Intelligent adaptive multi-channel current time division multiplexing detection method and system

Abstract

The invention relates to the technical field of current acquisition, and provides an intelligent adaptive multi-channel current time division multiplexing detection method and system. The method comprises the following steps: acquiring historical sampling data; calculating a current change slope according to the historical sampling data; determining a sampling priority according to the current change slope; generating a non-uniform sampling sequence of the next sampling period according to the sampling priority; and controlling the plurality of sampling channels to collect current information according to the non-uniform sampling sequence in the next sampling period. The current change trend of each channel is evaluated according to the historical sampling data, the sampling sequence of the next sampling period is dynamically adjusted according to the current change trend, and limited sampling resources are intelligently allocated to the channels with violent current change, so that the sampling efficiency is improved on the premise that the hardware cost is not increased. The capturing capability and the monitoring precision of the transient current change of the key channel are effectively improved, and the problem that important fault information may be lost due to the fixed sampling rate in a traditional time division multiplexing scheme is solved.

Description

Technical Field

[0001] This application relates to the field of current acquisition technology, and in particular to an intelligent adaptive multi-channel current time-division multiplexing detection method and system. Background Technology

[0002] In the current computing power economy era centered on large AI models (such as the GPT series) and cloud computing, the interconnection speed within data centers is experiencing explosive growth. 800G and 1.6T fiber optic communication modules have gradually become standard configurations for server clusters. These modules integrate multiple laser drivers and receiver control circuits, and each channel requires real-time current sampling to maintain dynamic balance of optical power and ensure equipment safety.

[0003] In traditional analog circuit design, multi-channel current sensing typically employs a "one-to-one" parallel architecture, where each current sampling point is configured with an independent sampling resistor, precision operational amplifier, and analog-to-digital converter (ADC). While this architecture enables synchronous sampling, it reveals serious drawbacks in high-performance scenarios. First, the hardware overhead is enormous. In 8-channel or 16-channel optical modules, repetitive analog front-end circuitry occupies valuable chip area, and the power consumption of high-level ADCs increases linearly with multiple parallel connections, which is unacceptable for optical modules with extremely demanding heat dissipation environments.

[0004] To address these issues, related technologies have proposed a time-division multiplexing-based detection scheme, which uses analog switches to alternately connect multiple signals to the same amplifier and ADC. However, current time-division multiplexing schemes typically employ a fixed-slot polling mechanism. This mechanism is highly susceptible to aliasing when dealing with high-speed transient currents (such as current steps caused by data bursts) due to uneven sampling rate distribution, resulting in the loss of crucial fault traces. Summary of the Invention

[0005] In order to at least partially solve the above-mentioned technical problems in the related technologies, this application provides an intelligent adaptive multi-channel current time-division multiplexing detection method and system.

[0006] On the one hand, this application provides an intelligent adaptive multi-channel current time-division multiplexing detection method, which adopts the following technical solution: A smart adaptive multi-channel current time-division multiplexing detection method includes the following steps: S1. Obtain historical sampling data from multiple sampling channels.

[0007] S2. Calculate the slope of the current change in the sampling channel based on the historical sampling data.

[0008] S3. Determine the sampling priority of the sampling channel based on the slope of the current change.

[0009] S4. Generate a non-uniform sampling sequence for the next sampling period according to the sampling priority. The non-uniform sampling sequence includes multiple channel identifiers arranged in order. The channel identifiers are used to identify the sampling channels, wherein the channel identifiers corresponding to higher sampling priorities appear more often than the channel identifiers corresponding to lower sampling priorities.

[0010] S5. In the next sampling period, control multiple sampling channels to collect current information according to the non-uniform sampling sequence.

[0011] By adopting the above technical solution, the current change trend of each channel is evaluated based on historical sampling data, and the sampling sequence of the next sampling period is dynamically adjusted accordingly. Limited sampling resources are intelligently allocated to channels with drastic current changes, thereby effectively improving the ability to capture and monitor transient current changes in key channels without increasing hardware costs. This solves the problem that important fault information may be lost due to the fixed sampling rate in traditional time-division multiplexing schemes.

[0012] Optionally, in step S1, the N sampling channels are sequentially polled for sampling, wherein the sampling period T is... cycle The process is divided into M time slices, and each channel is allocated M / N time slices for sampling.

[0013] By adopting the above technical solution, a uniform polling sampling method is used in the initial stage or when there is no specific priority information, which provides reliable and unbiased benchmark data for subsequent slope calculation and priority determination, ensuring the stability and accuracy of the adaptive algorithm startup.

[0014] Optionally, in step S2, for sampling channel i, the last sampled value I within the current sampling period k is compared. i (k end The last sampled value I within the previous sampling period k-1 i (k-1 end ), calculate the slope of the current change corresponding to the sampling channel i. i Slope is expressed as the following formula: i =|I i (k end )-I i (k-1 end )| / T cycle Alternatively, the slope of the current change can be evaluated by calculating the linear regression slope of multiple sampling points within the sampling period.

[0015] By adopting the above technical solution, a slope evaluation method with low computational complexity and effective reflection of current change trend is provided by comparing the last sampled values ​​between different sampling periods or by using multi-point linear regression within the period. This enables the system to quickly and efficiently determine the activity level of each channel, laying the foundation for real-time priority adjustment.

[0016] Optionally, in step S3, the slope threshold is compared with the calculated current change slope. i A comparison is made to determine the sampling priority of the sampling channel, and the slope threshold includes a low slope threshold Th. low and high slope threshold Th high The sampling priority is positively correlated with the slope of the current change; wherein, when the slope i >Th high When Th..., the sampling channel is divided into a high-priority channel; when Th... low <Slope i <=Th high When the sampling channel is divided into a medium-priority channel, the sampling channel is divided into a medium-priority channel; when the slope i <=Th low When this happens, the sampling channel is divided into a low-priority channel.

[0017] By adopting the above technical solution, a clear and quantitative priority determination mechanism is achieved by setting multiple slope thresholds to prioritize channels. This mechanism can effectively distinguish channels with different current changes, thus providing a clear basis for subsequent differentiated allocation of sampling resources.

[0018] Optionally, step S4 includes: allocating sampling time slices to the sampling channel according to the sampling priority, wherein the number of sampling time slices corresponds to the number of times the channel identifier appears in the non-uniform sampling sequence; and filling the non-uniform sampling sequence with the channel identifier in an interleaved distribution manner.

[0019] By adopting the above technical solution, by allocating different numbers of time slices to channels with different priorities and arranging the sampling points of high-priority channels in an interleaved manner in the sampling sequence, it is ensured that the monitoring of channels with drastic current changes is not only improved in frequency, but also that the real-time response is guaranteed, thus avoiding missing key changes at other moments in the cycle due to the concentration of sampling points.

[0020] Optionally, steps S1 to S5 can be executed repeatedly.

[0021] By adopting the above technical solution, a closed-loop control system is constructed by cyclically executing the entire adaptive sampling process. This enables the sampling strategy to continuously and dynamically track the load changes of each channel, keeping the system's detection resources focused on the channels that require the most attention, thus achieving a dynamic balance between high efficiency and high precision.

[0022] On the other hand, this application also provides an intelligent adaptive multi-channel current time-division multiplexing detection system, which adopts the following technical solution: An intelligent adaptive multi-channel current time-division multiplexing detection system includes: Multiple sampling channels, each of which includes a current sampling resistor and a current sampling signal switch. The first end of the current sampling resistor is used to connect to a functional module to obtain a current sampling voltage, and the current sampling signal switch is connected to the second end of the current sampling resistor. Operational amplifier module is used to receive the corresponding current sampling voltage in a time-division multiplexing manner through the current sampling signal switch, and convert the current sampling voltage into an amplified voltage signal; A microcontroller, connected to the operational amplifier module, is used to calculate the current information of the functional module based on the amplified voltage signal; The control module includes a sampling unit, a slope calculation unit, a priority calculation unit, a sequence generation unit, and a channel switching unit. The sampling unit is connected to a microcontroller and is used to acquire the current information as historical sampling data for multiple sampling channels. The slope calculation unit is connected to the sampling unit and is used to calculate the current change slope of the sampling channel based on the historical sampling data. The priority calculation unit is connected to the slope calculation unit and is used to determine the sampling priority of the sampling channel based on the current change slope. The sequence generation unit is connected to the priority calculation unit and is used to generate a non-uniform sampling sequence for the next sampling period based on the sampling priority. The non-uniform sampling sequence includes multiple sequentially arranged channel identifiers, which are used to identify the sampling channel. The channel identifier corresponding to a higher sampling priority appears more frequently than the channel identifier corresponding to a lower sampling priority. The channel switching unit is connected to the sequence generation unit and the current sampling signal switch and is used to switch the current sampling signal switch according to the non-uniform sampling sequence in the next sampling period to control the multiple sampling channels to acquire current information.

[0023] By adopting the above technical solution, the functions of sampling, slope calculation, priority determination, sequence generation and channel switching are integrated into the control module, and work together with the microcontroller, operational amplifier module and multiple sampling channels to build a complete hardware implementation architecture. This architecture enables the implementation of the adaptive sampling method and achieves efficient, accurate and highly consistent detection of multiple current signals at a low hardware cost.

[0024] Optionally, the sampling unit is configured to sequentially poll the N sampling channels, wherein the sampling period T is... cycle The sampling is divided into M time slices, and each channel is allocated M / N time slices for sampling. The slope calculation unit is configured to compare the last sampled value I of sampling channel i within the current sampling period k. i (k end The last sampled value I within the previous sampling period k-1 i (k-1 end ), calculate the slope of the current change corresponding to the sampling channel i. i Slope is expressed as the following formula: i =|I i (k end )-I i (k-1 end )| / T cycle Alternatively, the slope of the current change can be evaluated by calculating the linear regression slope of multiple sampling points within the sampling period.

[0025] By adopting the above technical solution and configuring specific execution methods for the sampling unit and slope calculation unit, the implementation path for the system to acquire reference data and evaluate current change trends is clarified, ensuring that the system can stably start the adaptive process and efficiently produce key parameters for priority determination.

[0026] Optionally, the priority calculation unit is configured to compare the slope threshold with the calculated current change slope. i A comparison is made to determine the sampling priority of the sampling channel, and the slope threshold includes a low slope threshold Th. low and high slope threshold Th high The sampling priority is positively correlated with the slope of the current change; wherein, when the slope i >Th high When Th..., the sampling channel is divided into a high-priority channel; when Th... low <Slope i <=Th high When the sampling channel is divided into a medium-priority channel, the sampling channel is divided into a medium-priority channel; when the slope i <=Th low When this happens, the sampling channel is divided into a low-priority channel.

[0027] By adopting the above technical solution, and configuring the priority calculation unit with comparison logic based on multi-level thresholds, a structured and quantifiable priority classification standard is provided for the system, making the system's division of channels with different activity levels clearer and more reliable, thereby enhancing the accuracy of subsequent resource allocation.

[0028] Optionally, the sequence generation unit is configured to allocate sampling time slices to the sampling channels according to the sampling priority, the number of sampling time slices corresponding to the number of times the channel identifier appears in the non-uniform sampling sequence; and to fill the non-uniform sampling sequence with the channel identifier in an interleaved distribution manner.

[0029] By adopting the above technical solution and configuring the sequence generation unit with an interleaved distribution-based arrangement strategy, it is ensured that the generated non-uniform sampling sequence can not only meet the sampling number requirements of the high-priority channel, but also make its sampling points uniformly distributed throughout the entire period, thereby maximizing the system's ability to capture high-speed transient signals and its real-time response speed.

[0030] In summary, this application includes at least one of the following beneficial technical effects: 1. By evaluating the current change trend of each channel based on historical sampling data and dynamically adjusting the sampling sequence of the next sampling period accordingly, the limited sampling resources are intelligently allocated to channels with drastic current changes. This effectively improves the ability to capture and monitor transient current changes in key channels without increasing hardware costs, and solves the problem that important fault information may be lost due to the fixed sampling rate in traditional time-division multiplexing schemes.

[0031] 2. By setting multiple slope thresholds to prioritize channels, a clear and quantitative priority determination mechanism is achieved, which can effectively distinguish channels with different current changes, thus providing a clear basis for subsequent differentiated allocation of sampling resources.

[0032] 3. By allocating different numbers of time slices to channels with different priorities and arranging the sampling points of high-priority channels in an interleaved manner in the sampling sequence, the monitoring of channels with drastic current changes is not only improved in frequency, but also guaranteed in real-time response, avoiding missing key changes at other moments in the cycle due to the concentration of sampling points.

[0033] 4. By cyclically executing the entire adaptive sampling process, a closed-loop control system is constructed, enabling the sampling strategy to continuously and dynamically track the load changes of each channel, and always concentrate the system's detection resources on the channels that require the most attention, thus achieving a dynamic balance between high efficiency and high precision. Attached Figure Description

[0034] Figure 1 This paper illustrates a circuit diagram of an intelligent adaptive multi-channel current time-division multiplexing detection system according to an embodiment of this application. Figure 2 A schematic diagram of the control module of an intelligent adaptive multi-channel current time-division multiplexing detection system according to an embodiment of this application is shown. Figure 3 A flowchart illustrating an intelligent adaptive multi-channel current time-division multiplexing detection method according to an embodiment of this application is shown.

[0035] Explanation of reference numerals in the attached figures: 10, Control module; 11, Sampling unit; 12, Slope calculation unit; 13, Priority calculation unit; 14, Sequence generation unit; 15, Channel switching unit; 20, Operational amplifier module; 30, Microcontroller. Detailed Implementation

[0036] The following combination Figures 1-3 This application will be described in further detail.

[0037] This application discloses an intelligent adaptive multi-channel current time-division multiplexing detection system.

[0038] Figure 1 A circuit diagram of an intelligent adaptive multi-channel current time-division multiplexing detection system according to an embodiment of this application is shown. (Refer to...) Figure 1 The system includes multiple sampling channels, a control module 10, an operational amplifier module 20, and a microcontroller 30. Each sampling channel consists of current sampling resistors (Rs1-Rsn), main current loop switches (Sa1-San), and current sampling signal switches (Sb1-Sbn). The operational amplifier module 20 includes a differential operational amplifier U1 and operational amplifier module resistors (including input resistor Ra, feedback resistor Rf, and output resistor Rs). The microcontroller 30 has an analog-to-digital converter (ADC) and a microcontroller unit (MCU).

[0039] Multiple current sampling resistors for each sampling channel are connected in series with each functional module. Multiple main current loop switches and multiple current sampling signal switches correspond one-to-one with the multiple current sampling resistors. The main current loop switches are positioned between the current sampling resistors and the functional modules, and the current sampling signal switches are positioned between the current sampling resistors and the operational amplifier module 20. The main current loop switches are connected to the enable signal terminal EN (not shown in the figure), and the multiple current sampling signal switches are connected to the control module 10. The operational amplifier module 20 is connected to the microcontroller 30. The enable signal from the enable signal terminal EN controls the on / off state of the main current loop switches, and the control module 10 controls the on / off state of the current sampling signal switches. This sequentially switches the voltage signals from the current sampling resistors to the operational amplifier module 20. The analog-to-digital converter (ADC) acquires the digital signals from each channel in a time-division multiplexing manner. This allows one operational amplifier module 20 and one ADC to detect the current of multiple functional modules, avoiding the cost and matching discrepancy issues associated with using multiple operational amplifier modules 20 and multiple ADCs.

[0040] The main current loop switches and current sampling signal switches can be transistor switches, such as MOSFET switches, or relay switches can be used as alternatives. Each main current loop switch connects a current sampling resistor to its corresponding functional module. After power-on at the power input terminal VIN and the bias voltage terminal VBAIS, when the enable signal is low (EN=0V), all main current loop switches are in the off state, and the output voltage Vs=0V. When the enable signal is high (EN=high), all main current loop switches are in the on state, and all functional modules are powered on and ready to operate, forming current sampling voltages on the multiple current sampling resistors. For example, a first sampling voltage Vrs1 is formed on the first current sampling resistor Rs1. The current sampling signal switch connects the current sampling resistor to the differential operational amplifier U1 of the operational amplifier module 20.

[0041] Figure 2 This diagram illustrates a control module of an intelligent adaptive multi-channel current time-division multiplexing detection system according to an embodiment of this application. (Refer to...) Figure 2 The control module 10 includes a sampling unit 11, a slope calculation unit 12, a priority calculation unit 13, a sequence generation unit 14, and a channel switching unit 15. This control module 10 can employ a Complex Programmable Logic Device (CPLD) controller or a Field Programmable Gate Array (FPGA) controller. These controllers are well-suited for high-speed, precise timing control tasks, ensuring high-speed and precise switching of the sampling channels.

[0042] The sampling unit 11 is connected to the microcontroller 30 and is used to acquire current information as historical sampling data for multiple sampling channels. In some embodiments, the sampling unit 11 is configured to perform sequential polling sampling on N sampling channels, wherein the sampling period T is... cycle The process is divided into M time slices, and each channel is allocated M / N time slices for sampling.

[0043] The slope calculation unit 12 is connected to the sampling unit 11 and is used to calculate the slope of the current change of the sampling channel based on historical sampling data. In some embodiments, the slope calculation unit 12 is configured to compare the last sampled value I of sampling channel i within the current sampling period k. i (k end The last sampled value I within the previous sampling period k-1 i (k-1 end ), calculate the slope of the current change corresponding to sampling channel i. i Slope is expressed as the following formula: i =|I i (k end )-I i (k-1 end )| / T cycle Alternatively, the slope of the current change can be assessed by calculating the linear regression slope of multiple sampling points within the sampling period.

[0044] Priority calculation unit 13 is connected to slope calculation unit 12 and is used to determine the sampling priority of the sampling channel based on the current change slope. In some embodiments, priority calculation unit 13 is configured to compare a slope threshold with the calculated current change slope. i A comparison is made to determine the sampling priority of the sampling channel, and the slope threshold includes a low slope threshold Th. low and high slope threshold Th high The sampling priority is positively correlated with the slope of the current change; where, when the slope i >Th high When Th is in use, the sampling channel is divided into high-priority channels; when Th is in use, the sampling channel is divided into high-priority channels. low <Slope i <=Th high When the slope is high, the sampling channel is divided into a medium-priority channel; when the slope is low, the sampling channel is divided into a medium-priority channel. i <=Th low At that time, the sampling channels are divided into low-priority channels.

[0045] The sequence generation unit 14 is connected to the priority calculation unit 13 and is used to generate a non-uniform sampling sequence for the next sampling period according to the sampling priority. The non-uniform sampling sequence includes multiple channel identifiers arranged in order. The channel identifiers are used to identify the sampling channels, wherein the channel identifiers corresponding to higher sampling priorities appear more often than the channel identifiers corresponding to lower sampling priorities. In some embodiments, the sequence generation unit 14 is configured to allocate sampling time slices to the sampling channels according to the sampling priority, and the number of sampling time slices corresponds to the number of times the channel identifiers appear in the non-uniform sampling sequence; the channel identifiers are filled into the non-uniform sampling sequence in an interleaved distribution manner.

[0046] The channel switching unit 15 is connected to the sequence generation unit 14 and the current sampling signal switch, and is used to switch the current sampling signal switch according to the non-uniform sampling sequence in the next sampling period to control multiple sampling channels to collect current information.

[0047] Operational amplifier module 20 is used to amplify the sampled voltage to generate an amplified voltage signal. The differential operational amplifier U1 is connected to the bias voltage terminal VBAIS. The inverting input (-) of the differential operational amplifier U1 is connected to the corresponding current sampling resistor through multiple current sampling signal switches. The non-inverting input (+) of the differential operational amplifier U1 is connected to ground through the input resistor Ra. The output terminal of the differential operational amplifier U1 is connected to the first end of the output resistor Rs, the second end of the output resistor Rs is connected to ground, and the output terminal of the differential operational amplifier U1 is connected to the non-inverting input (+) of the differential operational amplifier U1 through the feedback resistor Rf. Taking the first current sampling signal switch Sb1 being turned on as an example, the first sampled voltage Vrs1 is amplified by A times (A=Rf / Ra) by the differential operational amplifier U1, forming an amplified voltage signal Vs (Vs=A*Vrs1) on the output resistor Rs.

[0048] The microcontroller 30 is connected to the operational amplifier module 20 and is used to calculate current information based on the amplified voltage signal and execute corresponding operations. The analog-to-digital converter (ADC) converts the amplified voltage signal Vs into a digital signal. The MCU's operational unit performs relevant logic operations based on the digital signal to calculate the current information and execute overcurrent protection and other actions related to the functional modules. This digital signal is also acquired by the sampling unit 11 of the control module 10 to serve as historical sampling data for the sampling channel in the next sampling cycle.

[0049] The implementation principle of the intelligent adaptive multi-channel current time-division multiplexing detection system in this application embodiment is as follows: by integrating sampling, slope calculation, priority determination, sequence generation and channel switching functions into the control module 10, and working in conjunction with the microcontroller 30, operational amplifier module 20 and multiple sampling channels, a complete hardware implementation architecture is constructed, which can implement the adaptive sampling method and achieve efficient, accurate and highly consistent detection of multiple current signals with low hardware cost.

[0050] This application also discloses an intelligent adaptive multi-channel current time-division multiplexing detection method. This method is mainly implemented through the control module of the intelligent adaptive multi-channel current time-division multiplexing detection system described in the above embodiments. The traditional static polling logic is replaced by an adaptive scheduling algorithm based on current slope prediction. By monitoring the changing trends of each current channel in real time, the time slice allocation is dynamically adjusted to ensure that active channels receive high-frequency sampling, while silent channels perform down-frequency scanning. This achieves equivalent multi-channel parallel tracking capability under a single ADC architecture.

[0051] Figure 3 A flowchart illustrating an intelligent adaptive multi-channel current time-division multiplexing detection method according to an embodiment of this application is shown. (Refer to...) Figure 3 The method includes the following steps: S1. Obtain historical sampling data from multiple sampling channels.

[0052] S2. Calculate the slope of the current change in the sampling channel based on historical sampling data.

[0053] S3. Determine the sampling priority of the sampling channel based on the slope of the current change.

[0054] S4. Generate a non-uniform sampling sequence for the next sampling period based on the sampling priority. The non-uniform sampling sequence includes multiple channel identifiers arranged in order. The channel identifiers are used to identify the sampling channels. The channel identifiers corresponding to higher sampling priorities appear more often than the channel identifiers corresponding to lower sampling priorities.

[0055] S5. In the next sampling period, control multiple sampling channels to collect current information according to the non-uniform sampling sequence.

[0056] The steps of this method are explained in detail below.

[0057] In step S1, historical sampling data from multiple sampling channels are acquired. Within one sampling period, the current sampling signal switches of multiple sampling channels are sequentially controlled to connect the current sampling voltages of multiple functional modules to the operational amplifier module 20 in a time-division manner. The operational amplifier module 20 converts the current sampling voltages into amplified voltage signals, and the microcontroller 30 calculates the current information of the functional modules based on the amplified voltage signals. This current information is the current sampling value of the corresponding sampling channel. Initially, the sampling process adopts a uniform sampling strategy, sequentially polling and sampling all N sampling channels. For example, the sampling period T... cycle The process is divided into M time slices, and each channel is allocated M / N time slices for sampling. This step allows us to obtain the initial current values ​​of all sampled channels, providing a benchmark for subsequent current change slope calculations.

[0058] In step S2, the current change slope of the sampling channel is calculated based on historical sampling data. Specifically, after the reference sampling period or any sampling period ends, for each sampling channel i, the current change slope is calculated by comparing it with the last sampled value I within the current sampling period k. i (k end The last sampled value I within the previous sampling period k-1 i (k-1 end ), calculate the slope of the current change corresponding to sampling channel i. i Slope is expressed as the following formula: i =|I i (k end )-I i (k-1 end )| / T cycle Alternatively, the slope of the current change can be evaluated by calculating the slope of a linear regression at multiple sampling points within a sampling period. For example, the slope of the regression line obtained by least-squares fitting of multiple data points collected within a period can be used as the evaluation value.

[0059] In step S3, the sampling priority of the sampling channel is determined based on the slope of the current change. This involves setting or adaptively adjusting multiple slope thresholds (e.g., a low slope threshold Th). low and high slope threshold Th high (and the calculated slope of the current change) i The sampling channels are compared and prioritized according to their sampling priority, which is positively correlated with the slope of the current change. For example, when the slope... i >Th high When Th..., the sampling channel is classified as a high-priority channel, indicating that the current in that sampling channel has changed drastically and requires close monitoring. low <Slope i <=Thhigh When the sampling channel is divided into a medium-priority channel, it indicates that the current of that sampling channel fluctuates to some extent. When the slope... i <=Th low When the sampling channel is divided into a low-priority channel, it indicates that the current state of the sampling channel is stable.

[0060] In step S4, a non-uniform sampling sequence for the next sampling period is generated based on the sampling priority. This non-uniform sampling sequence includes multiple sequentially arranged channel identifiers, configured to identify sampling channels. The channel identifiers corresponding to higher sampling priorities appear more frequently than those corresponding to lower sampling priorities. Specifically, based on the priority evaluation results of the sampling channels, a non-uniform sampling sequence is dynamically generated for the next sampling period k+1. This non-uniform sampling sequence is a one-dimensional array or list of length M (i.e., one large period contains M time slices), where each element is a channel identifier to be sampled. First, different numbers of time slices are allocated to the sampling channels according to their priorities. The number of sampling time slices corresponds to the number of times the channel identifier appears in the non-uniform sampling sequence. For example, N is allocated to each high-priority channel. high N sampling time slices are allocated to each medium-priority channel. mid N are allocated to each low-priority channel. low There are N, of which N high >N mid >N lowThe total number of time slices allocated to all channels should be equal to M. The allocation weights for time slices can be a fixed mapping table or a continuous function positively correlated with the slope value. For channel identifiers with allocated time slices, they are filled into a non-uniform sampling sequence of length M according to specific rules. To ensure real-time monitoring of high-priority channels, the arrangement rule preferably adopts an interleaved distribution rather than concentrating sampling time slices of the same channel together. Compared to concentrating sampling points of the same channel, interleaved distribution avoids long-term monitoring blind spots within the sampling period, thereby maximizing the real-time response capability to transient events. In some embodiments, the arrangement algorithm uses a weighted round-robin algorithm. The sampling frequency of each channel is used as a weight. In each round of selection, the channel with the highest weight is selected first for sampling, and its weight decreases accordingly after sampling until the sampling frequency of all channels has been filled. This ensures that high-priority channels are accessed uniformly and frequently throughout the entire sampling period. For example, consider a system with four sampling channels Ch1-Ch4, and 16 time slices in one sampling period. Calculations show that Ch2 has high priority and is allocated 8 time slices; Ch1 has medium priority and is allocated 4; and Ch3 and Ch4 have low priority and are each allocated 2. The generated non-uniform sampling sequence can be [2, 1, 3, 2, 4, 1, 2, 1, 2, 3, 1, 2, 4, 2, 1, 2]. In this sequence, Ch2 has the highest frequency of occurrence, and its sampling points are evenly distributed throughout the entire period, thus enabling timely capture of rapid changes in current.

[0061] In step S5, multiple sampling channels are controlled to acquire current information according to the non-uniform sampling sequence in the next sampling period. Specifically, at the beginning of the next sampling period k+1, the current sampling signal switches of multiple sampling channels are controlled to connect the current sampling voltages of multiple functional modules to the operational amplifier module 20 in a time-division manner, according to the channel order and time slice in the non-uniform sampling sequence. The operational amplifier module 20 converts the current sampling voltages into amplified voltage signals, and the microcontroller 30 calculates the current information of the functional modules based on the amplified voltage signals. After one sampling period ends, steps S1-S5 can be repeated to form a closed-loop adaptive control, enabling the sampling strategy to continuously and dynamically adapt to changes in the load of each channel, concentrating limited system detection resources on the channels requiring the most attention, thus achieving a balance between high efficiency and high accuracy.

[0062] The implementation principle of the intelligent adaptive multi-channel current time-division multiplexing detection method in this application is as follows: by evaluating the current change trend of each channel based on historical sampling data, and dynamically adjusting the sampling sequence of the next sampling period accordingly, the limited sampling resources are intelligently allocated to the channels with drastic current changes. Thus, without increasing hardware costs, the ability to capture transient current changes in key channels and the monitoring accuracy are effectively improved, solving the problem that important fault information may be lost due to the fixed sampling rate in traditional time-division multiplexing schemes.

[0063] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A smart adaptive multi-channel current time-division multiplexing detection method, characterized in that, Includes the following steps: S1. Obtain historical sampling data from multiple sampling channels; S2. Calculate the slope of the current change in the sampling channel based on the historical sampling data; S3. Determine the sampling priority of the sampling channel based on the slope of the current change; S4. Generate a non-uniform sampling sequence for the next sampling period according to the sampling priority. The non-uniform sampling sequence includes multiple channel identifiers arranged in order. The channel identifiers are used to identify the sampling channels, wherein the channel identifiers corresponding to higher sampling priorities appear more often than the channel identifiers corresponding to lower sampling priorities. S5. In the next sampling period, control multiple sampling channels to collect current information according to the non-uniform sampling sequence.

2. The intelligent adaptive multi-channel current time-division multiplexing detection method according to claim 1, characterized in that, In step S1, sequential polling sampling is performed on the N sampling channels, wherein the sampling period T is... cycle The process is divided into M time slices, and each channel is allocated M / N time slices for sampling.

3. The intelligent adaptive multi-channel current time-division multiplexing detection method according to claim 1, characterized in that, In step S2, for sampling channel i, the last sampled value I within the current sampling period k is compared. i (k end The last sampled value I within the previous sampling period k-1 i (k-1 end ), calculate the slope of the current change corresponding to the sampling channel i. i Slope is expressed as the following formula: i =|I i (k end )-I i (k-1 end )| / T cycle Alternatively, the slope of the current change can be evaluated by calculating the linear regression slope of multiple sampling points within the sampling period.

4. The intelligent adaptive multi-channel current time-division multiplexing detection method according to claim 3, characterized in that, In step S3, the slope threshold is compared with the calculated current change slope. i A comparison is made to determine the sampling priority of the sampling channel, and the slope threshold includes a low slope threshold Th. low and high slope threshold Th high The sampling priority is positively correlated with the slope of the current change; wherein, when the slope i >Th high When Th..., the sampling channel is divided into a high-priority channel; when Th... low <Slope i <=Th high When the sampling channel is divided into a medium-priority channel, the sampling channel is divided into a medium-priority channel; when the slope i <=Th low When this happens, the sampling channel is divided into a low-priority channel.

5. The intelligent adaptive multi-channel current time-division multiplexing detection method according to claim 4, characterized in that, Step S4 includes: allocating sampling time slices to the sampling channel according to the sampling priority, wherein the number of sampling time slices corresponds to the number of times the channel identifier appears in the non-uniform sampling sequence; and filling the non-uniform sampling sequence with the channel identifier in an interleaved distribution manner.

6. The intelligent adaptive multi-channel current time-division multiplexing detection method according to any one of claims 1-5, characterized in that, Repeat steps S1 to S5.

7. An intelligent adaptive multi-channel current time-division multiplexing detection system, characterized in that, include: Multiple sampling channels, each of which includes a current sampling resistor and a current sampling signal switch. The first end of the current sampling resistor is used to connect to a functional module to obtain a current sampling voltage, and the current sampling signal switch is connected to the second end of the current sampling resistor. Operational amplifier module (20) is used to receive the corresponding current sampling voltage in a time-division multiplexing manner through the current sampling signal switch, and convert the current sampling voltage into an amplified voltage signal; A microcontroller (30) is connected to the operational amplifier module (20) and is used to calculate the current information of the functional module based on the amplified voltage signal; The control module (10) includes a sampling unit (11), a slope calculation unit (12), a priority calculation unit (13), a sequence generation unit (14), and a channel switching unit (15). The sampling unit (11) is connected to the microcontroller (30) and is used to acquire the current information as historical sampling data for multiple sampling channels. The slope calculation unit (12) is connected to the sampling unit (11) and is used to calculate the current change slope of the sampling channel based on the historical sampling data. The priority calculation unit (13) is connected to the slope calculation unit (12) and is used to determine the sampling priority of the sampling channel based on the current change slope. The sequence generation unit (14) is connected to the microcontroller (30). (14) is connected to the priority calculation unit (13) and is used to generate a non-uniform sampling sequence for the next sampling period according to the sampling priority. The non-uniform sampling sequence includes multiple channel identifiers arranged in order. The channel identifiers are used to identify the sampling channels. The number of occurrences of the channel identifier corresponding to the higher sampling priority is greater than the number of occurrences of the channel identifier corresponding to the lower sampling priority. The channel switching unit (15) is connected to the sequence generation unit (14) and the current sampling signal switch and is used to switch the current sampling signal switch according to the non-uniform sampling sequence in the next sampling period to control the multiple sampling channels to collect current information.

8. The intelligent adaptive multi-channel current time-division multiplexing detection system according to claim 7, characterized in that, The sampling unit (11) is configured to sequentially poll the N sampling channels, wherein the sampling period T is... cycle The sampling is divided into M time slices, and each channel is allocated M / N time slices for sampling; the slope calculation unit (12) is configured to compare the last sampled value I of sampling channel i within the current sampling period k. i (k end The last sampled value I within the previous sampling period k-1 i (k-1 end ), calculate the slope of the current change corresponding to the sampling channel i. i Slope is expressed as the following formula: i =|I i (k end )-I i (k-1 end )| / T cycle Alternatively, the slope of the current change can be evaluated by calculating the linear regression slope of multiple sampling points within the sampling period.

9. The intelligent adaptive multi-channel current time-division multiplexing detection system according to claim 8, characterized in that, The priority calculation unit (13) is configured to combine the slope threshold with the calculated current change slope. i A comparison is made to determine the sampling priority of the sampling channel, and the slope threshold includes a low slope threshold Th. low and high slope threshold Th high The sampling priority is positively correlated with the slope of the current change; wherein, when the slope i >Th high When Th..., the sampling channel is divided into a high-priority channel; when Th... low <Slope i <=Th high When the sampling channel is divided into a medium-priority channel, the sampling channel is divided into a medium-priority channel; when the slope i <=Th low When this happens, the sampling channel is divided into a low-priority channel.

10. The intelligent adaptive multi-channel current time-division multiplexing detection system according to claim 9, characterized in that, The sequence generation unit (14) is configured to allocate sampling time slices to the sampling channel according to the sampling priority, wherein the number of sampling time slices corresponds to the number of times the channel identifier appears in the non-uniform sampling sequence; The channel identifiers are filled into the non-uniform sampling sequence according to an interleaved distribution method.

Images