Method and device for detecting household energy storage system

Abstract

The invention discloses a method and device for detecting a household energy storage system, and relates to the technical field of energy control, and the method comprises the steps: receiving a monitoring request which is transmitted by user equipment and contains detailed feature identification information, the feature identification information covers a user geographic position, an energy storage equipment model, installation time, an initial performance parameter, a daily use mode and surrounding environment data, and the surrounding environment data comprises temperature, humidity and altitude. Through an advanced algorithm, target equipment is accurately obtained, multi-dimensional data is comprehensively collected, data processing is optimized, the service life state of the equipment is accurately predicted, detailed early warning is generated, early warning is transmitted in time by means of multiple communication modes, data safety is guaranteed through a block chain, backtracking analysis is facilitated by constructing a database, safety accidents can be reduced, the user experience is improved, and the user experience is improved. The industry development is promoted.

Description

Technical Field

[0001] The present invention relates to the field of energy control technology, and in particular to a detection method and device for a household energy storage system. Background Art

[0002] With the continuous growth of energy demand and the widespread application of renewable energy, household energy storage systems are becoming increasingly important in daily life. However, during the use of household energy storage equipment, due to various factors, its performance and lifespan may gradually decline, or even fail, causing safety hazards and economic losses to users.

[0003] At present, there are many deficiencies in the detection methods for household energy storage systems. Traditional detection methods often rely on users' regular self-inspections or simple equipment monitoring, which cannot accurately obtain the operating status and potential problems of the equipment in real time. Moreover, these methods perform poorly in terms of the comprehensiveness and accuracy of data collection and the ability to predict the life of the equipment, making it difficult to meet users' needs for safe and reliable operation of energy storage equipment.

[0004] In addition, the existing detection technology is relatively simple in the application of algorithms, lacks effective processing and in-depth analysis of complex data, and is unable to accurately judge the abnormal status of equipment and predict life changes. At the same time, the communication methods and early warning mechanisms are not perfect, resulting in users being unable to obtain key information in a timely manner and take corresponding measures. Summary of the invention

[0005] The object of the present invention is to provide a method and device for detecting a household energy storage system, so as to solve at least any one of the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solution: a method and device for detecting a household energy storage system, comprising the following steps:

[0007] Receiving a monitoring request sent by a user device containing detailed feature identification information, wherein the feature identification information includes the user's geographic location, energy storage device model, installation time, initial performance parameters, daily usage patterns and surrounding environment data, wherein the surrounding environment data includes, for example, temperature, humidity and altitude;

[0008] The characteristic identification information in the monitoring request is classified using a preset classification algorithm based on hierarchical clustering analysis and combined with genetic algorithm optimization. The specific formula is: Where D ij represents the distance between two data points i and j, x ik and jk They represent the values ​​of data points i and j in the kth dimension respectively, and n is the number of dimensions, so as to accurately obtain the target household energy storage equipment.

[0009] Preferably, multiple high-precision, high-sensitivity and self-calibration sensors installed on the target household energy storage device can collect real-time detection data covering multiple dimensions such as current, voltage, temperature, humidity, charge and discharge times, battery internal resistance, energy conversion efficiency, battery health status and device vibration frequency;

[0010] The data fusion algorithm based on extended Kalman filter and fused particle filter is used to fuse the collected multi-dimensional detection data to eliminate the noise and outliers in the data. The state update equation of the extended Kalman filter is: in is the estimated value of the state at time k, K k is the Kalman gain, z k is the measurement value at time k, and H is the measurement matrix.

[0011] Preferably, a prediction algorithm based on a combination of a deep neural network and a recurrent neural network and an attention mechanism is used to input the processed detection data into a pre-trained neural network model to determine whether the detection data is greater than or equal to the preset detection data; the training loss function of the neural network model is: Where N is the number of samples, y i is the true label, is the predicted label.

[0012] Preferably, when the detection data is greater than or equal to the preset detection data, it is confirmed that the service life of the target household energy storage device is in the first abnormal state, and based on the combination of fuzzy logic reasoning and expert system, and at the same time incorporating deep learning into the abnormal diagnosis algorithm, detailed warning information is generated, and the membership function of the fuzzy logic reasoning algorithm is: Where a and b are the set parameters, and x is the input variable;

[0013] The early warning information includes accurate assessment of the degree of abnormality and possible causes of failure, in-depth analysis, and actionable and targeted processing suggestions.

[0014] Preferably, the warning information is sent to the user equipment through a variety of communication methods, such as Bluetooth, Wi-Fi, mobile network, satellite communication and low-power wide area network, so that the user corresponding to the user equipment can process the target household energy storage device in a timely and accurate manner according to the warning information;

[0015] Build a database with data storage and backtracking functions to classify and store all data in the detection process.

[0016] Preferably, a receiving module, a classification module, a collection module, a data processing module, a judgment module and an early warning module;

[0017] The receiving module is used to receive a monitoring request including detailed feature identification information sent by a user equipment;

[0018] The classification module uses a preset classification algorithm based on hierarchical clustering analysis and combined with genetic algorithm optimization to classify the feature identification information in the monitoring request to accurately locate the target household energy storage device.

[0019] Preferably, the acquisition module collects multi-dimensional detection data of the target household energy storage device in real time through multiple high-precision, high-sensitivity sensors with self-calibration function;

[0020] The data processing module uses a data fusion algorithm based on extended Kalman filtering and fused with particle filtering to fuse the collected multi-dimensional detection data to improve the accuracy and reliability of the data.

[0021] Preferably, the judgment module adopts a prediction algorithm based on a combination of a deep neural network and a recurrent neural network and introduces an attention mechanism, inputs the processed detection data into a pre-trained neural network model, and accurately judges whether the detection data is greater than or equal to the preset detection data;

[0022] When detecting data abnormalities, the early warning module generates comprehensive and detailed early warning information based on the combination of fuzzy logic reasoning and expert system, and incorporates deep learning abnormal diagnosis algorithms, and sends it to user devices through multiple communication channels to assist users in timely and effective handling of abnormal situations of target household energy storage equipment.

[0023] The beneficial effects of the present invention are as follows:

[0024] In the present invention, through advanced cluster analysis algorithms and optimization methods, the target household energy storage equipment can be accurately obtained, the pertinence and efficiency of detection can be improved, and multiple high-precision sensors are used to collect multi-dimensional detection data in real time to fully reflect the operation status of the equipment. Then, the data fusion processing method that integrates multiple filtering algorithms is combined to eliminate noise and outliers, thereby providing accurate and reliable detection data. At the same time, the prediction algorithm that combines deep neural networks, recurrent neural networks and attention mechanisms can accurately judge the life status of the equipment, discover potential problems in advance, and generate detailed early warning information based on multiple advanced algorithms to provide users with abnormal degree assessment, fault cause analysis and targeted processing suggestions, which is convenient for users to handle in time. With the help of multiple communication methods, it is ensured that the early warning information is delivered to the user in time, blockchain technology is introduced to ensure data security, and a database is constructed to realize data storage and backtracking, which is convenient for subsequent analysis and optimization. This series of innovative measures can not only monitor and accurately predict the life status of the equipment in real time, discover anomalies in time, significantly reduce the occurrence of safety accidents, and protect the safety of users' lives and property, but also greatly improve the user experience, allowing users to more conveniently manage and maintain household energy storage equipment, and help promote the technological progress and development of the household energy storage system industry and improve the overall industry level. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a partial flow chart of the detection method of the present invention;

[0026] Figure 2 It is a schematic diagram of some modules of the detection device of the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] The present invention provides Figure 1-2 A method and device for detecting a household energy storage system shown includes the following steps:

[0029] Receiving a monitoring request sent by a user device containing detailed feature identification information, wherein the feature identification information includes the user's geographic location, energy storage device model, installation time, initial performance parameters, daily usage patterns and surrounding environment data, wherein the surrounding environment data includes, for example, temperature, humidity and altitude;

[0030] The characteristic identification information in the monitoring request is classified using a preset classification algorithm based on hierarchical clustering analysis and combined with genetic algorithm optimization. The specific formula is: Where D ij represents the distance between two data points i and j, x ik and jk They represent the values ​​of data points i and j in the kth dimension respectively, and n is the number of dimensions, so as to accurately obtain the target household energy storage equipment.

[0031] Through the installation of multiple high-precision, high-sensitivity and self-calibration sensors on the target household energy storage equipment, real-time detection data covering multiple dimensions such as current, voltage, temperature, humidity, charge and discharge times, battery internal resistance, energy conversion efficiency, battery health status and equipment vibration frequency can be collected;

[0032] The data fusion algorithm based on extended Kalman filter and fused particle filter is used to fuse the collected multi-dimensional detection data to eliminate the noise and outliers in the data. The state update equation of the extended Kalman filter is: in is the estimated value of the state at time k, K k is the Kalman gain, z k is the measurement value at time k, and H is the measurement matrix.

[0033] A prediction algorithm based on a combination of a deep neural network and a recurrent neural network and an attention mechanism is used to input the processed test data into a pre-trained neural network model to determine whether the test data is greater than or equal to the preset test data; the training loss function of the neural network model is: Where N is the number of samples, y i is the true label, is the predicted label.

[0034] When the detection data is greater than or equal to the preset detection data, it is confirmed that the service life of the target household energy storage device is in the first abnormal state, and based on the combination of fuzzy logic reasoning and expert system, the abnormal diagnosis algorithm of deep learning is integrated to generate detailed warning information. The membership function of the fuzzy logic reasoning algorithm is: Where a and b are the set parameters, and x is the input variable;

[0035] The early warning information includes accurate assessment of the degree of abnormality and possible causes of failure, in-depth analysis, and actionable and targeted processing suggestions.

[0036] Send warning information to user devices through various communication methods, such as Bluetooth, Wi-Fi, mobile networks, satellite communications, and low-power wide area networks, so that users of the user devices can handle the target household energy storage devices in a timely and accurate manner according to the warning information;

[0037] Build a database with data storage and backtracking functions to classify and store all data in the detection process.

[0038] A detection device for a household energy storage system, comprising: a receiving module, a classification module, a collection module, a data processing module, a judgment module and an early warning module;

[0039] The receiving module is used to receive a monitoring request including detailed feature identification information sent by a user equipment;

[0040] The classification module uses a preset classification algorithm based on hierarchical clustering analysis and combined with genetic algorithm optimization to classify the feature identification information in the monitoring request to accurately locate the target household energy storage device.

[0041] The acquisition module collects multi-dimensional detection data of the target household energy storage device in real time through multiple high-precision, high-sensitivity sensors with self-calibration function;

[0042] The data processing module uses a data fusion algorithm based on extended Kalman filtering and fused with particle filtering to perform fusion processing on the collected multi-dimensional detection data, thereby improving the accuracy and reliability of the data.

[0043] The judgment module adopts a prediction algorithm based on a combination of a deep neural network and a recurrent neural network and introduces an attention mechanism, inputs the processed detection data into a pre-trained neural network model, and accurately determines whether the detection data is greater than or equal to the preset detection data;

[0044] When detecting data abnormalities, the early warning module generates comprehensive and detailed early warning information based on the combination of fuzzy logic reasoning and expert system, and incorporates deep learning abnormal diagnosis algorithms, and sends it to user devices through multiple communication channels to assist users in timely and effective handling of abnormal situations of target household energy storage equipment.

[0045] The following embodiments are included:

[0046] Our detection system first receives the monitoring request sent by the user through the mobile phone application, which contains the above detailed feature identification information, such as the latitude and longitude of the user's geographical location, the specific model and number of the energy storage device, the installation time, the initial performance parameters, the daily usage mode and the surrounding environment data; the system uses a classification algorithm based on hierarchical clustering analysis and combined with genetic algorithm optimization to process this information. By calculating the distance between data points Among them, n covers multiple dimensions, such as device model, usage mode, geographic location, etc.; combined with the cluster center and number optimized by the genetic algorithm, the user's energy storage device is accurately classified into the corresponding category and determined as the target household energy storage device. Then, the high-precision, high-sensitivity and self-calibration sensors installed on the device, including current sensors, voltage sensors, temperature sensors, humidity sensors, equipment vibration frequency sensors, etc., collect multi-dimensional detection data in real time, such as current, voltage, temperature, humidity, charge and discharge times, battery internal resistance, energy conversion efficiency, equipment vibration frequency, etc.; then, a data fusion algorithm based on extended Kalman filtering and fused particle filtering is used to process these data; the state update equation of the extended Kalman filter is: The Kalman gain K k According to the changes in data, the measurement matrix H is determined according to the characteristics of the sensor and the measurement principle. The particle filter is used to deal with nonlinear and non-Gaussian situations, further eliminating noise and outliers in the data, and obtaining more accurate and reliable fused data. After that, the processed data is input into the prediction algorithm model based on the combination of deep neural network and recurrent neural network, and the attention mechanism is introduced. The training loss function of the neural network model is: The model has been trained in advance using a large amount of historical test data, and determines that the current test data is close to or exceeds the preset test data, indicating that the equipment may be in an abnormal life state. When the equipment life is confirmed to be abnormal, the abnormal diagnosis algorithm based on fuzzy logic reasoning combined with the expert system and deep learning is used to generate detailed warning information. The membership function of the fuzzy logic reasoning algorithm is: According to the performance and historical data settings of the equipment, the warning information indicates that the abnormality of the equipment is moderate. The possible cause of the failure is that the internal resistance increases due to battery aging. It is recommended that the user contact professionals as soon as possible to replace the battery, and specific processing steps and precautions are provided; finally, the warning information is sent to the user's mobile phone application through multiple communication methods such as Bluetooth, Wi-Fi and mobile network. After receiving the warning, the user contacted the maintenance personnel in time for processing, avoiding possible more serious failures. During the entire detection process, all data is stored in a database with data storage and backtracking functions, and blockchain technology is used to ensure the security and non-tamperability of the data. These data can be further analyzed and model optimized in the future to improve the performance and accuracy of the detection system. Through the above detailed embodiments, the household energy storage system detection method of this patent can effectively realize real-time and accurate detection and warning of household energy storage equipment, ensure the safe and stable operation of the equipment, and improve the user experience.

[0047] The following is a detailed explanation of each part of the above formula:

[0048] Distance formula in hierarchical clustering analysis algorithm: D ij : represents the distance between data points i and j. The larger the distance, the greater the difference in features between the two points; the smaller the distance, the more similar they are.

[0049] x ik and jk : Represents the values ​​of data points i and j in the kth dimension. For example, if we use geographic location (longitude, latitude) and device model to describe an energy storage device, then x i1 is the longitude of device i, x i2 is the latitude, y j1 and j2 is the corresponding value of device j;

[0050] n: indicates the number of dimensions. For example, the geographic location mentioned above has 2 dimensions (longitude and latitude), plus the device model, which is 3 dimensions, so n is 3;

[0051] The entire formula obtains the comprehensive distance between two data points by calculating the square root of the sum of the squares of the differences in each dimension.

[0052] The state update equation of the extended Kalman filter algorithm is:

[0053] Represents the estimated value of the system state at time . This is the latest estimate of the state of the energy storage device that we want to obtain.

[0054] It is the estimated value of the system state at the previous moment (k-1).

[0055] K k : It is called the Kalman gain, which determines the new measurement value z k The degree of influence on the estimated value. If the Kalman gain is large, it means that the new measurement value is more credible and the correction to the estimated value is large; if the Kalman gain is small, it means that the previous estimate is more trusted and the correction is small.

[0056] z k : It is the measured value at the moment, that is, the data of the energy storage device actually detected by the sensor.

[0057] H: is the measurement matrix, which relates the state variables to the measurements.

[0058] Represents the current measured value predicted based on the estimated value at the previous moment.

[0059] The difference between the measured and predicted values ​​is called the measurement residual.

[0060] The purpose of this equation is to comprehensively consider the estimate at the previous moment, the new measurement, and the Kalman gain to obtain a more accurate state estimate at the current moment.

[0061] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting a household energy storage system, characterized in that: The following steps are involved: Receiving a monitoring request sent by a user device containing detailed feature identification information, wherein the feature identification information includes the user's geographic location, energy storage device model, installation time, initial performance parameters, daily usage patterns and surrounding environment data, wherein the surrounding environment data includes, for example, temperature, humidity and altitude; The characteristic identification information in the monitoring request is classified using a preset classification algorithm based on hierarchical clustering analysis and combined with genetic algorithm optimization. The specific formula is: Where D ij represents the distance between two data points i and j, x ik and jk They represent the values ​​of data points i and j in the kth dimension respectively, and n is the number of dimensions, so as to accurately obtain the target household energy storage equipment.

2. A method for detecting a household energy storage system according to claim 1, characterized in that: Through the installation of multiple high-precision, high-sensitivity and self-calibration sensors on the target household energy storage equipment, real-time detection data covering multiple dimensions such as current, voltage, temperature, humidity, charge and discharge times, battery internal resistance, energy conversion efficiency, battery health status and equipment vibration frequency can be collected; The data fusion algorithm based on extended Kalman filter and fused particle filter is used to fuse the collected multi-dimensional detection data to eliminate the noise and outliers in the data. The state update equation of the extended Kalman filter is: in is the estimated value of the state at time k, K k is the Kalman gain, z k is the measurement value at time k, and H is the measurement matrix.

3. A method for detecting a household energy storage system according to claim 1, characterized in that: A prediction algorithm based on a combination of a deep neural network and a recurrent neural network and an attention mechanism is used to input the processed test data into a pre-trained neural network model to determine whether the test data is greater than or equal to the preset test data; the training loss function of the neural network model is: Where N is the number of samples, y i is the true label, is the predicted label.

4. A method for detecting a household energy storage system according to claim 1, characterized in that: When the detection data is greater than or equal to the preset detection data, it is confirmed that the service life of the target household energy storage device is in the first abnormal state, and based on the combination of fuzzy logic reasoning and expert system, the abnormal diagnosis algorithm of deep learning is integrated to generate detailed warning information. The membership function of the fuzzy logic reasoning algorithm is: Where a and b are the set parameters, and x is the input variable; The early warning information includes accurate assessment of the degree of abnormality and possible causes of failure, in-depth analysis, and actionable and targeted processing suggestions.

5. A method for detecting a household energy storage system according to claim 1, characterized in that: Send warning information to user devices through various communication methods, such as Bluetooth, Wi-Fi, mobile networks, satellite communications, and low-power wide area networks, so that users of the user devices can handle the target household energy storage devices in a timely and accurate manner according to the warning information; Build a database with data storage and backtracking functions to classify and store all data in the detection process.

6. A detection device for a household energy storage system, characterized in that: include: Receiving module, classification module, acquisition module, data processing module, judgment module and early warning module; The receiving module is used to receive a monitoring request including detailed feature identification information sent by a user equipment; The classification module uses a preset classification algorithm based on hierarchical clustering analysis and combined with genetic algorithm optimization to classify the feature identification information in the monitoring request to accurately locate the target household energy storage device.

7. A detection device for household energy storage system according to claim 6, characterized in that: The acquisition module collects multi-dimensional detection data of the target household energy storage device in real time through multiple high-precision, high-sensitivity sensors with self-calibration function; The data processing module uses a data fusion algorithm based on extended Kalman filtering and fused with particle filtering to perform fusion processing on the collected multi-dimensional detection data, thereby improving the accuracy and reliability of the data.

8. The detection device for household energy storage system according to claim 6, characterized in that: The judgment module adopts a prediction algorithm based on a combination of a deep neural network and a recurrent neural network and introduces an attention mechanism, inputs the processed detection data into a pre-trained neural network model, and accurately determines whether the detection data is greater than or equal to the preset detection data; When detecting data abnormalities, the early warning module generates comprehensive and detailed early warning information based on the combination of fuzzy logic reasoning and expert system, and incorporates deep learning abnormal diagnosis algorithms, and sends it to user devices through multiple communication channels to assist users in timely and effective handling of abnormal situations of target household energy storage equipment.

Images