Intelligent non-intrusive charging pile cluster load online monitoring device and method

Abstract

The invention relates to the technical field of charging pile cluster load monitoring, in particular to an intelligent non-intrusive charging pile cluster load online monitoring device and method. The device comprises a charging pile cluster non-intrusive intelligent acquisition device, a charging pile load intelligent sensing network big data operation server and a remote client monitoring platform. Wherein the charging pile cluster non-intrusive intelligent acquisition device is connected with a direct current charging pile, the charging pile cluster non-intrusive intelligent acquisition device is connected with an alternating current charging pile, and the charging pile load intelligent sensing network big data operation server is connected with the charging pile cluster non-intrusive intelligent acquisition device. And the remote client monitoring platform is connected with the charging pile load intelligent sensing network big data operation server. According to the method, an improved wavelet threshold denoising and multi-scene data fusion algorithm is used, accurate monitoring and intelligent management of the charging pile cluster load can be achieved, and the method has the advantages of being efficient, non-invasive and high in real-time performance.

Description

Technical Field

[0001] The present invention relates to the technical field of charging pile cluster load monitoring, and in particular to an intelligent non-intrusive charging pile cluster load online monitoring device and method. Background Art

[0002] In recent years, the development of domestic electric vehicles has shown a rapid trend. Driven by policy support and market demand, the production and sales of electric vehicles have continued to grow, and the technical level has also been continuously improved. The government has created favorable conditions for the development of electric vehicles by implementing new energy vehicle subsidy policies and building charging infrastructure. At the same time, domestic automobile enterprises have also increased their R & D efforts and launched a series of competitive electric vehicles. In addition, with the continuous progress of battery technology and the increasing improvement of charging facilities, the driving range and charging convenience of electric vehicles have been significantly improved, further meeting the needs of consumers. The domestic electric vehicle industry is entering a new stage of rapid development with broad prospects for the future.

[0003] In recent years, with the country's attention and policy support for the new energy vehicle industry, remarkable achievements have been made in the construction of charging pile infrastructure. At present, the number of charging piles across the country continues to increase, covering two major categories: public charging piles and private charging piles. These charging piles are distributed in every corner of the city, providing convenience for electric vehicle charging. At the same time, with the continuous progress of technology, the charging efficiency of charging piles is also constantly improving, with faster charging speed and more convenient charging process. In addition, the government has also increased its investment in the construction of charging infrastructure and further promoted the development of electric vehicle charging piles by building charging stations and promoting charging piles. With the continuous expansion of the new energy vehicle market and the continuous progress of technology, the construction of electric vehicle charging piles in China will continue to maintain a rapid growth trend, providing a strong guarantee for the popularization and development of electric vehicles.

[0004] The development of the monitoring method for electric vehicle charging piles is mainly reflected in aspects such as intelligence, integration, automation, real-time, visualization, sharing, and standardization. With the rapid development of technologies such as artificial intelligence, the Internet of Things, and big data, the charging pile monitoring technology is gradually achieving intelligence and integration. By introducing advanced sensing technologies, image processing, and data analysis, etc., it can comprehensively and accurately evaluate the physical structure, electrical performance, and safety of charging piles. At the same time, the monitoring system has realized a fully automated monitoring process, from data collection, processing to analysis, all of which are automatically completed by the system, greatly improving the monitoring efficiency. In addition, the monitoring system also has the characteristics of real-time, capable of real-time monitoring of the operating status of charging piles, promptly discovering abnormal situations and giving early warnings. Through technologies such as cloud computing and big data, the monitoring data is processed visually and applied in a shared manner, facilitating users to intuitively understand the usage of charging piles and supporting remote monitoring and data sharing. In order to promote the healthy development of the charging pile industry, the country is actively promoting the standardization and normalization of monitoring standards, formulating unified monitoring technology standards, equipment standards, and data interface standards, standardizing the charging pile market, and improving the quality of charging services and user experience. Summary of the Invention

[0005] Aiming at the deficiencies existing in the above-mentioned prior art, the present invention provides an intelligent non-intrusive on-line monitoring device and method for the load of a charging pile cluster. Its purpose is to achieve the invention purpose of being able to effectively promote the popularization of the non-intrusive on-line monitoring method for the load of a charging pile cluster in an electric vehicle charging station.

[0006] The technical solution adopted by the present invention to achieve the above purpose is as follows:

[0007] An intelligent non-intrusive on-line monitoring device for the load of a charging pile cluster includes: a non-intrusive intelligent acquisition device for the charging pile cluster, a big data operation server for the intelligent perception network of the charging pile load, and a remote client monitoring platform; wherein, the non-intrusive intelligent acquisition device for the charging pile cluster is connected to the DC charging piles and the AC charging piles, the big data operation server for the intelligent perception network of the charging pile load is connected to the non-intrusive intelligent acquisition device for the charging pile cluster, and the remote client monitoring platform is connected to the big data operation server for the intelligent perception network of the charging pile load; the non-intrusive intelligent acquisition device for the charging pile cluster is used to obtain non-intrusive information of the load data of the charging pile cluster, filter, denoise, and normalize the non-intrusively obtained load data, perform fusion operation on the processed data, and then transmit the data after the fusion operation to the big data operation server for the intelligent perception network of the charging pile load through the remote client monitoring platform.

[0008] Furthermore, the non-intrusive intelligent acquisition device for the charging pile cluster includes: a charging pile cluster load perception unit, a charging pile cluster load data preprocessing unit, a charging pile cluster load data fusion unit, and a communication unit; the charging pile cluster load perception unit realizes the acquisition of non-intrusive information of the charging pile cluster load data; the charging pile cluster load data preprocessing unit is used to filter, denoise, and normalize the non-intrusively acquired load data, and transmit the processed data to the charging pile cluster load data fusion unit; the charging pile cluster load data fusion unit performs fusion operations on the preprocessed data; the communication unit is used to transmit the data after the fusion operation to the big data operation server of the charging pile load intelligent perception network through wireless transmission; the charging pile cluster load data preprocessing unit uses an improved wavelet threshold denoising method to preprocess the non-intrusively acquired load data; the charging pile cluster load data fusion unit uses a multi-scenario data fusion algorithm to perform fusion operations on the preprocessed data; the charging pile cluster load data fusion unit uses a charging pile cluster clustering algorithm to manage the charging pile cluster.

[0009] Furthermore, the big data operation server of the charging pile load intelligent perception network includes: a charging pile cluster load fusion database, a charging pile cluster load fusion data operation platform, a charging pile cluster adjustable resource evaluation unit, and a communication unit; among them, the charging pile cluster load fusion database is used to store the fused load data transmitted from the communication unit of the non-intrusive intelligent acquisition device for the charging pile cluster; the charging pile cluster load fusion data operation platform uses a similarity clustering algorithm and a charging pile load intelligent perception network data training method to perform operations on the fused load data; the charging pile cluster adjustable resource evaluation unit is used to evaluate and predict the resource scheduling potential in the charging pile cluster; the communication unit is used to transmit the operation result to the remote client monitoring platform through wireless transmission.

[0010] Furthermore, the remote client monitoring platform includes: a data visualization unit, a user interaction unit, a remote control unit, and a communication unit; the data visualization unit is used to display the real-time load data and historical data of the charging pile cluster in a graphical interface; the user interaction unit is used to receive the instructions and parameter settings input by the user, and transmit the instructions and parameters to the big data operation server of the charging pile load intelligent perception network; the remote control unit remotely controls and manages the charging pile cluster according to the user instructions; the communication unit is used to receive the operation result from the big data operation server of the charging pile load intelligent perception network.

[0011] An intelligent non-intrusive on-line monitoring method for the load of a charging pile cluster includes:

[0012] Using the charging pile cluster load perception unit to obtain non-intrusive information of the charging pile cluster load data;

[0013] The non-intrusive load data obtained is filtered, denoised, and normalized by the charging pile cluster load data preprocessing unit, and the processed data is transmitted to the charging pile cluster load data fusion unit;

[0014] The charging pile cluster load data fusion unit performs a fusion operation on the preprocessed data;

[0015] The charging pile cluster load data fusion unit manages the charging pile cluster based on the obtained fusion operation data;

[0016] The communication unit transmits the data after the fusion operation to the big data operation server of the charging pile load intelligent perception network through wireless transmission.

[0017] Furthermore, the charging pile cluster load data preprocessing unit uses an improved wavelet threshold denoising method to preprocess the non-intrusively obtained load data; the charging pile cluster load data fusion unit uses a multi-scenario data fusion algorithm to perform a fusion operation on the preprocessed data; the charging pile cluster load data fusion unit uses a charging pile cluster clustering algorithm to manage the charging pile cluster.

[0018] Furthermore, the data preprocessing using the improved wavelet threshold denoising method includes:

[0019] Obtain load data from the charging pile cluster load perception unit;

[0020] Use the wavelet basis function to decompose the noise of the obtained load data;

[0021] Calculate the noise variance μ n , the noise variance μ n is expressed as:

[0022]

[0023] where ω j,k is the wavelet coefficient, median(·) is the median function, and ∣·∣ is the absolute value operation;

[0024] Calculate the threshold λ j , the threshold λ j is expressed as:

[0025]

[0026] where j is the number of decomposition layers, and N j is the sample number;

[0027] According to the calculated threshold λ j , update the wavelet coefficient as follows:

[0028]

[0029] Among them, α is an adjustment coefficient, sgn(·) is a step function, and e is the natural constant;

[0030] According to the obtained wavelet coefficients Obtain the denoised data;

[0031] Furthermore, the multi-scenario data fusion algorithm includes:

[0032] Calculate the Euclidean distance d between the load data of the charging pile cluster nm_T As follows:

[0033]

[0034] Among them, n represents the nth charging pile, n' represents the n'th charging pile, N is the total number of charging piles, m is the number of groups of charging pile cluster load data, M is the total number of groups of charging pile cluster load data, T is the time, and S nm-T is the data of the mth group of the nth charging pile at time T, and S n'm-T is the data of the mth group of the n'th charging pile at time T;

[0035] Calculate the average value of each group of data of the charging piles As follows:

[0036]

[0037] Calculate the weight value w of the load data of the nth charging pile in the mth group in the charging pile cluster at time T nm_T As follows:

[0038]

[0039] According to the calculated Euclidean distance d of the charging pile cluster load data nm-T and the weight value w nm_T , calculate the charging pile cluster load fusion data S m_T As follows:

[0040]

[0041] Among them, s m_T is the load data of a charging pile in the charging pile cluster;

[0042] Furthermore, the management of the charging pile cluster by using the charging pile cluster clustering algorithm includes:

[0043] Randomly assign values between 0 and 1 as initial values to different charging piles;

[0044] Calculate the cluster head determination threshold T n As follows:

[0045]

[0046] Where p is the probability that a charging pile is selected as the cluster head, n is the total number of nodes in the sensing network, and d A→B is the distance from charging pile A to charging pile B, and d max is the maximum value of the charging pile distance, and e is the natural constant;

[0047] Based on the cluster head determination threshold T n classify the charging pile cluster. The charging piles higher than the cluster head determination threshold T n are used as cluster head charging piles, and the charging piles lower than the cluster head determination threshold T n are used as ordinary charging piles and are added to the charging pile grouping.

[0048] A computer device includes a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor. When the processor executes the computer program, it implements the steps of any one of the intelligent non-intrusive charging pile cluster load online monitoring methods.

[0049] A computer storage medium stores a computer program. When the computer program is executed by a processor, it implements the steps of any one of the intelligent non-intrusive charging pile cluster load online monitoring methods.

[0050] The present invention has the following beneficial effects and advantages:

[0051] The present invention relates to an intelligent non-intrusive charging pile cluster load online monitoring device and method. The device includes a non-intrusive intelligent acquisition device for the charging pile cluster, a big data operation server for the intelligent perception network of the charging pile load, and a remote client monitoring platform.

[0052] The non-intrusive intelligent acquisition device for the charging pile cluster is connected to DC and AC charging piles. Load data is obtained through the load perception unit of the charging pile cluster, and after being filtered, denoised, and normalized by the load data preprocessing unit of the charging pile cluster, fusion operations are performed by the load data fusion unit of the charging pile cluster. The fused data is transmitted to the big data operation server for the intelligent perception network of the charging pile load through the communication unit. The big data operation server for the intelligent perception network of the charging pile load includes a load fusion database for the charging pile cluster, a load fusion data operation platform for the charging pile cluster, an adjustable resource evaluation unit for the charging pile cluster, and a communication unit, and data processing and resource scheduling evaluation are performed through a similarity clustering algorithm and a data training method for the intelligent perception network of the charging pile load. The remote client monitoring platform provides data visualization, user interaction, and remote control functions.

[0053] The method of the present invention uses a non-invasive design, which ensures that the device can complete data collection without affecting the normal operation of the charging pile, avoiding interference and loss to the device by traditional methods. The adopted wavelet threshold denoising algorithm can more efficiently eliminate noise and achieve sensitive capture of the load changes of the electric vehicle charging pile cluster. In addition, the multi-scenario data fusion algorithm improves the load monitoring accuracy in complex environments. Through the collaborative work of the intelligent perception network and the big data operation server, the real-time performance and resource scheduling efficiency are significantly improved, and the adjustable resources of the charging pile can be accurately evaluated, so as to realize the intelligent management of the load of the electric vehicle charging pile cluster. The method of the present invention has the characteristics of high efficiency, non-invasiveness and strong real-time performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0055] Figure 1 It is a schematic diagram of the deployment of an intelligent non-invasive charging pile cluster load online monitoring device of the present invention;

[0056] Figure 2 It is a structural diagram of a non-invasive intelligent acquisition device for a charging pile cluster of the present invention;

[0057] Figure 3 It is a structural diagram of a big data operation server of an intelligent perception network for a charging pile load of the present invention;

[0058] Figure 4 It is a structural diagram of a remote client monitoring platform of the present invention;

[0059] Figure 5 It is a flow chart of an improved wavelet threshold denoising method of the present invention;

[0060] Figure 6 It is a flow chart of a multi-scenario fusion algorithm of the present invention;

[0061] Figure 7 It is a flow chart of a clustering algorithm for a charging pile cluster of the present invention;

[0062] Figure 8 It is a flow chart of a similarity clustering algorithm of the present invention;

[0063] Figure 9 It is a flow chart of an intelligent non-invasive charging pile cluster load online monitoring method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0064] In order to more clearly understand the above-mentioned objects, features, and advantages of the present invention, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, without conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0065] In the following description, many specific details are set forth in order to provide a thorough understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited by the specific embodiments disclosed below.

[0066] The following refers to Figures 1-9 Describe the technical solutions of some embodiments of the present invention.

[0067] The present invention provides an embodiment, which is an intelligent non-intrusive charging pile cluster load online monitoring device and method, which can effectively promote the popularization of the non-intrusive charging pile cluster load online monitoring method in electric vehicle charging stations.

[0068] Embodiment 1

[0069] The present invention discloses an intelligent non-intrusive charging pile cluster load online monitoring device, as Figure 1 shown, Figure 1 is a deployment schematic diagram of an intelligent non-intrusive charging pile cluster load online monitoring device of the present invention.

[0070] The intelligent non-intrusive charging pile cluster load online monitoring device includes:

[0071] A non-intrusive intelligent acquisition device for charging pile clusters, a big data operation server for an intelligent perception network of charging pile loads, and a remote client monitoring platform. Among them, the non-intrusive intelligent acquisition device for charging pile clusters is connected to DC charging piles, the non-intrusive intelligent acquisition device for charging pile clusters is connected to AC charging piles, the big data operation server for the intelligent perception network of charging pile loads is connected to the non-intrusive intelligent acquisition device for charging pile clusters, and the remote client monitoring platform is connected to the big data operation server for the intelligent perception network of charging pile loads. The non-intrusive intelligent acquisition device for charging pile clusters is used to obtain non-intrusive information of the charging pile cluster load data, filter, denoise, and normalize the non-intrusively obtained load data, perform fusion operations on the processed data, and then transmit the data after the fusion operation to the big data operation server for the intelligent perception network of charging pile loads through the remote client monitoring platform.

[0072] As Figure 2 shown, Figure 2This is the structural diagram of the non-intrusive intelligent acquisition device for the charging pile cluster of the present invention. The non-intrusive intelligent acquisition device for the charging pile cluster of the present invention includes: a charging pile cluster load perception unit, a charging pile cluster load data preprocessing unit, a charging pile cluster load data fusion unit, and a communication unit. Among them, the charging pile cluster load perception unit realizes the acquisition of non-intrusive information of the charging pile cluster load data. The acquired information includes active power, reactive power, power factor, time information, etc. The charging pile cluster load data preprocessing unit is used to filter, denoise, and normalize the non-intrusively acquired load data, and transmit the processed data to the charging pile cluster load data fusion unit. The charging pile cluster load data fusion unit uses a data fusion algorithm to perform fusion operations on the preprocessed data. The communication unit is used to transmit the data after the fusion operation to the big data operation server of the charging pile load intelligent perception network through a wireless transmission method.

[0073] The charging pile cluster load data preprocessing unit uses an improved wavelet threshold denoising method to preprocess the non-intrusively acquired load data;

[0074] The charging pile cluster load data fusion unit uses a multi-scenario data fusion algorithm to perform fusion operations on the preprocessed data;

[0075] The charging pile cluster load data fusion unit uses a charging pile cluster clustering algorithm to manage the charging pile cluster.

[0076] As Figure 3 shown, Figure 3 This is the structural diagram of the big data operation server of the charging pile load intelligent perception network of the present invention. The big data operation server of the charging pile load intelligent perception network of the present invention includes: a charging pile cluster load fusion database, a charging pile cluster load fusion data operation platform, a charging pile cluster adjustable resource evaluation unit, and a communication unit. Among them, the charging pile cluster load fusion database is used to store the fusion load data transmitted from the communication unit of the non-intrusive intelligent acquisition device of the charging pile cluster.

[0077] The charging pile cluster load fusion data operation platform uses a similarity clustering algorithm and a charging pile load intelligent perception network data training method to perform operations on the fusion load data.

[0078] The charging pile cluster adjustable resource evaluation unit is used to evaluate and predict the resource scheduling potential in the charging pile cluster.

[0079] The communication unit is used to transmit the operation result to the remote client monitoring platform through a wireless transmission method.

[0080] As Figure 4 shown, Figure 4This is the structural diagram of the remote client monitoring platform of the present invention. The remote client monitoring platform of the present invention includes: a data visualization unit, a user interaction unit, a remote control unit, and a communication unit. Among them, the data visualization unit is used to display the real-time load data and historical data of the charging pile cluster in a graphical interface. The user interaction unit is used to receive the instructions and parameter settings input by the user, and transmit the instructions and parameters to the big data operation server of the charging pile load intelligent perception network. The remote control unit is used to remotely control and manage the charging pile cluster according to the user instructions. The communication unit is used to receive the operation results from the big data operation server of the charging pile load intelligent perception network.

[0081] Embodiment 2

[0082] The present invention further provides an embodiment, which is an intelligent non-intrusive online monitoring method for the load of a charging pile cluster, as Figure 9 shown, Figure 9 is the flowchart of an intelligent non-intrusive online monitoring method for the load of a charging pile cluster of the present invention, including the following steps:

[0083] Step 1. Use the charging pile cluster load perception unit to obtain non-intrusive information of the charging pile cluster load data; the obtained information includes active power, reactive power, power factor, and time information, etc.;

[0084] Step 2. Use the charging pile cluster load data preprocessing unit to perform filtering, denoising, and normalization processing on the obtained non-intrusive load data, and transmit the processed data to the charging pile cluster load data fusion unit;

[0085] The charging pile cluster load data preprocessing unit uses an improved wavelet threshold denoising method to perform data preprocessing on the non-intrusively obtained load data;

[0086] Step 3. The charging pile cluster load data fusion unit uses a multi-scenario data fusion algorithm to perform fusion operations on the preprocessed data;

[0087] Step 4. The charging pile cluster load data fusion unit uses a charging pile cluster clustering algorithm to manage the charging pile cluster based on the fusion operation data obtained in Step 3.

[0088] Step 5. Use the communication unit to transmit the fusion operation data to the big data operation server of the charging pile load intelligent perception network through a wireless transmission method.

[0089] Embodiment 3

[0090] The present invention further provides an embodiment, which is an intelligent non-intrusive charging pile cluster load online monitoring method. The charging pile cluster load data preprocessing unit uses an improved wavelet threshold denoising method to preprocess the load data obtained non-intrusively; as Figure 5 shown, Figure 5 is the flow chart of the improved wavelet threshold denoising method of the present invention.

[0091] The data preprocessing by the improved wavelet threshold denoising method specifically includes the following steps:

[0092] Step 1. Obtain load data from the charging pile cluster load sensing unit, including active power, reactive power, power factor, time information, etc.

[0093] Step 2. Decompose the noise of the obtained load data using wavelet basis functions.

[0094] Step 3. Calculate the noise variance μ n , the noise variance μ n can be expressed as:

[0095]

[0096] where ω j,k is the wavelet coefficient, median(·) is the median function, and ∣·∣ is the absolute value operation.

[0097] Step 4. Calculate the threshold λ j , the threshold λ j can be expressed as:

[0098]

[0099] where j is the number of decomposition layers, and N j is the sample number.

[0100] Step 5. According to the calculated threshold λ j , update the wavelet coefficient as follows:

[0101]

[0102] where α is the adjustment coefficient, sgn(·) is the step function, and e is the natural constant.

[0103] Step 6. Obtain the denoised data according to the obtained wavelet coefficient .

[0104] Embodiment 4

[0105] The present invention further provides an embodiment, which is an intelligent non-intrusive charging pile cluster load online monitoring method. The charging pile cluster load data fusion unit uses a multi-scenario data fusion algorithm to perform fusion operations on the preprocessed data, as Figure 6 shown Figure 6 in the flowchart of the multi-scenario fusion algorithm of the present invention.

[0106] The multi-scenario data fusion algorithm specifically includes the following steps:

[0107] Step 1. Calculate the Euclidean distance d between the load data of the charging pile cluster nm_T as follows:

[0108]

[0109] where n represents the nth charging pile, n' represents the n'th charging pile, N is the total number of charging piles, m is the number of groups of charging pile cluster load data, M is the total number of groups of charging pile cluster load data, T is time, S nm-T is the mth group of data of the nth charging pile at time T, and S n'm-T is the mth group of data of the n'th charging pile at time T;

[0110] Next, calculate the average value of each group of data of the charging piles as follows:

[0111]

[0112] Step 2. Calculate the weight value w of the mth group of load data of the nth charging pile in the charging pile cluster at time T nm_T as follows:

[0113]

[0114] Step 3. According to the Euclidean distance d nm_T and the weight value w nm_T of the charging pile cluster load data obtained in Step 1 and Step 2, calculate the charging pile cluster load fusion data S m_T as follows:

[0115]

[0116] where s m_T is the load data of a charging pile in the charging pile cluster.

[0117] Embodiment 5

[0118] The present invention further provides an embodiment, which is an intelligent non-intrusive charging pile cluster load online monitoring method. The charging pile cluster load data fusion unit manages the charging pile cluster by using the charging pile cluster clustering algorithm based on the obtained fusion operation data, as Figure 7 shown Figure 7 is the flowchart of the charging pile cluster clustering algorithm of the present invention.

[0119] The management of the charging pile cluster by using the charging pile cluster clustering algorithm specifically includes the following steps:

[0120] Step 1. Randomly assign values between 0 and 1 to different charging piles as initial values.

[0121] Step 2. Calculate the cluster head determination threshold T n as follows:

[0122]

[0123] where p is the probability of a charging pile being selected as a cluster head, n is the total number of nodes in the sensing network, d A→B is the distance from charging pile A to charging pile B, and d max is the maximum value of the charging pile distance, and e is the natural constant.

[0124] Step 3. Classify the charging pile cluster based on the cluster head determination threshold T n Charging piles higher than the cluster head determination threshold T n are used as cluster head charging piles, and charging piles lower than the cluster head determination threshold T n are used as ordinary charging piles and are added to the charging pile grouping.

[0125] Embodiment 6

[0126] The present invention further provides an embodiment, which is an intelligent non-intrusive charging pile cluster load online monitoring method. The charging pile cluster load fusion data operation platform of the present invention performs operations on the fusion load data by using the similarity clustering algorithm and the charging pile load intelligent perception network data training method. Among them, the flowchart of the similarity clustering algorithm is as Figure 8 shown.

[0127] The similarity clustering algorithm specifically includes the following steps:

[0128] Step 1. Randomly determine k initial centers and calculate the Euclidean distance from each power point to the initial center. The Euclidean distance is the shortest distance between two power points in the PQ two-dimensional space as follows:

[0129]

[0130] where dist(m i, x) is the Euclidean distance between point m i and point x, k is the total number of initial centers, m i is the i-th initial center, and x is the power point excluding the initial centers.

[0131] Step 2. Calculate the average distance until the squared error meets the requirement. Calculate the squared error minE as follows:

[0132]

[0133] where min(·) is the function to take the minimum value, ||·|| 2 is the operation to take the two-norm, t i is the total number of data points, x j is the x j -th data point, k is the total number of similarity clustering center points, m i is the i-th similarity clustering center point.

[0134] Step 3. According to the calculation results of Step 1 and Step 2, calculate the total number of clustering centers N clust as follows:

[0135]

[0136] where N represents the number of aggregated electric vehicle types, M represents the number of charging piles, represents the number of switching actions of the i-th charging pile in the M charging pile clusters.

[0137] The intelligent perception network data training method for charging pile load specifically includes the following steps:

[0138] Step 1. Select a suitable activation function.

[0139] Step 2. Obtain the load data of charging piles in different clusters, including total active power, power factor, timestamp, etc. Decompose the load data according to the characteristics of each similarity center, and perform normalization processing on the decomposed load data. The calculation formula is as follows:

[0140]

[0141] In the formula, P norm is the normalized power matrix, P is the matrix of all measured powers, P min and P max are the minimum power and the maximum power in the power matrix respectively.

[0142] Step 3. Calculate the hidden layer error E hid as follows:

[0143]

[0144] Among them, d k is the expected output, f(·) is the activation function, and w jk is the weight from the hidden layer to the output layer, and y j is the output of the hidden layer neurons.

[0145] Step 4. Calculate the input layer error E input as follows:

[0146]

[0147] Among them, v ij is the weight from the input layer to the hidden layer, and x i is the input of the input layer neurons.

[0148] Step 5. Calculate the error partial derivatives, and use the obtained partial derivatives to update the weight coefficients. The calculation formula is as follows:

[0149]

[0150] Among them, η is the learning rate, and Δw jk is the weight adjustment amount between the j-th layer and the k-th layer, is the partial derivative of the error E with respect to the weight w jk and Δv ij is the weight adjustment amount between the i-th node of the input layer and the j-th node of the hidden layer, is the partial derivative of the error E with respect to the weight v ij of the partial derivative

[0151] Step 6. Repeat the above steps until the error meets the requirements.

[0152] Example 7

[0153] The present invention further provides an embodiment, which is an intelligent non-intrusive charging pile cluster load online monitoring device, including:

[0154] An acquisition module, configured to use a charging pile cluster load sensing unit to acquire non-intrusive information of the charging pile cluster load data; the acquired information includes active power, reactive power, power factor, and time information, etc.;

[0155] A data preprocessing module, configured to use an improved wavelet threshold denoising method by a charging pile cluster load data preprocessing unit to perform data preprocessing on the charging pile cluster load data information non-intrusively acquired by the charging pile cluster load sensing unit;

[0156] A data fusion module, configured to use a multi-scenario data fusion algorithm by a charging pile cluster load data fusion unit to perform data fusion on the data preprocessed by the data preprocessing module;

[0157] A management module, configured to manage the charging pile cluster by using the charging pile cluster clustering algorithm based on the obtained fusion operation data by the charging pile cluster load data fusion unit;

[0158] A transmission module, configured to transmit the data after fusion operation to the big data operation server of the charging pile load intelligent perception network in a wireless transmission manner by using the communication unit.

[0159] The intelligent non-intrusive charging pile cluster load online monitoring device described in the present invention is used to implement the steps of the intelligent non-intrusive charging pile cluster load online monitoring method described in any one of Embodiments 2-6.

[0160] Embodiment 8

[0161] Based on the same inventive concept, an embodiment of the present invention further provides a computer device, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor. When the processor executes the computer program, the steps of the intelligent non-intrusive charging pile cluster load online monitoring method described in any one of Embodiments 2-6 are implemented.

[0162] Embodiment 9

[0163] Based on the same inventive concept, an embodiment of the present invention further provides a computer storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of any one of the intelligent non-intrusive charging pile cluster load online monitoring methods described in Embodiments 2-6 are implemented.

[0164] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0165] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing device generate a means for implementing the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0166] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including an instruction means that implements the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0167] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in the Figure 1 one or more flows and / or blocks Figure 1 one or more blocks.

[0168] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: it is still possible to modify the specific implementation manners of the present invention or make equivalent replacements, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. An intelligent non-intrusive charging pile cluster load online monitoring device, characterized by: include: Charging pile cluster non-intrusive intelligent collection device, charging pile load intelligent perception network big data computing server and remote client monitoring platform; wherein, the charging pile cluster non-intrusive intelligent collection device is connected to the DC charging pile, the charging pile cluster non-intrusive intelligent collection device is connected to the AC charging pile, the charging pile load intelligent perception network big data computing server is connected to the charging pile cluster non-intrusive intelligent collection device, and the remote client monitoring platform is connected to the charging pile load intelligent perception network big data computing server; the charging pile cluster non-intrusive intelligent collection device is used to obtain non-intrusive information of the charging pile cluster load data, and filter, denoise and normalize the non-intrusive acquired load data, perform fusion operation on the processed data, and then transmit the fusion operation data to the charging pile load intelligent perception network big data computing server through the remote client monitoring platform.

2. According to claim 1, the intelligent non-intrusive charging pile cluster load online monitoring device is characterized by: The non-intrusive intelligent collection device for charging pile clusters includes: a charging pile cluster load sensing unit, a charging pile cluster load data preprocessing unit, a charging pile cluster load data fusion unit and a communication unit; the charging pile cluster load sensing unit realizes the acquisition of non-intrusive information of the charging pile cluster load data; the charging pile cluster load data preprocessing unit is used to filter, denoise and normalize the non-intrusive acquired load data, and transmit the processed data to the charging pile cluster load data fusion unit; the charging pile cluster load data fusion unit performs fusion operation on the preprocessed data; the communication unit is used to transmit the fusion operation data to the charging pile load intelligent perception network big data operation server through wireless transmission; The charging pile cluster load data preprocessing unit uses an improved wavelet threshold denoising method to perform data preprocessing on the non-intrusive acquired load data; The charging pile cluster load data fusion unit uses a multi-scenario data fusion algorithm to perform fusion operations on the pre-processed data; The charging pile cluster load data fusion unit adopts a charging pile cluster clustering algorithm to manage the charging pile cluster.

3. According to claim 1, the intelligent non-intrusive charging pile cluster load online monitoring device is characterized by: The charging pile load intelligent perception network big data computing server includes: a charging pile cluster load fusion database, a charging pile cluster load fusion data computing platform, a charging pile cluster adjustable resource evaluation unit and a communication unit; wherein the charging pile cluster load fusion database is used to store fused load data transmitted from the communication unit of the non-intrusive intelligent collection device of the charging pile cluster; the charging pile cluster load fusion data computing platform uses a similarity clustering algorithm and a charging pile load intelligent perception network data training method to calculate the fused load data; the charging pile cluster adjustable resource evaluation unit is used to evaluate and predict the resource scheduling potential in the charging pile cluster; the communication unit is used to transmit the calculation results to the remote client monitoring platform via wireless transmission.

4. According to claim 1, the intelligent non-intrusive charging pile cluster load online monitoring device is characterized by: The remote client monitoring platform includes: a data visualization unit, a user interaction unit, a remote control unit and a communication unit; the data visualization unit is used to display the real-time load data and historical data of the charging pile cluster in a graphical interface; the user interaction unit is used to receive the instructions and parameter settings input by the user, and transmit the instructions and parameters to the charging pile load intelligent perception network big data computing server; the remote control unit remotely controls and manages the charging pile cluster according to the user instructions; the communication unit is used to receive the calculation results from the charging pile load intelligent perception network big data computing server.

5. An intelligent non-intrusive charging pile cluster load online monitoring method, characterized by: include: Using the charging pile cluster load sensing unit to obtain non-intrusive information of the charging pile cluster load data; The acquired non-intrusive load data is filtered, denoised and normalized by using the charging pile cluster load data preprocessing unit, and the processed data is transmitted to the charging pile cluster load data fusion unit; The charging pile cluster load data fusion unit is used to perform fusion operation on the pre-processed data; The charging pile cluster load data fusion unit manages the charging pile cluster based on the obtained fusion operation data; The communication unit is used to transmit the fused data to the charging pile load intelligent perception network big data computing server through wireless transmission.

6. The method for online monitoring of a cluster load of an intelligent non-intrusive charging pile according to claim 5 is characterized in that: The charging pile cluster load data preprocessing unit uses an improved wavelet threshold denoising method to perform data preprocessing on the non-intrusive acquired load data; The charging pile cluster load data fusion unit adopts a multi-scenario data fusion algorithm to perform fusion operations on the pre-processed data; the charging pile cluster load data fusion unit adopts a charging pile cluster clustering algorithm to manage the charging pile cluster.

7. The method for online monitoring of a cluster load of an intelligent non-intrusive charging pile according to claim 6 is characterized by: The data preprocessing using the improved wavelet threshold denoising method includes: Obtain load data from the load sensing unit of the charging pile cluster; Use wavelet basis functions to decompose the noise of the acquired load data; Calculate the noise variance μ n , noise variance μ n It is expressed as: Among them, ω j,k is the wavelet coefficient, median(·) is the median function, |·| is the absolute value operation; Calculate the threshold λ j , threshold λ j It is expressed as: Where j is the number of decomposition layers, N j Number the sample; According to the calculated threshold λ j , update the wavelet coefficients as follows: Among them, α is the adjustment coefficient, sgn(·) is the step function, and e is the natural constant; According to the obtained wavelet coefficients Get the denoised data.

8. The method for online monitoring of a cluster load of an intelligent non-intrusive charging pile according to claim 6 is characterized by: The multi-scenario data fusion algorithm includes: Calculate the Euclidean distance d between the charging pile cluster load data nmT as follows: Where n represents the nth charging pile, n' represents the n'th charging pile, N is the total number of charging piles, m is the number of charging pile cluster load data groups, M is the total number of charging pile cluster load data groups, T is time, S nm-T is the mth group of data of the nth charging pile at time T, S n'm-T is the mth group of data of the n'th charging pile at time T; Calculate the average value of each group of charging pile data as follows: Calculate the weight value w of the mth group of load data of the nth charging pile in the charging pile cluster at time T nm_T as follows: According to the calculated charging pile cluster load data Euclidean distance d nm-T and weight value w nm_T , calculate the charging pile cluster load fusion data S m_T as follows: Among them, s m_T It is the load data of a charging pile in the charging pile cluster.

9. The method for online monitoring of a cluster load of an intelligent non-intrusive charging pile according to claim 6 is characterized by: The method of managing the charging pile cluster by using the charging pile cluster clustering algorithm includes: Different charging piles are randomly assigned values ​​between 0 and 1 as initial values; Calculate the cluster head decision threshold T n as follows: Where p is the probability of a charging pile being selected as a cluster head, n is the total number of nodes in the sensor network, and d A→B is the distance from charging pile A to charging pile B, d max is the maximum distance of the charging pile, and e is a natural constant; Based on cluster head determination threshold T n Classify the charging pile cluster, which is higher than the cluster head judgment threshold T n The charging pile with a value lower than the cluster head judgment threshold T n The charging piles are treated as ordinary charging piles and added to the charging pile group.

10. A computer device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, characterized in that: When the processor executes the computer program, the steps of an intelligent non-intrusive charging pile cluster load online monitoring method as described in any one of claims 5 to 9 are implemented.

11. A computer storage medium, characterized in that: The computer storage medium stores a computer program, which, when executed by a processor, implements the steps of an intelligent non-intrusive charging pile cluster load online monitoring method as described in any one of claims 5 to 9.

Images