Multi-way switch dynamic control platform based on multi-source data

Abstract

The present invention discloses a multi-way switch dynamic control platform based on multi-source data, which relates to the field of electric switch technology. The system includes: obtaining multi-source data streams of the environment and multi-source data streams of the equipment operation, and simultaneously obtaining target scene behavior data, and encrypting and transmitting to a data processing module; mapping and preprocessing the collected data with multi-channel data preprocessing to obtain an application scene fusion data stream; obtaining key feature information of the application scene, constructing a switch control strategy solution space, matching and optimizing to obtain a target switch control strategy solution; and dynamically executing and controlling the multi-way switch device based on the target switch control strategy solution through a switch execution module. The system solves the technical problem that the existing multi-way switch control relies on a single data source and a fixed control strategy, is difficult to meet the needs of complex and changeable application scenarios, and results in insufficient flexibility and accuracy in the control of the switch state, and achieves the technical effect of improving the flexibility and accuracy of the multi-way switch control.

Description

Technical Field

[0001] The present application relates to the technical field related to electric switches, and in particular to a multi-way switch dynamic control platform based on multi-source data. Background Art

[0002] Compared with traditional single-way switches, multi-way switches can control the switch status of two or more electrical appliances or devices at the same time. In the circuit, the internal structure of the multi-way switch is composed of multiple switches, each of which can independently control the power connection of one or more electrical appliances or devices. With the growing demand for efficient, flexible and precise control in modern industrial automation, smart home, network communication and other fields, multi-way switches are becoming more and more widely used, especially in the fields of smart home and lighting control. Through multi-way switches, users can achieve centralized control of different electrical appliances in the room, such as lights, air conditioners, televisions, etc. However, traditional multi-way switch control is often based on fixed logic or preset programs to perform control tasks, which is powerless when dealing with complex and changing environments. It is difficult to dynamically adjust the switch status according to current environmental conditions and user needs, and thus it is impossible to achieve real-time, precise and flexible control of the equipment.

[0003] Therefore, the current multi-way switch control related technologies rely on a single data source and a fixed control strategy, which makes it difficult to meet the needs of complex and changing application scenarios, resulting in technical problems such as insufficient flexibility and precision in controlling the switch state. Summary of the invention

[0004] This application provides a multi-way switch dynamic control platform based on multi-source data, and adopts technical means such as building data processing channels and extracting key features to solve the technical problem that the existing multi-way switch control relies on a single data source and fixed control strategy, is difficult to meet the needs of complex and changing application scenarios, and leads to insufficient flexibility and precision in the control of the switch state, thereby achieving the technical effect of improving the flexibility and accuracy of multi-way switch control.

[0005] The present application provides a multi-way switch dynamic control platform based on multi-source data, the system includes: a data acquisition module, the data acquisition module is wirelessly connected to a sensor group and a user application device, and obtains an environmental multi-source data stream and a device operation multi-source data stream through the sensor group, and simultaneously obtains target scene behavior data through the user application device, and encrypts and transmits the environmental multi-source data stream, the device operation multi-source data stream and the target scene behavior data to a data processing module; a data processing module, through which a data preprocessing multi-channel is constructed, and the environmental multi-source data stream, the device operation multi-source data stream and the target scene behavior data are encrypted and transmitted to the data processing module; The invention relates to a method for implementing a switch control strategy scheme space, a switch execution module, a switch execution module and a multi-channel mapping preprocessing method for the data to obtain a fusion data stream of the application scenario; a control decision module, a key feature extraction of the fusion data stream of the application scenario to obtain key feature information of the application scenario, a switch control strategy scheme space is constructed according to the control decision module, and a matching optimization is performed with the switch control strategy scheme space based on the key feature information of the application scenario to obtain a target switch control strategy scheme; a switch execution module, the switch execution module is communicatively connected with the multi-way switch device, and the multi-way switch device is dynamically executed and controlled based on the target switch control strategy scheme through the switch execution module.

[0006] In the implementation of the present application, the construction of the data preprocessing multi-channel also performs the following processing: obtaining data attribute identification elements through the data processing module, the data attribute identification elements including data type identification elements and data structured format identification elements; based on the data attribute identification elements, structured classification is performed on the environment multi-source data stream, the device operation multi-source data stream, and the target scenario behavior data to obtain a data type attribute set and a data structured classification factor set; preprocessing requirements analysis is performed on the data type attribute set and the data structured classification factor set to determine a data structured factor preprocessing program set; data preprocessing channel configuration is performed based on the data structured factor preprocessing program set to construct the data preprocessing multi-channel.

[0007] In the implementation method of the present application, the key feature information of the application scenario is obtained, and the following processing is also performed: the application scenario fusion data stream is standardized to obtain a standard fusion data stream of the application scenario; decentralized processing is performed based on the standard fusion data stream of the application scenario to obtain a covariance matrix, and the eigenvalues ​​and eigenvectors of the covariance matrix are obtained by solving; principal component information greater than a preset threshold is selected according to the eigenvalue, and the standard fusion data stream of the application scenario is projected onto the eigenvector based on the principal component information to determine the key feature information of the application scenario.

[0008] In the implementation of the present application, the target switch control strategy scheme is obtained by further performing the following processing: using the key feature information of the application scenario as a constraint parameter, performing similarity calculation with the switch control strategy scheme space, and obtaining an application scenario feature similarity set; setting a similarity threshold, and filtering and matching the application scenario feature similarity set based on the similarity threshold, and obtaining a switch control strategy scheme search library; defining a switch control target, and constructing a switch cost function according to the switch control target; and performing global evaluation and optimization on the switch control strategy scheme search library based on the switch cost function, and obtaining the target switch control strategy scheme.

[0009] In the implementation of the present application, the target switch control strategy scheme is obtained by further performing the following processing: a global evaluation calculation is performed on the switch control strategy scheme search library through the switch cost function to obtain cost information of multiple switch control strategy schemes; strategy schemes are optimized based on the cost information of the multiple switch control strategy schemes to determine a matching control strategy scheme, and a key feature matching degree is determined based on the matching control strategy scheme; a strategy parameter fine-tuning step is determined based on the key feature matching degree, and the matching control strategy scheme is adaptively adjusted based on the strategy parameter fine-tuning step to obtain multiple fine-tuning control strategy schemes; the multiple fine-tuning control strategy schemes are evaluated and optimized based on the switch cost function to determine the target switch control strategy scheme.

[0010] In the implementation of the present application, the switch cost function is constructed according to the switch control target, and the following processing is also performed: based on the switch control target, the cost function is fitted to the switch control strategy scheme space respectively to obtain a set of sub-control target cost functions; according to the switch control requirements, the importance of each sub-target in the switch control target is evaluated to obtain a set of sub-target weight factors; based on the sub-target weight factor set, the sub-control target cost function set is weighted and fused to construct the switch cost function.

[0011] In the implementation method of the present application, the dynamic execution control of the multiple switch devices based on the target switch control strategy scheme also performs the following processing: execution control monitoring of the multiple switch devices based on the target switch control strategy scheme to obtain a multiple switch operation monitoring data stream; performance optimization direction analysis of the multiple switch operation monitoring data stream to obtain a control strategy variation direction; dynamic optimization of the target switch control strategy scheme based on the control strategy variation direction to determine the target switch control optimization strategy scheme, and dynamic execution control of the multiple switch devices based on the target switch control optimization strategy scheme.

[0012] In the implementation of the present application, the determination of the target switch control optimization strategy scheme also performs the following processing: performing correlation analysis on the multi-way switch operation monitoring data stream based on the control strategy variation direction to obtain the control strategy variation distribution; performing mutation update on the target switch control strategy scheme according to the control strategy variation distribution to obtain a set of switch control variation strategy schemes; and performing evaluation and optimization based on the set of switch control variation strategy schemes to determine the target switch control optimization strategy scheme.

[0013] It is proposed to connect the performance analysis platform through the multi-way switch dynamic control platform based on multi-source data proposed in this application to obtain the performance analysis data of engineering materials; perform data stability probability analysis based on the performance analysis data, and determine the constraint performance based on the data trend stability probability; perform parameter decomposition on the extrusion control process, and use the process target results to perform support relationship calculations on the decomposed parameters to obtain the core control parameters of the process; establish a mapping relationship between the constraint performance and the core control parameters of the process, and configure the process optimization variables; based on the process optimization variables, optimize the control parameters with the maximum control target, obtain the extrusion control information, and send it to the control center for extrusion operation process control. It solves the technical problems that the existing multi-way switch control relies on a single data source and a fixed control strategy, which is difficult to meet the needs of complex and changeable application scenarios, resulting in insufficient flexibility and accuracy in the control of the switch state, and achieves the technical effect of improving the flexibility and accuracy of the multi-way switch control. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0015] Figure 1 A schematic diagram of the structure of a multi-way switch dynamic control platform based on multi-source data provided in an embodiment of the present application;

[0016] Figure 2 A schematic diagram of the execution process of a data processing module in a multi-way switch dynamic control platform based on multi-source data provided in an embodiment of the present application.

[0017] Description of reference numerals: data acquisition module 10 , data processing module 20 , control decision module 30 , switch execution module 40 . DETAILED DESCRIPTION

[0018] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0019] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0020] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0021] The embodiment of the present application provides a multi-way switch dynamic control platform based on multi-source data, such as Figure 1 As shown, the system comprises:

[0022] The data acquisition module 10 is wirelessly connected to the sensor group and the user application device, and obtains the environment multi-source data stream and the device operation multi-source data stream through the sensor group, and obtains the target scene behavior data through the user application device, and encrypts and transmits the environment multi-source data stream and the device operation multi-source data stream and the target scene behavior data to the data processing module. The data acquisition module, the sensor group, the user application device and the data processing module together constitute an intelligent control or monitoring system, which realizes the real-time collection and transmission of the environment multi-source data stream, the device operation multi-source data stream and the target scene behavior data through wireless connection to support the dynamic control of the target environment or equipment, such as the smart home scene, which collects indoor environmental parameters (such as temperature, humidity, light, etc.), equipment status (such as air conditioning, lighting, curtains, etc.) and user demand scene behavior such as playing games in real time through multiple sensors. Through control strategy analysis, dynamic switch control solutions such as turning on the master bedroom lights and opening and closing curtains are executed. Specifically, the data acquisition module is responsible for collecting data from various sources (such as sensor groups, user application devices, etc.), including wireless communication interfaces (such as Wi-Fi, Bluetooth, ZigBee, etc.), which are used to wirelessly connect to the sensor group and user application devices. The sensor group contains multiple different types of sensors for real-time monitoring and measurement of various parameters of the target environment or equipment. For example, temperature sensors are used to measure ambient temperature, humidity sensors are used to measure ambient humidity, and current sensors, voltage sensors, etc. are used to monitor the operating status of the equipment. The various environmental data collected by the sensors constitute the environmental multi-source data stream, and the collected equipment operating status-related data constitute the equipment operation multi-source data stream. Among them, user application devices (such as smart phones, tablets, etc.) are the interaction interfaces between users and the system. Users use these The target scene behavior data is input by the application device, such as setting temperature, brightness, mode, etc. The target scene behavior data refers to the data input by the user, which describes the state or behavior that the user wants the system to achieve, and encrypts the multi-source data stream of the environment, the multi-source data stream of the device operation, and the target scene behavior data and transmits them to the data processing module. The data encryption transmission is to ensure the security and integrity of the data. The data acquisition module encrypts the data before transmitting it to the data processing module. For example, encryption algorithms (such as AES, RSA, etc.) and key management are used to encrypt the data to ensure that only authorized devices or systems can decrypt and access it. The data processing module is responsible for receiving, processing and analyzing the data from the data acquisition module. The data processing module may include high-performance processors, memories, and algorithm libraries to perform complex computing tasks such as data analysis, pattern recognition, decision making, etc.

[0023] The data processing module 20 constructs a data preprocessing multi-channel through the data processing module, maps and preprocesses the environment multi-source data stream, the device operation multi-source data stream and the target scene behavior data with the data preprocessing multi-channel to obtain an application scenario fusion data stream. The data processing module realizes the preprocessing and mapping of the multi-source data stream of the environment, the multi-source data stream of the equipment operation, and the target scene behavior data by constructing a data preprocessing multi-channel, thereby generating an application scenario fusion data stream. Specifically, the data processing module is designed with multiple preprocessing channels, each of which contains specific data processing algorithms and rules to adapt to the preprocessing requirements of different types of data and is responsible for processing a certain type of data. The data preprocessing multi-channel can work in parallel to improve data processing efficiency. In the data preprocessing stage, the data processing module will map the received multi-source data stream of the environment, the multi-source data stream of the equipment operation, and the target scene behavior data with the corresponding preprocessing channel, which means that each type of data will be assigned to the corresponding preprocessing channel for targeted processing. In each preprocessing channel, the data will undergo a series of preprocessing operations, such as data cleaning, data integration, data transformation, feature extraction, and data reduction, etc., aiming to improve the quality of the data, unify the format and structure of the data, reduce the redundancy and noise of the data, and extract useful features that are helpful for subsequent analysis. The preprocessed data will be integrated into a high-quality, unified format and structure fusion data stream, namely, the application scenario fusion data stream.

[0024] The control decision module 30 is used to extract key features from the application scenario fusion data stream, obtain key feature information of the application scenario, construct a switch control strategy solution space according to the control decision module, match and optimize the switch control strategy solution space based on the key feature information of the application scenario, and obtain the target switch control strategy solution. The main task of the control decision module is to formulate and optimize the switch control strategy according to the application scenario fusion data stream. The control decision module extracts key features from the application scenario fusion data stream, and matches and optimizes the control strategy solution space based on the extracted key feature information of the application scenario to obtain the optimal switch control strategy solution, that is, the target switch control strategy solution. Specifically, a specific feature algorithm (such as statistical feature extraction, etc.) is used to extract key feature information from the application scenario fusion data stream. The key features may include various parameters, indicators or patterns, which can reflect the current state of the system or equipment, environmental changes or user behavior, etc. The control decision module constructs a switch control strategy solution space according to the system requirements and design, which includes all possible switch control strategy solutions. Each solution corresponds to different control parameters and settings. These strategy solutions may be based on expert knowledge, historical data or machine learning models. The control decision module is generated based on the key feature information of the application scenario, aiming to cover various possible application scenarios and situations. The control decision module matches the switch control strategy scheme space according to the key feature information of the application scenario, that is, it is achieved by calculating the similarity or fitness between the feature information and each strategy scheme. The goal of matching optimization is to find one or more strategy schemes that best match or are optimal with the key feature information through search and optimization to ensure that the switch control can meet the needs and conditions of the current application scenario. After matching optimization, the control decision module will determine a target switch control strategy scheme, which is obtained based on the matching results of the key feature information and the strategy scheme space. It contains specific control instructions for each switch device, such as the on / off state of the switch, switching time, priority, etc., which is considered to be the most suitable for the current application scenario. The control decision module will transmit the target switch control strategy scheme to the execution module or device to achieve real-time control of the switch.

[0025] The switch execution module 40 is connected to the multi-way switch device in communication, and the multi-way switch device is dynamically executed and controlled by the switch execution module based on the target switch control strategy scheme. The switch execution module is an important component of the control system, which is responsible for communicating with the multi-way switch device and dynamically executing and controlling these switch devices according to the target switch control strategy scheme. Specifically, the switch execution module establishes a communication connection with the multi-way switch device through a certain communication protocol (such as Modbus, CAN bus, Ethernet, etc.), allowing the execution module to send control instructions to the switch device and receive status information from the switch device. The switch execution module receives the target switch control strategy scheme and dynamically controls the status of the multi-way switch device according to it, including opening or closing a specific switch, adjusting the switching frequency of the switch, changing the priority of the switch, etc., wherein dynamic execution control refers to adjusting the status of the switch device in real time according to the changes in real-time data and application scenarios to meet the needs and conditions of the system, and can achieve accurate and flexible control of the switch status.

[0026] The multi-way switch dynamic control platform based on multi-source data according to the embodiment of the present invention is used to solve the technical problem that the existing multi-way switch control relies on a single data source and a fixed control strategy, which is difficult to meet the needs of complex and changeable application scenarios, resulting in insufficient flexibility and accuracy in the control of the switch state, thereby achieving the technical effect of improving the flexibility and accuracy of the multi-way switch control. The multi-way switch dynamic control platform based on multi-source data includes: a data acquisition module 10, a data processing module 20, a control decision module 30, and a switch execution module 40.

[0027] The specific configuration of the data processing module 20 will be described in detail below. Figure 2As shown, the data processing module 20 may further include: obtaining data attribute identification elements through the data processing module, and the data attribute identification elements include data type identification elements and data structured format identification elements. Obtaining data attribute identification elements through the data processing module refers to extracting metadata about the data. Data attribute identification elements are important elements used to describe and distinguish different data features, including data type identification elements and data structured format identification elements. Specifically, data structured format identification elements are used to describe how data is organized and stored, including structured, semi-structured and unstructured data. Data type identification elements refer to attributes used to describe the types of data items or data sets, including environmental data, device status data, and user behavior data. It also includes, based on the data attribute identification elements, structured classification of the environmental multi-source data stream, the device operation multi-source data stream, and the target scenario behavior data is performed to obtain a data type attribute set and a data structured classification factor set. Based on data attribute identification elements, the multi-source data streams of the environment, the multi-source data streams of the equipment operation, and the target scene behavior data are structured and classified, aiming to convert the original and diverse data streams into a data set with clear data type attributes and structured classification factors, so as to facilitate subsequent data management and analysis, and obtain a data type attribute set and a data structured classification factor set. Specifically, the data processing module will analyze each data item in the data stream and identify its data type. Each data item will be assigned a clear data type and classified into a data type attribute set; the data stream may be unstructured (such as plain text), semi-structured (such as data in JSON or XML format) or structured (such as data in a relational database). The data processing module will identify the structural features of the data stream and classify it into a data structured classification factor set. It also includes pre-processing requirements analysis of the data type attribute set and the data structured classification factor set to determine a data structured factor pre-processing program set. Perform preprocessing demand analysis on the data type attribute set and the data structured classification factor set, determine the specific preprocessing requirements based on the analysis results, and then select or design appropriate preprocessing programs to form a data structured factor preprocessing program set. For example, structured data requires data cleaning, unstructured data requires filtering and denoising, and the standardized formats of various types of data such as environmental data and equipment status data are also different. Configure different preprocessing channels according to the preprocessing program to improve data processing efficiency. It also includes configuring data preprocessing channels based on the data structured factor preprocessing program set to construct the data preprocessing multi-channel. According to the different attributes, types and structures of the data, design and implement multiple independent data preprocessing channels, that is, data preprocessing multi-channels, each channel contains a series of preprocessing steps and procedures for specific data types to ensure that the data meets specific formats, standards and requirements before entering the analysis or machine learning model.

[0028] The specific configuration of the control decision module 30 will be described in detail below. The control decision module 30 may further include: performing standard processing on the application scenario fusion data stream to obtain an application scenario standard fusion data stream. The application scenario fusion data streams from different sources and with different formats and characteristics are integrated into a fusion data stream with a unified format, structure and quality standard through a series of standardized processing steps to obtain an application scenario standard fusion data stream. It also includes performing decentralized processing based on the application scenario standard fusion data stream to obtain a covariance matrix, and solving to obtain the eigenvalues ​​and eigenvectors of the covariance matrix. For the application scenario standard fusion data stream, the purpose of decentralization is to eliminate the average trend of the data so that the change of the data only reflects the fluctuation relative to the mean. The covariance matrix is ​​a symmetric matrix whose elements represent the covariance between different features in the data set. Covariance is an indicator of how two variables change together. Positive values ​​indicate that the two variables change in the same direction, negative values ​​indicate changes in the opposite direction, and the size of the value indicates the degree of change. The eigenvalue indicates the size of the data variance in the direction of the corresponding eigenvector, that is, the degree of dispersion of the data in this direction. The larger eigenvalue corresponds to the direction in which the data changes more; the eigenvector indicates the main direction of data change. In data processing, it is usually associated with the intrinsic structure or pattern of the data. By solving the eigenvalues ​​and eigenvectors of the covariance matrix, the most important change direction and pattern in the data set can be identified. It also includes selecting principal component information greater than a preset threshold according to the eigenvalue, projecting the application scenario standard fusion data stream onto the eigenvector based on the principal component information, and determining the key feature information of the application scenario. In principal component analysis, each principal component corresponds to an eigenvalue. The size of the eigenvalue reflects the importance of the principal component in describing the changes in the original data. The preset threshold is a value set according to the specific application scenario and requirements. Selecting an eigenvalue greater than this threshold means retaining only those principal components that can describe the main changes in the original data. The purpose is to reduce the dimension of the data and simplify the data set by reducing the number of principal components while retaining the important information in the original data as much as possible. After selecting the important principal components, the original data (that is, the standard fusion data stream of the application scenario) is projected onto the eigenvectors corresponding to these principal components through matrix multiplication. Through this projection process, the n-dimensional features of the original data are converted to k-dimensional, where k is the number of selected principal components. This k-dimensional data is the projection of the original data in the k main change directions, which is regarded as a new feature representation of the original data, and finally determines the key feature information of the application scenario.

[0029] The specific configuration of the control decision module 30 will be described in detail below. The control decision module 30 may further include: using the key feature information of the application scenario as a constraint parameter, performing similarity calculation with the switch control strategy scheme space, and obtaining an application scenario feature similarity set. Using similarity calculation methods such as cosine similarity and Euclidean distance, the similarity between the key feature information of the application scenario and each control strategy scheme is compared to find the control strategy scheme that best meets the application scenario requirements. The result of the similarity calculation will form a set, namely the application scenario feature similarity set. Each element in this set represents a similarity score between a control strategy scheme and the key feature information of the application scenario, and the control strategy scheme that best matches the application scenario can be found. It also includes setting a similarity threshold, screening and matching the application scenario feature similarity set based on the similarity threshold, and obtaining a switch control strategy scheme search library. The similarity threshold is a preset value used to determine whether the similarity between the key feature information of the application scenario and the switch control strategy solution is high enough. The similarity threshold is used to screen the application scenario feature similarity set. If the similarity score between a certain control strategy solution and the key feature information of the application scenario is higher than or equal to the similarity threshold, the control strategy solution is added to the switch control strategy solution search library. After screening and matching, a switch control strategy solution search library is integrated. It also includes defining a switch control target and constructing a switch cost function according to the switch control target. The switch control target refers to a specific state or performance indicator that the control system expects to achieve, such as switch energy consumption, device response time, switch stability, etc. In a smart home system, the switch control target may be to ensure that the indoor temperature is maintained within a comfortable range; in an automated production line, the switch control target may be to ensure the stable operation of the machine equipment and the product quality; the switch cost function (or cost function, loss function) is used to quantify the gap between the current system state and the control target. By constructing the switch cost function, the control problem can be converted into an optimization problem. The goal of the control system is to find a control strategy that minimizes the cost function, that is, to achieve the optimization of the control target. The method further includes performing a global evaluation and optimization on the switch control strategy solution search library based on the switch cost function to obtain the target switch control strategy solution.

[0030] The specific configuration of the control decision module 30 will be described in detail below. The control decision module 30 may further include: performing a global evaluation calculation on the switch control strategy solution search library through the switch cost function to obtain multiple switch control strategy solution cost information. Applying the defined switch cost function to each solution in the switch control strategy solution search library means substituting the specific parameters of each solution (such as the switch on time, off time, switch sequence, etc.) into the cost function, calculating a corresponding cost value, and obtaining a complete cost information set, which includes the cost values ​​of all solutions in the search library, and finally obtaining multiple switch control strategy solution cost information. It also includes optimizing the strategy solution based on the multiple switch control strategy solution cost information, determining the matching control strategy solution, and determining the key feature matching degree according to the matching control strategy solution. According to the cost information of the obtained multiple switch control strategy solutions, the solutions are optimized, and the solution with the lowest cost or the best performance is selected. One or more matching control strategy solutions are determined, which are selected according to the key feature information of the application scenario, similarity calculation and cost information evaluation, and have high matching degree and performance. After the matching control strategy solution is determined, it is necessary to further evaluate the matching degree between the solution and the key feature information of the application scenario. The key feature matching degree is used to measure the similarity or consistency between the control strategy solution and the key features of the application scenario. It also includes, according to the key feature matching degree, determining the fine-tuning step length of the strategy parameter, and adaptively adjusting the matching control strategy solution based on the fine-tuning step length of the strategy parameter to obtain multiple fine-tuning control strategy solutions. According to the key feature matching degree, the fine-tuning step length of the strategy parameter is determined, and based on the determined fine-tuning step length, the strategy parameters in the matching control strategy solution are adaptively adjusted to obtain multiple fine-tuning control strategy solutions, and the matching degree with the key features of the application scenario is further improved. It also includes, based on the switch cost function, evaluating and optimizing the multiple fine-tuning control strategy solutions to determine the target switch control strategy solution. By comparing the cost of different fine-tuning control strategy solutions and evaluating the performance differences between them, after evaluating the performance of all fine-tuning control strategy solutions, an optimization process is required, that is, selecting the solution with the best performance or that best meets the application scenario requirements, and finally determining the target switch control strategy solution.

[0031] The specific configuration of the control decision module 30 will be described in detail below. The control decision module 30 may further include: fitting the cost function of the switch control strategy solution space based on the switch control target, and obtaining a set of sub-control target cost functions. Decompose the switch control target into multiple sub-control targets, and find or construct a corresponding cost function in the control strategy solution space for each sub-control target to obtain a set of sub-control target cost functions. Specifically, decompose the total switch control target into multiple sub-control targets, find or construct a cost function for each sub-control target, and the function can quantitatively evaluate the performance of different control strategy solutions in achieving the sub-control target. The sub-control target cost function set contains the cost functions corresponding to all sub-control targets, and each function evaluates a specific sub-control target. It also includes evaluating the importance of each sub-target in the switch control target according to the switch control requirements to obtain a set of sub-target weight factors. For multiple sub-control objectives, each sub-objective is quantitatively evaluated according to the needs and priorities of the actual application scenario to determine their importance or contribution to the overall control objective. The evaluation result is usually expressed as a set of weight factors, where each weight factor corresponds to a sub-objective and reflects the relative importance of the sub-objective in the overall control objective. It also includes weighted fusion of the sub-control objective cost function set based on the sub-objective weight factor set to construct the switch cost function. The cost functions of multiple sub-control objectives are weighted and combined according to their importance (represented by the sub-objective weight factors) to form a comprehensive switch cost function to evaluate the advantages and disadvantages of different control strategy schemes in terms of global performance.

[0032] The specific configuration of the control decision module 30 will be described in detail below. The control decision module 30 may further include: executing control monitoring on the multiple switch devices based on the target switch control strategy scheme, and obtaining the multiple switch operation monitoring data stream. According to the target switch control strategy scheme that has been determined and optimized, multiple switch devices (for example, switches of electrical equipment, circuit switches, etc.) are controlled, and the operating status of these switch devices is continuously monitored at the same time, and the multiple switch operation monitoring data stream is obtained, including the current state of the switch (on / off), voltage, current, temperature, humidity, power consumption and other parameters. It also includes analyzing the performance optimization direction of the multiple switch operation monitoring data stream to obtain the control strategy variation direction. Based on the collected multiple switch operation monitoring data stream, the performance of the switch control system is evaluated, and possible optimization points or improvement directions are identified, so as to determine which adjustments or variations need to be made to the control strategy, such as excessive switch energy consumption and low user satisfaction with home control, and corresponding associated optimization direction determination, such as reducing the associated light on time. It also includes dynamically optimizing the target switch control strategy scheme based on the control strategy variation direction, determining the target switch control optimization strategy scheme, and dynamically executing control on the multiple switch devices based on the target switch control optimization strategy scheme. Based on the analysis results of the control strategy variation direction, the target switch control strategy scheme is dynamically optimized, for example, the control algorithm is modified, the control parameters are adjusted, a new control logic is designed, etc., the target switch control optimization strategy scheme is determined, and based on the determined target switch control optimization strategy scheme, the multiple switch devices are dynamically executed and controlled, which means that the control system will automatically adjust the state or parameters of the switch device according to the real-time operation data and the optimization strategy scheme to achieve the best control effect.

[0033] The specific configuration of the control decision module 30 will be described in detail below. The control decision module 30 may further include: performing correlation analysis on the multi-way switch operation monitoring data stream based on the control strategy variation direction to obtain the control strategy variation distribution. Using statistical, data mining or machine learning methods, the collected multi-way switch operation monitoring data stream is deeply analyzed to explore the impact of different control strategy variations on the switch operation performance and its distribution. Specifically, correlation analysis, cluster analysis, regression analysis, etc. are used to perform correlation analysis on the multi-way switch operation monitoring data stream and the control strategy variation to find out the correlation between different control strategy variations and switch operation performance, as well as the strength and direction of these correlations, and determine the distribution of control strategy variation, which may include the frequency of variation type, the magnitude of variation degree, the range of variation on performance, etc. It also includes mutating and updating the target switch control strategy scheme according to the control strategy variation distribution to obtain a set of switch control variation strategy schemes. According to the control strategy variation distribution obtained by analysis, the original target switch control strategy scheme is modified and updated in a targeted manner, thereby generating a series of new and potential optimization strategy schemes, which constitute a set, namely, the switch control variation strategy scheme set. The method further includes evaluating and optimizing the switch control variation strategy scheme set to determine the target switch control optimization strategy scheme. The generated multiple switch control variation strategy schemes are systematically evaluated, and the optimal strategy scheme or the strategy scheme that best meets the specific requirements is selected as the target switch control optimization strategy scheme based on the evaluation results. Specifically, each scheme in the switch control variation strategy scheme set is evaluated based on simulation tests, laboratory experiments or actual operation data, and the performance of different schemes is quantitatively evaluated by comparing the performance of different schemes on various evaluation indicators, and the optimal switch control strategy scheme or the switch control strategy scheme that best meets the specific requirements is selected as the target switch control optimization strategy scheme.

[0034] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0035] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. A multi-way switch dynamic control platform based on multi-source data, characterized in that: The platform includes: A data acquisition module, which is wirelessly connected to a sensor group and a user application device, obtains an environment multi-source data stream and a device operation multi-source data stream through the sensor group, and simultaneously obtains target scene behavior data through the user application device, and encrypts and transmits the environment multi-source data stream, the device operation multi-source data stream and the target scene behavior data to a data processing module; A data processing module, through which a data preprocessing multi-channel is constructed, the environment multi-source data stream, the device operation multi-source data stream, and the target scene behavior data are mapped and preprocessed with the data preprocessing multi-channel to obtain an application scene fusion data stream; A control decision module, the control decision module is used to extract key features from the application scenario fusion data stream to obtain key feature information of the application scenario, construct a switch control strategy solution space according to the control decision module, and perform matching and optimization based on the key feature information of the application scenario and the switch control strategy solution space to obtain a target switch control strategy solution; A switch execution module, the switch execution module is in communication with the multi-way switch device, and the multi-way switch device is dynamically executed and controlled based on the target switch control strategy scheme by the switch execution module; The control decision module executes the following steps: Standardizing the application scenario fusion data stream to obtain an application scenario standard fusion data stream; Based on the application scenario standard, the fused data stream is decentralized processed to obtain a covariance matrix, and the eigenvalues ​​and eigenvectors of the covariance matrix are obtained by solving; Selecting principal component information greater than a preset threshold value according to the eigenvalue, projecting the application scenario standard fusion data stream onto the eigenvector based on the principal component information, and determining key feature information of the application scenario; The control decision module performs the following steps: Using the key feature information of the application scenario as a constraint parameter, and performing similarity calculation with the switch control strategy solution space, to obtain an application scenario feature similarity set; Setting a similarity threshold, and screening and matching the application scenario feature similarity set based on the similarity threshold to obtain a switch control strategy solution search library; defining a switch control target, and constructing a switch cost function according to the switch control target; Based on the switch cost function, a global evaluation is performed on the switch control strategy solution search library to obtain the target switch control strategy solution; The control decision module executes the following steps: Performing a global evaluation calculation on the switch control strategy solution search library by using the switch cost function to obtain cost information of multiple switch control strategy solutions; Optimizing the strategy scheme based on the cost information of the multiple switch control strategy schemes, determining a matching control strategy scheme, and determining a key feature matching degree according to the matching control strategy scheme; Determining a fine-tuning step length of a strategy parameter according to the key feature matching degree, and adaptively adjusting the matching control strategy scheme based on the strategy parameter fine-tuning step length to obtain a plurality of fine-tuning control strategy schemes; Evaluate and optimize the multiple fine-tuning control strategy solutions based on the switch cost function to determine the target switch control strategy solution; The control decision module executes the following steps: Based on the switch control target, cost functions are fitted to the switch control strategy solution space respectively to obtain a sub-control target cost function set; According to the switch control requirements, importance evaluation is performed on each sub-goal in the switch control goal to obtain a sub-goal weight factor set; Based on the sub-objective weight factor set, the sub-control objective cost function set is weightedly integrated to construct the switch cost function; The control decision module executes the following steps: Based on the target switch control strategy, the multi-way switch equipment is executed and controlled to monitor, and a multi-way switch operation monitoring data stream is obtained; Performing performance optimization direction analysis on the multi-way switch operation monitoring data stream to obtain the control strategy variation direction; Dynamically optimizing the target switch control strategy scheme based on the control strategy variation direction, determining the target switch control optimization strategy scheme, and dynamically executing control on the multiple switch devices based on the target switch control optimization strategy scheme; The control decision module executes the following steps: Performing correlation analysis on the multi-way switch operation monitoring data stream based on the control strategy variation direction to obtain the control strategy variation distribution; Mutating and updating the target switch control strategy scheme according to the control strategy variation distribution to obtain a set of switch control variation strategy schemes; An evaluation and optimization is performed based on the set of switch control variation strategies to determine the target switch control optimization strategy.

2. The multi-way switch dynamic control platform based on multi-source data according to claim 1, characterized in that: The data processing module performs the following steps: Acquiring data attribute identification elements through the data processing module, wherein the data attribute identification elements include data type identification elements and data structured format identification elements; Based on the data attribute identification elements, the environment multi-source data stream, the device operation multi-source data stream, and the target scenario behavior data are structurally classified to obtain a data type attribute set and a data structured classification factor set; Performing preprocessing requirement analysis on the data type attribute set and the data structured classification factor set to determine a data structured factor preprocessing program set; The data preprocessing channel configuration is performed based on the data structuring factor preprocessing program set to construct the data preprocessing multi-channel.

Images