Minimum power utilization sensing method and system for multi-element load of industrial park

Abstract

The invention relates to an industrial park multi-element load minimum power utilization sensing method and system, and the method comprises the following steps: obtaining a plurality of industrial park key load cluster division results, generating a set of magneto-dependent sensor deployment scheme for each division result, and taking the scheme as a feasible solution based on the division result, further constructing a feasible solution set for minimizing power utilization perception; considering the comprehensive cost of the deployment scheme and the power utilization perception income to establish an evaluation model for minimizing the power utilization perception, and solving an optimal solution for minimizing the power utilization perception based on the evaluation model; and after the magnetic sensor deployment scheme corresponding to the optimal solution is deployed, generating power utilization sensing data assets of the key load cluster. According to the invention, power utilization sensing data assets of the whole industrial park can be obtained by installing a small number of intelligent circuit breakers (containing magnetic sensitive electrical quantity sensors), and technical support and data basis are provided for deduction of multi-element load power utilization regulation potential and accurate user portraits in the industrial park.

Description

Technical Field

[0001] The present invention belongs to the technical field of load power consumption perception, and more specifically, relates to a method for minimizing power consumption perception of multiple loads in an industrial park. The method can be applied to the fields of critical load cluster identification of multiple loads in an industrial park and minimized accurate perception of their power consumption status. Background Art

[0002] As an important entity on the load side, the observability and measurability of the power consumption status of industrial parks is of great significance for the state perception and flexible regulation of the load side of the power system. However, the current way of obtaining power consumption data in industrial parks is generally based on the overall measurement data of gateway meters, which makes it difficult to accurately obtain the detailed power consumption status of multiple loads within the park. Therefore, studying the power consumption perception technology of multiple loads behind the gateway meter of the industrial park can obtain the refined power consumption data of multiple loads, which will help improve the demand response and flexible regulation capabilities of the industrial park, promote friendly interaction between power grid supply and demand, and support lean production and upgrading of industrial parks.

[0003] At present, domestic and foreign research teams focus their research on load power consumption perception on two areas: load power consumption data measurement and load cluster division.

[0004] In the field of load power consumption data measurement, representative methods include direct measurement method and non-invasive load decomposition method, both of which aim to obtain the electrical quantity of the load by direct or indirect means, and then infer the load power consumption status. 1) Direct measurement method is suitable for the power consumption perception of a single load. For example, a smart meter is installed for a single load to obtain load power consumption data, but its cost is high, the installation process is cumbersome, and the installation location must be dry and ventilated. It is mainly suitable for sensing the power consumption status of high-power equipment, and it is difficult to deploy it on a large scale in industrial parks with low load rated power, large data volume, and large differences in load types. The magnetic sensitive electrical quantity sensing technology that has emerged in recent years can also realize the power consumption perception of a single load and overcome the above problems to a certain extent, but the high economic cost also restricts the large-scale promotion of load power consumption perception methods based on magnetic sensitive electrical quantity sensing technology. 2) Based on big data analysis, the non-invasive load decomposition method can decompose the load superposition signal from the current data to extract the load operation status. However, the types of multiple loads in industrial parks are complex, and the non-invasive load decomposition method generally has the defect of low accuracy in the power consumption perception of multiple loads in the park.

[0005] In the field of load cluster identification, domestic and foreign research teams have proposed a method for constructing industrial park load clusters based on algorithms such as rough sets, principal component analysis, grey correlation analysis, and cluster analysis, providing new technical ideas for the perception of power consumption of multiple loads in industrial parks. This type of method first identifies the multi-dimensional core power consumption characteristics of multiple loads in industrial parks, and then uses clustering algorithms to group multiple loads to obtain a series of key load clusters, and finally conducts power consumption perception for key load clusters. However, this type of method will cause problems with randomness in load cluster division and lack of evaluation methods for the pros and cons of different load cluster division results, and lacks research on industrial park power consumption perception data assets.

[0006] The electricity consumption sensing method for industrial parks disclosed in the existing technical documents focuses on the division of clusters, and only briefly introduces the work after the division. The post-division scheme fails to take into account the economy and accuracy of electricity consumption sensing, and only infers other loads from the power consumption data of representative loads in space. Summary of the invention

[0007] In order to solve the deficiencies in the prior art, in view of the randomness of the load cluster division method described in the above background technology and the technical defects that the existing industrial park load power consumption sensing method fails to finely sense the power consumption status of multiple loads, the present invention proposes a method for minimizing power consumption sensing of multiple loads in an industrial park. The core concept at least includes: based on the identification results of the key load clusters in the industrial park, taking the key load clusters as the power consumption sensing objects, first designing multiple magnetic sensor deployment schemes according to all possible key load cluster division results, and using these deployment schemes as the feasible solution set of the minimization power consumption sensing method; then, considering the comprehensive cost and power consumption sensing benefits of various deployment schemes, constructing an evaluation model for minimizing power consumption sensing, and using the model to calculate the optimal solution from the feasible solution set; finally, implementing the magnetic sensor deployment scheme corresponding to the optimal solution, and generating power consumption sensing data assets of the key load clusters based on the magnetic electrical quantity sensing technology.

[0008] The present invention adopts the following technical solution.

[0009] A first aspect of the present invention provides a method for minimizing power consumption of multiple loads in an industrial park, comprising the following steps:

[0010] Obtain the key load cluster division results of various industrial parks, generate a set of magnetic sensor deployment solutions for each division result as a feasible solution based on the division result, and then construct a feasible solution set for minimizing power consumption perception;

[0011] Taking into account the comprehensive cost of the deployment plan and the benefits of power consumption perception, an evaluation model for minimizing power consumption perception is established, and the optimal solution for minimizing power consumption perception is solved based on the evaluation model;

[0012] After the magnetic sensor deployment plan corresponding to the optimal solution is deployed, the power consumption perception data assets of the key load cluster are generated.

[0013] Preferably, generating a set of magnetic sensor deployment solutions for each division result includes:

[0014] Magnetic sensors are configured for representative loads of all key load clusters within the division result, and a set of magnetic sensor deployment plans is formed by the deployment positions of the magnetic sensors configured for representative loads of all key clusters within the division result.

[0015] Preferably, the step of selecting a representative load of a critical load cluster comprises:

[0016] For each critical load cluster, the similarity between each load and all other loads in the same critical load cluster is calculated, which is expressed as follows:

[0017]

[0018] Where:

[0019] is the similarity between load i and other loads in the critical load cluster l, is the number of key load clusters in deployment scheme k, k∈[1,α], α is the number of key load cluster division results in the industrial park,

[0020] M is the number of loads in the lth critical load cluster,

[0021] S i，j is the Euclidean distance between load i and load j, and j is all loads in the cluster except load i.

[0022] R is the radius of the critical load cluster l, which is equal to half of the Euclidean distance between the two farthest loads in the critical load cluster l;

[0023] The similarity value array is obtained by the similarity between each load and all other loads in the same critical load cluster, which is expressed as follows:

[0024]

[0025] The load with the smallest value in the array is compared and selected as the representative load of the key load cluster l, and the position is determined as the deployment position of the magnetic sensor.

[0026] Preferably, the steps of establishing an evaluation model for minimizing power consumption perception by taking into account the comprehensive cost of the deployment solution and the power consumption perception benefit include:

[0027] The ratio of comprehensive cost to power consumption perception benefit is used as the evaluation indicator; the function of comprehensive cost includes: sensor price cost, communication cost between sensor and edge smart energy unit; power consumption perception benefit includes: variance and accuracy of power consumption similarity of each load in the cluster.

[0028] Preferably, the evaluation model is as shown in formula (3):

[0029]

[0030] Where:

[0031] To evaluate the objective function;

[0032] f cost (d v ) is the comprehensive cost function, f gain (d v ) is the electricity consumption perception benefit function;

[0033] x l The communication distance between the key load and the edge smart energy unit in the cluster cannot exceed x max ;

[0034] λ k is the cost coefficient, λ c is the communication cost coefficient, λ s is the similarity coefficient, λ p is the accuracy coefficient;

[0035] J l is the cost of the critical load sensor equipment in the critical load cluster l;

[0036] C l is the communication cost of the critical loads in the critical load cluster l;

[0037] S l is the power consumption similarity between the key load and other loads in the key load cluster l,

[0038] D(S) means calculating the variance of array S;

[0039] l max is the number of key clusters in the v-th clustering method;

[0040] p l is the accuracy of the critical load in the critical load cluster l;

[0041] lThe superscript indicates the sensor deployment location;

[0042] f is the frequency;

[0043] The probability that the critical load has inaccurate measurement results due to the deviation between the power consumption data measured by the sensor itself and the power consumption data actually generated by the sensor itself;

[0044] The probability that the calculation result of the key load is inaccurate due to the deviation between the power consumption data of other loads in the cluster calculated by the power consumption data measured by the representative load in the cluster and the actual power consumption data generated by the key load itself;

[0045] It is the probability that the critical load is inaccurate due to the inconsistency between the power consumption data received by the smart energy unit and the power consumption data actually sent by the representative load.

[0046] Preferably, the evaluation values ​​of the feasible solutions are calculated according to formula (3), and the feasible solution with the smallest calculation result is taken as the optimal solution.

[0047] Preferably, the step of generating the power consumption perception data asset of the critical load cluster includes:

[0048] Based on the representative load power consumption data measured in real time by the magnetic sensor, the power consumption deduction data of other loads in the cluster is calculated, as shown in formula (5).

[0049]

[0050] Where:

[0051] F v,j is the jth load in the vth critical load cluster, and the power consumption deduction data of other loads in the critical load cluster except the magnetic sensor deployment location i,

[0052] F v,i The power consumption sensing data of the magnetic sensor deployment location i is obtained.

[0053] is the power load similarity relationship function between load i and load j,

[0054] Δ v,i,j Calculate the error in load power consumption data between load i and load j.

[0055] Preferably, on a spatial scale, the power consumption data of other loads in the cluster are calculated by using the power consumption sensing data of representative loads collected in real time by magnetic sensors;

[0056] In terms of time scale, electricity consumption data at different times are collected to form data assets. Based on the electricity consumption data of key loads in the industrial park at a certain moment, they are extended to multiple time scales to form the data assets of the entire industrial park.

[0057] Preferably, the deduced data assets can provide technical support for the deduction of multi-load control potential in industrial parks and refined user profiling technology.

[0058] A second aspect of the present invention provides a system for minimizing power consumption of multiple loads in an industrial park, and a method for minimizing power consumption of multiple loads in an industrial park, including:

[0059] A feasible solution set construction module is used to obtain the key load cluster division results of various industrial parks, generate a set of magnetic sensor deployment solutions for each division result as a feasible solution based on the division result, that is, a potential minimized power consumption perception solution, and then construct a feasible solution set for minimized power consumption perception;

[0060] The optimal solution obtaining module is used to establish a minimum power consumption perception evaluation model by comprehensively considering the cost and power consumption perception benefits, and to obtain the optimal solution of the minimum power consumption perception method based on the evaluation model;

[0061] The data asset generation module is used to obtain the electricity consumption data assets of the entire industrial park after the magnetic sensor deployment plan corresponding to the optimal solution is deployed.

[0062] Compared with the prior art, the beneficial effects of the present invention include at least: the present invention proposes a method for minimizing electricity consumption perception of multiple loads in an industrial park, takes key electricity load clusters as the objects of minimizing electricity consumption perception, and selects the optimal solution from a series of feasible solutions by establishing an evaluation model that takes into account comprehensive costs and electricity consumption perception benefits, thereby achieving the installation of a small number of intelligent circuit breakers (containing magnetic sensitive electrical quantity sensors) to obtain the electricity consumption perception data assets of the entire industrial park, providing technical support and data foundation for the deduction of the potential for electricity consumption regulation of multiple loads in industrial parks and accurate user portraits.

[0063] More specifically, (1) the present invention proposes for the first time a minimization power consumption perception evaluation model that takes into account comprehensive costs and power consumption perception benefits, where the comprehensive cost takes into account the price of magnetic sensors and communication costs, and the power consumption perception benefits take into account the power consumption similarity between loads within the key cluster, sensor measurement accuracy, similarity function calculation accuracy, and data transmission accuracy. This ensures the accuracy of power consumption perception in industrial parks while reducing economic costs.

[0064] (2) Aiming at the problem of refined power consumption perception of multiple loads in industrial parks, the present invention proposes a method for minimizing power consumption perception of load clusters in industrial parks that considers comprehensive costs and power consumption perception benefits. This method follows the technical route of "minimized collection + digital deduction", takes each key load cluster as the analysis object, and determines the minimized collection scheme of perception data by establishing a profit-loss ratio evaluation model of each alternative minimized collection scheme (feasible solution). Then, the power consumption perception data of representative loads in each load cluster are measured in real time through magnetic-sensitive electrical quantity sensors, and the power consumption deduction data of other loads in each load cluster are deduced based on the power consumption perception data to form perception-deduction fusion data. This method realizes refined power consumption perception of loads in industrial parks by using a small number of optimally deployed magnetic-sensitive electrical quantity sensors and perception-deduction fusion data, and selects the optimal scheme from the perspectives of comprehensive cost and power consumption perception benefits based on the proposed evaluation model, thereby improving the accuracy of power consumption perception in the park and effectively reducing the economic cost of power consumption perception in the park, taking into account both the economy and accuracy of power consumption perception.

[0065] (3) The present invention proposes for the first time to further expand the perception-calculation fusion data to the time domain, forming data assets for the key loads of the industrial park, and correcting the similarity relationship of the power load within the cluster through historical power consumption perception data, thereby improving the accuracy of minimizing power consumption perception, which helps to support advanced applications such as load regulation potential deduction and user power consumption profiling. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0067] Figure 1 The technical framework and workflow diagram of the method of the present invention;

[0068] Figure 2 A corresponding relationship diagram between the target problem and the mathematical problem of the method for minimizing power consumption perception;

[0069] Figure 3 This is a schematic diagram of the data assets of the industrial park;

[0070] Figure 4 The multi-load distribution and power supply topology of the electric tool factory in the embodiment of the present invention;

[0071] Figure 5 This is a schematic diagram of the first critical load cluster division result in an embodiment of the present invention;

[0072] Figure 6 This is a schematic diagram of the second critical load cluster division result in an embodiment of the present invention;

[0073] Figure 7 This is a schematic diagram of the third critical load cluster division result in an embodiment of the present invention. DETAILED DESCRIPTION

[0074] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The described embodiments are only embodiments of a part of the present invention, not all embodiments. Based on the spirit of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the protection scope of the present invention.

[0075] In order to precisely perceive the power consumption status of multiple loads in industrial parks, this patent proposes a method for minimizing power consumption perception for multiple loads in industrial parks. This method takes a series of key load clusters in the industrial park as power consumption perception objects, deploys magnetic sensors for representative loads of each key load cluster to achieve minimized collection of power consumption perception data, and then calculates the power consumption deduction data of other loads based on the power consumption similarity between these representative loads, and generates power consumption perception data assets for key load clusters based on magnetic-sensitive electrical quantity sensing technology, thereby realizing power consumption perception of multiple loads in the park based on "minimized collection + digital deduction". This method achieves a balance between the comprehensive cost of power consumption perception and the benefits of power consumption perception, and promotes the participation of industrial parks in friendly interaction between power grid supply and demand.

[0076] like Figure 1 As shown, embodiment 1 of the present invention provides a method for minimizing power consumption of multiple loads in an industrial park, comprising the following steps:

[0077] Step 1: Establish a technical framework for the power consumption minimization perception method. This step establishes a technical framework for the power consumption minimization perception method and clarifies the work of each step. The technical framework for the power consumption minimization perception method includes three parts: constructing a feasible solution set for the power consumption minimization perception method, solving the optimal solution for the power consumption minimization perception method based on the evaluation model, and generating power consumption perception data assets for key load clusters. The technical framework is as follows: Figure 1 shown.

[0078] Step 2: Construct a feasible solution set for the method of minimizing power consumption perception. First, design multiple magnetic sensor deployment schemes based on all possible key load clustering results, and use the magnetic sensor deployment scheme as the feasible solution set; second, infer the representative load based on the power consumption similarity between the loads in the key load cluster, and use the representative load as the magnetic sensor deployment point of the key load cluster; finally, based on the load power consumption similarity of the key load cluster, use the representative load to calculate the power consumption data of other loads in the cluster.

[0079] Preferably, the step 2 specifically includes:

[0080] Step 2.1: Construct a magnetic sensor deployment plan. According to the existing industrial park key load cluster identification method, a series of key load cluster division results are obtained, and a total of α key load cluster division results are set {..., G (i) ,...}, i∈[1,α], each deployment scheme G (i) Include Key load clusters, is the upper limit of the number of critical load clusters for each deployment scheme; and then, according to the division results, corresponding α magnetic sensor deployment schemes are designed.

[0081] Step 2.2: Infer the representative load of each key load cluster. According to the similarity of load power consumption within the key load cluster, infer the representative load of each cluster, including the following steps:

[0082] 1) Assume that the lth cluster contains a total of M loads, take half of the Euclidean distance between the two farthest loads as the radius R of the cluster, and the similarity between the i-th load and all other loads As shown in formula (1):

[0083]

[0084] Where:

[0085] S i,j is the Euclidean distance between the i-th load and the j-th load,

[0086] j is all the loads in the cluster except load i.

[0087] 2) Calculate the similarity value array of all loads in cluster l by formula (1): The load with the smallest value in the array is compared and selected as the representative load of the critical load cluster, that is, the deployment location of the magnetic sensor.

[0088] Step 2.3: Use the representative load to calculate the power consumption data of other loads in the cluster. According to the representative load power consumption data measured in real time by the magnetic sensor, the power consumption data of other loads in the cluster are calculated, as shown in formula (2).

[0089]

[0090] Where:

[0091] F v，j The power consumption deduction data of the jth load in the vth critical load cluster, except for the magnetic sensor deployment location i in the critical load cluster;

[0092] F v，i Power consumption sensing data for the magnetic sensor deployment location i;

[0093] is the power load similarity relationship function between load i and load j;

[0094] Δ v，i，j Calculate the error in load power consumption data between load i and load j.

[0095] At this point, the deployment locations of magnetic sensors of key load clusters in each feasible solution can be determined, that is, d′=(d′1, d′2, ..., d′ v , ..., d′ n ).

[0096] Step 3: Solve the optimal solution of the method for minimizing power consumption perception based on the evaluation model. Through the above steps, a set of feasible solutions for minimizing power consumption perception can be obtained, and the deployment location of the magnetic sensor in each feasible solution can be determined. Next, this step considers the comprehensive cost of various deployment schemes and the benefits of power consumption perception to establish an evaluation model for minimizing power consumption perception. The model is used to calculate the optimal solution from the feasible solution set, providing technical support for the subsequent steps to obtain the data assets of key power loads in the industrial park. The correspondence between the target problem and the mathematical problem of the method for minimizing power consumption perception is as follows: Figure 2 shown.

[0097] Preferably, the step 3 specifically includes:

[0098] Step 3.1: Establish a minimization power consumption perception evaluation model that considers comprehensive cost and power consumption perception benefits. This evaluation model uses the ratio of comprehensive cost to power consumption perception benefits as an evaluation indicator. The smaller the evaluation indicator of the feasible solution, the more power consumption perception benefits the feasible solution can obtain at a lower comprehensive cost. The function of the comprehensive cost consists of the sensor price cost and the communication cost between the sensor and the edge smart energy unit, where the communication cost depends on the communication distance and the selected frequency. The function of power consumption perception benefits focuses on reflecting the variance and accuracy of the power consumption similarity of each load in each cluster. The accuracy p is determined by the measurement accuracy of the sensor, the accuracy of the similarity function calculation, and the accuracy of the data transmitted between the sensor and the edge smart energy unit. The evaluation model is shown in formula (3):

[0099]

[0100] Where:

[0101] To evaluate the objective function;

[0102] f cost (d v ) is the comprehensive cost function, f gain (d v ) is the electricity consumption perception benefit function;

[0103] x l The communication distance between the key load and the edge smart energy unit in the cluster (in km), which cannot exceed x max ;

[0104] λ k is the cost coefficient, λ c is the communication cost coefficient, λ s is the similarity coefficient, λ p is the accuracy coefficient;

[0105] J l is the cost of the key load sensor equipment in the lth cluster;

[0106] C l is the communication cost of the critical load in the lth cluster;

[0107] S l is the electricity consumption similarity between the key load and other loads in the lth cluster, D(S) represents the variance of the calculation array S;

[0108] l max is the number of key clusters in the v-th clustering method,

[0109] p l is the accuracy of the critical load in the lth cluster;

[0110] The probability that the critical load has inaccurate measurement results due to the deviation between the power consumption data measured by the sensor itself and the power consumption data actually generated by the sensor itself;

[0111] The probability that the calculation result of the key load is inaccurate due to the deviation between the power consumption data of other loads in the cluster calculated by the power consumption data measured by the representative load in the cluster and the actual power consumption data generated by the key load itself;

[0112] The probability that the critical load is inaccurate due to the inconsistency between the power consumption data received by the smart energy unit and the power consumption data actually sent by the representative load;

[0113] f is the frequency (in MHz);

[0114] The l subscript in formula (3) indicates the sensor deployment location.

[0115] Step 3.2: Calculate the evaluation value of each feasible solution. Obtain the optimal solution of the method of minimizing power consumption. v Substitute it into formula (3) and take the feasible solution with the smallest calculated result as the optimal solution.

[0116] As a prominent substantive feature of the present invention and one of the significant advances it brings to the prior art, this patent focuses more on the optimization of solutions after cluster division and the expansion of power load data assets; it takes into account a variety of cluster division results, and selects the optimal solution from the two perspectives of comprehensive cost and power consumption perception benefits based on the proposed evaluation model. The division scheme of the present invention takes into account both the economy and accuracy of power consumption perception.

[0117] Step 4: Generate power consumption data assets of key load clusters. After determining the key power consumption load clusters and their magnetic sensor deployment locations in the industrial park, the power consumption data assets of the entire industrial park can be obtained by calculating the power consumption data of other loads in each cluster through the power consumption similarity relationship. The specific process of obtaining data assets in the industrial park is as follows: Figure 3 As shown, step 4 consists of the following steps:

[0118] Step 4.1: On a spatial scale, the power consumption data of other loads in the cluster are calculated through the power consumption data of representative loads collected in real time by magnetic sensors. That is, at the same time, the power consumption data of representative loads in each cluster are actually measured by magnetic sensors, and then the power consumption data of other loads in each cluster are calculated according to step 2.3.

[0119] Step 4.2: Collect electricity consumption data at different times to form data assets on a time scale. According to step 4.1, the electricity consumption data of key loads in the industrial park at a certain time is obtained, and it is extended to multiple time scales to form the data assets of the entire industrial park. In addition, some historical electricity consumption data are used as samples, and the other part of the data is used to train the model. Formula (2) is optimized and corrected by data fitting or machine learning methods to obtain a specific expression, which can further improve the accuracy of the deduced data in the cluster.

[0120] Step 4.3: Use power consumption sensing data assets to support the power consumption situation prediction technology and load regulation potential deduction technology of multiple loads in industrial parks. The data assets obtained in step 4.2 can provide technical support for the deduction of multiple load regulation potential and refined user profiling technology in industrial parks.

[0121] As a prominent substantive feature of the present invention and one of the significant advances it brings to the prior art, the present invention further expands the perception-inference fusion data to the time domain. The higher-dimensional data assets are conducive to supporting advanced applications such as load profiling.

[0122] Embodiment 2 of the present invention provides a system for minimizing power consumption of multiple loads in an industrial park, and runs a method for minimizing power consumption of multiple loads in an industrial park described in embodiment 1, including:

[0123] A feasible solution set construction module is used to obtain the key load cluster division results of various industrial parks, generate a set of magnetic sensor deployment solutions for each division result as a feasible solution based on the division result, that is, a potential minimized power consumption perception solution, and then construct a feasible solution set for minimized power consumption perception;

[0124] The optimal solution obtaining module is used to establish a minimum power consumption perception evaluation model by comprehensively considering the cost and power consumption perception benefits, and to obtain the optimal solution of the minimum power consumption perception method based on the evaluation model;

[0125] The data asset generation module is used to obtain the electricity consumption data assets of the entire industrial park after the magnetic sensor deployment plan corresponding to the optimal solution is deployed.

[0126] Embodiment 3 of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is loaded into the processor, the method for minimizing power consumption perception of multiple loads in an industrial park according to Embodiment 1 is implemented.

[0127] Embodiment 4 of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and is characterized in that when the computer program is executed by a processor, a method for minimizing power consumption perception of multiple loads in an industrial park according to Embodiment 1 is implemented.

[0128] In order to more clearly introduce the outstanding essential features of the present invention and the significant progress it brings to the prior art, an example of applying the present invention is given below. The present invention provides a method for minimizing power consumption of multiple loads in an industrial park, the steps of which are as follows: Figure 1 The following describes the method flow of the present invention to achieve refined power consumption perception in industrial parks in combination with on-site data of a power tool factory.

[0129] First, according to step 1 of the method of the present invention, a technical framework of the method for minimizing power consumption perception is established. The technical framework includes three parts: constructing a feasible solution set for the method for minimizing power consumption perception, solving the optimal solution for the method for minimizing power consumption perception based on the evaluation model, and generating power consumption perception data assets for key load clusters.

[0130] Then, according to step 2 of the method of the present invention, each feasible solution for accurate perception of load minimization in the park is determined, that is, a potential perception scheme for minimizing electricity consumption. First, a set of feasible solutions for the division of key load clusters is obtained. Secondly, the sensor deployment location of each key load cluster is selected. Finally, the power consumption deduction data of other loads is calculated based on the similarity relationship within each key load cluster. Considering that the total load power of the electric tool factory in the embodiment of the present invention is 6423kW, with reference to the production process and production schedule of the electric tool factory, injection molding machines, air conditioners, and water heaters are selected as key loads. The total power of the key loads is 4421kW (accounting for 68.8%), see for details. Figure 4 .

[0131] According to step 2.1 of the method of the present invention, the deployment location of the magnetic sensor in each feasible solution is selected. According to the existing industrial park key load cluster identification method, multiple key load cluster division results are obtained, and representative loads are selected based on the power consumption similarity between loads, and the representative loads are used as the deployment location of the magnetic sensor. The cluster division results d1', d'2, d3' of the sensor deployment location in the key load cluster are clarified. Figure 5 , Figure 6 , Figure 7 and Table 1. Figure 5 The intelligent circuit breaker (containing a magnetic sensitive electrical quantity sensor that can measure electrical quantities such as current and voltage) transmits the collected representative load power consumption perception data to the smart energy unit (edge ​​node). The smart energy unit uses step 2.2 to calculate the power consumption deduction data of other loads in the key load cluster from multiple time and space scales based on the collected power consumption perception data, and then obtains the power consumption perception data assets of the industrial park. Figure 6 and Figure 7 They represent other feasible solutions for the deployment of magnetic sensors.

[0132] Table 1 Three feasible solutions for key load cluster division

[0133]

[0134] Afterwards, an evaluation model for minimizing power consumption perception is constructed according to step 3 of the method of the present invention. Through the aforementioned steps, various feasible solutions for minimizing power consumption perception can be obtained, and the deployment location of the intelligent circuit breaker in each feasible solution is determined. Next, the optimal solution will be selected according to the evaluation model for minimizing power consumption perception, providing data foundation and technical support for obtaining data assets of key load clusters in industrial parks.

[0135] According to step 3.1 of the method of the present invention, a minimum power consumption perception evaluation model considering comprehensive cost and power consumption perception benefit is established. In order to facilitate the feasibility of the present invention, it is assumed that the unit cost of the intelligent circuit breaker is 15 yuan, the parameter f = 500, λ k =0.5;λ c =0.5;λ s =0.5;λ p =0.5. The parameters of other key load clusters are shown in Table 2.

[0136] Table 2 Key load cluster division parameters

[0137]

[0138]

[0139] Substitute the key load cluster division results d1', d'2, d3' into formula (3) to calculate the evaluation objective function value

[0140] According to step 3.2 of the method of the present invention, the optimal solution is obtained. Therefore, d3' is the optimal solution for minimizing the power consumption sensing method.

[0141] According to step 4 of the method of the present invention, the power consumption perception data assets of the power tool factory are generated. After determining the key power consumption load clusters and their representative loads in the industrial park, the power consumption data of other loads in each cluster are calculated through the power consumption similarity relationship to obtain the power consumption perception data assets of the entire industrial park. The process of obtaining power consumption perception data assets is as follows: Figure 3 shown.

[0142] According to step 4.1 of the method of the present invention, on a spatial scale, the representative load power consumption sensing data collected by the smart circuit breaker is used to calculate the power consumption deduction data of other loads in the cluster. According to the cluster division result, the actual measurement of cluster C1 at time t1 shows that the power consumption of the smart circuit breaker deployment location is V C1,L1,t1 At the same time, according to formula (2), the power consumption of other loads in the cluster is calculated to be V C1,L*,t1 . Integrate actual measured data with estimated data to obtain data assets at a spatial scale [V C1,L1,t1 ,V C1,L*,t1 ].

[0143] According to step 4.2 of the method of the present invention, on a time scale, the power consumption data at different times are collected to form data assets. According to the cluster division result, the actual measurement of cluster C1 at time t1 shows that the power consumption of the intelligent circuit breaker deployment location is V C1,L1,t1 At time t2, the actual measurement shows that the power consumption at the deployment location of the intelligent circuit breaker is V C1,L1,t2 , integrate the actual measured data or estimated data at each moment to obtain the data assets on the time scale of cluster C1 [V C1,L1,t1 ,V C1,L1,t* ] or [V C1,L*,t1 ,V C1,L*,t* ].

[0144] From the above-mentioned process of minimizing power consumption perception in the electric tool factory, it can be seen that the minimizing power consumption perception proposed in the present invention patent includes two characteristics: "minimizing collection" and "digital deduction". By selecting the most representative load in the key load cluster, deploying magnetic sensors on the load and measuring its power consumption perception data in real time, the power consumption of other loads is calculated based on the power consumption similarity within the key load cluster.

[0145] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0146] Finally, 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 above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for minimizing power consumption of multiple loads in an industrial park, characterized in that: The following steps are involved: Obtain the key load cluster division results of various industrial parks, generate a set of magnetic sensor deployment solutions for each division result as a feasible solution based on the division result, and then construct a feasible solution set for minimizing power consumption perception; Taking into account the comprehensive cost of the deployment plan and the benefits of power consumption perception, an evaluation model for minimizing power consumption perception is established, and the optimal solution for minimizing power consumption perception is solved based on the evaluation model; After the magnetic sensor deployment plan corresponding to the optimal solution is deployed, the power consumption perception data assets of the key load cluster are generated.

2. The method for minimizing power consumption of multiple loads in an industrial park according to claim 1, characterized in that: A set of magnetic sensor deployment solutions is generated for each division result, including: Magnetic sensors are configured for representative loads of all key load clusters within the division result, and a set of magnetic sensor deployment plans is formed by the deployment positions of the magnetic sensors configured for representative loads of all key clusters within the division result.

3. The method for minimizing power consumption of multiple loads in an industrial park according to claim 2, characterized in that: The steps to select representative loads of the critical load cluster include: For each critical load cluster, the similarity between each load and all other loads in the same critical load cluster is calculated, which is expressed as follows: Where: is the similarity between load i and other loads in the critical load cluster L, is the number of key load clusters in deployment scheme k, K∈[1,α], α is the number of key load cluster division results in the industrial park, M is the number of loads in the lth critical load cluster, S i,j is the Euclidean distance between load i and load j, and j is all loads in the cluster except load i. R is the radius of the critical load cluster l, which is equal to half of the Euclidean distance between the two farthest loads in the critical load cluster l; The similarity value array is obtained by the similarity between each load and all other loads in the same critical load cluster, which is expressed as follows: The load with the smallest value in the array is compared and selected as the representative load of the key load cluster l, and the position is determined as the deployment position of the magnetic sensor.

4. The method for minimizing power consumption of multiple loads in an industrial park according to claim 2, characterized in that: The steps of establishing an evaluation model for minimizing power consumption awareness by taking into account the comprehensive cost of the deployment solution and the benefits of power consumption awareness include: The ratio of comprehensive cost to power consumption perception benefit is used as the evaluation indicator; the function of comprehensive cost includes: sensor price cost, communication cost between sensor and edge smart energy unit; power consumption perception benefit includes: variance and accuracy of power consumption similarity of each load in the cluster.

5. The method for minimizing power consumption of multiple loads in an industrial park according to claim 4, characterized in that: The evaluation model is shown in formula (3): Where: To evaluate the objective function; f cost (d v ) is the comprehensive cost function, f gain (d v ) is the electricity consumption perception benefit function; x l The communication distance between the key load and the edge smart energy unit in the cluster cannot exceed x max ; λ k is the cost coefficient, λ c is the communication cost coefficient, λ s is the similarity coefficient, λ p is the accuracy coefficient; J l is the cost of the critical load sensor equipment in the critical load cluster l; C l is the communication cost of the critical loads in the critical load cluster l; S l is the electricity consumption similarity between the critical load and other loads in the critical load cluster l, D(S) represents the variance of the calculated array S; l max is the number of key clusters in the v-th clustering method; p l is the accuracy of the critical load in the critical load cluster l; f is the frequency; The probability that the critical load has inaccurate measurement results due to the deviation between the power consumption data measured by the sensor itself and the power consumption data actually generated by the sensor itself; The probability that the calculation result of the key load is inaccurate due to the deviation between the power consumption data of other loads in the cluster calculated by the power consumption data measured by the representative load in the cluster and the actual power consumption data generated by the key load itself; The probability that the critical load is inaccurate due to the inconsistency between the power consumption data received by the smart energy unit and the power consumption data actually sent by the representative load; The l superscript indicates the sensor deployment location.

6. The method for minimizing power consumption of multiple loads in an industrial park according to claim 5, characterized in that: The evaluation value of each feasible solution is calculated according to formula (3), and the feasible solution with the smallest calculation result is taken as the optimal solution.

7. The method for minimizing power consumption of multiple loads in an industrial park according to claim 4, characterized in that: The steps to generate the power consumption-aware data assets of the critical load cluster include: Based on the representative load power consumption data measured in real time by the magnetic sensor, the power consumption deduction data of other loads in the cluster is calculated, as shown in formula (6). Where: F v,j is the jth load in the vth critical load cluster, and the power consumption deduction data of other loads in the critical load cluster except the magnetic sensor deployment location i, F v,i The power consumption sensing data of the magnetic sensor deployment location i is obtained. is the power load similarity relationship function between load i and load j, Δ v,i,j Calculate the error in load power consumption data between load i and load j.

8. The method for minimizing power consumption of multiple loads in an industrial park according to claim 7, characterized in that: On a spatial scale, the power consumption data of other loads in the cluster are calculated through the representative load power consumption data collected in real time by magnetic sensors; In terms of time scale, electricity consumption data at different times are collected to form data assets. Based on the electricity consumption data of key loads in the industrial park at a certain moment, they are extended to multiple time scales to form the data assets of the entire industrial park.

9. The method for minimizing power consumption of multiple loads in an industrial park according to claim 8, characterized in that: The deduced data assets can provide technical support for the deduction of multi-load control potential in industrial parks and refined user profiling technology.

10. A system for minimizing power consumption of multiple loads in an industrial park, which runs a method for minimizing power consumption of multiple loads in an industrial park as claimed in claims 1 to 9, characterized in that: include: A feasible solution set construction module is used to obtain the key load cluster division results of various industrial parks, generate a set of magnetic sensor deployment solutions for each division result as a feasible solution based on the division result, that is, a potential minimized power consumption perception solution, and then construct a feasible solution set for minimized power consumption perception; The optimal solution obtaining module is used to establish a minimum power consumption perception evaluation model by comprehensively considering the cost and power consumption perception benefits, and to obtain the optimal solution of the minimum power consumption perception method based on the evaluation model; The data asset generation module is used to obtain the electricity consumption data assets of the entire industrial park after the magnetic sensor deployment plan corresponding to the optimal solution is deployed.

Images