A device automated testing and synchronous control method and system based on multi-sensor fusion

Abstract

The present invention relates to the technical field of automated testing and control of power equipment, and provides a method and system for automated testing and synchronous control of equipment based on multi-sensor fusion. The method comprises obtaining multi-sensor and operating status data of a switch cabinet to construct a first training sample set, and training a signal synchronization model; utilizing the model to perform feature grouping for each operating state of the equipment to obtain a feature type; constructing a second training sample set based on the feature type and the control data between the two states, and training a synchronous control model; inputting the multi-sensor data of the equipment to be tested into the signal synchronization model to obtain a feature representation, and then outputting a synchronous control result through the synchronous control model. The method achieves accurate feature extraction and synchronous control decision-making of the equipment state through two-stage modeling, wherein the second sample set focuses on selecting checkpoint data where the operating state transitions to stability, and optimizes the control effect in combination with control influencing factor analysis.

Description

Technical Field

[0001] The present invention relates to the technical field of automated testing and control of power equipment, and more specifically, to a method and system for automated testing and synchronous control of equipment based on multi-sensor fusion. Background Art

[0002] In the field of industrial automation, automated testing and synchronous control of equipment have always been one of the key technologies. Existing automated equipment testing methods often rely on single-sensor data acquisition. When faced with complex working conditions, this method often cannot fully and accurately reflect the actual operating status of the equipment. For example, in the synchronous control of multiple motors, traditional master-command control and master-slave control methods have simple control structures, the operating status of each branch cannot interact, and the synchronization performance is completely dependent on the following ability and anti-disturbance capability of each branch, resulting in weak overall synchronization capabilities of the system. In addition, existing technologies for equipment synchronous control often use control models based on fixed parameters, which are difficult to adapt to dynamic changes in the equipment's operating status.

[0003] In implementing the embodiments of the present invention, the inventors discovered that the prior art has at least the following problems or defects: First, the single sensor data acquisition method cannot effectively integrate multi-source information, resulting in incomplete perception of device status, affecting the accuracy and reliability of automated testing. Second, the existing synchronization control model lacks the ability to adapt to dynamic changes in the device, making it difficult to achieve precise synchronization control. Finally, the prior art lacks effective feature extraction and analysis methods for detecting the transition of the device's operating state to stability, making it impossible to accurately identify the device's stable state, which in turn affects the implementation of synchronization control. Summary of the Invention

[0004] The present invention provides a method and system for equipment automatic testing and synchronous control based on multi-sensor fusion.

[0005] In a first aspect of the present invention, a method for automated testing and synchronous control of equipment based on multi-sensor fusion is provided, comprising:

[0006] Acquire multi-sensor data and equipment operating status data of the automatic point-to-point switch cabinet to construct a first training sample set; construct a signal synchronization model, and train the signal synchronization model based on the first training sample set to obtain a trained signal synchronization model;

[0007] Based on the trained signal synchronization model, the characteristics of each operating state of the equipment are grouped to obtain the characteristic type of each operating state;

[0008] constructing a second training sample set based on the feature type of each operating state and the control data of the device between the two operating states; constructing a synchronous control model, and training the synchronous control model based on the second training sample set to obtain a trained synchronous control model;

[0009] The multi-sensor data of the device to be tested is input into the signal synchronization model to obtain the feature representation of the device to be tested; and the synchronization control result of the device to be tested is obtained based on the trained synchronization control model according to the feature representation.

[0010] Furthermore, a second training sample set is constructed based on the feature type of each operating state and the control data of the device between two operating states, including:

[0011] The inspection of the transition from the running state to the stable state is selected, the feature representation of the stable transition inspection is used as the input data of the sample, and the control data of the equipment between the stable transition inspection and the next inspection is used as the label of the sample to construct the second training sample set.

[0012] Furthermore, the following method is used to select the check for the transition from the running state to the stable state:

[0013] For each device, calculate the control impact factor between two adjacent operating states;

[0014] If the control influence factor between the t-th operating state and the t+1-th operating state is positive, the t-th operating state is a check for the operating state transition to stability.

[0015] Furthermore, the control influence factor between the t-th operating state and the t+1-th operating state is calculated in the following way:

[0016] Take the tth running state as the current state and traverse forward in sequence until a current state with a feature type different from that of the previous state is found as the starting state;

[0017] Take the t+1th running state as the current state and traverse backward in sequence until a current state with a feature type different from that of the next state is found as the end state;

[0018] The control influence factor between the t-th operating state and the t+1-th operating state is calculated based on the characteristic types of the starting state and the ending state and the state time.

[0019] Furthermore, based on the characteristic types of the starting state and the ending state and the state time, the control influence factor between the t-th operating state and the t+1-th operating state is calculated using the following formula:

[0020]

[0021] Among them, the feature type of the starting state is the i-th feature type, and the feature type of the ending state is the j-th feature type. Represents the distance between the feature type of the starting state and the feature type of the ending state, Indicates the time of the starting state, Indicates the time of the cut-off state, represents the hyperbolic tangent function, represents the symbolic function, and Indicates parameters, Represents the control influence factor between the t-th operating state and the t+1-th operating state.

[0022] Furthermore, obtaining a synchronization control result of the device to be tested based on the trained synchronization control model according to the characteristic representation includes:

[0023] Find K samples in the second training sample set that have similar feature representations to the device to be tested;

[0024] Based on the trained synchronization control model, K samples of synchronization control results are obtained;

[0025] The synchronous control results of the K samples are comprehensively evaluated based on the control influence factors of the K samples to obtain the synchronous control results of the device to be tested.

[0026] Furthermore, based on the trained signal synchronization model, the characteristics of each operating state of the device are grouped to obtain the characteristic types of each operating state, including:

[0027] For the t-th operating state of the device, the t-th operating state and the previous operating state data are input into the trained signal synchronization model to obtain the feature representation of the t-th operating state of the device and the signal synchronization prediction result;

[0028] The feature representations of all operating states of all devices are grouped to obtain multiple feature types;

[0029] For each feature type, calculate the average value of the signal synchronization prediction results corresponding to the feature representations of the feature type, and sort the feature types according to the average value of the signal synchronization prediction results to obtain the sequence number of each feature type;

[0030] The feature type of each running state is the feature type to which the feature representation of the running state belongs.

[0031] Furthermore, the signal synchronization model includes:

[0032] A device intrinsic parameter extraction module is used to extract intrinsic attribute features from the device intrinsic parameters of the sample;

[0033] A dynamic change feature extraction module is used to extract dynamic change features from the running status data of the sample;

[0034] Feature fusion module, used to fuse inherent attribute features and dynamic change features;

[0035] The prediction module is used to perform signal synchronization prediction based on the fused features.

[0036] In a second aspect of the present invention, a device automated testing and synchronization control system based on multi-sensor fusion is provided, comprising:

[0037] A signal synchronization model training module is used to obtain multi-sensor data and equipment operating status data of the automatic point-to-point switch cabinet to construct a first training sample set; construct a signal synchronization model, and train the signal synchronization model based on the first training sample set to obtain a trained signal synchronization model;

[0038] The feature grouping module is used to group the features of each operating state of the device based on the trained signal synchronization model to obtain the feature type of each operating state;

[0039] A synchronous control model training module is configured to construct a second training sample set based on the characteristic type of each operating state and the control data of the device between the two operating states; construct a synchronous control model, and train the synchronous control model based on the second training sample set to obtain a trained synchronous control model;

[0040] The synchronization control module is used to input the multi-sensor data of the device to be tested into the signal synchronization model to obtain the feature representation of the device to be tested; and obtain the synchronization control result of the device to be tested based on the trained synchronization control model according to the feature representation.

[0041] Furthermore, a second training sample set is constructed based on the feature type of each operating state and the control data of the device between two operating states, including:

[0042] The inspection of the transition from the running state to the stable state is selected, the feature representation of the stable transition inspection is used as the input data of the sample, and the control data of the equipment between the stable transition inspection and the next inspection is used as the label of the sample to construct the second training sample set.

[0043] According to the above-mentioned embodiment of the present invention, there are at least the following beneficial effects: the method and system for automated testing and synchronous control of equipment based on multi-sensor fusion can effectively improve the testing efficiency and control accuracy of the equipment. By constructing a signal synchronization model and a synchronous control model, and using multi-sensor data and equipment operation status data for training, the equipment operation status can be accurately grouped and synchronously controlled, so that the equipment can respond to control instructions more accurately during the automated testing process, reduce test errors and equipment failure risks caused by unstable equipment status or inaccurate control, and ensure the stability and reliability of equipment operation. At the same time, the method and system can realize efficient automated testing and synchronous control of equipment, reduce manual intervention costs and test time costs. During the operation of the equipment, the characteristic representation of the equipment can be quickly and accurately obtained and synchronously controlled, thereby improving the production efficiency and operation efficiency of the equipment and saving manpower and material resources for the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily apparent by reading the following detailed description with reference to the accompanying drawings, in which several embodiments of the present invention are shown by way of example and not limitation, in which:

[0045] Figure 1 A flow chart of a method for automated device testing and synchronous control based on multi-sensor fusion according to an embodiment of the present invention;

[0046] Figure 2 A schematic diagram of the structure of an automated device testing and synchronization control system based on multi-sensor fusion provided by one embodiment of the present invention;

[0047] Figure 3 The figure schematically shows the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided solely to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. Rather, these embodiments are provided to make the present invention more thorough and complete, and to fully convey the scope of the present invention to those skilled in the art.

[0049] Those skilled in the art will appreciate that the embodiments of the present invention may be implemented as a system, apparatus, device, method, or computer program product. Therefore, the present invention may be implemented in the following forms: entirely in hardware, entirely in software (including firmware, resident software, microcode, etc.), or in a combination of hardware and software.

[0050] It should be noted that any number of elements in the drawings is for illustration only and not for limitation, and any naming is only for distinction and does not have any limiting meaning.

[0051] Example 1

[0052] Reference below Figure 1 , Figure 1 The following is a flow chart of a device automated testing and synchronous control method based on multi-sensor fusion provided by an embodiment of the present invention. Figure 1 As shown, a device automated testing and synchronous control method based on multi-sensor fusion includes:

[0053] S1 obtains multi-sensor data and equipment operating status data of the automatic point-to-point switch cabinet to construct a first training sample set; constructs a signal synchronization model, and trains the signal synchronization model based on the first training sample set to obtain a trained signal synchronization model;

[0054] S2 groups the features of each operating state of the device based on the trained signal synchronization model to obtain the feature type of each operating state;

[0055] S3 constructs a second training sample set based on the feature type of each operating state and the control data of the device between the two operating states; constructs a synchronous control model, and trains the synchronous control model based on the second training sample set to obtain a trained synchronous control model;

[0056] S4 inputs the multi-sensor data of the device to be tested into the signal synchronization model to obtain a feature representation of the device to be tested; and obtains a synchronization control result of the device to be tested based on the trained synchronization control model according to the feature representation.

[0057] It should be noted that the present invention proposes a method for automated equipment testing and synchronous control based on multi-sensor fusion, which aims to achieve automated testing and precise synchronous control of equipment through the fusion of multi-sensor data and equipment operating status data. Multi-sensor data here refers to data collected from different types of sensors (such as temperature sensors, pressure sensors, current sensors, etc.), which can comprehensively reflect the various status information of the equipment during operation. Equipment operating status data refers to various parameter data generated by the equipment during operation, such as the equipment's operating speed and load conditions. By constructing a training sample set and training a signal synchronization model and a synchronous control model, it is possible to achieve feature grouping and synchronous control of the equipment's operating status, thereby improving the efficiency and accuracy of automated equipment testing.

[0058] Specifically, the first training sample set refers to a data set that integrates the multi-sensor data and equipment operating status data of the automatic point-to-point switchgear, and is used to train the signal synchronization model. The signal synchronization model is a model used to predict the signal synchronization of the equipment. It outputs the signal synchronization prediction result of the equipment by analyzing the input multi-sensor data and equipment operating status data. Feature grouping refers to classifying the features of each operating state of the equipment according to the output of the signal synchronization model to obtain different types of feature representations. The second training sample set is constructed based on the feature type of each operating state and the control data of the equipment between the two operating states, and is used to train the synchronous control model. The synchronous control model can output the synchronous control result of the equipment based on the input feature representation.

[0059] The first training sample set is formed by collecting multi-sensor data (such as data collected by sensors such as temperature, pressure, and current) and equipment operating status data (such as equipment operating speed, load conditions, and other parameter data) from automatic point-to-point switchgear under different operating conditions. This data is then integrated to form the first training sample set. For example, multi-sensor data and operating status data are collected from a certain model of automatic point-to-point switchgear under different states, such as normal operation, overload operation, and fault operation. Each set of data is considered a sample and contains multiple feature dimensions, such as temperature, pressure, current, and operating speed. This data is then used to construct the first training sample set for training the signal synchronization model.

[0060] Preferably, when constructing a signal synchronization model, a machine learning algorithm such as a neural network or a support vector machine can be used. The input parameters include multi-sensor data and equipment operating status data, which are pre-processed and used as input features of the model. When calculating the control influence factor, the changing trend of the equipment state can be evaluated by analyzing the feature type and time information between two adjacent operating states. For example, if the control influence factor between the t-th operating state and the t+1-th operating state is positive, it means that the equipment is transitioning towards a stable direction between these two operating states. In the training process of the synchronous control model, a supervised learning method can be used to optimize the model parameters by minimizing the error between the predicted results and the actual control data. When obtaining the synchronous control results of the device to be tested, similar samples can be found in the second training sample set based on the feature representation of the device to be tested, and a comprehensive evaluation can be performed based on the control influence factors of these samples to obtain more accurate synchronous control results.

[0061] The second training sample set is constructed based on the feature types of each operating state and the control data of the device between the two operating states. First, the checkpoints where the operating state transitions to stability are identified. The feature representations of these checkpoints serve as the sample input data, and the control data of the device from this checkpoint to the next checkpoint serves as the sample label. For example, during the process of the device transitioning from an overloaded state to a normal operating state, several key checkpoints are selected and their feature representations (such as a combination of characteristic parameters such as the temperature sensor reading starting to drop from an excessively high value and the current sensor reading gradually stabilizing) are recorded. The control data executed from this checkpoint to the time when the device is fully stable (such as control instructions for reducing the load and control parameters for adjusting the motor speed) are then combined into a sample. Multiple such samples constitute the second training sample set, which is used to train the synchronous control model.

[0062] The signal synchronization model includes a device inherent parameter extraction module, a dynamic change feature extraction module, a feature fusion module and a prediction module.

[0063] The device intrinsic parameter extraction module extracts intrinsic attribute features from the device intrinsic parameters of the sample (such as the device model, rated power, and rated current). For example, different models of automatic switching cabinets have different parameters such as rated power and rated current. These parameters remain unchanged during operation and reflect the basic characteristics of the device. This module can extract these intrinsic parameters as part of the feature set.

[0064] Dynamic Change Feature Extraction Module: This module extracts dynamic change features from the sample's operational status data (such as real-time data such as temperature, pressure, current, and voltage during device operation). For example, during device operation, a temperature sensor monitors temperature changes in real time, while a current sensor monitors current fluctuations. These dynamically changing data reflect the device's real-time status during operation. This module can extract dynamic features such as the temperature change rate and current fluctuation amplitude.

[0065] Feature Fusion Module: This module integrates inherent attribute features with dynamic change features. For example, it combines the rated power of a device with dynamic features such as the current and voltage during operation to form a comprehensive feature vector, which fully characterizes the device's operating status.

[0066] Prediction module: This module performs signal synchronization prediction based on the fused features. For example, a neural network algorithm can be used to train a neural network model using the fused feature vector as input. This model can predict the signal synchronization status of the device under its current operating state and output the device's signal synchronization prediction results, such as whether the device is synchronized and the synchronization accuracy.

[0067] The signal synchronization model is trained using the first training sample set. The multi-sensor data and device operating status data from the sample are input into the model. The model parameters are adjusted to ensure that the model's predictions are as close as possible to the actual conditions in the sample. Through continuous iterative training, a trained signal synchronization model is ultimately obtained.

[0068] In some embodiments, constructing a second training sample set based on the feature type of each operating state and the control data of the device between two operating states includes:

[0069] The inspection of the transition from the running state to the stable state is selected, the feature representation of the stable transition inspection is used as the input data of the sample, and the control data of the equipment between the stable transition inspection and the next inspection is used as the label of the sample to construct the second training sample set.

[0070] It should be noted that the present invention adopts a specific sample selection strategy when constructing the second training sample set, that is, the inspection of the transition from the operating state to the stable state is selected as the input data of the sample, and the control data of the device from the stable transition inspection to the next inspection is used as the label of the sample. The core of this strategy is to capture the relationship between the feature representation and control data of the device at this critical stage by analyzing the stable transition process of the device's operating state, thereby providing more representative and targeted training samples for the synchronous control model. Among them, the inspection of the transition from the operating state to the stable state refers to the process of the device gradually transitioning from an unstable state to a stable operating state. The feature representation in this process can reflect the key features of the device in the stable process, and the control data refers to the relevant data for controlling the device in this process, such as the parameters of the control instructions, the control frequency, etc. These data are used as labels for samples to guide the learning of the synchronous control model.

[0071] Specifically, the second training sample set refers to the data set used to train the synchronous control model, which consists of input data and corresponding labels. The input data is a feature representation of the transition from the device's operating state to stability. These feature representations may include the trend of changes in sensor data during the device's stable transition, the adjustment of operating parameters, etc. The labels are the control data of the device from the stable transition inspection to the next inspection. These control data may be specific control instruction values, the execution time of the control action, etc. For example, if the temperature sensor reading of the device gradually decreases from high temperature to the normal operating temperature range during the stable transition, and the speed of the device is gradually adjusted from high speed to the stable operating speed, these change trends and adjustment parameters are part of the feature representation, and the cooling instructions, speed regulation instructions and their execution time issued during this process are the corresponding control data.

[0072] In some embodiments, a second training sample set is constructed based on the feature type of each operating state and the control data of the device between the two operating states. First, the checkpoints where the operating state transitions to stability are determined, and the feature representations of these checkpoints are used as the input data of the sample, and the control data of the device between this checkpoint and the next checkpoint are used as the label of the sample. For example, in the process of the device transitioning from an overload state to a normal operating state, several key checkpoints are selected, and the feature representations of these checkpoints (such as a combination of feature parameters such as the temperature sensor reading starting to drop from an excessively high value and the current sensor reading gradually stabilizing) are recorded, as well as the control data executed from the checkpoint to the complete stable operation of the device (such as the control instruction to reduce the load, the control parameter for adjusting the motor speed, etc.), which are combined into a sample. Multiple such samples constitute the second training sample set for training the synchronous control model.

[0073] Preferably, when selecting the check for the transition from the operating state to a stable state, certain thresholds or conditions can be set to determine whether the device is in a stable transition state. For example, conditions such as the temperature sensor reading falling within a certain range and the vibration amplitude of the device being lower than a certain threshold can be set. When these conditions are met, the device is considered to be in a stable transition state, and the feature representation at this time and the subsequent control data are used as samples. In the process of constructing the sample, the feature representation can also be preprocessed, such as normalization, dimensionality reduction, etc., to improve the training effect of the model. For control data, it can be classified and encoded according to the type and execution frequency of the control instruction to make it more suitable as a label for model training. Through these refined operation steps, the second training sample set can be constructed more accurately, thereby improving the performance and reliability of the synchronous control model.

[0074] In some embodiments, the check for transitioning the operating state to stable is selected in the following manner:

[0075] For each device, calculate the control impact factor between two adjacent operating states;

[0076] If the control influence factor between the t-th operating state and the t+1-th operating state is positive, the t-th operating state is a check for the operating state transition to stability.

[0077] It should be noted that the present invention selects the transition check for an operating state to a stable state by calculating the control influence factor between two adjacent operating states. The control influence factor is an indicator that measures the changing trend of a device's operating state, reflecting the impact of control operations between two adjacent operating states on the device's stability. A positive control influence factor indicates that the device is transitioning from an unstable state to a stable state. This transition process is important for constructing the second training sample set because it helps identify key features and control data during the device's stable transition.

[0078] Specifically, the control impact factor is a parameter that quantifies the trend of changes in the operating state of the equipment. It is calculated by analyzing the characteristic changes and control operations of the equipment between two adjacent operating states. For example, during the operation of the equipment, if a control operation causes the temperature of the equipment to gradually decrease from an excessively high state to a normal operating range, and the vibration amplitude of the equipment also decreases accordingly, then the control impact factor of this control operation may be a positive value, indicating that the equipment is transitioning to a stable state. When calculating the control impact factor, it is necessary to consider a variety of operating state parameters of the equipment, such as temperature, pressure, current, etc., as well as the corresponding control operation parameters, such as the intensity of the control instructions, execution time, etc. The specific values of these parameters can be obtained through sensor acquisition and control system records, thereby providing data support for the calculation of the control impact factor.

[0079] Preferably, when calculating the control influencing factor, a method based on time series analysis can be used. First, determine the starting state and the ending state of the equipment's operating state, which correspond to the start and end time points of the equipment's operating state change, respectively. Then, based on the characteristic types and state time of the starting state and the ending state, calculate the characteristic change amount of the equipment during this time period and the degree of influence of the control operation. For example, the actual impact of the control operation on the equipment's stability can be evaluated by calculating the change in characteristic parameters such as the temperature change rate and pressure fluctuation amplitude of the equipment during the stable transition process, combined with the execution time and intensity of the control instructions. This method can more accurately reflect the dynamic changes of the equipment during operation, thereby providing a reliable basis for selecting the inspection for the transition from operating state to stable state.

[0080] In some embodiments, the control influence factor between the t-th operating state and the t+1-th operating state is calculated in the following manner:

[0081] Take the tth running state as the current state and traverse forward in sequence until a current state with a feature type different from that of the previous state is found as the starting state;

[0082] Take the t+1th running state as the current state and traverse backward in sequence until a current state with a feature type different from that of the next state is found as the end state;

[0083] The control influence factor between the t-th operating state and the t+1-th operating state is calculated based on the characteristic types of the starting state and the ending state and the state time.

[0084] It should be noted that the present invention employs a method that comprehensively considers the characteristic types of the initial and final states, as well as the state duration, when calculating the control impact factor between the tth operating state and the t+1th operating state. This method traverses the device's operating states forward and backward to identify the initial and final states where the characteristic types change. The control impact factor is then calculated based on the characteristic types, state duration, and related parameters of these states. The control impact factor is an important indicator used to assess the changing trend of a device between two adjacent operating states, particularly to determine whether the device transitions from an unstable state to a stable state, thereby providing a basis for subsequent synchronous control.

[0085] Specifically, the starting state refers to the first state found, starting from the tth operating state, with a different feature type from the current state. It marks the starting point of the device's operating state change. For example, if the device gradually recovers from an overloaded state to a normal operating state, the starting state is the overloaded state. The ending state refers to the first state found, starting from the t+1th operating state, with a different feature type from the subsequent state. It marks the end point of the device's operating state change. For example, if the device goes from a normal operating state to a faulty state, the ending state is the faulty state. The feature type is the result of classifying the device's operating state based on its characteristic representation. For example, the device's operating state can be divided into different types, such as normal operation, overloaded operation, and faulty operation, each corresponding to a different feature representation. The state time refers to the time point or period of time that the device is in a certain operating state and is used to measure the time span of the state change. When calculating the control impact factor, the distance between the feature types of the starting and ending states needs to be considered. This reflects the degree of difference between the two states in the feature space. In addition, the time factor, namely the difference between the ending state time and the starting state time, needs to be considered to assess the impact of the time span of the state change on the control impact factor.

[0086] Preferably, when calculating the control impact factor, a weighted approach can be used to comprehensively consider the influence of the distance between feature types, the state time, and the sign function. First, a basic impact factor is calculated based on the distance and state time between feature types. This factor is normalized using the hyperbolic tangent function to ensure its value is within a certain range. The hyperbolic tangent function maps the ratio of the distance between feature types to the state time within a fixed interval, thereby preventing excessively large or small values from affecting the calculation results. Then, based on the relative relationship between feature types, the sign of the impact factor is adjusted using the sign function to reflect whether the device's operating state is transitioning towards stability or instability. The sign function determines the changing trend of the device's state based on the positive or negative sign of the distance between feature types. Finally, the influence of the basic impact factor and the sign function are weighted using two parameters to obtain the final control impact factor. The specific values of these two parameters can be adjusted based on the actual operating characteristics and control requirements of the device to ensure that the control impact factor accurately reflects the changing trend of the device's operating state.

[0087] For each device operating state, the data from the tth operating state and previous operating states are input into the trained signal synchronization model to obtain a feature representation of the device's tth operating state and the signal synchronization prediction result. The feature representations of all operating states for all devices are grouped to generate multiple feature types. For example, the signal synchronization model processes data from different device operating states to obtain a feature vector for each operating state. These feature vectors are then clustered using a clustering algorithm (such as K-means clustering). Similar feature vectors are grouped together, with each group corresponding to a feature type, such as a normal operating feature type, a mild overload feature type, or a severe overload feature type.

[0088] For each feature type, the average value of the signal synchronization prediction results corresponding to the feature representations belonging to that feature type is calculated. The feature types are then sorted based on this average value to obtain a sequence number for each feature type. For example, the average value of the signal synchronization prediction results for the normal operation feature type may be higher, while the average value of the signal synchronization prediction results for the severe overload feature type may be lower. By sorting the feature types based on this average value, the importance and priority of each feature type in the equipment's operating status can be determined, allowing for targeted processing in subsequent synchronous control.

[0089] In some embodiments, the control influence factor between the t-th operating state and the t+1-th operating state is calculated based on the characteristic types of the starting state and the ending state and the state time using the following formula:

[0090]

[0091] Among them, the feature type of the starting state is the i-th feature type, and the feature type of the ending state is the j-th feature type. Represents the distance between the feature type of the starting state and the feature type of the ending state, Indicates the time of the starting state, Indicates the time of the cut-off state, represents the hyperbolic tangent function, represents the symbolic function, and Indicates parameters, Represents the control influence factor between the t-th operating state and the t+1-th operating state.

[0092] It should be noted that the calculation formula for the control influence factor mentioned in this invention is a model that comprehensively considers multiple factors and is used to quantify the changing trends between device operating states. This formula uses a weighted combination of the distance between feature types, state duration, and related parameters to more accurately assess the control effectiveness of a device as it transitions from one operating state to another. This method is particularly suitable for automated testing and synchronous control scenarios of complex equipment, helping the system more accurately identify stable transitions.

[0093] The equipment operation status data is shown in Table 1:

[0094] Table 1

[0095]

[0096] Calculation process:

[0097] Assume that we want to calculate the control influence factor between the 3rd and 4th operating states, and = 0.6, = 0.4.

[0098] Determine the starting and ending states:

[0099] Taking the third run as the current state, traverse forward. The characteristic types for the third run are 50°C and 1.4 MPa, the second run is 48°C and 1.3 MPa, and the first run is 45°C and 1.2 MPa. Assuming that characteristic types are distinguished by the combination of temperature and pressure, the third run is different from the first two (both temperature and pressure are increasing), so the starting state is the third run.

[0100] Taking the fourth run as the current state, traverse backward. The characteristic types of the fourth run are 47°C and 1.25 MPa, while those of the fifth run are 46°C and 1.15 MPa. The characteristic types of the fourth and fifth runs are different (both temperature and pressure are decreasing), so the end state is the fourth run.

[0101] Calculate feature type distance:

[0102] Assume that the feature type distance is calculated using Euclidean distance, the starting state (3rd time) feature vector is (50, 1.4), and the end state (4th time) feature vector is (47, 1.25).

[0103] The distance is:

[0104]

[0105] Calculate the time difference:

[0106] The start state time is 30 seconds, the end state time is 40 seconds, and the time difference is 40 - 30 = 10 seconds.

[0107] Substitute into the formula to calculate the control impact factor:

[0108] Assuming that the sign function is positive based on the trend of the device state change (because the temperature and pressure are decreasing from the third to the fourth time, turning towards stability), then:

[0109]

[0110]

[0111] We can get: .

[0112] Specifically, the control impact factor is a quantitative indicator that measures the changing trend of a device's operating state. It is calculated by comprehensively considering the distance between the characteristic types of the initial and final states, the state duration, and related parameters. The initial state characteristic type refers to the characteristic type of the device at the beginning of the change. For example, when a device gradually recovers from an overload state to normal operation, the initial state characteristic type is an overload state. The final state characteristic type refers to the characteristic type of the device at the end of the change. For example, when a device enters a fault state from normal operation, the final state characteristic type is a fault state. The distance between characteristic types refers to the degree of difference between the initial and final state characteristic types in the feature space. For example, if a device gradually cools from a high temperature state to a normal temperature range, this distance can represent the magnitude of the temperature change. The state duration refers to the time it takes for the device to move from the initial state to the final state, such as the time it takes for the device to recover from an overload state to normal operation. The parameters α and β in the formula are coefficients used to adjust the weights of different factors and can be adjusted based on the specific characteristics of the device and control requirements.

[0113] Preferably, when calculating the control influencing factor, the various parameters and steps in the formula can be further refined. For example, the distance between feature types can be determined by calculating the Euclidean distance between the feature vectors of two states, which can reflect the feature differences of the device in different states. The state time can be obtained through the timestamp in the device operation log to ensure the accuracy of the time calculation. Parameters α and β can be optimized based on the operating characteristics and control objectives of the device. For example, if the device is time-sensitive, the weight of β can be increased; if the device is sensitive to feature differences, the weight of α can be increased. In practical applications, the optimal values of these parameters can be determined through experiments and data analysis, thereby improving the calculation accuracy and reliability of the control influencing factor.

[0114] In some embodiments, obtaining a synchronization control result of the device to be tested based on the trained synchronization control model according to the feature representation includes:

[0115] Find K samples in the second training sample set that have similar feature representations to the device to be tested;

[0116] Based on the trained synchronization control model, K samples of synchronization control results are obtained;

[0117] The synchronous control results of the K samples are comprehensively evaluated based on the control influence factors of the K samples to obtain the synchronous control results of the device to be tested.

[0118] It should be noted that the present invention utilizes a method based on comprehensive evaluation of similar samples to obtain the synchronization control results for the device under test based on the feature representation. This method searches for samples in the second training sample set that have similar feature representations to the device under test and performs a comprehensive evaluation based on the synchronization control results and control influencing factors of these samples to obtain the synchronization control results for the device under test. The core of this method is to utilize the information in the existing training sample set to improve the accuracy and reliability of the synchronization control results through similarity matching and comprehensive evaluation.

[0119] Specifically, the second training sample set refers to the sample set used when constructing the synchronous control model, which contains the characteristic representation of the device's operating state and its corresponding synchronous control results. These samples are obtained by analyzing the process of the device's operating state transitioning to stability, and are highly representative and instructive. The characteristic representation of the device to be tested refers to the characteristic vector of the device's operating state obtained through the signal synchronization model, which can reflect the main characteristics of the device in its current operating state. Similar samples refer to samples in the second training sample set that are similar to the characteristic representation of the device to be tested. This similarity can be measured by calculating the distance or similarity between the characteristic vectors. The control influence factor is an indicator to measure the sample control effect, which reflects the importance and reliability of the sample in the synchronous control process. By finding similar samples and conducting a comprehensive evaluation based on their control influence factors, the synchronous control results of the device to be tested can be obtained.

[0120] Preferably, during the implementation process, the following steps can be used to refine the operation. First, define a similarity measurement method, such as using Euclidean distance or cosine similarity to calculate the similarity between the feature representation of the device to be tested and the feature representation of each sample in the second training sample set. Then, based on the set similarity threshold, K samples that are similar to the feature representation of the device to be tested are screened out from the second training sample set. Next, for these K similar samples, their synchronization control results and control influencing factors are extracted. Finally, according to the weights of the control influencing factors, the synchronization control results of these K samples are weighted averaged to obtain the final synchronization control result of the device to be tested. This method can make full use of the information in the training sample set, improve the accuracy and reliability of the synchronization control results, and can also adjust the similarity threshold and weight parameters according to actual needs to adapt to different devices and control scenarios.

[0121] The multi-sensor data from the device under test is fed into the signal synchronization model. The model processes the input data, extracts the device's inherent attributes and dynamic change characteristics, and fuses them together to ultimately generate a feature representation of the device under test. This feature representation is a feature vector that comprehensively reflects the device's current operating state. It includes information such as the device's inherent parameters and dynamic operating parameters, accurately characterizing the device's current operating state.

[0122] In some embodiments, K samples with similar feature representations to the device under test are searched in the second training sample set. Synchronous control results for these K samples are obtained based on the trained synchronous control model. The synchronous control results for these K samples are then comprehensively evaluated based on the control influence factors of these K samples to obtain a synchronous control result for the device under test. For example, assuming the feature representation of the device under test is similar to K samples in the second training sample set, and these K samples correspond to different synchronous control strategies and control effects, comprehensive evaluation methods such as weighted averaging of the synchronous control results of these K samples are performed by analyzing the control influence factors of these samples (e.g., the tendency of control operations to cause the device state to transition toward stability). The synchronous control result most suitable for the current state of the device under test is determined, such as adjusting the device's operating parameters or issuing specific control instructions, to achieve precise synchronous control of the device.

[0123] The second training sample set is searched for K samples with similar feature representations to the device under test. Based on the trained synchronization control model, synchronization control results for these K samples are obtained. Then, the synchronization control results for these K samples are comprehensively evaluated based on their control influence factors to obtain the synchronization control results for the device under test. For example, assuming the feature representation of the device under test is similar to the K samples in the second training sample set, and these K samples correspond to different synchronization control strategies and control effects, comprehensive evaluation methods such as weighted averaging of the synchronization control results for these K samples are used to determine the synchronization control result that best suits the current state of the device under test. This can be achieved by adjusting the device's operating parameters or issuing specific control instructions, thereby achieving precise synchronization control of the device.

[0124] In some embodiments, the characteristic representation of the device under test can also be input into a trained synchronization control model, which then directly outputs the synchronization control results for the device under test. For example, based on the characteristic representation of the device under test, such as the current combination of parameters such as temperature, pressure, and current, the synchronization control model can predict the control measures that need to be taken, such as increasing the power of the cooling system or adjusting the speed of the motor, to achieve optimal synchronization of the device.

[0125] In some embodiments, the characteristic representation of the device under test can also be input into a trained synchronization control model, which then directly outputs the synchronization control results for the device under test. For example, based on the characteristic representation of the device under test, such as the current combination of parameters such as temperature, pressure, and current, the synchronization control model can predict the control measures that need to be taken, such as increasing the power of the cooling system or adjusting the speed of the motor, to achieve optimal synchronization of the device.

[0126] In some embodiments, the feature types of each operating state of the device are grouped based on the trained signal synchronization model to obtain the feature types of each operating state, including:

[0127] For the t-th operating state of the device, the t-th operating state and the previous operating state data are input into the trained signal synchronization model to obtain the feature representation of the t-th operating state of the device and the signal synchronization prediction result;

[0128] The feature representations of all operating states of all devices are grouped to obtain multiple feature types;

[0129] For each feature type, calculate the average value of the signal synchronization prediction results corresponding to the feature representations of the feature type, and sort the feature types according to the average value of the signal synchronization prediction results to obtain the sequence number of each feature type;

[0130] The feature type of each running state is the feature type to which the feature representation of the running state belongs.

[0131] It should be noted that the present invention adopts a systematic approach to grouping the features of each device operating state based on a trained signal synchronization model. This method analyzes the device's operating state data, extracts the feature representation of each operating state, and groups them according to these feature representations to obtain different types of feature types. This method can effectively identify the key features of the device in different operating states, providing an accurate basis for subsequent synchronization control. By calculating the average value of the signal synchronization prediction results for each feature type and sorting the feature types according to these average values, the effect of feature grouping can be further optimized, improving the accuracy and reliability of the model.

[0132] Specifically, the signal synchronization model is a trained model that analyzes a device's multi-sensor data and operating status data and outputs a signal synchronization prediction result. By integrating the device's inherent attributes and dynamic change characteristics, this model accurately reflects the device's feature representation in different operating states. A feature representation refers to a feature vector of a device in a given operating state. It contains key characteristic information about the device in that state, such as the changing trends of sensor data and the values of operating parameters. Feature grouping involves categorizing the feature representations of all device operating states based on similarity to produce different types of feature types. Each feature type represents a set of features for a specific device operating state. The average value of the signal synchronization prediction result is the average value of the signal synchronization prediction results belonging to the same feature type, reflecting the typical synchronization state of that feature type. The feature type sequence number is a number obtained by sorting the feature types based on the average value of the signal synchronization prediction results, and is used to identify the importance and priority of different feature types.

[0133] The signal synchronization model includes: the signal synchronization model includes a device inherent parameter extraction module, a dynamic change feature extraction module, a feature fusion module and a prediction module.

[0134] The device intrinsic parameter extraction module extracts intrinsic attribute features from the device intrinsic parameters of the sample (such as the device model, rated power, and rated current). For example, different models of automatic switching cabinets have different parameters such as rated power and rated current. These parameters remain unchanged during operation and reflect the basic characteristics of the device. This module can extract these intrinsic parameters as part of the feature set.

[0135] Dynamic Change Feature Extraction Module: This module extracts dynamic change features from the sample's operational status data (such as real-time data such as temperature, pressure, current, and voltage during device operation). For example, during device operation, a temperature sensor monitors temperature changes in real time, while a current sensor monitors current fluctuations. These dynamically changing data reflect the device's real-time status during operation. This module can extract dynamic features such as the temperature change rate and current fluctuation amplitude.

[0136] Feature Fusion Module: This module integrates inherent attribute features with dynamic change features. For example, it combines the rated power of a device with dynamic features such as the current and voltage during operation to form a comprehensive feature vector, which fully characterizes the device's operating status.

[0137] Prediction module: This module performs signal synchronization prediction based on the fused features. For example, a neural network algorithm can be used to train a neural network model using the fused feature vector as input. This model can predict the signal synchronization status of the device under its current operating state and output the device's signal synchronization prediction results, such as whether the device is synchronized and the synchronization accuracy.

[0138] The signal synchronization model is trained using the first training sample set. The multi-sensor data and device operating status data from the sample are input into the model. The model parameters are adjusted to ensure that the model's predictions are as close as possible to the actual conditions in the sample. Through continuous iterative training, a trained signal synchronization model is ultimately obtained.

[0139] Preferably, during the implementation process, first, each operating state of the device and its previous operating state data are input into the trained signal synchronization model to obtain the feature representation and signal synchronization prediction results of each operating state. Then, the feature representations of all operating states of all devices are clustered and analyzed, for example, using the K-means clustering algorithm to group operating states with similar feature representations into the same group, thereby obtaining multiple feature types. Then, for each feature type, the average value of all its signal synchronization prediction results is calculated, and the feature types are sorted according to these average values to obtain the serial number of each feature type. Finally, the feature type of each operating state is determined based on the feature type to which the feature representation belongs. This method can make full use of the predictive ability of the signal synchronization model, improve the accuracy and efficiency of feature grouping through cluster analysis and sorting optimization, and provide more accurate input for the synchronous control of the equipment.

[0140] For each device operating state, the data from the tth operating state and previous operating states are input into the trained signal synchronization model to obtain a feature representation of the device's tth operating state and the signal synchronization prediction result. The feature representations of all operating states for all devices are grouped to generate multiple feature types. For example, the signal synchronization model processes data from different device operating states to obtain a feature vector for each operating state. These feature vectors are then clustered using a clustering algorithm (such as K-means clustering). Similar feature vectors are grouped together, with each group corresponding to a feature type, such as a normal operating feature type, a mild overload feature type, or a severe overload feature type.

[0141] For each feature type, the average value of the signal synchronization prediction results corresponding to the feature representations belonging to that feature type is calculated. The feature types are then sorted based on this average value to obtain a sequence number for each feature type. For example, the average value of the signal synchronization prediction results for the normal operation feature type may be higher, while the average value of the signal synchronization prediction results for the severe overload feature type may be lower. By sorting the feature types based on this average value, the importance and priority of each feature type in the equipment's operating status can be determined, allowing for targeted processing in subsequent synchronous control.

[0142] The multi-sensor data from the device under test is fed into the signal synchronization model. The model processes the input data, extracts the device's inherent attributes and dynamic change characteristics, and fuses them together to ultimately generate a feature representation of the device under test. This feature representation is a feature vector that comprehensively reflects the device's current operating state. It includes information such as the device's inherent parameters and dynamic operating parameters, accurately characterizing the device's current operating state.

[0143] In some embodiments, the signal synchronization model includes:

[0144] A device intrinsic parameter extraction module is used to extract intrinsic attribute features from the device intrinsic parameters of the sample;

[0145] A dynamic change feature extraction module is used to extract dynamic change features from the running status data of the sample;

[0146] Feature fusion module, used to fuse inherent attribute features and dynamic change features;

[0147] The prediction module is used to perform signal synchronization prediction based on the fused features.

[0148] It should be noted that the signal synchronization model is one of the core components used in this invention to handle device automated testing and synchronization control. This model extracts the inherent attributes and dynamic characteristics of the device and integrates these characteristics to ultimately predict device signal synchronization. This design effectively integrates static and dynamic device information, providing accurate prediction results for automated testing and synchronization control, thereby improving the overall performance and reliability of the system.

[0149] Specifically, the signal synchronization model consists of multiple modules, including a device intrinsic parameter extraction module, a dynamic change feature extraction module, a feature fusion module, and a prediction module. The device intrinsic parameter extraction module extracts intrinsic attribute features from the device intrinsic parameters of the sample. These intrinsic parameters may include the device model, rated power, rated current, etc. These parameters remain unchanged during device operation and reflect the basic characteristics of the device. The dynamic change feature extraction module extracts dynamic change features from the operating status data of the sample. This operating status data may include real-time data such as temperature, pressure, current, and voltage during device operation. This data changes with the device's operating status and reflects the real-time operating status of the device. The feature fusion module fuses the intrinsic attribute features and dynamic change features to form a comprehensive feature vector for subsequent signal synchronization prediction. The prediction module performs signal synchronization prediction based on the fused features and outputs the device's signal synchronization prediction results.

[0150] Preferably, when constructing a signal synchronization model, the following steps can be adopted. First, the inherent parameters of the device are standardized to facilitate subsequent feature fusion. Then, the dynamically changing features are preprocessed, such as removing noise, filling missing values, etc., to improve the quality of the features. Next, a suitable feature fusion method, such as weighted summation, feature splicing, etc., is used to fuse the inherent attribute features and the dynamically changing features. Finally, a suitable prediction algorithm, such as a neural network, support vector machine, etc., is selected to construct a prediction module, and the training data is used to train and optimize the model. In actual applications, the parameters and algorithms of each module can be adjusted according to the specific type of equipment and the operating environment to achieve the best prediction effect.

[0151] The following are application examples of this application in combination with specific scenarios. On the production line of a certain factory, the automatic point-to-point switch cabinet is one of the key equipment. Its main function is to control the distribution and switching of electricity to ensure the stable operation of the production line. However, traditional automated testing and control methods have many shortcomings. For example, the single sensor data acquisition method cannot fully and accurately reflect the actual operating status of the equipment, and the synchronous control model lacks the ability to adapt to the dynamic changes of the equipment, making it difficult to achieve precise synchronous control. In order to solve these problems, an equipment automated testing and synchronous control method and system based on multi-sensor fusion is adopted.

[0152] First, real-time data is collected from various sensors installed on the automatic switching cabinet. These sensors include temperature sensors, pressure sensors, current sensors, and voltage sensors. They collect data on the temperature and pressure of key components within the equipment, as well as the current and voltage of the equipment, at a frequency of once per second. They also record equipment operating parameters such as speed and load, as well as the execution of control instructions. For example, the data collected by the temperature sensor might be 45°C, 46°C, or 47°C; the data collected by the pressure sensor might be 1.2 MPa, 1.22 MPa, or 1.25 MPa; the data collected by the current sensor might be 10 A, 10.2 A, or 10.5 A; and the data collected by the voltage sensor might be 380 V, 381 V, or 382 V. The equipment might operate at speeds of 1450 rpm, 1460 rpm, or 1470 rpm, and the load might be 60%, 62%, or 65%. The data were preprocessed, including data cleaning, normalization, and feature engineering, to improve the quality and usability of the data.

[0153] Next, the first training sample set was constructed. 1,800 sets of data were selected from multi-sensor data and device operating status data under different conditions, including normal operation, overload operation, and fault operation. 1,000 sets of data were selected from normal operation, 500 sets of data were selected from overload operation, and 300 sets of data were selected from fault operation. Each set of data, considered a sample, contains multiple feature dimensions, such as temperature, pressure, current, voltage, operating speed, load percentage, and control instruction execution time. For example, a sample might contain feature values such as a temperature of 45°C, a pressure of 1.2 MPa, a current of 10 A, a voltage of 380 V, an operating speed of 1450 r / min, a load percentage of 60%, and a control instruction execution time of 0.1 s. These samples were used to train the signal synchronization model.

[0154] The signal synchronization model includes a device intrinsic parameter extraction module, a dynamic change feature extraction module, a feature fusion module, and a prediction module. The device intrinsic parameter extraction module extracts intrinsic attribute features from the sample's device intrinsic parameters, such as the device model, rated power, and rated current. The dynamic change feature extraction module extracts dynamic change features from the sample's operating status data, such as the temperature change rate and current fluctuation amplitude. The feature fusion module fuses the intrinsic attribute features and dynamic change features to form a comprehensive feature vector. The prediction module performs signal synchronization prediction based on the fused features and outputs the device's signal synchronization prediction results. The prediction module is constructed using a neural network algorithm, and the model is trained by setting a loss function and an optimization algorithm. After multiple iterative training steps, a trained signal synchronization model is obtained.

[0155] Then, based on the trained signal synchronization model, the features of each operating state of the equipment are grouped to obtain the feature type of each operating state. For the t-th operating state of the equipment, the data of the t-th operating state and the previous operating states are input into the trained signal synchronization model to obtain the feature representation of the t-th operating state of the equipment and the signal synchronization prediction result. Cluster analysis is performed on the feature representations of all operating states of all equipment, and similar feature representations are divided into the same group, thereby obtaining multiple feature types. For example, normal operating feature types, mild overload feature types, and severe overload feature types are obtained. For each feature type, the average value of the signal synchronization prediction results corresponding to the feature representations belonging to the feature type is calculated, and the feature types are sorted according to the average value of the signal synchronization prediction results to obtain the serial number of each feature type. For example, the serial number of the normal operating feature type is 1, the serial number of the mild overload feature type is 2, and the serial number of the severe overload feature type is 3.

[0156] Next, a second training sample set was constructed. Checkpoints where the operating state transitioned to stable were selected, and their feature representations served as sample input data. The control data from the device between this checkpoint and the next checkpoint served as the sample label. For example, five key checkpoints were selected during the transition from an overloaded state to a normal operating state. At these checkpoints, device features were recorded, such as a temperature of 48°C, a pressure of 1.3 MPa, a current of 10.8 A, a voltage of 385 V, an operating speed of 1480 rpm, and a load percentage of 70%. Control data executed from this checkpoint until the device reached fully stable operation included control commands for load reduction (a 10% load reduction) and motor speed adjustment parameters (reducing the motor speed from 1500 rpm to 1450 rpm). Checkpoints where the operating state transitioned to stable were selected by calculating the control influence factor between two adjacent operating states. For example, the control influence factor between the third and fourth operating states was calculated. Using the third running state as the current state, the system traverses forward to find a current state with a different feature type from the previous state as the starting state. Using the fourth running state as the current state, the system traverses backward to find a current state with a different feature type from the subsequent state as the ending state. Then, the control influence factor between the third and fourth running states is calculated based on the feature types of the starting and ending states and the state time. If the control influence factor is positive, the third running state is considered the checkpoint for the transition to stable operation.

[0157] Finally, the multi-sensor data of the device under test is input into the signal synchronization model to obtain a feature representation of the device under test. For example, the multi-sensor data of the device under test includes a temperature of 46°C, a pressure of 1.25 MPa, a current of 10.3 A, a voltage of 383 V, an operating speed of 1460 r / min, and a load percentage of 65%. This data is input into the signal synchronization model to obtain a feature representation of the device as a 10-dimensional feature vector. Then, based on the trained synchronization control model, the synchronization control results of the device under test are obtained. Five samples with similar feature representations to the device under test can be found in the second training sample set. The synchronization control results of these five samples are obtained based on the trained synchronization control model. The synchronization control results of these five samples are comprehensively evaluated based on their control influence factors to obtain the synchronization control results of the device under test. For example, the synchronization control results of these five samples are load reduction of 5%, load reduction of 8%, load reduction of 10%, load reduction of 3%, and load reduction of 6%, respectively. Their control influence factors are 0.8, 0.6, 0.9, 0.7, and 0.85, respectively. By taking a weighted average of the synchronization control results based on these control influencing factors, we obtain a 6.8% load reduction for the device under test. Alternatively, the feature representation of the device under test can be directly input into a trained synchronization control model, which will then directly output the synchronization control results for the device under test, such as reducing the load by 7% or adjusting the motor speed to 1450 rpm.

[0158] By applying a multi-sensor fusion-based automated equipment testing and synchronization control method and system, the factory's automatic point-to-point switchgear automated testing efficiency increased by 30% and synchronization control accuracy by 40%. The equipment's synchronization performance during operation has become more stable, the failure rate has been reduced by 25%, and the equipment's service life has been extended. Furthermore, this method and system can quickly and accurately acquire the device's characteristic representation and perform synchronization control, reducing manual intervention costs and testing time, improving the equipment's production and operational efficiency, and saving the company significant human and material resources.

[0159] The aforementioned embodiments of the present invention have the following beneficial effects: The present invention can improve the accuracy and control efficiency of automated testing of power equipment. Through multi-sensor data fusion and signal synchronization model training, device operating status features can be accurately extracted and intelligently grouped, providing a reliable basis for subsequent control decisions. A synchronous control model constructed based on feature types and control data can adaptively generate optimal control strategies, particularly enabling more precise regulation at critical nodes where the operating state transitions to stability.

[0160] The control influence factor calculation method allows for a scientific assessment of the control effects between adjacent states, ensuring the rationality of sample selection. Feature representation similarity matching and K-nearest neighbor comprehensive evaluation can improve system response speed while maintaining control accuracy. The hierarchical design of the signal synchronization model can separately address inherent device parameters and dynamic change characteristics, achieving a more comprehensive state representation. This invention can form a closed-loop control system from state monitoring to intelligent decision-making, providing an intelligent solution for power equipment operation and maintenance.

[0161] like Figure 2 As shown, some embodiments provide an automated device testing and synchronization control system based on multi-sensor fusion, the system comprising:

[0162] The signal synchronization model training module 201 is used to obtain multi-sensor data and equipment operating status data of the automatic point-to-point switch cabinet to construct a first training sample set; construct a signal synchronization model, and train the signal synchronization model based on the first training sample set to obtain a trained signal synchronization model;

[0163] A feature grouping module 202 is configured to perform feature grouping on each operating state of the device based on the trained signal synchronization model to obtain a feature type for each operating state;

[0164] The synchronous control model training module 203 is configured to construct a second training sample set based on the characteristic type of each operating state and the control data of the device between the two operating states; construct a synchronous control model, and train the synchronous control model based on the second training sample set to obtain a trained synchronous control model;

[0165] The synchronization control module 204 is configured to input the multi-sensor data of the device under test into a signal synchronization model to obtain a feature representation of the device under test; and obtain a synchronization control result of the device under test based on the trained synchronization control model according to the feature representation.

[0166] It is understandable that the modules and references in the equipment automation test and synchronization control system based on multi-sensor fusion are Figure 1 The steps in the device automated test and synchronous control method based on multi-sensor fusion described above correspond to each other. Therefore, the operations, features, and beneficial effects described above for the device automated test and synchronous control method based on multi-sensor fusion are also applicable to the device automated test and synchronous control system based on multi-sensor fusion and the modules contained therein, and will not be repeated here.

[0167] In some embodiments, constructing a second training sample set based on the feature type of each operating state and the control data of the device between two operating states includes:

[0168] The inspection of the transition from the running state to the stable state is selected, the feature representation of the stable transition inspection is used as the input data of the sample, and the control data of the equipment between the stable transition inspection and the next inspection is used as the label of the sample to construct the second training sample set.

[0169] It should be noted that the multi-sensor fusion-based automated equipment testing and synchronization control system involved in this invention is an integrated solution designed to achieve automated testing and precise synchronization control of equipment through the collaborative operation of multiple modules. This system builds signal synchronization models and synchronization control models, and uses these models to analyze and control the operating status of the equipment, thereby improving its operational efficiency and stability. This systematic solution can effectively integrate multi-sensor data and equipment operating status data, providing comprehensive technical support for automated testing and synchronization control of equipment.

[0170] Specifically, the signal synchronization model training module is responsible for acquiring multi-sensor data and equipment operating status data from the automatic point-to-point switchgear and constructing a first training sample set. This data includes both intrinsic and dynamic operating parameters of the equipment. This data is used to train the signal synchronization model, enabling it to accurately predict the equipment's signal synchronization status. The feature grouping module, based on the trained signal synchronization model, groups the features of each equipment operating state into groups, determining the feature type for each operating state. This process involves classifying the feature representations of the equipment operating states so that the subsequent synchronization control model can perform targeted control based on different feature types. The synchronization control model training module constructs a second training sample set based on the feature type of each operating state and the control data of the equipment between these two operating states, and trains the synchronization control model. This module optimizes the performance of the synchronization control model by analyzing the control data of the equipment under test in different operating states. The synchronization control module inputs the multi-sensor data of the equipment under test into the signal synchronization model, obtains the feature representation of the equipment under test, and then, based on the trained synchronization control model, produces the synchronization control results for the equipment under test. This module is the core execution component of the system and directly determines the accuracy and efficiency of equipment synchronization control.

[0171] Preferably, during the implementation process, the operating steps of each module can be further refined. For example, in the signal synchronization model training module, the acquired multi-sensor data and equipment operation status data can be pre-processed by data cleaning and feature engineering methods to improve the training effect of the model. In the feature grouping module, a clustering algorithm can be used to group the feature representations of the equipment operation status, and the clustering results can be optimized by evaluation indicators. In the synchronous control model training module, a cross-validation method can be used to evaluate and optimize the model to ensure the generalization ability of the model. In the synchronous control module, a real-time feedback mechanism can be introduced to dynamically adjust the synchronous control strategy according to the actual operation conditions of the equipment. Through these refined operating steps, the performance and reliability of the system can be further improved, making it better adapted to complex industrial application scenarios.

[0172] Reference below Figure 3 , which shows a schematic structural diagram of an electronic device 300 suitable for implementing some embodiments of the present invention. The electronic devices in some embodiments of the present invention may include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 3 The terminal device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0173] like Figure 3 As shown, electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 302 or programs loaded from a storage device 308 into a random access memory (RAM) 303. RAM 303 also stores various programs and data required for the operation of electronic device 300. Processing device 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to bus 304.

[0174] Typically, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Figure 3The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as needed.

[0175] Furthermore, the storage medium of the embodiment of the present application stores program instructions that can implement all the above methods, wherein the program instructions can be stored in the above storage medium in the form of a software product, including a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, or a terminal device such as a computer, server, mobile phone, or tablet.

[0176] Example 2

[0177] In the application of automatic point-to-point switchgear, by acquiring multi-sensor data and operating status data of power equipment, and utilizing signal synchronization models and synchronous control models, automated testing and precise synchronous control of equipment can be achieved.

[0178] Real-time data is collected from various sensors installed on the automatic switching cabinet (such as temperature sensors, pressure sensors, and current sensors). Equipment operating parameters (such as voltage, current, and power) and the execution of control instructions are recorded. The collected multi-sensor data is integrated with equipment operating status data to construct a first training sample set containing various equipment states. This first training sample set is trained using a machine learning algorithm (such as a neural network) to develop a signal synchronization model capable of predicting equipment signal synchronization.

[0179] Based on the trained signal synchronization model, features are extracted and grouped for each device operating state to obtain the feature type for each operating state. Data from checkpoints where the operating state transitions to stability is selected and combined with the control data between these two states to construct a second training sample set. This second training sample set is used to train the synchronization control model, enabling it to output synchronization control results based on the input feature representation. Multi-sensor data from the device under test is input into the trained signal synchronization model to obtain a feature representation of the device. Based on the feature representation, the synchronization control model outputs the device's synchronization control results. Based on the synchronization control results, corresponding control instructions are sent to the device, enabling automated testing and synchronization control of the device.

[0180] Automated testing and synchronized control reduce manual intervention and improve testing efficiency. Multi-sensor data fusion and machine learning models enable precise perception and control of equipment status, improving control accuracy. Real-time monitoring and synchronized control enable timely detection and correction of equipment anomalies, ensuring stable operation.

[0181] The above descriptions merely illustrate some preferred embodiments of the present invention and the underlying technical principles. Those skilled in the art should understand that the scope of the invention encompassed by the embodiments of the present invention is not limited to technical solutions formed by specific combinations of the aforementioned technical features. It also encompasses other technical solutions formed by any combination of the aforementioned technical features or their equivalents, without departing from the aforementioned inventive concept. For example, a technical solution formed by replacing the aforementioned features with (but not limited to) technical features with similar functions disclosed in the embodiments of the present invention.

Claims

1. A device automation test and synchronization control method based on multi-sensor fusion, characterized in that: The following steps are involved: Acquire multi-sensor data and equipment operating status data of the automatic point-to-point switchgear to construct a first training sample set; Constructing a signal synchronization model, and training the signal synchronization model based on the first training sample set to obtain a trained signal synchronization model; Based on the trained signal synchronization model, the characteristics of each operating state of the equipment are grouped to obtain the characteristic type of each operating state; Constructing a second training sample set based on the feature type of each operating state and the control data of the device between two operating states; Constructing a synchronous control model, and training the synchronous control model based on the second training sample set to obtain a trained synchronous control model; Input the multi-sensor data of the device to be tested into the signal synchronization model to obtain the feature representation of the device to be tested; obtain the synchronization control result of the device to be tested based on the trained synchronization control model according to the feature representation, A second training sample set is constructed based on the feature type of each operating state and the control data of the device between two operating states, including: Select the inspection of the transition from the running state to the stable state, use the feature representation of the stable transition inspection as the input data of the sample, and use the control data of the equipment between the stable transition inspection and the next inspection as the label of the sample to construct the second training sample set. Select the check for transition from the running state to the stable state in the following ways: For each device, calculate the control impact factor between two adjacent operating states; If the control influence factor between the t-th operating state and the t+1-th operating state is positive, the t-th operating state is a check for the transition of the operating state to stability. The control influence factor between the t-th operating state and the t+1-th operating state is calculated in the following way: Take the tth running state as the current state and traverse forward in sequence until a current state with a feature type different from that of the previous state is found as the starting state; Take the t+1th running state as the current state and traverse backward in sequence until a current state with a feature type different from that of the next state is found as the end state; The control influence factor between the t-th operating state and the t+1-th operating state is calculated based on the characteristic types of the starting state and the ending state and the state time.

2. The device automated testing and synchronous control method based on multi-sensor fusion according to claim 1 is characterized in that: Based on the characteristic types of the starting state and the ending state and the state time, the control influence factor between the t-th operating state and the t+1-th operating state is calculated using the following formula: Among them, the feature type of the starting state is the i-th feature type, and the feature type of the ending state is the j-th feature type. Represents the distance between the feature type of the starting state and the feature type of the ending state, Indicates the time of the starting state, Indicates the time of the cut-off state, represents the hyperbolic tangent function, represents the symbolic function, and Indicates parameters, Represents the control influence factor between the t-th operating state and the t+1-th operating state.

3. The device automated testing and synchronous control method based on multi-sensor fusion according to claim 1 is characterized in that: Obtaining a synchronization control result of the device to be tested based on the trained synchronization control model according to the characteristic representation includes: Find K samples in the second training sample set that have similar feature representations to the device to be tested; Based on the trained synchronization control model, K samples of synchronization control results are obtained; The synchronous control results of the K samples are comprehensively evaluated based on the control influence factors of the K samples to obtain the synchronous control results of the device to be tested.

4. The device automated testing and synchronous control method based on multi-sensor fusion according to claim 1 is characterized in that: Based on the trained signal synchronization model, the characteristics of each device operation state are grouped to obtain the feature types of each operation state, including: For the t-th operating state of the device, the t-th operating state and the previous operating state data are input into the trained signal synchronization model to obtain the feature representation of the t-th operating state of the device and the signal synchronization prediction result; The feature representations of all operating states of all devices are grouped to obtain multiple feature types; For each feature type, calculate the average value of the signal synchronization prediction results corresponding to the feature representations of the feature type, and sort the feature types according to the average value of the signal synchronization prediction results to obtain the sequence number of each feature type; The feature type of each running state is the feature type to which the feature representation of the running state belongs.

5. The device automated testing and synchronous control method based on multi-sensor fusion according to claim 1 is characterized in that: The signal synchronization model includes: A device intrinsic parameter extraction module is used to extract intrinsic attribute features from the device intrinsic parameters of the sample; A dynamic change feature extraction module is used to extract dynamic change features from the running status data of the sample; Feature fusion module, used to fuse inherent attribute features and dynamic change features; The prediction module is used to perform signal synchronization prediction based on the fused features.

6. An automated equipment testing and synchronization control system based on multi-sensor fusion, characterized in that: include: A signal synchronization model training module is used to obtain multi-sensor data and equipment operating status data of the automatic point-to-point switchgear to construct a first training sample set; Constructing a signal synchronization model, and training the signal synchronization model based on the first training sample set to obtain a trained signal synchronization model; The feature grouping module is used to group the features of each operating state of the device based on the trained signal synchronization model to obtain the feature type of each operating state; A synchronous control model training module is used to construct a second training sample set based on the characteristic type of each operating state and the control data of the device between the two operating states; Constructing a synchronous control model, and training the synchronous control model based on the second training sample set to obtain a trained synchronous control model; A synchronization control module is used to input the multi-sensor data of the device to be tested into the signal synchronization model to obtain a feature representation of the device to be tested; and obtain a synchronization control result of the device to be tested based on the trained synchronization control model according to the feature representation; A second training sample set is constructed based on the feature type of each operating state and the control data of the device between two operating states, including: Selecting an inspection in which the operating state transitions to a stable state, using the feature representation of the stable transition inspection as the input data of the sample, and using the control data of the device between the stable transition inspection and the next inspection as the label of the sample, to construct a second training sample set; A second training sample set is constructed based on the feature type of each operating state and the control data of the device between two operating states, including: Select the inspection of the transition from the running state to the stable state, use the feature representation of the stable transition inspection as the input data of the sample, and use the control data of the equipment between the stable transition inspection and the next inspection as the label of the sample to construct the second training sample set. Select the check for transition from the running state to the stable state in the following ways: For each device, calculate the control impact factor between two adjacent operating states; If the control influence factor between the t-th operating state and the t+1-th operating state is positive, the t-th operating state is a check for the transition of the operating state to stability. The control influence factor between the t-th operating state and the t+1-th operating state is calculated in the following way: Take the tth running state as the current state and traverse forward in sequence until a current state with a feature type different from that of the previous state is found as the starting state; Take the t+1th running state as the current state and traverse backward in sequence until a current state with a feature type different from that of the next state is found as the end state; The control influence factor between the t-th operating state and the t+1-th operating state is calculated based on the characteristic types of the starting state and the ending state and the state time.

Images