unit of product is usually not a constant but varies with the company's production status and scale. Therefore, how to more accurately predict a company's production load rate based on the characteristics of electricity data has become an urgent problem to be solved. Summary of the Invention

[0004] To address the problems existing in the prior art, in a first aspect, this application provides a method for predicting enterprise production load rate, comprising:

[0005] Obtain the target company's electricity consumption during the forecast period;

[0006] The predicted value of the production load rate for the period to be predicted is determined based on the electricity consumption and the pre-generated production load rate prediction model; wherein, the production load rate prediction model is obtained by cluster analysis based on multiple historical electricity data and corresponding multiple historical production load rate data of the target enterprise.

[0007] In one embodiment, the step of establishing the production load rate prediction model includes:

[0008] Obtain multiple historical electricity consumption data for the target company;

[0009] Cluster analysis was performed on the multiple historical electricity data to obtain multiple cluster data and the center electricity value of each cluster data;

[0010] The production load rate prediction model is generated based on the power consumption values ​​of each center and their corresponding production load rates.

[0011] In one embodiment, the step of performing cluster analysis on the multiple historical electricity data to obtain multiple sets of cluster data and corresponding multiple central electricity values ​​includes:

[0012] Initial clustering step: Use k electricity values ​​from the multiple historical electricity data as initial center values ​​to cluster the multiple historical electricity data, resulting in k initial cluster data; the k electricity values ​​are randomly selected from the multiple historical electricity data; k is a positive integer;

[0013] The steps for determining the center are as follows: K corresponding center power values ​​are determined based on the historical power data in each cluster data group.

[0014] Re-clustering step: Based on the k center power values, the multiple historical power data are clustered again to obtain k sets of updated clustered data;

[0015] Repeat the center determination step and the re-clustering step until the k center power values ​​remain unchanged.

[0016] In one embodiment, the method for predicting enterprise production load factor further includes:

[0017] The clustering error is determined based on the multiple historical power data and the k central power values ​​when the clustered data is divided into k groups.

[0018] The optimal number of clusters K is determined based on the clustering error when k takes different values.

[0019] In one embodiment, generating the production load rate prediction model based on the power consumption values ​​of each center and their corresponding production load rates includes:

[0020] Obtain the power values ​​of the k centers and their corresponding production load rates when k is the optimal cluster size K;

[0021] Based on the k central power values ​​and their corresponding production load rates, a functional relationship between power values ​​and production load rates is established to obtain the production load rate prediction model.

[0022] Secondly, this application also provides a device for predicting enterprise production load rate, comprising:

[0023] The electricity consumption acquisition module is used to acquire the electricity consumption of the target enterprise during the forecast period;

[0024] The production load rate prediction module is used to determine the predicted value of the production load rate for the period to be predicted based on the electricity consumption and the pre-generated production load rate prediction model; wherein, the production load rate prediction model is obtained by cluster analysis based on multiple historical electricity data and corresponding multiple historical production load rate data of the target enterprise.

[0025] In one embodiment, the enterprise production load rate prediction device further includes:

[0026] The historical data acquisition module is used to acquire multiple historical electricity consumption data of the target enterprise;

[0027] The cluster center determination module is used to perform cluster analysis on the multiple historical electricity data to obtain multiple sets of cluster data and the center electricity value of each set of cluster data;

[0028] The model building module is used to generate the production load rate prediction model based on the power consumption values ​​of each center and their corresponding production load rates.

[0029] In one embodiment, the cluster center determination module includes a cluster center determination unit, used for:

[0030] Initial clustering: Using k electricity values ​​from the multiple historical electricity data as initial center values, the multiple historical electricity data are clustered to obtain k initial cluster data; the k electricity values ​​are randomly selected from the multiple historical electricity data; k is a positive integer;

[0031] Center determination: Determine the corresponding k center power values ​​based on the historical power data in each group of cluster data;

[0032] Re-clustering: Based on the k central power values, the multiple historical power data are clustered again to obtain k sets of updated clustered data;

[0033] Repeat the center determination step and the re-clustering step until the k center power values ​​remain unchanged.

[0034] In one embodiment, the cluster center determination module further includes an optimal cluster number determination unit, used for:

[0035] The clustering error is determined based on the multiple historical power data and the k central power values ​​when the clustered data is divided into k groups.

[0036] The optimal number of clusters K is determined based on the clustering error when k takes different values.

[0037] In one embodiment, the model building module is specifically used for:

[0038] Obtain the power values ​​of the k centers and their corresponding production load rates when k is the optimal cluster size K;

[0039] Based on the k central power values ​​and their corresponding production load rates, a functional relationship between power values ​​and production load rates is established to obtain the production load rate prediction model.

[0040] Thirdly, this application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any enterprise production load rate prediction method provided in this application.

[0041] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the enterprise production load rate prediction method provided in this application.

[0042] The method and apparatus for predicting enterprise production load rate in this application have the following advantages:

[0043] (1) This application can obtain a quantitative value of the enterprise's production load rate based on power data, which is a significant improvement compared to the prior art, which can usually only determine the enterprise's production stoppage or restriction based on power data.

[0044] (2) The existing technology for establishing the relationship between electricity and output usually requires a large amount of enterprise output data as support, but it is usually difficult to obtain a large amount of enterprise output data. This application only requires the output data corresponding to the characteristic electricity of the enterprise, which effectively solves this problem.

[0045] (3) This application can be applied to various industrial enterprises, has wide applicability, and has low technical cost, making it easy to promote and apply. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0047] Figure 1 A schematic diagram illustrating the enterprise production load rate prediction method provided in this application.

[0048] Figure 2 A schematic diagram illustrating the steps involved in establishing a production load rate prediction model provided in this application.

[0049] Figure 3 This is a schematic diagram illustrating another step in establishing a production load rate prediction model provided in this application.

[0050] Figure 4 This is a schematic diagram illustrating another step in establishing a production load rate prediction model provided in this application.

[0051] Figure 5This is a schematic diagram illustrating another step in establishing a production load rate prediction model provided in this application.

[0052] Figure 6 This is a schematic diagram of the kS curve provided in this application.

[0053] Figure 7 This is a schematic diagram of the clustering results of the historical electricity data provided in this application.

[0054] Figure 8 This is a schematic diagram of the production load rate prediction model provided in this application.

[0055] Figure 9 A schematic diagram of a device for predicting enterprise production load rate provided in this application.

[0056] Figure 10 Another schematic diagram of the enterprise production load rate prediction device provided in this application.

[0057] Figure 11 Another schematic diagram of the enterprise production load rate prediction device provided in this application.

[0058] Figure 12 A schematic diagram of a computer device provided in this application. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0060] It should be noted that the enterprise production load rate prediction method and device of this application can be used in the field of power big data technology, or in any field other than the field of power big data technology. This application does not limit the application field of the enterprise production load rate prediction method and device.

[0061] Firstly, this application provides a method for predicting enterprise production load rate. For example... Figure 1 As shown, the method includes:

[0062] Step S101: Obtain the electricity consumption of the target enterprise during the forecast period;

[0063] Step S102: Determine the predicted value of the production load rate for the period to be predicted based on the electricity consumption and the pre-generated production load rate prediction model.

[0064] The production load rate prediction model is derived through cluster analysis of multiple historical electricity consumption data and corresponding historical production load rate data for the target enterprise. The production load rate prediction model can output the corresponding production load rate based on the input electricity consumption.

[0065] Therefore, this application can obtain a quantitative value of the enterprise's production load rate based on power data. The technical solution of this application can be applied to various industrial enterprises, has wide applicability, and has low technical cost, making it easy to promote and apply.

[0066] In one embodiment, the enterprise production load rate prediction method of this application further includes the step of establishing the production load rate prediction model. For example... Figure 2 As shown, the steps for establishing the production load rate prediction model include:

[0067] Step S201: Obtain multiple historical electricity data of the target enterprise.

[0068] Specifically, based on the time resolution of the target enterprise's production load rate to be predicted, historical electricity data with the same time resolution are collected to form a sample set X = {x1, x2, ..., x}. n}, x i (1≤i≤n, where n is a positive integer) represents the historical electricity consumption data, where x n This represents the highest historical electricity consumption. The time resolution for electricity consumption data can be daily, hourly, 15-minute, or any other time resolution available from the power grid company.

[0069] It is important to note that, in order to ensure the accuracy of the model, the historical electricity data should cover the electricity consumption corresponding to the company's typical production load rate, and the historical electricity data should be consistent with the company's production process during the period to be predicted.

[0070] Step S202: Perform cluster analysis on the multiple historical power data to obtain multiple cluster data and the center power value of each cluster data.

[0071] Specifically, the multiple historical electricity data X = {x1, x2, ..., x...} n Clustering is performed to obtain k clusters of data, where k is a positive integer. Preferably, the value of k in this application ranges from 1 to 10. Then, for each cluster, the central electricity value of the cluster is determined based on the historical electricity data contained in that cluster. Thus, k central electricity values ​​are ultimately obtained, represented as M = {m1, m2, ..., m...} k}

[0072] Step S203, based on the power values ​​of each center M={m1,m2,…,m k} and its corresponding production load factor Y = {y1, y2, ..., y kGenerate the production load rate prediction model.

[0073] Specifically, the production load rate corresponding to the electricity consumption of each center is obtained through on-site enterprise surveys, questionnaires, or voluntary reporting by enterprises. Based on the correspondence between the electricity consumption of each center and the production load rate, a functional model of enterprise electricity consumption (M) and production load rate (Y) is established, which is the production load rate prediction model obtained in this step.

[0074] In one embodiment, such as Figure 3 As shown, in step S202, cluster analysis is performed on the multiple historical electricity data to obtain multiple sets of cluster data and corresponding multiple central electricity values, including:

[0075] Step S2021: Randomly select k power values ​​from the plurality of historical power data, denoted as m = {m 10 m 20 , ..., m k0};

[0076] Step S2022, Initial Clustering Step: Using the k electricity values ​​selected in step S2021 as initial center values, cluster the multiple historical electricity data X = {x1, x2, ..., x...} n Clustering is performed to obtain k initial cluster data sets, where k is greater than n;

[0077] Specifically, for the sample set X = {x1, x2, ..., x...} n A historical battery data x i (i = 1, 2, ..., n), calculate x respectively. i With m={m 10 ,m 20 ,…,m k0 The central values ​​m in} j0 The distances to (j = 1, 2, ..., k) yield the distance set D = {d} i1 ,d i2 ,…,d ik}, distance d ij The calculation formula is as follows:

[0078] d ij =|x i -m j0 | 2

[0079] Where, x i ∈X, m j0 ∈m, i=1,2,…,n, j=1,2,…,k.

[0080] Determine the minimum value from the distance set D, let's say it's d. ij Then x iDivide to m j0 In the corresponding clustered data group.

[0081] Using the above method, the historical electricity data in sample set X can be divided into k clusters {C1, C2, ..., C...} k At this point, there are k clustered data sets {C1, C2, ..., C}. k The corresponding center power value is m={m 10 m 20 , ..., m k0}

[0082] Step S2023, Center Determination Step: Determine the corresponding k center power values ​​based on the historical power data in each group of cluster data.

[0083] For each clustering data C obtained in step S021 j (j=1,2,…,k), recalculate C j The central energy value of all historical energy data is used to update the previous central energy value. The specific formula is as follows:

[0084]

[0085] Where, x ji ∈C j N is the clustering data group C j The total number of historical electricity data in (j=1,2,…,k), m j For clustered data group C j The updated value of the central power value corresponding to (j=1,2,…,k).

[0086] Using the above method, we can obtain k clustered data sets {C1, C2, ..., C...}. k The updated center power value is denoted as M = {m1, m2, ..., m}. k}

[0087] Step S2024: Determine the updated k center power values ​​M = {m1, m2, ..., m} k If the current value corresponds to the previous central power value, proceed to step S2026; otherwise, proceed to steps S2025 and S2023 in sequence.

[0088] Step S2025, re-clustering step: Based on the updated k center power values ​​M = {m1, m2, ..., m kThe multiple historical electricity data are clustered again to obtain k sets of updated cluster data. The process of clustering again to obtain k sets of updated cluster data is similar to the initial clustering step in step S2021. The only difference is that the initial center value is replaced with the updated center electricity value in step S2022. They can be referred to each other during implementation.

[0089] Step S2026: The latest k clustering data and their corresponding k central power values ​​M = {m1, m2, ..., m} are then processed. k Output.

[0090] In one embodiment, such as Figure 4 As shown, according to steps S2021 to S2026 above, k central electrical values ​​M = {m1, m2, ..., m} corresponding to different values ​​of k are determined respectively. k Following this, the method for predicting enterprise production load rate further includes:

[0091] Step S2027: Determine the clustering error corresponding to the k groups of clustered data based on the multiple historical power data and the k central power values.

[0092] Specifically, the clustering error S is calculated according to the following formula:

[0093]

[0094] Among them, C j For the j-th cluster data, j = 1, 2, ..., k, x ji ∈C j Let i = 1, 2, ..., N, where N is the clustering data set C. j The total number of historical electricity data in (j=1,2,…,k), m j Let be the central electricity value corresponding to the j-th cluster data.

[0095] The clustering error S for different values ​​of k can be obtained using the above method.

[0096] Step S2028: Determine the optimal number of clusters K based on the clustering error when k takes different values.

[0097] Specifically, plot the graph with k as the horizontal axis and S as the vertical axis (see...). Figure 6 The kS curve is obtained. The k value corresponding to the elbow of the kS curve is selected as the optimal number of clusters K.

[0098] In one embodiment, such as Figure 5 As shown, step S203, generating the production load rate prediction model based on the power consumption values ​​of each center and their corresponding production load rates, includes:

[0099] Step S2031: Obtain the electrical values ​​M = {m1, m2, ..., m} of the K centers when k is the optimal number of clusters. K} and its corresponding production load factor Y = {y1, y2, ..., y K};

[0100] Step S2032, based on the K central power values ​​M = {m1, m2, ..., m K} and its corresponding production load factor Y = {y1, y2, ..., y K Establish a functional relationship between the electricity consumption value x and the production load rate y to obtain the production load rate prediction model.

[0101] The specific formula is as follows:

[0102] y = α·x + b;

[0103]

[0104] b = y j -α·m j ;

[0105] x∈[m j ,m j+1 ];j∈[0,K];

[0106] Where x is the electricity consumption value for the period to be predicted, y is the production load rate for the period to be predicted, and m j,j＝0 =0,m j,j≠0 ∈M,m j,j＝K+1 =x n x n This is the highest battery capacity in history. j,j＝0 =0,y j,j≠0 ∈Y, y j,j＝K+1 =1.

[0107] Substituting the electricity consumption value x for the period to be predicted into the above formula yields the production load rate y for the period to be predicted. This formula is the production load rate prediction model of this application. Figure 8 This is a schematic diagram of the production load rate prediction model.

[0108] The following is a specific embodiment provided in this application, which demonstrates the production load rate prediction process of a cement enterprise in city A based on big data on electricity.

[0109] Step 1: The cement company has been in normal production for the past three years, and its production process has not changed. Three years of historical daily electricity consumption data for a cement production company in city A were collected from the power grid company, forming an electricity consumption dataset X. The maximum daily electricity consumption over the three years was 30,000 kWh.

[0110] The second step is to perform cluster analysis on the electricity consumption data. The technical methods are as described above, and in practice, this process can be implemented using computer programs such as Python. Through cluster analysis, the functional relationship between the number of clusters and the clustering error S is obtained, as follows: Figure 6 As shown. By Figure 6 It can be determined that the number of clusters corresponding to the elbow of the curve is 4. Therefore, the optimal number of clusters for the electricity consumption of this cement company is 4. The classification of electricity consumption is as follows: Figure 7 As shown in the figure, based on the above method, the four power center values ​​corresponding to the cluster number of 4 are obtained, which are 7000 kWh, 12000 kWh, 17000 kWh and 25000 kWh respectively.

[0111] Step 3: Conduct on-site surveys of enterprises to find that the production load rate of enterprises is 20% when the electricity consumption is 7000 kWh, 50% when the electricity consumption is 12000 kWh, 70% when the electricity consumption is 17000 kWh, and 85% when the electricity consumption is 25000 kWh.

[0112] Step 4: Establish the functional relationship between the company's electricity consumption (x) and production load rate (y). Based on the data in Step 3, the functional relationship can be obtained as follows:

[0113]

[0114]

[0115]

[0116]

[0117]

[0118] For details, please refer to Figure 8 .

[0119] The fifth step is to calculate the production load rate of the enterprise corresponding to the given electricity consumption. Assuming the electricity consumption is 15500 kWh, substituting it into the formula in the fourth step, the enterprise's production load rate can be calculated to be 64%.

[0120] When the enterprise's electricity consumption is measured in hourly data, a similar method can be used to calculate it in the above embodiments.

[0121] In summary, the enterprise production load rate prediction method of this application has the following advantages:

[0122] (1) This application can obtain a quantitative value of the enterprise's production load rate based on power data, which is a significant improvement compared to the prior art, which can usually only determine the enterprise's production stoppage or restriction based on power data.

[0123] (2) The existing technology for establishing the relationship between electricity and output usually requires a large amount of enterprise output data as support, but it is usually difficult to obtain a large amount of enterprise output data. This application only requires the output data corresponding to the characteristic electricity of the enterprise, which effectively solves this problem.

[0124] (3) This application can be applied to various industrial enterprises, has wide applicability, and has low technical cost, making it easy to promote and apply.

[0125] Based on the same inventive concept, this application also provides an apparatus for predicting enterprise production load rate, which can be used to implement the method described in the above embodiments, as described in the following embodiments. Since the principle of the enterprise production load rate prediction apparatus is similar to that of the enterprise production load rate prediction method, the implementation of the enterprise production load rate prediction apparatus can refer to the implementation of the enterprise production load rate prediction method, and repeated details will not be elaborated further. As used below, the terms "unit" or "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the system described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0126] Secondly, this application also provides a device for predicting enterprise production load rate, such as... Figure 9 As shown, the device includes:

[0127] Electricity consumption acquisition module 301 is used to acquire the electricity consumption of the target enterprise during the forecast period;

[0128] The production load rate prediction module 302 is used to determine the predicted value of the production load rate for the period to be predicted based on the electricity consumption and the pre-generated production load rate prediction model; wherein, the production load rate prediction model is obtained by cluster analysis based on multiple historical electricity data and corresponding multiple historical production load rate data of the target enterprise.

[0129] In one embodiment, such as Figure 10 As shown, the enterprise production load rate prediction device further includes:

[0130] Historical data acquisition module 303 is used to acquire multiple historical electricity data of the target enterprise;

[0131] The cluster center determination module 304 is used to perform cluster analysis on the multiple historical electricity data to obtain multiple sets of cluster data and the center electricity value of each set of cluster data;

[0132] The model building module 305 is used to generate the production load rate prediction model based on the power values ​​of each center and their corresponding production load rates.

[0133] In one embodiment, such as Figure 11 As shown, the cluster center determination module 304 includes a cluster center determination unit 3041, used for:

[0134] Initial clustering: Using k electricity values ​​from the multiple historical electricity data as initial center values, the multiple historical electricity data are clustered to obtain k initial cluster data; the k electricity values ​​are randomly selected from the multiple historical electricity data; k is a positive integer;

[0135] Center determination: Determine the corresponding k center power values ​​based on the historical power data in each group of cluster data;

[0136] Re-clustering: Based on the k central power values, the multiple historical power data are clustered again to obtain k sets of updated clustered data;

[0137] Repeat the center determination step and the re-clustering step until the k center power values ​​remain unchanged.

[0138] In one embodiment, please continue to see Figure 11 The cluster center determination module 304 further includes an optimal cluster number determination unit 3042, used for:

[0139] The clustering error is determined based on the multiple historical power data and the k central power values ​​when the clustered data is divided into k groups.

[0140] The optimal number of clusters K is determined based on the clustering error when k takes different values.

[0141] In one embodiment, the model building module 305 is specifically used for:

[0142] Obtain the power values ​​of the k centers and their corresponding production load rates when k is the optimal cluster size K;

[0143] Based on the k central power values ​​and their corresponding production load rates, a functional relationship between power values ​​and production load rates is established to obtain the production load rate prediction model.

[0144] The enterprise production load rate prediction device of this application has the following advantages: (1) This application can obtain a quantitative value of the enterprise production load rate based on power data, which is a significant improvement compared with the prior art, which can usually only judge the enterprise's production stoppage or restriction based on power data; (2) The prior art establishes the relationship between power and output, which usually requires a large amount of enterprise output data as support, and it is usually difficult to obtain a large amount of enterprise output data. This application only requires the output data corresponding to the characteristic power of the enterprise, which effectively solves this problem; (3) This application can be applied to various industrial enterprises, has wide applicability, and has low technical cost, making it easy to promote and apply.

[0145] In one embodiment, this application also provides a computer device, see [link to previous document]. Figure 12 The electronic device 100 specifically includes:

[0146] The system includes a central processing unit (CPU) 110, a memory 120, a communication module 130, an input unit 140, an output unit 150, and a power supply 160.

[0147] The memory 120, communication module 130, input unit 140, output unit 150, and power supply 160 are all connected to the central processing unit 110. The memory 120 stores a computer program, which the central processing unit 110 can call. When the central processing unit 110 executes the computer program, it implements all the steps in the enterprise production load rate prediction method described in the above embodiments.

[0148] In one embodiment, the present application also provides a computer-readable storage medium for storing a computer program that can be executed by a processor. When executed by a processor, the computer program implements the enterprise production load rate prediction method provided by the present invention.

[0149] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0150] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0151] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0152] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0153] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for predicting enterprise production load rate, characterized in that, include: Obtain the target company's electricity consumption during the forecast period; The predicted value of the production load rate for the period to be predicted is determined based on the electricity consumption and the pre-generated production load rate prediction model; wherein, the production load rate prediction model is obtained by cluster analysis based on multiple historical electricity consumption data and corresponding multiple historical production load rate data of the target enterprise; The steps for establishing the production load rate prediction model include: Obtain multiple historical electricity consumption data for the target company; Cluster analysis was performed on the multiple historical electricity data to obtain multiple cluster data and the center electricity value of each cluster data; The production load rate prediction model is generated based on the electricity values ​​of each center and their corresponding production load rates. The cluster analysis of the multiple historical electricity data yields multiple sets of cluster data and corresponding multiple central electricity values, including: Initial clustering step: Clustering the data from the multiple historical electricity data sets... k Using individual electricity values ​​as initial center values, the multiple historical electricity data are clustered to obtain... k Initial clustering data for the group; k Each power consumption value is randomly selected from the multiple historical power consumption data; k It is a positive integer; The center determination steps are as follows: The corresponding center is determined based on the historical electricity data in each cluster of data. k Individual center power value; Re-clustering step: based on the described k The central power values ​​are then clustered again based on the multiple historical power data to obtain... k Group update cluster data; Repeat the center determination step and the re-clustering step until the... k The power value of each center remains unchanged; The method further includes: based on the plurality of historical power data and the... k The clustering data is determined by the individual power values. k The clustering error corresponding to the group; according to k Determining the optimal number of clusters based on the clustering error with different values K； The step of generating the production load rate prediction model based on the power consumption values ​​of each center and their corresponding production load rates includes: obtaining... k Find the optimal number of clusters K time k Each center's power value and its corresponding production load rate; according to the above k A functional relationship between the power consumption value and the corresponding production load rate is established for each central power consumption value and its corresponding production load rate, thus obtaining the production load rate prediction model.

2. A device for predicting enterprise production load rate, characterized in that, include: The electricity consumption acquisition module is used to acquire the electricity consumption of the target enterprise during the forecast period; The production load rate prediction module is used to determine the predicted value of the production load rate for the period to be predicted based on the electricity consumption and the pre-generated production load rate prediction model; wherein, the production load rate prediction model is obtained by cluster analysis based on multiple historical electricity data and corresponding multiple historical production load rate data of the target enterprise. The historical data acquisition module is used to acquire multiple historical electricity consumption data of the target enterprise; The cluster center determination module is used to perform cluster analysis on the multiple historical electricity data to obtain multiple sets of cluster data and the center electricity value of each set of cluster data; The model building module is used to generate the production load rate prediction model based on the power values ​​of each center and their corresponding production load rates. The cluster center determination module includes a cluster center determination unit, used for: Initial clustering: Clustering the data from the multiple historical electricity data sets... k Using individual electricity values ​​as initial center values, the multiple historical electricity data are clustered to obtain... k Initial clustering data for the group; k Each power consumption value is randomly selected from the multiple historical power consumption data; k It is a positive integer; The center determined: based on the historical electricity data in each cluster of data, the corresponding... k Individual center power value; Re-clustering: based on the described k The central power values ​​are then clustered again based on the multiple historical power data to obtain... k Group update cluster data; Repeat the center determination step and the re-clustering step until the... k The power value of each center remains unchanged; The cluster center determination module further includes an optimal cluster number determination unit, used for: Based on the aforementioned historical power data and the k The clustering data is determined by the individual power values. k Clustering error corresponding to grouping time; according to k The optimal number of clusters K is determined by taking different values ​​of the clustering error. The model building module is specifically used for: Get k Find the optimal number of clusters K time k The power consumption values ​​of each center and their corresponding production load rates; According to the above k A functional relationship between the power consumption value and the corresponding production load rate is established for each central power consumption value and its corresponding production load rate, thus obtaining the production load rate prediction model.

3. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the enterprise production load rate prediction method of claim 1.

4. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the enterprise production load rate prediction method of claim 1.

Images