Multi-modal information self-adaptive acquisition method and acquisition device for power equipment

Abstract

The invention discloses a multi-modal information adaptive acquisition method and acquisition device for power equipment. The method comprises the following steps: acquiring monitoring data of the power equipment; according to the monitoring data, state prediction is carried out based on a preset prediction model, and a prediction value sequence of the adaptive quantity in a preset time domain is obtained; performing calculation based on a preset adaptive frequency adjustment strategy according to the predicted value sequence to obtain an optimal acquisition frequency adjustment amount required at the current moment; and according to the optimal acquisition frequency adjustment amount required at the current moment, in combination with the acquisition frequency of the adaptive quantity at the current moment, obtaining the optimal acquisition frequency of the adaptive quantity at the next moment, and updating the acquisition frequency setting of the adaptive quantity so as to realize adaptive dynamic frequency acquisition of the adaptive quantity. And for the slowly changing adaptive quantity, the data transmission quantity and the storage quantity can be effectively reduced. And when an abnormal condition occurs, the sampling frequency can be improved in time, and the timeliness and the accuracy of monitoring the running state of the power equipment are enhanced.

Description

Technical Field

[0001] This invention relates to a multimodal adaptive acquisition method and acquisition device for power equipment, belonging to the field of power equipment information acquisition technology. Background Technology

[0002] The safe and stable operation of power equipment is crucial for power installations. Health monitoring of power equipment requires the collection of information across multiple modalities, including electrical quantities (such as current, voltage, and power) and non-electrical quantities (such as vibration, insulating oil chromatography, insulating oil temperature, and gas content in the insulating oil). Non-electrical quantities may change in response to variations in electrical quantities or be affected by environmental factors (such as temperature). Depending on the operating conditions of the power equipment, both electrical and non-electrical quantities may exhibit gradual changes or rapid fluctuations.

[0003] In traditional monitoring methods, the sensor's acquisition frequency is set at a fixed frequency. This fixed acquisition frequency cannot adapt to rapidly changing parameters. For parameters with gradual changes, reducing the sensor's acquisition frequency can help reduce data transmission and storage, thus alleviating the burden on the device. Conversely, when anomalies cause sudden parameter changes, increasing the sensor's acquisition frequency becomes crucial; otherwise, a fixed acquisition frequency cannot capture sufficient details of the changes, affecting the accuracy of fault diagnosis. Furthermore, the correlation between different parameters is not fully utilized, resulting in a lack of flexibility and specificity in the acquisition strategy. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a multimodal information adaptive acquisition method and device for power equipment. It employs a dynamic acquisition method based on state prediction and adaptive frequency adjustment strategies to adaptively acquire adaptive quantities. For slowly changing adaptive quantities, it can effectively reduce data transmission and storage volume, alleviating the data processing burden on the device. Furthermore, in the event of abnormal conditions, it can promptly increase the sampling frequency, enhancing the timeliness and accuracy of monitoring the operating status of power equipment.

[0005] To achieve the above objectives, the present invention is implemented using the following technical solution:

[0006] On one hand, this invention discloses an adaptive acquisition method for multimodal information of power equipment, comprising the following steps:

[0007] Acquire monitoring data of power equipment, the monitoring data including monitoring data of basic quantities and monitoring data of adaptive quantities;

[0008] Based on the monitoring data, state prediction is performed using a preset prediction model to obtain a sequence of predicted values ​​for the adaptive quantity within a preset time domain.

[0009] Based on the predicted value sequence, the optimal acquisition frequency adjustment amount required at the current moment is calculated using a preset adaptive frequency adjustment strategy.

[0010] Based on the optimal acquisition frequency adjustment required at the current moment, and combined with the acquisition frequency of the adaptive quantity at the current moment, the optimal acquisition frequency of the adaptive quantity at the next moment is obtained, and the acquisition frequency setting of the adaptive quantity is updated to achieve adaptive dynamic frequency acquisition of the adaptive quantity.

[0011] Furthermore, the basic quantities include one or more of current, voltage, and ambient temperature, and the monitoring data of the basic quantities are periodically collected using a preset fixed acquisition frequency.

[0012] The adaptive quantity includes one or more of insulating oil chromatography and equipment temperature, and the monitoring data of the adaptive quantity is adaptively acquired using a dynamic acquisition method based on state prediction and adaptive frequency adjustment strategy.

[0013] Furthermore, the expression for the prediction model is as follows:

[0014]

[0015] In the formula, Indicates the first The predicted value of the adaptive quantity at time;

[0016] Indicates the forgetting factor; Indicates the first The predicted value of the adaptive quantity at time;

[0017] Indicates the first The monitored value of the adaptive quantity at any given time;

[0018] Indicates the first The monitored value of the adaptive quantity at any given time;

[0019] Indicates the first The basic monitoring value at any given time;

[0020] Indicates the first The basic monitoring value at any given time;

[0021] Indicates the number of historical monitoring values;

[0022] This represents the prediction function.

[0023] Furthermore, the prediction function includes a multiple linear regression function, the expression of which is as follows:

[0024]

[0025] In the formula, This represents the weighting coefficient of the monitored value of the i-th adaptive quantity; Indicates the first The monitored value of the adaptive quantity at any given time;

[0026] This represents the weighting coefficient of the monitoring value of the j-th basic quantity; Indicates the first Monitoring values ​​of basic quantities at any given time;

[0027] This indicates the bias term.

[0028] Furthermore, obtaining the optimal acquisition frequency adjustment amount required at the current moment includes the following steps:

[0029] Based on the predicted value sequence and combined with the monitoring data, calculate the prediction deviation sequence of the adaptive quantity within the preset time domain;

[0030] Based on the predicted deviation sequence and a preset adaptive frequency adjustment strategy, the initial acquisition frequency adjustment amount required at the current moment is calculated.

[0031] If the initial acquisition frequency adjustment amount is greater than or equal to a preset minimum acquisition frequency adjustment threshold and less than or equal to a preset maximum acquisition frequency adjustment threshold, then the initial acquisition frequency adjustment amount is taken as the optimal acquisition frequency adjustment amount required at the current moment.

[0032] If the initial acquisition frequency adjustment amount is less than the preset minimum acquisition frequency adjustment threshold, then the minimum acquisition frequency adjustment threshold is taken as the optimal acquisition frequency adjustment amount required at the current moment.

[0033] If the initial acquisition frequency adjustment amount is greater than the preset maximum acquisition frequency adjustment threshold, then the maximum acquisition frequency adjustment threshold is taken as the optimal acquisition frequency adjustment amount required at the current moment.

[0034] Furthermore, the adaptive frequency adjustment strategy includes a model predictive control method, which solves for minimizing the accumulation of the prediction deviation sequence as the objective function to calculate the initial acquisition frequency adjustment amount required at the current moment.

[0035] The objective function is expressed as follows:

[0036]

[0037] In the formula, Represent the objective function;

[0038] Indicates the first The predicted deviation between the adaptive quantity and the monitored quantity at time t is expressed as: , Indicates the first The predicted value of the adaptive quantity at time; Indicates the first The monitored value of the adaptive quantity at any given time;

[0039] Indicates the sequence number for the preset time. Indicates the range of the preset time domain;

[0040] Indicates the first The adaptive sampling frequency adjustment value at any given time;

[0041] Indicates the sequential index for controlling the time domain; Indicates the range of control over the time domain, and ;

[0042] Represents the L2 norm; The weight matrix represents the adjustment for prediction bias; This represents the weight matrix used to adjust the sampling frequency.

[0043] Furthermore, the adaptive frequency adjustment strategy includes a proportional-integral (PI) control method. This PI control method calculates the required acquisition frequency adjustment at the current moment based on a preset proportional-integral-derivative (PID) controller, as expressed below:

[0044]

[0045] In the formula, Indicates the first The amount of time-adaptive measurement frequency adjustment;

[0046] Indicates the proportionality coefficient; Indicates the first The predicted value of the adaptive quantity at any given time is the predicted deviation from the monitored value.

[0047] Indicates the integral coefficient; Indicates the first The predicted value of the adaptive quantity at any given time is the predicted deviation from the monitored value.

[0048] Represents the differential coefficient; Indicates the first The predicted value of the adaptive quantity at time point is the deviation from the predicted value of the monitored quantity.

[0049] Furthermore, obtaining the optimal acquisition frequency of the adaptive quantity at the next moment includes the following steps:

[0050] Based on the optimal acquisition frequency adjustment required at the current moment, combined with the acquisition frequency of the adaptive quantity at the current moment, and based on the smooth update strategy, the preliminary acquisition frequency of the adaptive quantity at the next moment is calculated.

[0051] If the initial acquisition frequency is greater than or equal to a preset minimum acquisition frequency threshold and less than or equal to a preset maximum acquisition frequency threshold, then the initial acquisition frequency is taken as the optimal acquisition frequency for the adaptive quantity at the next moment.

[0052] If the initial acquisition frequency is less than a preset minimum acquisition frequency threshold, then the minimum acquisition frequency threshold is used as the optimal acquisition frequency for the adaptive quantity at the next moment.

[0053] If the initial acquisition frequency is greater than the preset maximum acquisition frequency threshold, then the maximum acquisition frequency threshold is used as the optimal acquisition frequency for the adaptive quantity at the next moment.

[0054] Furthermore, the calculation steps of the smooth update strategy are as follows:

[0055]

[0056] In the formula, Indicates the first The sampling frequency of the adaptive quantity at any given time; Indicates the smoothing coefficient; Indicates the first The amount of time-adaptive measurement frequency adjustment; Indicates the first The sampling frequency of the adaptive quantity at any given time.

[0057] On the other hand, the present invention discloses a multimodal information adaptive acquisition device for power equipment, applicable to the aforementioned multimodal information adaptive acquisition method for power equipment, comprising:

[0058] The data acquisition module is used to acquire monitoring data of power equipment, including monitoring data of basic quantities and monitoring data of adaptive quantities;

[0059] The state prediction module is used to predict the state based on the monitoring data and a preset prediction model to obtain a sequence of predicted values ​​of the adaptive quantity in a preset time domain.

[0060] The frequency adjustment module is used to calculate the optimal acquisition frequency adjustment amount required at the current moment based on the predicted value sequence and a preset adaptive frequency adjustment strategy.

[0061] The adaptive acquisition module is used to adjust the optimal acquisition frequency according to the current time, combine the current time's adaptive acquisition frequency to obtain the next time's optimal acquisition frequency, and update the adaptive acquisition frequency setting to achieve adaptive dynamic frequency acquisition of the adaptive quantity.

[0062] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0063] This invention discloses a multimodal information adaptive acquisition method and device for power equipment. It employs a dynamic acquisition method based on state prediction and adaptive frequency adjustment strategies to adaptively acquire adaptive quantities. For slowly changing adaptive quantities, the sampling frequency can be reduced, effectively decreasing data transmission and storage volume, alleviating the data processing burden on the device, and improving data storage and transmission efficiency. Conversely, in abnormal situations, such as sudden changes in equipment load, the sampling frequency can be increased promptly to capture more detailed changes, helping to quickly and accurately identify potential equipment problems and enhancing the timeliness and accuracy of power equipment operating status monitoring. Attached Figure Description

[0064] Figure 1 This is a flowchart illustrating the adaptive acquisition method for multimodal information of power equipment provided in Embodiment 1 of the present invention.

[0065] Figure 2 This is a schematic diagram of the structure of the multimodal information adaptive acquisition device for power equipment provided in Embodiment 4 of the present invention. Detailed Implementation

[0066] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0067] Example 1

[0068] This embodiment 1 provides a method for adaptive acquisition of multimodal information of power equipment, such as... Figure 1 As shown, it includes the following steps:

[0069] Acquire monitoring data of power equipment, including monitoring data of basic quantities and monitoring data of adaptive quantities;

[0070] Based on the monitoring data, state prediction is performed using a preset prediction model to obtain a sequence of predicted values ​​for the adaptive quantity within a preset time domain.

[0071] Based on the predicted value sequence, the optimal acquisition frequency adjustment amount required at the current moment is calculated using a preset adaptive frequency adjustment strategy.

[0072] Based on the optimal acquisition frequency adjustment required at the current moment, and combined with the acquisition frequency of the adaptive quantity at the current moment, the optimal acquisition frequency of the adaptive quantity at the next moment is obtained, and the acquisition frequency setting of the adaptive quantity is updated to achieve adaptive dynamic frequency acquisition of the adaptive quantity.

[0073] The technical concept of this invention is as follows: This application employs a dynamic acquisition method based on state prediction and adaptive frequency adjustment strategies to adaptively acquire adaptive quantities. For slowly changing adaptive quantities, the sampling frequency can be reduced, effectively reducing data transmission and storage volume, alleviating the data processing burden of the device, and improving the efficiency of data storage and transmission. However, in abnormal situations, such as sudden changes in equipment load, the sampling frequency can be increased promptly to capture more detailed changes, helping to quickly and accurately identify potential equipment problems and enhancing the timeliness and accuracy of monitoring the operating status of power equipment.

[0074] like Figure 1 As shown, the specific steps are as follows:

[0075] Step 1: Obtain monitoring data of the power equipment. The monitoring data includes monitoring data of basic quantities and monitoring data of adaptive quantities.

[0076] The basic quantities include one or more of current, voltage and ambient temperature. The monitoring data of the basic quantities are collected periodically at a preset fixed acquisition frequency.

[0077] Adaptive quantities include one or more of insulating oil chromatography and equipment temperature. The monitoring data of adaptive quantities are acquired using a dynamic acquisition method based on state prediction and adaptive frequency adjustment strategy to reduce data transmission and storage while meeting the requirements of acquisition accuracy.

[0078] Specifically, data is collected through pre-installed current sensors, voltage sensors, temperature sensors, oil chromatography sensors, etc. in the power equipment, and the collected analog signals are converted into digital signals and transmitted to the host computer.

[0079] In actual power equipment operation monitoring, such as monitoring a transformer, parameters such as current, voltage, and ambient temperature can be determined as basic quantities. These basic quantities directly and stably reflect the basic operating state of the transformer, and their sampling frequency can be set to a fixed value, such as sampling once every 10 seconds. Parameters such as insulating oil chromatography and equipment temperature are used as adaptive quantities. Insulating oil chromatography changes relatively slowly, and high-frequency sampling is not required under normal circumstances. However, when potential equipment malfunctions occur, it is necessary to capture changes promptly, so its sampling frequency needs to be adaptively adjusted according to the situation. Equipment temperature may be affected by various factors such as current and ambient temperature, and its variation pattern is complex, making an adaptive sampling frequency suitable for it as well.

[0080] Step 2: Based on the monitoring data, perform state prediction using a preset prediction model to obtain the predicted value sequence of the adaptive quantity within a preset time domain.

[0081] For different adaptive quantities, a prediction model is built and its parameters are trained using its own historical monitoring values ​​and the historical monitoring values ​​of related basic quantities.

[0082] The expression for the prediction model is as follows:

[0083]

[0084] In the formula, Indicates the first The predicted value of the adaptive quantity at time;

[0085] Indicates the forgetting factor; Indicates the first The predicted value of the adaptive quantity at time;

[0086] Indicates the first The monitored value of the adaptive quantity at any given time;

[0087] Indicates the first The monitored value of the adaptive quantity at any given time;

[0088] Indicates the first The basic monitoring value at any given time;

[0089] Indicates the first The basic monitoring value at any given time;

[0090] Indicates the number of historical monitoring values;

[0091] This represents the prediction function.

[0092] It should be noted here that each monitoring data is divided into basic quantity and adaptive quantity. For every basic quantity monitoring value, there is one adaptive quantity monitoring value. Therefore, the number of basic quantity monitoring values ​​and adaptive quantity monitoring values ​​is n.

[0093] Furthermore, the prediction function includes a multiple linear regression function, the expression of which is as follows:

[0094]

[0095] In the formula, This represents the weighting coefficient of the monitored value of the i-th adaptive quantity; Indicates the first The monitored value of the adaptive quantity at any given time; Indicates the number of historical monitoring values;

[0096] This represents the weighting coefficient of the monitoring value of the j-th basic quantity; Indicates the first Monitoring values ​​of basic quantities at any given time;

[0097] This indicates the bias term.

[0098] It should be noted here that, These are the relevant basic quantities, and may involve one or more basic quantities depending on actual needs.

[0099] Forgetting factor The range of values ​​for is: The value should be selected based on the actual situation.

[0100] Step 3: Based on the predicted value sequence, calculate the optimal acquisition frequency adjustment amount required at the current moment according to the preset adaptive frequency adjustment strategy.

[0101] To obtain the optimal acquisition frequency adjustment required at the current moment, the following steps are included:

[0102] Based on the predicted value sequence and combined with the monitoring data, calculate the prediction deviation sequence of the adaptive quantity within the preset time domain;

[0103] Based on the predicted deviation sequence and a preset adaptive frequency adjustment strategy, the initial acquisition frequency adjustment amount required at the current moment is calculated.

[0104] If the initial acquisition frequency adjustment is greater than or equal to the preset minimum acquisition frequency adjustment threshold and less than or equal to the preset maximum acquisition frequency adjustment threshold, then the initial acquisition frequency adjustment is taken as the optimal acquisition frequency adjustment required at the current moment.

[0105] If the initial acquisition frequency adjustment is less than the preset minimum acquisition frequency adjustment threshold, then the minimum acquisition frequency adjustment threshold is taken as the optimal acquisition frequency adjustment required at the current moment.

[0106] If the initial acquisition frequency adjustment amount is greater than the preset maximum acquisition frequency adjustment threshold, then the maximum acquisition frequency adjustment threshold is taken as the optimal acquisition frequency adjustment amount required at the current moment.

[0107] This method avoids excessively frequent sampling frequency adjustments by setting the maximum allowable step size for sampling frequency adjustment between adjacent sampling times, i.e., the maximum sampling frequency adjustment threshold.

[0108] This method also suppresses invalid adjustment disturbances and prevents mechanical fatigue of the sensor / actuator due to high-frequency micro-adjustments by setting the minimum allowable step size for the sampling frequency adjustment between adjacent sampling times, i.e., the minimum acquisition frequency adjustment threshold.

[0109] Specifically, adaptive frequency adjustment strategies include model predictive control and proportional-integral control, which should be selected according to the actual situation.

[0110] The model predictive control method uses minimizing the accumulation of the prediction deviation sequence as the objective function to calculate the initial acquisition frequency adjustment required at the current moment.

[0111] The objective function is expressed as follows:

[0112]

[0113] In the formula, Represent the objective function;

[0114] Indicates the first The predicted deviation between the adaptive quantity and the monitored quantity at time t is expressed as: , Indicates the first The predicted value of the adaptive quantity at time; Indicates the first The monitored value of the adaptive quantity at any given time;

[0115] Indicates the sequence number for the preset time. Indicates the range of the preset time domain;

[0116] Indicates the first The adaptive sampling frequency adjustment value at any given time;

[0117] Indicates the sequential index for controlling the time domain; Indicates the range of control over the time domain, and ;

[0118] Represents the L2 norm; The weight matrix represents the adjustment for prediction bias; This represents the weight matrix used to adjust the sampling frequency.

[0119] The proportional-integral (PI) control method calculates the required frequency adjustment for the current moment based on a preset proportional-integral-derivative (PID) controller, as shown in the following expression:

[0120]

[0121] In the formula, Indicates the first The amount of time-adaptive measurement frequency adjustment;

[0122] Indicates the proportionality coefficient; Indicates the first The predicted value of the adaptive quantity at any given time is the predicted deviation from the monitored value.

[0123] Indicates the integral coefficient; Indicates the first The predicted value of the adaptive quantity at any given time is the predicted deviation from the monitored value.

[0124] Represents the differential coefficient; Indicates the first The predicted value of the adaptive quantity at time point is the deviation from the predicted value of the monitored quantity.

[0125] The proportional, integral, and derivative coefficients are tuned according to the characteristics of the adaptive quantity and the control requirements.

[0126] Step 4: Based on the optimal acquisition frequency adjustment required at the current moment, and combined with the acquisition frequency of the adaptive quantity at the current moment, obtain the optimal acquisition frequency of the adaptive quantity at the next moment, and update the acquisition frequency setting of the adaptive quantity to achieve adaptive dynamic frequency acquisition of the adaptive quantity.

[0127] Specifically, the following steps are included:

[0128] Based on the optimal acquisition frequency adjustment required at the current moment, combined with the acquisition frequency of the adaptive quantity at the current moment, and based on the smooth update strategy, the preliminary acquisition frequency of the adaptive quantity at the next moment is calculated.

[0129] If the initial acquisition frequency is greater than or equal to the preset minimum acquisition frequency threshold and less than or equal to the preset maximum acquisition frequency threshold, then the initial acquisition frequency is taken as the optimal acquisition frequency of the adaptive quantity at the next moment.

[0130] If the initial acquisition frequency is less than the preset minimum acquisition frequency threshold, then the minimum acquisition frequency threshold is used as the optimal acquisition frequency for the adaptive quantity at the next moment.

[0131] If the initial acquisition frequency is greater than the preset maximum acquisition frequency threshold, then the maximum acquisition frequency threshold will be used as the optimal acquisition frequency for the adaptive quantity at the next moment.

[0132] The smooth update strategy smooths out the sampling frequency adjustment process by using methods such as low-pass filtering or weighted averaging to make the sampling frequency changes more stable and reduce the impact on the data acquisition device. The calculation steps in this embodiment are as follows:

[0133]

[0134] In the formula, Indicates the first The sampling frequency of the adaptive quantity at any given time; Indicates the smoothing coefficient; Indicates the first The amount of time-adaptive measurement frequency adjustment; Indicates the first The sampling frequency of the adaptive quantity at any given time.

[0135] This invention divides the physical quantities monitored by the sensor into basic quantities and adaptive quantities, and adaptively adjusts the sampling frequency of the adaptive quantities. For slowly changing adaptive quantities, the sampling frequency can be reduced, effectively reducing data transmission and storage volume, alleviating the data processing burden of the device, and improving the efficiency of data storage and transmission. Conversely, in abnormal situations, such as sudden changes in equipment load, the sampling frequency can be increased promptly to capture more detailed changes, helping to quickly and accurately identify potential equipment problems and enhancing the timeliness and accuracy of monitoring the operating status of power equipment.

[0136] By establishing an adaptive prediction model and updating it continuously, combined with calculation of prediction bias, the predictions become more accurate. An adaptive frequency adjustment strategy based on control theory minimizes prediction bias and further optimizes the sampling frequency. Sampling frequency adjustment constraints and smoothing processes ensure reasonable and stable adjustments, avoiding the impact of frequent changes on the equipment. Overall, this improves the adaptability and reliability of multimodal information acquisition for power equipment, helps to more accurately grasp the equipment's operating status, prevent faults in advance, ensure the safe and stable operation of power installations, reduce equipment maintenance costs, and enhance the intelligence level of power equipment management.

[0137] Example 2

[0138] This embodiment 2 provides a method for adaptive acquisition of multimodal information of power equipment. Taking the equipment temperature collected by a temperature sensor as an example, it obtains historical monitoring data of the equipment temperature over the past 10 time periods. The current, along with the current collected by the current sensor, will be used as the basic quantity. Historical monitoring data collected during the same period .

[0139] The prediction function in the prediction model is constructed based on a combination of the physical principles of heat conduction and data-driven approaches. Through training with a large amount of historical data, the parameters in the function are determined, thereby obtaining the device temperature at time [time value missing]. Predicted value .

[0140] At each sampling time, such as the 50th sampling time, the latest device temperature measurement value is obtained. and current measurement value .

[0141] Setting a forgetting factor Based on the previously established prediction model and update formula The prediction model for equipment temperature is updated to obtain new predicted values. This is so that the prediction bias can be calculated later.

[0142] When the device temperature is obtained at time... Predicted value and actual measured value Then, the prediction deviation is calculated. This deviation value will serve as an important basis for subsequent adjustments to the sampling frequency.

[0143] The adaptive frequency adjustment strategy used in this embodiment 2 is a model predictive control method: assuming we set the prediction time domain... , Weight matrix and The sampling frequency is set based on the importance of the equipment temperature and the sensitivity to adjustments. At a given sampling time, the sampling frequency is adjusted based on existing predicted equipment temperatures. and actual measured value And the previously calculated sampling frequency adjustment amount Substitute this into the objective function.

[0144] Simultaneously consider the physical limitations of equipment temperature, such as the temperature range corresponding to the sensor's measurement range limiting the sampling frequency, and the data acquisition capabilities, such as the maximum sampling frequency allowed by the data transmission bandwidth, and set upper and lower limits for the sampling frequency. Hz, which means an interval of 20 seconds. Hz, i.e., an interval of 100ms; and sampling frequency adjustment rate constraints. Hz, Hz. By solving the above optimization problem, the initial acquisition frequency adjustment required at the current moment is obtained. .

[0145] Set the maximum allowable step size for sampling frequency adjustment between adjacent sampling times, i.e., the preset maximum sampling frequency adjustment threshold. Hz. If the sampling frequency adjustment is calculated at a certain sampling time... Hz, then limit it to Hz. A weighted average method is used for smoothing, and a smoothing coefficient is set. Assuming the current sampling frequency Hz, Hz, then the current sampling frequency Hz.

[0146] The model predictive control method employed in Example 2 achieves intelligent frequency adjustment ahead of equipment state changes by continuously optimizing the predicted deviation sequence in the future time domain. This method, centered on an objective function, balances acquisition accuracy and system cost to obtain the globally optimal solution within a preset time domain, effectively overcoming the short-sightedness of traditional methods and ensuring long-term control quality. As a feedforward-feedback composite control mechanism, it can act in advance based on predicted information, resulting in a faster and more intelligent response. Simultaneously, this optimization framework naturally incorporates the physical constraints of frequency and its rate of change, fundamentally guaranteeing the stability and engineering feasibility of the control process. This strategy is particularly suitable for acquisition scenarios with complex dynamics and inertia, such as equipment temperature, highlighting not only the advanced nature and intelligent advantages of this invention but also marking a creative application of predictive control theory in the field of power equipment information acquisition.

[0147] Example 3

[0148] This embodiment 3 provides a multimodal information adaptive acquisition device for power equipment. Unlike embodiment 2, this embodiment 3 uses a proportional-integral control method for adaptive frequency adjustment, while the other steps are the same.

[0149] For the adaptive variable of equipment temperature, the proportional coefficient is tuned according to its characteristics and control requirements. =0.8, integral coefficient Differential coefficients At the sampling time The initial acquisition frequency adjustment amount was calculated.

[0150] Assuming the calculation yields Hz, then perform amplitude limiting, because it meets the requirements. , Minimum sampling frequency adjustment threshold This represents the maximum acquisition frequency adjustment threshold, so the initial acquisition frequency adjustment amount can be directly used as the optimal acquisition frequency adjustment amount required at the current moment.

[0151] The proportional-integral (PI) control method employed in Example 3 uses feedback correction based on the instantaneous deviation between the current and past states, achieving rapid response and precise correction to changes in equipment status. This method, with its simple algorithm structure, boasts extremely high response speed, enabling immediate responses to sudden deviations and ensuring timely control. Furthermore, this strategy does not rely on a precise system mathematical model, exhibiting strong robustness to parameter changes and model mismatch, and good adaptability. In addition, the PID controller, as the most classic and widely used control algorithm in industry, has significant advantages in terms of ease of understanding, implementation, and debugging, effectively reducing system implementation and maintenance costs. This design is particularly suitable for rapidly changing data acquisition scenarios requiring immediate response, or for providing reliable control when models are difficult to construct accurately, demonstrating strong practicality and wide applicability.

[0152] Example 4

[0153] This embodiment 4 provides a multimodal information adaptive acquisition device for power equipment, such as... Figure 2 As shown, the adaptive acquisition method for multimodal information of the aforementioned power equipment includes:

[0154] The data acquisition module is used to acquire monitoring data of power equipment, including monitoring data of basic quantities and monitoring data of adaptive quantities.

[0155] The state prediction module is used to predict the state based on the monitoring data and a preset prediction model to obtain the predicted value sequence of the adaptive quantity in the preset time domain.

[0156] The frequency adjustment module is used to calculate the optimal acquisition frequency adjustment amount required at the current moment based on the predicted value sequence and a preset adaptive frequency adjustment strategy.

[0157] The adaptive acquisition module is used to adjust the amount according to the optimal acquisition frequency required at the current moment. Combined with the acquisition frequency of the adaptive quantity at the current moment, it obtains the optimal acquisition frequency of the adaptive quantity at the next moment and updates the acquisition frequency setting of the adaptive quantity to realize the adaptive dynamic frequency acquisition of the adaptive quantity.

[0158] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

[0159] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0160] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0161] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0162] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0163] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for adaptive acquisition of multimodal information of power equipment, characterized in that, Includes the following steps: Acquire monitoring data of power equipment, the monitoring data including monitoring data of basic quantities and monitoring data of adaptive quantities; Based on the monitoring data, state prediction is performed using a preset prediction model to obtain a sequence of predicted values ​​for the adaptive quantity within a preset time domain. Based on the predicted value sequence, the optimal acquisition frequency adjustment amount required at the current moment is calculated using a preset adaptive frequency adjustment strategy. Based on the optimal acquisition frequency adjustment required at the current moment, and combined with the acquisition frequency of the adaptive quantity at the current moment, the optimal acquisition frequency of the adaptive quantity at the next moment is obtained, and the acquisition frequency setting of the adaptive quantity is updated to achieve adaptive dynamic frequency acquisition of the adaptive quantity.

2. The adaptive acquisition method for multimodal information of power equipment according to claim 1, characterized in that, The basic quantities include one or more of current, voltage and ambient temperature, and the monitoring data of the basic quantities are periodically collected using a preset fixed collection frequency. The adaptive quantity includes one or more of insulating oil chromatography and equipment temperature, and the monitoring data of the adaptive quantity is adaptively acquired using a dynamic acquisition method based on state prediction and adaptive frequency adjustment strategy.

3. The adaptive acquisition method for multimodal information of power equipment according to claim 1, characterized in that, The expression for the prediction model is as follows: ； In the formula, Indicates the first The predicted value of the adaptive quantity at time; Indicates the forgetting factor; Indicates the first The predicted value of the adaptive quantity at time; Indicates the first The monitored value of the adaptive quantity at any given time; Indicates the first The basic monitoring value at any given time; Indicates the first The basic monitoring value at any given time; Indicates the number of historical monitoring values; This represents the prediction function.

4. The adaptive acquisition method for multimodal information of power equipment according to claim 3, characterized in that, The prediction function includes a multiple linear regression function, the expression of which is as follows: ； In the formula, This represents the weighting coefficient of the monitored value of the i-th adaptive quantity; Indicates the first The monitored value of the adaptive quantity at any given time; This represents the weighting coefficient of the monitoring value of the j-th basic quantity; Indicates the first Monitoring values ​​of basic quantities at any given time; This indicates the bias term.

5. The adaptive acquisition method for multimodal information of power equipment according to claim 1, characterized in that, Obtaining the optimal acquisition frequency adjustment amount required at the current moment includes the following steps: Based on the predicted value sequence and combined with the monitoring data, calculate the prediction deviation sequence of the adaptive quantity within the preset time domain; Based on the predicted deviation sequence and a preset adaptive frequency adjustment strategy, the initial acquisition frequency adjustment amount required at the current moment is calculated. If the initial acquisition frequency adjustment amount is greater than or equal to a preset minimum acquisition frequency adjustment threshold and less than or equal to a preset maximum acquisition frequency adjustment threshold, then the initial acquisition frequency adjustment amount is taken as the optimal acquisition frequency adjustment amount required at the current moment. If the initial acquisition frequency adjustment amount is less than the preset minimum acquisition frequency adjustment threshold, then the minimum acquisition frequency adjustment threshold is taken as the optimal acquisition frequency adjustment amount required at the current moment. If the initial acquisition frequency adjustment amount is greater than the preset maximum acquisition frequency adjustment threshold, then the maximum acquisition frequency adjustment threshold is taken as the optimal acquisition frequency adjustment amount required at the current moment.

6. The adaptive acquisition method for multimodal information of power equipment according to claim 5, characterized in that, The adaptive frequency adjustment strategy includes a model predictive control method, which uses minimizing the accumulation of the prediction deviation sequence as the objective function to calculate the initial acquisition frequency adjustment amount required at the current moment. The objective function is expressed as follows: ； In the formula, Represent the objective function; Indicates the first The predicted deviation between the adaptive quantity and the monitored quantity at time t is expressed as: , Indicates the first The predicted value of the adaptive quantity at time; Indicates the first The monitored value of the adaptive quantity at any given time; Indicates the sequence number for the preset time. Indicates the range of the preset time domain; Indicates the first The adaptive sampling frequency adjustment value at any given time; Indicates the sequential index for controlling the time domain; Indicates the range of control over the time domain, and ; Represents the L2 norm; The weight matrix represents the adjustment for prediction bias; This represents the weight matrix used to adjust the sampling frequency.

7. The adaptive acquisition method for multimodal information of power equipment according to claim 5, characterized in that, The adaptive frequency adjustment strategy includes a proportional-integral (PI) control method. This PI control method calculates the required sampling frequency adjustment at the current moment based on a preset proportional-integral-derivative (PID) controller, as shown in the following expression: ； In the formula, Indicates the first The amount of time-adaptive measurement frequency adjustment; Indicates the proportionality coefficient; Indicates the first The predicted value of the adaptive quantity at any given time is the predicted deviation from the monitored value. Indicates the integral coefficient; Indicates the first The predicted value of the adaptive quantity at any given time is the predicted deviation from the monitored value. Represents the differential coefficient; Indicates the first The predicted value of the adaptive quantity at time point is the deviation from the predicted value of the monitored quantity.

8. The adaptive acquisition method for multimodal information of power equipment according to claim 1, characterized in that, Obtaining the optimal acquisition frequency for the adaptive quantity at the next moment includes the following steps: Based on the optimal acquisition frequency adjustment required at the current moment, combined with the acquisition frequency of the adaptive quantity at the current moment, and based on the smooth update strategy, the preliminary acquisition frequency of the adaptive quantity at the next moment is calculated. If the initial acquisition frequency is greater than or equal to a preset minimum acquisition frequency threshold and less than or equal to a preset maximum acquisition frequency threshold, then the initial acquisition frequency is taken as the optimal acquisition frequency for the adaptive quantity at the next moment. If the initial acquisition frequency is less than a preset minimum acquisition frequency threshold, then the minimum acquisition frequency threshold is used as the optimal acquisition frequency for the adaptive quantity at the next moment. If the initial acquisition frequency is greater than the preset maximum acquisition frequency threshold, then the maximum acquisition frequency threshold is used as the optimal acquisition frequency for the adaptive quantity at the next moment.

9. The adaptive acquisition method for multimodal information of power equipment according to claim 8, characterized in that, The calculation steps of the smooth update strategy are as follows: ； In the formula, Indicates the first The sampling frequency of the adaptive quantity at any given time; Indicates the smoothing coefficient; Indicates the first The amount of time-adaptive measurement frequency adjustment; Indicates the first The sampling frequency of the adaptive quantity at any given time.

10. A multimodal information adaptive acquisition device for power equipment, applicable to the multimodal information adaptive acquisition method for power equipment as described in any one of claims 1-9, characterized in that, include: The data acquisition module is used to acquire monitoring data of power equipment, including monitoring data of basic quantities and monitoring data of adaptive quantities; The state prediction module is used to predict the state based on the monitoring data and a preset prediction model to obtain a sequence of predicted values ​​of the adaptive quantity in a preset time domain. The frequency adjustment module is used to calculate the optimal acquisition frequency adjustment amount required at the current moment based on the predicted value sequence and a preset adaptive frequency adjustment strategy. The adaptive acquisition module is used to adjust the optimal acquisition frequency according to the current time, combine the current time's adaptive acquisition frequency to obtain the next time's optimal acquisition frequency, and update the adaptive acquisition frequency setting to achieve adaptive dynamic frequency acquisition of the adaptive quantity.

Images