A multi-source status data acquisition and processing system for AI control terminals

Abstract

This invention relates to the field of state data monitoring technology, specifically to an AI control terminal multi-source state data acquisition and processing system. The system includes a channel coupling detection module, a signal timing alignment module, a state trend construction module, a transition path locking module, and a state response result generation module. In this invention, the channel response trend is determined by the direction of impedance change, allowing for continuous division of passable and restricted channels. Combined with trigger timing alignment, a unified starting signal set is formed. Under time-consistent conditions, the channel change direction is extracted and collaborative change relationships are identified. A stable path structure is constructed based on the synchronous trend within a continuous period and correlated with the terminal task structure. This ensures a continuous connection between channel state evolution, trend direction, and control objectives, enhancing the collaborative adaptability of multi-source state data in three dimensions: spatial distribution, temporal process, and trend change. It also improves the directional aggregation of channel selection and the task fit of control response.

Description

Technical Field

[0001] This invention relates to the field of status data monitoring technology, and in particular to an AI control terminal multi-source status data acquisition and processing system. Background Technology

[0002] The field of state data monitoring technology involves the continuous observation and processing of various state parameters generated during the operation of systems or equipment. Its core aspects include real-time acquisition of state information, multi-source coordination of data content, identification of patterns in parameter changes, and analysis and judgment of state evolution processes. This technology typically involves deploying sensing devices to collect parameter signals such as voltage, current, temperature, and vibration frequency, combining this with preset timing control logic for data capture, relying on rule-based judgment mechanisms to conduct state screening and data integration, and using this to support subsequent state judgment and trend analysis. Among these, the traditional AI control terminal multi-source state data acquisition and processing system refers to a system based on artificial intelligence... This system is a data processing system that uses basic control methods to synchronously collect and uniformly process status parameters from multiple data sources. The technical issues it addresses are the inconsistent sampling frequencies, large differences in data formats, and asynchronous collection times of multi-source data. Traditional AI control terminal multi-source status data acquisition and processing systems typically collect parameters such as voltage, current, and temperature values ​​by connecting multiple sensing devices to the terminal, poll the data at a fixed period, initially remove abnormal data through numerical comparison, integrate parameter information from different sources in a time sequence using timestamp matching, and perform unified standardized processing on the data based on judgment rules.

[0003] Existing technologies rely on preset rules for data screening and integration. When faced with inconsistent sampling frequencies or misaligned trigger times, it is difficult to establish a dynamic identification mechanism for the response trend of input channels. The data polling method is limited by fixed cycles and single-point judgment, resulting in a lack of continuity in tracking the direction of state changes. In scenarios involving multi-channel collaborative response, problems such as unclear channel division or missing state trends may occur, leading to fuzzy path construction and inaccurate positioning of control commands. This can easily cause information mismatch and response deviations in the decision-making and execution process of the control terminal. Summary of the Invention

[0004] To address the technical problems existing in the prior art, this invention provides an AI control terminal multi-source status data acquisition and processing system.

[0005] On the one hand, an AI-controlled terminal multi-source status data acquisition and processing system is provided, which includes:

[0006] The channel coupling detection module acquires the status signal coupling interface input channel, records the impedance change direction within a continuous period, compares the change trend of adjacent periods, determines the channel status, outputs the passable and restricted status according to the number, and obtains the channel activation tag mapping table.

[0007] The signal timing alignment module extracts the trigger time point based on the active channel number in the channel activation tag mapping table, aligns the channel start time, and obtains a set of time synchronization channels.

[0008] The status trend composition module extracts the change direction of the same time node based on the time synchronization channel set, identifies the channel combination with the same direction, outputs the combination correspondence according to the synchronization situation, and obtains the trend channel combination directory structure.

[0009] The jump path locking module, based on the trend channel combination directory structure, determines the state direction within a continuous period, filters combinations with consistent directions, adds the channel number and trend direction to the path set, and obtains a trend stable channel path identifier set.

[0010] The status response result generation module extracts the target channel of the path based on the trend stable channel path identifier set and the corresponding task structure of the artificial intelligence control terminal, associates the number with the trend information, and obtains the multi-source status data response control result.

[0011] As a further embodiment of the present invention, the channel activation tag mapping table includes channel number, access status, and restricted status; the time synchronization channel set includes activation channel number, unified start time, and synchronization trigger time point; the trend channel combination directory structure includes channel combination, state change direction, consistency change relationship, and synchronization change status; the trend stable channel path identifier set includes path number, channel number, and trend direction information; and the multi-source state data response control result includes target channel number, trend direction, and task tag.

[0012] As a further aspect of the present invention, the determination of channel status refers to the process of distinguishing whether the channel is in a passable or restricted state by analyzing whether the direction of state change of the input channel remains consistent within a continuous period.

[0013] The direction of change at the same time point refers to the direction of increase or decrease in the channel status values ​​extracted at the same time point among all channels that have completed time alignment.

[0014] As a further aspect of the present invention, the direction-consistent channel combination refers to a set of cooperative change channels formed by merging multiple channels that exhibit the same direction of state change within a time node and a continuous time interval.

[0015] The combination correspondence refers to the association description of the synchronous changes of channels with the same direction in time, and the determination of the mapping relationship between channel combination, change direction and duration interval.

[0016] As a further aspect of the present invention, the channel coupling detection module includes:

[0017] The channel impedance acquisition submodule acquires the input channel of the status signal coupling interface, sequentially loads the channel according to the number order, acquires the impedance value sequence in multiple consecutive time periods, extracts the impedance reference value of adjacent time periods, and differentially calculates adjacent values ​​to obtain the impedance change trend sequence.

[0018] The change direction judgment submodule calls the impedance change trend sequence, identifies the change direction based on the positive and negative directions of the impedance difference for each channel time period, compares the continuous change directions point by point, analyzes the continuous change state between directions, and obtains a channel response consistency mark list.

[0019] The state mapping output submodule calls the channel response consistency tag list, and according to the directional continuity status of each channel, maps the number to the passage status to obtain the channel activation tag mapping table.

[0020] As a further aspect of the present invention, the signal timing alignment module includes:

[0021] The activation channel extraction submodule retrieves the original sampling signals of the channels in the current acquisition period based on the channel numbers in the channel activation tag mapping table that are accessible. It then indexes and retrieves the data sequence of each channel, activates the channel, and binds the time data to obtain the activation channel signal set.

[0022] The trigger time recognition submodule calls the activation channel signal set, detects the starting position of the signal sequence in the channel, collects the position where the signal first exceeds the trigger threshold, extracts the trigger time point and marks it with a number, compares the order of the channel trigger points according to the time axis, and obtains the trigger time sorting list.

[0023] The start time alignment submodule, based on the trigger time point ranked first in the trigger time sorting list, shifts the signal data time in the activation channel to synchronize the start time and reference time of the signal, thus obtaining a set of time synchronization channels.

[0024] As a further aspect of the present invention, the state trend construction module includes:

[0025] The state direction extraction submodule reads adjacent state values ​​sequentially from consecutive time nodes based on the state sequence of channels in the time synchronization channel set at a unified start time. According to the increase or decrease between the values, it writes the corresponding direction results in the order of channel number to obtain the channel state change direction set.

[0026] The channel direction filtering submodule calls the channel state change direction set, extracts the direction results for each channel at each time node, writes the channel numbers with the same direction at the current time point into the same index item, expands the written content across nodes, and obtains the channel direction synchronization sequence set.

[0027] The synchronous change tracking submodule scans the direction results of each channel group sequentially along the time node according to the distribution of the channel number in the channel direction synchronization sequence, marks the repeated number items between adjacent nodes, and outputs the time interval and the corresponding channel number at the same time to obtain the trend channel combination directory structure.

[0028] As a further aspect of the present invention, the jump path locking module includes:

[0029] The direction continuity recognition submodule reads the channel direction results within a continuous period based on the time series direction information of each group of channels in the trend channel combination directory structure, judges the continuity performance between channels based on the direction consistency between adjacent periods, records the continuous direction status in chronological order, and obtains the channel direction continuity sequence set.

[0030] The channel path writing submodule calls the channel direction continuation sequence set, scans the channel number and direction continuation, adds the channel number with direction continuation to the corresponding path position, and writes the path set entries in the combination order to obtain the trend channel path set.

[0031] The trend information appending submodule adds the direction information data of the channel to the corresponding path position based on the channel number index already included in the trend channel path set, and completes the association injection of channel number and direction to obtain the trend stable channel path identifier set.

[0032] As a further aspect of the present invention, during the direction continuity comparison process, when the direction information in the three cycles is consistent, the direction state of the continuation channel is extracted; when the direction is inconsistent, the extraction of direction information is restarted from the current cycle.

[0033] During the process of constructing the channel direction continuation sequence set: when the number of consecutive extractions of the channel direction state reaches a preset number standard, the channel number is placed into the corresponding path position, and the path writing is completed in the order of the trend channel combination directory structure.

[0034] As a further aspect of the present invention, the state response result generation module includes:

[0035] The path channel retrieval submodule scans the number information in each path based on the channel number sequence in the trend stable channel path identifier set, reads the path number and direction content, writes the channel number into the index table, and maps it according to the path affiliation to obtain the path channel mapping structure.

[0036] The channel task matching submodule calls the path channel mapping structure, extracts the task binding channel number item by item from the channel task binding structure of the artificial intelligence control terminal, identifies the number content that appears in both the path structure and the binding structure, and obtains the target channel association index table.

[0037] The task tag fusion submodule writes the corresponding task tag field according to the channel number in the target channel association index table, and introduces path direction content as additional information, and associates and binds the channel number, trend direction and task tag to obtain the multi-source status data response control result.

[0038] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0039] In this invention, the channel response trend is determined by the direction of impedance change, which can continuously divide the passable and restricted channels. Combined with the trigger timing alignment, a unified starting signal set is formed. Under the condition of consistent time, the channel change direction is extracted and the cooperative change relationship is identified. Based on the synchronous trend within the continuous period, a stable path structure is constructed and correspondingly associated with the terminal task structure, so that the channel state evolution, trend direction and control target are continuously connected. This enhances the cooperative adaptation capability of multi-source state data in three dimensions: spatial distribution, time process and trend change, and improves the directional aggregation of channel selection and the task fit of control response. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a system flowchart of the present invention;

[0042] Figure 2 This is a flowchart of the channel coupling detection module in this invention;

[0043] Figure 3 This is a flowchart of the signal timing alignment module in this invention;

[0044] Figure 4 This is a flowchart of the state trend construction module in this invention;

[0045] Figure 5 This is a flowchart of the jump path locking module in this invention;

[0046] Figure 6 This is a flowchart of the state response result generation module in this invention. Detailed Implementation

[0047] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0048] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0049] This invention provides a system for acquiring and processing multi-source status data of an AI control terminal, such as... Figure 1 The diagram shown illustrates a multi-source status data acquisition and processing system for an AI control terminal. This system includes:

[0050] The channel coupling detection module acquires the input channel of the status signal coupling interface, records the impedance change direction of the input channel in multiple consecutive time periods, compares the impedance change direction between adjacent time periods, determines the response trend of the input channel based on whether the change direction is consistent, classifies the input channel with the consistent change direction as passable, classifies the input channel with the reversed change direction as restricted, and outputs the passable and restricted states according to the input channel number to obtain the channel activation label mapping table.

[0051] The signal timing alignment module extracts the signal trigger time point within the current acquisition cycle of the channel based on the channel number activated in the channel activation tag mapping table, compares the trigger timing of the channel, selects the trigger time point of the channel with the earlier timing for alignment, unifies the start time of the channel signal to the time point, and synchronizes the time to obtain the time synchronization channel set;

[0052] The state trend composition module extracts the state change direction at the same time node based on the state change information of the channels in the time synchronization channel set at a unified start time, identifies the channel combination with the same change direction, tracks the synchronous change of the channel combination at consecutive time nodes, and outputs the channel combination and change relationship according to the synchronous change situation to obtain the trend channel combination directory structure.

[0053] The jump path locking module is based on the channel combination in the trend channel combination directory structure. It judges the continuity of the state change direction of the channel in multiple consecutive cycles, identifies the channel combination that maintains the same change direction in multiple cycles, fills the channel combination with the related trend into the path set, and attaches the corresponding channel number and trend direction information to the corresponding path to obtain the trend stable channel path identifier set.

[0054] The status response result generation module extracts the target channel group associated with the terminal control function based on the channel path identification information of the trend stable channel path identifier, corresponding to the channel task binding structure in the artificial intelligence control terminal. It then identifies the channels that appear in the trend path structure in the channel group and associates the channel identification number, trend direction information and task label to obtain the multi-source status data response control result.

[0055] The channel activation tag mapping table includes channel number, access status, and restricted status; the time synchronization channel set includes activation channel number, unified start time, and synchronization trigger time point; the trend channel combination directory structure includes channel combination, state change direction, consistency change relationship, and synchronization change status; the trend stable channel path identifier set includes path number, channel number, and trend direction information; and the multi-source state data response control result includes target channel number, trend direction, and task tag.

[0056] Specifically, such as Figure 2 As shown, the channel coupling detection module includes:

[0057] The channel impedance acquisition submodule acquires the input channel of the status signal coupling interface, sequentially loads the channel according to the number order, acquires the impedance value sequence in multiple consecutive time periods, extracts the impedance reference value of adjacent time periods, and differentially calculates adjacent values ​​to obtain the impedance change trend sequence.

[0058] First, the channel impedance acquisition submodule is configured with a high-frequency multiplexed switch matrix to directly scan the physical input line of the status signal coupling interface. This interface is connected to the front-end array impedance sensor network. The data acquisition unit, based on a preset channel physical address sequence (e.g., from channel 001 to channel 128), sequentially closes the relay contacts of the corresponding channels by sending a strobe level command, thus loading the channel under test. At the instant the channel is connected, the integrated analog front-end circuit within the module simultaneously acquires instantaneous analog signals of voltage and current at a sampling frequency of 2000 times per second within multiple set continuous time periods, for example, each period lasting 50 milliseconds. The acquired analog signals are filtered by a built-in anti-aliasing filter to remove high-frequency noise before being sent to an analog-to-digital converter to be converted into discrete digital quantities. Then, through division logic—that is, dividing the digital voltage value by the digital current value at each sampling moment—the instantaneous impedance value at that moment is calculated, thus forming the original impedance. The initial impedance value sequence is generated, and then the reference value extraction unit processes the data in the memory. In order to obtain a reference, the unit executes the sliding window mean calculation logic. For the Nth time period being processed, all impedance data from the previous time period are retrieved. Since the sampling frequency is 2000 Hz and the period is 50 milliseconds, the previous period contains 100 sampling points. The unit accumulates and sums these 100 impedance values. If the sum is 5000 ohms, it is divided by the number of sampling points, 100, to calculate 50 ohms as the impedance reference value for the adjacent time period. Then, the differential operation unit calls the real-time impedance value sequence of the current Nth period and the above impedance reference value, and performs a point-by-point subtraction operation. Assuming that the impedance value of the current sampling point is 52 ohms and the impedance reference value is 50 ohms, the operation logic subtracts the two to obtain a difference of positive 2 ohms. The unit performs this operation on all data points within the period, arranges the generated difference data in chronological order, and obtains the impedance change trend sequence.

[0059] The change direction judgment submodule calls the impedance change trend sequence, identifies the change direction based on the positive and negative directions of the impedance difference for each channel time period, compares the continuous change direction point by point, analyzes the continuous change state between directions, and obtains a list of channel response consistency markers.

[0060] First, the change direction determination submodule calls the impedance change trend sequence. The internally integrated sign logic analyzer performs numerical sign determination logic for each discrete data point in the sequence. If the impedance difference value at a certain time point is greater than zero, for example, +2.5 ohms, the logic marks the instantaneous change direction of that point as "positive." If the value is less than zero, for example, -1.2 ohms, it is marked as "negative." If the value is equal to zero, it is marked as "zero state." This logic traverses all sampling points within the current time period, generating a basic direction stream composed of direction signs. Based on this, the continuity analysis unit calls this basic direction stream and performs a sliding window comparison operation, setting a continuity verification window length of 5. Each sampling point is compared with the direction markers of adjacent data within the window. If the direction markers of 5 consecutive data points within the window are all "positive" or all "negative", then the local time period is determined to have directional continuity. For example, if the direction of data points with indices 10 to 15 in the sequence is all "positive", the cumulative value of the counter inside the unit reaches the threshold of 5, thus confirming that the interval is a valid change interval. The analysis unit further summarizes the verification results of the entire time period. For channels that meet the continuity condition, their status is recorded as "significant consistency"; otherwise, it is recorded as "chaotic fluctuation". The unit finally outputs the data structure containing the channel number, start and end time, and consistency status judgment to obtain a list of channel response consistency markers.

[0061] The state mapping output submodule calls the channel response consistency tag list, and according to the directional continuity status of each channel, it maps the number to the passage status to obtain the channel activation tag mapping table;

[0062] First, the state mapping output submodule calls the channel response consistency flag list. The configured hash mapping construction unit traverses each entry in the list, extracting the description field about the directional continuity state. The unit has a pre-set binary state transition logic. When the consistency state of a channel is detected as "significantly consistent", the passage state of the channel is defined as the number "1", representing that it is passable or active. When the state is detected as "chaotic fluctuation" or data is missing, the passage state of the channel is defined as the number "0", representing that it is blocked or suppressed. Then, the structured mapping unit uses the channel number as the primary key and the converted passage state as the value, writing it one by one into the lookup table structure in memory. For example, for channel number 055, if it is determined to be significantly consistent in the flag list, the unit will write the state value 1 at the index 055 of the mapping table. The unit continues to process until all channels have been traversed and performs integrity verification on the generated mapping table to ensure that there are no duplicate key values. Finally, this data structure containing the one-to-one correspondence between physical channel numbers and logical passage states is encapsulated to obtain the channel activation tag mapping table.

[0063] Specifically, such as Figure 3As shown, the signal timing alignment module includes:

[0064] The activation channel extraction submodule retrieves the original sampling signals of the channels in the current acquisition period based on the channel numbers in the channel activation tag mapping table that are accessible. It then indexes and retrieves the data sequence of each channel, activates the channel, and binds it with time data to obtain the activation channel signal set.

[0065] First, the channel extraction submodule, based on the channel numbers in the channel activation tag mapping table indicating their passable status, loads the mapping table using its integrated database retrieval engine and executes conditional filtering logic. It iterates through all key-value pairs in the table, identifying and locking all channel numbers with a passable status value of "1". For example, if only channels 012, 045, and 088 in the mapping table (channels 001 to 100) have a status value of 1, the retrieval engine stores these three numbers in a waiting queue. Subsequently, the signal retrieval unit sends a read command to the underlying large-capacity circular buffer based on the numbers in the waiting queue, calculating the memory address offset using the channel numbers. The system accurately locates and reads out the original voltage and current data sequences of the corresponding channels in batches. After the data is read out, the time binding unit immediately intervenes, obtains the current timestamp of the global clock as a reference, and assigns a precise absolute time tag to each sampling point in the data sequence based on a sampling frequency of 2000 Hz, i.e., one point every 0.5 milliseconds. Assuming the reference time is 10:00:00, the timestamp of the 10th point in the sequence is calculated and bound as 10:00:00:00:005 milliseconds. After the data packet set is processed by index positioning, data retrieval and timestamp binding, it is output uniformly to obtain the active channel signal set.

[0066] The trigger time recognition submodule calls the activation channel signal set, detects the starting position of the signal sequence in the channel, collects the position where the signal first exceeds the trigger threshold, extracts the trigger time point and marks it with a number, compares the order of the channel trigger points according to the time axis, and obtains the trigger time sorting list.

[0067] First, the trigger time recognition submodule calls the active channel signal set. The configured adaptive threshold comparator calculates the background noise baseline for the signal sequence of each channel in the set. The calculation logic involves selecting the first 50 points of the signal sequence as environmental noise samples, calculating their peak-to-peak values, and setting three times the peak-to-peak value as the trigger threshold. For example, if the peak-to-peak value of a certain channel's background noise is 2 millivolts, the trigger threshold is set to 6 millivolts. After setting the threshold, the comparator scans point by point from the beginning of the signal sequence, detecting the signal amplitude in real time. Once the absolute value of the amplitude at a sampling point exceeds 6 millivolts for the first time, the module immediately locks the point. The recorder identifies the trigger time point and extracts its absolute timestamp, marking it as the trigger time point for that channel. If channel 045 exceeds the threshold for the first time at 1002.5 milliseconds, and channel 012 exceeds the threshold for the first time at 1004.0 milliseconds, the recorder will save these time data respectively. Subsequently, the sorting unit uses a quicksort algorithm to sort all the extracted trigger time points in ascending order of their numerical values. The sorting process compares the time values ​​of each channel to determine their order, for example, placing 1002.5 milliseconds before 1004.0 milliseconds, thereby constructing a sequence reflecting the response speed of each channel and obtaining a sorted list of trigger times.

[0068] The start time alignment submodule, based on the trigger time point ranked first in the trigger time sorting list, shifts the signal data time in the activation channel to synchronize the start time of the signal with the reference time, thus obtaining a set of time synchronization channels;

[0069] First, the initial time alignment submodule uses the trigger time point ranked first in the trigger time sorting list. The embedded time axis translation operation unit reads this first time point as the global alignment reference. Assuming the first time point in the list is 1002.5 milliseconds, this time is set as T0. The translation operation unit then traverses the data of each channel in the active channel signal set and performs time difference calculation logic for each channel. It subtracts the global alignment reference time T0 from the trigger time point of the channel itself. For example, for channel 012 with a trigger time of 1004.0 milliseconds, the calculated time difference is 1.5 milliseconds. After obtaining the difference, the unit performs a time axis left shift operation on the entire signal data sequence of channel 012. That is, it uniformly subtracts 1.5 milliseconds from the time label of all data points in the sequence, or moves the data pointer to the left by the corresponding number of sampling points. If the sampling rate is 2000 Hz, it moves 3 sampling points. In this way, the trigger feature points of all channels are forced to be aligned to the same moment on the time axis, eliminating the time misalignment caused by physical transmission or response delay. After the translation processing, all channel data streams are finally converged to obtain the time synchronization channel set.

[0070] Specifically, such as Figure 4 As shown, the state trend construction module includes:

[0071] The state direction extraction submodule is based on the state sequence of channels in the time synchronization channel set at a unified start time. It reads adjacent state values ​​sequentially from consecutive time nodes, and writes the corresponding direction results according to the increase or decrease of values ​​before and after, in the order of channel number, to obtain the channel state change direction set.

[0072] First, the state direction extraction submodule, based on the state sequence of channels in the time synchronization channel set at a unified start time, operates on a unified relative time axis using a configured differential state analysis unit. A time stepper is set to access consecutive time nodes at fixed time intervals, such as 1 millisecond. At each time node, the impedance values ​​of all channels at that moment (the current value) and the value from the previous moment (the value before the previous moment) are read in parallel. The analysis unit internally performs increment / decrement logic operations, calculating the difference between the current and previous moment values. If the difference exceeds a set dead zone threshold, such as +0.0, the current value is considered lost. If the difference is 5 ohms, the state is determined to be "rising" and direction code 1 is written. If the difference is less than the negative dead zone threshold, such as -0.05 ohms, the state is determined to be "falling" and direction code -1 is written. If the difference is within the dead zone range, it is determined to be "stable" and direction code 0 is written. This unit stores the calculation results of each moment into the corresponding status register in the order of channel number. For example, if the difference of channel 012 at a relative time of 5 milliseconds is +0.2 ohms, the direction is recorded as "rising". After full-time scanning, a set containing the change states of all channels at each relative time point is generated, and the channel state change direction set is obtained.

[0073] The channel direction filtering submodule calls the channel state change direction set, extracts the direction results for each channel at each time node, writes the channel number with the same direction at the current time point into the same index item, expands the written content across nodes, and obtains the channel direction synchronization sequence set.

[0074] First, the channel direction filtering submodule calls the channel state change direction set. The configured multidimensional array classifier extracts the direction results of all channels one by one at each discrete time node, such as time point 1 and time point 2. The unit maintains temporary index buckets, corresponding to the three states of "rising", "falling" and "stable". At the current time node, if the classifier detects that the direction results of channels 001, 005 and 008 are all "rising", then these three channel numbers are written into the "rising" index bucket. If channels 002 and 006 are "falling", then they are written into the "falling" index bucket. After completing the classification of the current node, the unit solidifies the contents of the index bucket as the synchronization state record of that time point and clears the bucket to prepare for the processing of the next node. This process is repeated continuously along the time axis, expanding the written content across multiple time nodes, reconstructing the original discrete data with channels as the dimension into a data sequence with time points as the index and the same direction channel group as the content, thus obtaining the channel direction synchronization sequence set.

[0075] The synchronous change tracking submodule scans the direction results of each channel group sequentially along the time order based on the distribution of channel numbers in the channel direction synchronization sequence at the time nodes, continuously marks the duplicate numbers between adjacent nodes, and outputs the time interval and the corresponding channel number at the same time to obtain the trend channel combination directory structure.

[0076] First, the synchronization change tracking submodule, based on the distribution of channel numbers in the channel direction synchronization sequence set across time nodes, uses a configured sequence pattern matching unit to scan adjacent time nodes sequentially, such as nodes A and B. The unit extracts the set of channel numbers in the "rising" group from node A, such as channels 1, 3, and 5, and the set of channel numbers in the "rising" group from node B, such as channels 1, 3, 5, and 7, and performs an intersection operation. The result shows that channels 1, 3, and 5 maintain a synchronous rising trend between two consecutive nodes. The unit marks these duplicates consecutively and continues scanning to node C. If the intersection of node C still contains channels 1, 3, and 5, the effective time interval of this combination is extended. When the intersection result changes or is interrupted, the unit packages and outputs the currently recorded time interval, such as from node A to node Z, with the corresponding channel numbers. This logic effectively extracts the continuous trend from instantaneous synchronization, combines the time dimension and the spatial dimension to construct a hierarchical data object, and obtains a trend channel combination directory structure.

[0077] Specifically, such as Figure 5 As shown, the jump path locking module includes:

[0078] The direction continuity recognition submodule reads the channel direction results within a continuous period based on the time series direction information of each group of channels in the trend channel combination directory structure, judges the continuity performance between channels based on the direction consistency between adjacent periods, records the continuous direction status in chronological order, and obtains the channel direction continuity sequence set.

[0079] First, the direction continuity identification submodule, based on the time-series direction information of each channel group in the trend channel combination directory structure, is equipped with a long-term consistency verifier. For each recorded channel group and its corresponding time interval, the verifier backtracks and reads the original direction data. The verifier sets a consistency ratio threshold, for example, 90%. For a certain channel group recorded in the directory, the verifier counts its direction consistency within a specified continuous period. Assuming the time interval covers 100 sampling points, if a channel has 95 points whose direction is consistent with the recorded trend direction within the interval, the consistency ratio is calculated to be 95%, which is higher than the 90% threshold. The channel is then judged to have continuity. Conversely, if only 70 points are consistent, the channel is discarded. The verifier records all verified channels and their direction states in chronological order, forming a cleaned continuous state chain to ensure that only those channels with robust change patterns and no sporadic noise interference are retained, thus obtaining a channel direction continuity sequence set.

[0080] The channel path writing submodule calls the channel direction continuation sequence set, scans the channel number and direction continuation, adds the channel number with direction continuation to the corresponding path position, and writes the path set entries in the combination order to obtain the trend channel path set.

[0081] First, the channel path writing submodule calls the channel direction continuation sequence set. The configured graph theory path building engine treats channel numbers as nodes in the graph and directional continuation relationships as directed edges. The engine scans the sequence set, identifies the channel number with the earliest start time and continuation behavior, and uses it as the root node of the path. For example, if channel 001 is stable in the initial stage, the engine creates path entry 01 and writes 001 at the beginning. Then, the engine checks subsequent time segments. If it finds that channel 005 follows channel 001 in time and has a continuation relationship in space or logic, such as in terms of directional consistency, the engine appends channel 005 to path entry 01, forming a sequence of "001 to 005". The engine repeats this scanning and appending process, processes all channel numbers with continuation behavior, and assigns them to the corresponding path entries according to their appearance order and association logic. This connects discrete channel segments into a complete dynamic evolution trajectory, resulting in a set of trend channel paths.

[0082] The trend information appending submodule adds the direction information data of the channel to the corresponding path position based on the channel number index already included in the trend channel path set, and completes the association injection of channel number and direction to obtain the trend stable channel path identifier set.

[0083] First, the trend information appending submodule uses the channel number index already included in the trend channel path set and the configured metadata injection unit to semantically enhance the path set. The unit traverses the index of each channel node in the path set, and queries the previous direction continuation sequence set in reverse according to the index to extract the specific direction attributes of the channel at the corresponding path position, such as "rapid rise" or "slow decay", and quantitative features such as the slope of change. The injection unit appends the extracted direction information data to the corresponding node structure of the path set. For example, it updates the simple number "001" in the path to an object "001, direction is rising, slope is 2.5" which includes attributes. This operation completes the deep association injection of channel number and dynamic direction information, so that each node in the path carries a rich state description. After full injection processing, a trend stable channel path identifier set is obtained.

[0084] Specifically, such as Figure 6 As shown, the status response result generation module includes:

[0085] The path channel retrieval submodule scans the number information in each path based on the channel number sequence in the trend stable channel path identifier set, reads the path number and direction content, writes the channel number into the index table, and maps it according to the path affiliation to obtain the path channel mapping structure.

[0086] First, the path channel retrieval submodule, based on the channel number sequence in the trend-stable channel path identifier set, uses a configured bidirectional index builder to read the identifier set and parse each path. The builder first scans all channel numbers contained in the path, extracts their additional directional content, and then builds an efficient inverted index table. Using the channel number as the key, it stores the path identifier to which the channel belongs and the directional attribute in the path as data fields. At the same time, the builder establishes a forward mapping list according to the path affiliation, recording the complete set of channel numbers contained under each path. For example, the index table records that "channel 001" belongs to "path A" and "path B", and is "ascending" in path A and "stable" in path B. This structure allows the module to quickly look up paths by channels or channels by paths, optimizing data query efficiency and obtaining the path channel mapping structure.

[0087] The channel task matching submodule calls the path channel mapping structure, extracts the task binding channel number item by item from the channel task binding structure of the artificial intelligence control terminal, identifies the number content that appears in both the path structure and the binding structure, and obtains the target channel association index table.

[0088] First, the channel task matching submodule calls the path channel mapping structure. The configured task association processor acts as a hub connecting the underlying signals and the upper-layer applications. The processor first accesses the artificial intelligence control terminal through the network interface and retrieves the "channel task binding structure" in its memory. This structure contains a series of predefined key-value pairs that define the binding relationship between physical channels and specific tasks, such as "channel 001 is bound to the task of extending robotic arm A". Then, the processor calls the path channel mapping structure to perform set matching operations. The processor extracts the channel number from the task binding structure item by item and searches in the path mapping structure. If a channel number, such as 001, exists in both the task binding list and the valid trend path, the match is successful. The processor extracts the channel number and its index position in the path structure to ensure that only those channels that have been assigned tasks and show valid trend signals in the current probe are activated, thus obtaining the target channel association index table.

[0089] The task tag fusion submodule writes the corresponding task tag field according to the channel number in the target channel association index table, and introduces the path direction content as additional information, and associates and binds the channel number, trend direction and task tag to obtain the multi-source status data response control result.

[0090] First, the task tag fusion submodule generates the final control command based on the channel number in the target channel association index table, configured with a multi-source data fusion engine. The engine performs bidirectional data retrieval for each target channel number in the index table. On one hand, the engine retrieves the corresponding text-based "task tag" field from the AI ​​control terminal based on the number, such as "execute emergency braking." On the other hand, the engine extracts the "path direction content" and trend features of the channel from the path mapping structure, such as "pressure surge." The engine logically concatenates and encapsulates these three heterogeneous data types: channel number, trend direction, and task tag. For example, if the channel number is 101, the trend is "positive surge" with a quantization value of +10, and the task tag is "open safety valve," then a composite control command is generated: "ID is 101, trend is surge +10, task is open safety valve." This result serves as a high-level command package containing complete context information, yielding the multi-source state data response control result.

[0091] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A multi-source status data acquisition and processing system for an AI control terminal, characterized in that, The system includes: The channel coupling detection module acquires the status signal coupling interface input channel, records the impedance change direction within a continuous period, compares the change trend of adjacent periods, determines the channel status, outputs the passable and restricted status according to the number, and obtains the channel activation tag mapping table. The signal timing alignment module extracts the trigger time point based on the active channel number in the channel activation tag mapping table, aligns the channel start time, and obtains a set of time synchronization channels. The status trend grouping module extracts the direction of change of channel status values ​​of synchronized channels at the same time node based on the time synchronization channel set, identifies channel combinations with consistent directions, outputs the combination correspondence according to the synchronization status, and obtains the trend channel combination directory structure. The jump path locking module, based on the trend channel combination directory structure, determines the state direction within a continuous period, filters combinations with consistent directions, adds the channel number and trend direction to the path set, and obtains a trend stable channel path identifier set. The status response result generation module extracts the target channel of the path based on the trend stable channel path identifier set, corresponding to the task structure of the artificial intelligence control terminal, and associates the number with the trend information to obtain the multi-source status data response control result. The multi-source state data response control results include the target channel number, trend direction, and task label.

2. The AI ​​control terminal multi-source status data acquisition and processing system according to claim 1, characterized in that, The channel activation tag mapping table includes channel number, access status, and restricted status; the time synchronization channel set includes activation channel number, unified start time, and synchronization trigger time point; the trend channel combination directory structure includes channel combination, status change direction, consistency change relationship, and synchronization change status; and the trend stable channel path identifier set includes path number, channel number, and trend direction information.

3. The AI ​​control terminal multi-source status data acquisition and processing system according to claim 1, characterized in that, The determination of channel status refers to the process of distinguishing whether a channel is passable or restricted by analyzing whether the direction of state change of the input channel remains consistent within a continuous period. The direction of change of the channel status value of the synchronized channel at the same time point refers to the direction of increase or decrease of the channel status value extracted at the same time point among all channels that have completed time alignment.

4. The AI ​​control terminal multi-source status data acquisition and processing system according to claim 1, characterized in that, The so-called directional consistent channel combination refers to a set of coordinated change channels formed by merging multiple channels that exhibit the same direction of state change within a time node and a continuous time interval. The combination correspondence refers to the association description of the synchronous changes of channels with the same direction in time, and the determination of the mapping relationship between channel combination, change direction and duration interval.

5. The AI ​​control terminal multi-source status data acquisition and processing system according to claim 1, characterized in that, The channel coupling detection module includes: The channel impedance acquisition submodule acquires the input channel of the status signal coupling interface, sequentially loads the channel according to the number order, acquires the impedance value sequence in multiple consecutive time periods, extracts the impedance reference value of adjacent time periods, and differentially calculates adjacent values ​​to obtain the impedance change trend sequence. The change direction judgment submodule calls the impedance change trend sequence, identifies the change direction based on the positive and negative directions of the impedance difference for each channel time period, compares the continuous change directions point by point, analyzes the continuous change state between directions, and obtains a channel response consistency mark list. The state mapping output submodule calls the channel response consistency tag list, and according to the directional continuity status of each channel, maps the number to the passage status to obtain the channel activation tag mapping table.

6. The AI ​​control terminal multi-source status data acquisition and processing system according to claim 1, characterized in that, The signal timing alignment module includes: The activation channel extraction submodule retrieves the original sampling signals of the channels in the current acquisition period based on the channel numbers in the channel activation tag mapping table that are accessible. It then indexes and retrieves the data sequence of each channel, activates the channel, and binds the time data to obtain the activation channel signal set. The trigger time recognition submodule calls the activation channel signal set, detects the starting position of the signal sequence in the channel, collects the position where the signal first exceeds the trigger threshold, extracts the trigger time point and marks it with a number, compares the order of the channel trigger time points according to the time axis, and obtains the trigger time sorting list. The start time alignment submodule, based on the trigger time point ranked first in the trigger time sorting list, shifts the signal data time in the activation channel to synchronize the start time and reference time of the signal, thus obtaining a set of time synchronization channels.

7. The AI ​​control terminal multi-source status data acquisition and processing system according to claim 1, characterized in that, The state trend construction module includes: The state direction extraction submodule reads adjacent state values ​​sequentially from consecutive time nodes based on the state sequence of channels in the time synchronization channel set at a unified start time. According to the increase or decrease between the values, it writes the corresponding direction results in the order of channel number to obtain the channel state change direction set. The channel direction filtering submodule calls the channel state change direction set, extracts the direction results for each channel at each time node, writes the channel numbers with the same direction at the current time point into the same index item, expands the written content across nodes, and obtains the channel direction synchronization sequence set. The synchronous change tracking submodule scans the direction results of each channel group sequentially along the time node according to the distribution of the channel number in the channel direction synchronization sequence, marks the repeated number items between adjacent nodes, and outputs the time interval and the corresponding channel number at the same time to obtain the trend channel combination directory structure.

8. The AI ​​control terminal multi-source status data acquisition and processing system according to claim 1, characterized in that, The jump path locking module includes: The direction continuity recognition submodule reads the channel direction results within a continuous period based on the time series direction information of each group of channels in the trend channel combination directory structure, judges the continuity performance between channels based on the direction consistency between adjacent periods, records the continuous direction status in chronological order, and obtains the channel direction continuity sequence set. The channel path writing submodule calls the channel direction continuation sequence set, scans the channel number and direction continuation, adds the channel number with direction continuation to the corresponding path position, and writes the path set entries in the combination order to obtain the trend channel path set. The trend information appending submodule adds the direction information data of the channel at the corresponding path position based on the channel number index already included in the trend channel path set, and completes the association injection of channel number and direction to obtain the trend stable channel path identifier set.

9. The AI ​​control terminal multi-source status data acquisition and processing system according to claim 8, characterized in that, In the process of judging the continuity of channels based on the directional consistency between adjacent cycles and recording the continuous directional state in chronological order: when the directional information in three cycles is consistent, the directional state of the continuous channel is extracted; when the directional inconsistency occurs, the extraction of directional information is restarted from the current cycle. In the process of adding channel numbers with directional continuity to the corresponding path positions and writing them into the path set entries in the order of combination: when the number of consecutive extractions of channel directional status reaches a preset number standard, the channel number is placed into the corresponding path position, and the path writing is completed in the order of the trend channel combination directory structure.

10. The AI ​​control terminal multi-source status data acquisition and processing system according to claim 1, characterized in that, The status response result generation module includes: The path channel retrieval submodule scans the number information in each path based on the channel number sequence in the trend stable channel path identifier set, reads the path number and direction content, writes the channel number into the index table, and maps it according to the path affiliation to obtain the path channel mapping structure. The channel task matching submodule calls the path channel mapping structure, extracts the task binding channel number item by item from the channel task binding structure of the artificial intelligence control terminal, identifies the number content that appears in both the path structure and the binding structure, and obtains the target channel association index table. The task tag fusion submodule writes the corresponding task tag field according to the channel number in the target channel association index table, and introduces path direction content as additional information, and associates and binds the channel number, trend direction and task tag to obtain the multi-source status data response control result.

Images