Industrial information acquisition system based on environmental data acquisition and auxiliary data filtering

Abstract

This invention provides an industrial information acquisition system based on environmental data acquisition and auxiliary data filtering, belonging to the field of industrial data acquisition technology. It includes an acquisition layer consisting of a main matrix and a set of sub-matrices. The sub-matrices of the acquisition layer transmit data to a logic layer consisting of instantaneous partial order comparison logic, periodic detection logic, duration detection logic, regional resonance statistical logic, and a correction instruction lookup table. The main matrix data from the acquisition layer and the data from the logic layer are transmitted to an execution layer containing a data selector matrix, a sampling controller matrix, and a filter matrix of the same size as the main matrix. This invention completely avoids multiplication, division, and floating-point operations, making it suitable for pure hardware implementation. Each acquisition point makes independent decisions, achieving adaptive position compensation. No dedicated micro-vibration sensors are required. Resonance and disturbance are distinguished through triple verification of period, duration, and region, avoiding misjudgment. Data source replacement and filter switching replace numerical compensation, eliminating signal distortion.

Description

Technical Field

[0001] This invention relates to the field of industrial data acquisition technology, and in particular to an industrial information acquisition system based on environmental data acquisition-assisted data filtering. Background Technology

[0002] In the field of industrial automation and data acquisition, sensors are often deployed in harsh environments with high noise levels and long durations, such as large compressor rooms, stamping workshops, and crushing stations. The broadband mechanical vibrations generated by the operation of the equipment are transmitted to the sensors through the mounting structure and sound conduction. When the vibration frequency is close to the natural frequency of the sensor mounting base, it will cause structural resonance. Structural resonance causes continuous and periodic micro-deformation of the sensor mounting base, resulting in zero-point drift, sensitivity changes, or periodic deviations in the output signals of the main process sensors (such as pressure sensors, temperature sensors, and flow sensors), which seriously reduces the accuracy of data acquisition.

[0003] In existing technologies, methods for dealing with vibration interference mainly fall into the following categories:

[0004] The first type is fixed parameter filtering, which involves connecting a low-pass filter with a fixed cutoff frequency or a median filter with a fixed window length in series at the sensor output.

[0005] The second type relies on a single or a few environmental sensors for global compensation. For example, after monitoring the overall vibration level with an accelerometer, all main sensors are uniformly corrected.

[0006] The third type is to use a dedicated micro-vibration sensor for active compensation, which requires additional hardware costs and involves numerical calculations.

[0007] The fourth category is sensor fault detection and redundancy replacement technology, which determines whether a sensor is damaged by comparing the outputs of multiple redundant sensors;

[0008] The fifth category is the environment-process correlation matrix method, which maps environmental parameters to process parameters and uses this mapping to adjust production control quantities.

[0009] However, the aforementioned existing technologies have the following problems in practical applications:

[0010] Fixed-parameter filtering cannot adapt to dynamic changes in resonance intensity, resulting in insufficient filtering when resonance occurs and over-filtering when there is no vibration; global uniform compensation ignores the spatial distribution characteristics of vibration influence, leading to significant differences in the correction effect of sensors at different locations; dedicated micro-vibration sensor solutions increase hardware costs and computational complexity, making it difficult to meet real-time requirements in pure logic implementations; sensor fault detection technology cannot identify systematic measurement deviations caused by deformation of the mounting base, because the sensor itself is not damaged but the output is distorted; the environment-process correlation matrix adjusts production parameters rather than signal filtering parameters, which belongs to a different technical category than improving data acquisition quality.

[0011] In addition, existing schemes lack the ability to comprehensively judge the periodic oscillation characteristics, duration accumulation characteristics, and regional spatial propagation characteristics of resonance, which can easily lead to misjudging random disturbances as resonance or missing true resonance.

[0012] Therefore, an industrial information acquisition system is needed that can utilize the spatial redundancy and temporal patterns within the sensor matrix to detect structural resonance purely logically and adaptively correct data deviations to solve the above problems. Summary of the Invention

[0013] The purpose of this invention is to overcome the shortcomings of existing technologies and propose an industrial information acquisition system based on environmental data acquisition and auxiliary data filtering, which can stably provide highly reliable acquisition data in industrial environments with strong noise and long duration.

[0014] To achieve the above objectives, the present invention adopts the following technical solution: an industrial information acquisition system based on environmental data acquisition and auxiliary data filtering, comprising:

[0015] The acquisition layer comprises a main matrix and a set of sub-matrices; each unit of the main matrix is ​​a main process sensor; each main process sensor is associated with a sub-matrix in the set of sub-matrices, and the sub-matrix contains at least three auxiliary environmental sensors of the same type;

[0016] The logic layer includes instantaneous partial order comparison logic, periodic detection logic, duration detection logic, regional resonance statistics logic, and a correction instruction lookup table. The instantaneous partial order comparison logic generates an instantaneous micro-vibration mode flag based on the output levels of auxiliary environmental sensors within the same sub-matrix. The periodic detection logic outputs a periodic resonance flag based on the number of times the auxiliary environmental sensor levels reverse direction over multiple periods. The duration detection logic outputs a long-term resonance flag based on the number of consecutive abnormal periods indicated by the instantaneous micro-vibration mode flag. The regional resonance statistics logic outputs a regional resonance flag based on the resonance flags of adjacent units. The correction instruction lookup table outputs resonance correction instructions based on the combined state of the periodic resonance flag, long-term resonance flag, and regional resonance flag.

[0017] Execution layer: contains a data selector matrix, a sampling controller matrix, and a filter matrix with the same size as the main matrix; the data selector matrix, sampling controller matrix, and filter matrix all receive resonance correction instructions, which are used to select the data source, control the sampling mode, and select the filter parameter group, respectively.

[0018] Furthermore, the periodic detection logic includes: a comparator, a direction reversal detector, and a period counter; the comparator is used to compare the median level of the auxiliary environmental sensor in the current period with that in the previous period and output the direction of change; the direction reversal detector is used to detect the reversal of the direction of change from the previous period to the current period; the period counter is used to increment the count each time a reversal is detected, and reset the count when the direction of change remains unchanged for more than three consecutive periods; the period counter outputs a periodic resonance flag when it reaches a preset threshold.

[0019] Furthermore, the duration detection logic includes a long-time counter; the long-time counter increments when the instantaneous micro-vibration mode flag is abnormal, decrements when the instantaneous micro-vibration mode flag is normal, and outputs the long-time resonance flag when it reaches a preset high threshold.

[0020] Furthermore, the regional resonance statistical logic includes: each unit sends its own periodic resonance flag and long-term resonance flag to four neighboring units; each unit includes a neighborhood resonance counter, which is used to count the number of resonance flags received as true in the four neighboring units, and outputs the regional resonance flag when the number is greater than or equal to a preset threshold.

[0021] Furthermore, the resonance correction instruction includes a sampling control bit; the sampling controller matrix selects a normal sampling mode, a paused sampling mode, or a down-frequency sampling mode according to the sampling control bit; in the paused sampling mode, the data selector matrix output maintains the previous valid value; in the down-frequency sampling mode, the data selector matrix reads a new signal every multiple sampling cycles.

[0022] Furthermore, the execution layer also includes a history template register matrix, each unit of which is used to store the local main process signal of the last normal cycle before the resonance occurs; when the resonance correction instruction requires the use of the history template, the data selector matrix outputs the corresponding value in the history template register matrix.

[0023] Furthermore, the resonance correction instruction includes a data flag bit; the data flag bit is used to indicate that the output data is affected by resonance or the data is unavailable, and is output to the host system together with the purification signal output by the filter matrix.

[0024] Furthermore, the correction instruction lookup table is a read-only memory; the address of the read-only memory is formed by a combination of periodic resonance flags, long-term resonance flags and regional resonance flags; each address unit of the read-only memory stores a set of resonance correction instructions.

[0025] Furthermore, at least three auxiliary environmental sensors of the same type in the sub-matrix are arranged asymmetrically on the mounting base of the main process sensor, so that there is a preset partial order relationship between the output levels of the auxiliary environmental sensors under normal conditions; the instantaneous partial order comparison logic compares the current partial order relationship with the preset partial order relationship to generate an instantaneous micro-vibration mode flag.

[0026] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0027] This invention associates each main process sensor in the acquisition layer with a sub-matrix containing at least three similar auxiliary environmental sensors. Utilizing a pre-defined partial order relationship formed by asymmetric installation, and combining this with the collaborative judgment of four dimensions in the logic layer—instantaneous partial order comparison, periodic detection, duration detection, and regional resonance statistics—it can accurately identify the occurrence of structural resonance and its spatial range and temporal duration without any numerical calculations. Furthermore, it outputs resonance correction instructions containing data source selection, sampling control, filter mode switching, and data marking through a correction instruction lookup table. These instructions are executed in parallel by the data selector, sampling controller, historical template register, and filter in the execution layer. This enables flexible replacement from local signals to neighborhood medians or historical templates, adaptive control from normal sampling to paused or down-frequency sampling, and graded switching from normal filtering to strong or ultra-strong filtering. Finally, it adds a mark indicating that the output data is affected by resonance or is unusable.

[0028] Therefore, this invention completely avoids multiplication, division, and floating-point operations, making it suitable for pure hardware implementation; each acquisition point makes independent decisions, achieving position-adaptive resonance compensation; no dedicated micro-vibration sensors are required, reducing hardware costs; through periodicity, duration, and regional collaborative triple verification, it effectively distinguishes between structural resonance and random disturbances, avoiding erroneous corrections; by using data source replacement and filtering strategy switching instead of numerical compensation, it fundamentally eliminates signal distortion during resonance; the historical template register freezing mechanism ensures that reliable reference values ​​can still be output during long-term resonance; regional resonance statistical logic ensures that correction instructions of adjacent units maintain consistency, thereby guaranteeing the spatial continuity of output data;

[0029] This allows the present invention to stably provide highly reliable acquisition data in industrial environments with high noise and long duration. Attached Figure Description

[0030] Figure 1 This is an architecture diagram of the industrial information acquisition system based on environmental data acquisition and auxiliary data filtering according to the present invention.

[0031] Figure 2 This is a flowchart illustrating the deployment phase of the industrial information acquisition system based on environmental data acquisition and auxiliary data filtering according to the present invention.

[0032] Figure 3 This is a flowchart of the debugging phase of the industrial information acquisition system based on environmental data acquisition and auxiliary data filtering according to the present invention;

[0033] Figure 4 This is a flowchart illustrating the application phase of the industrial information acquisition system based on environmental data acquisition and auxiliary data filtering according to the present invention. Detailed Implementation

[0034] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] Example 1

[0036] like Figure 1 As shown, the present invention provides a technical solution: an industrial information acquisition system based on environmental data acquisition and auxiliary data filtering, including an acquisition layer. The acquisition layer contains a main matrix and a set of sub-matrices. The data acquired by the main matrix is ​​directly transmitted to the execution layer. The data acquired by the sub-matrices is transmitted to the logic layer. The data acquired by the sub-matrices is analyzed by the logic layer and then transmitted to the execution layer.

[0037] Industrial information acquisition systems specifically include:

[0038] 1) Acquisition Layer: The acquisition layer contains a main matrix and a set of sub-matrices, specifically:

[0039] 1.1) The main matrix is ​​defined as a two-dimensional array with A rows and B columns, where A and B are positive integers; each cell of the main matrix corresponds to a main process sensor at a physical location; the main process sensor is used to collect core parameters of the industrial process, including but not limited to pressure, temperature, flow rate or displacement; the main process sensor outputs raw analog or digital process signals.

[0040] 1.2) The sub-matrix set has the same spatial index structure as the main matrix. For each unit in the main matrix (i.e., each main process sensor), a dedicated sub-matrix is ​​assigned. The sub-matrix contains at least three auxiliary environmental sensors of the same type. The auxiliary environmental sensors are installed on the same mounting base of the main process sensor and arranged in an asymmetrical direction. Under normal conditions where there is no structural resonance or the micro-vibration is negligible, since the three auxiliary environmental sensors have different directions relative to the vibration source or the propagation path of the structure, there is a fixed partial order relationship between the output levels of each auxiliary environmental sensor. For example, when they are installed on the front, left, and top surfaces of the main sensor housing, the sensor level on the front surface ≥ the sensor level on the left surface ≥ the sensor level on the top surface. This partial order relationship is preset by the configuration register during system initialization and serves as the benchmark for subsequent logical comparisons.

[0041] The auxiliary environmental sensor converts the detected physical quantities into discrete level values, such as levels 0 to 7, through a threshold comparator. This quantization process uses only the comparator and a preset threshold and does not involve multiplication or division operations.

[0042] 2) Logic Layer: The logic layer equips each main matrix unit with an independent parallel processing submodule, and all submodules operate under the same clock cycle; each submodule contains the following logic components:

[0043] 2.1) Instantaneous Partial Sequence Comparison Logic: The input to the instantaneous partial sequence comparison logic is the current output level of three auxiliary environmental sensors within the same sub-matrix. Internally, the logic uses three comparators to determine whether the first level is greater than or equal to the second level, whether the second level is greater than or equal to the third level, and whether the first level is greater than or equal to the third level, respectively. This yields a three-bit binary number called the sequence status code. The sequence status code is compared bit-by-bit with a preset normal partial sequence code stored in the same unit. The comparison result generates an instantaneous micro-vibration mode flag. The specific rules are as follows: If the sequence status code is completely equal to the preset normal partial sequence code, the flag is 00, indicating no abnormal micro-vibration; if only one bit differs, the flag is 01, indicating slight micro-vibration; if two or three bits differ, or if a logical contradiction occurs (e.g., first level ≥ second level is true and second level ≥ third level is true but first level ≥ third level is false), the flag is 10, indicating severe micro-vibration or loose installation; if all three bits of the sequence status code are false, the flag is 11, indicating sensor malfunction. The flag is sent to the subsequent duration detection logic and also serves as part of the resonance state encoding.

[0044] 2.2) Periodic Detection Logic: The periodic detection logic is used to identify whether there are periodic fluctuations in the level of the auxiliary environmental sensors. Periodic fluctuations are a typical characteristic of structural resonance. The periodic detection logic includes a comparator, a direction reversal detector, and a period counter. Specifically:

[0045] The comparator compares the median of the three auxiliary environmental sensor levels in the current cycle with the median of the previous cycle in each cycle, and outputs the direction of change, which is either rising or falling. The median is obtained by sorting the levels of the three sensors through a sorting network that uses only the comparator. The direction of change is sent as a Boolean value (e.g., 0 for falling and 1 for rising) to the direction reversal detector.

[0046] The direction reversal detector stores the direction of change in the previous cycle and compares it with the direction of change in the current cycle. If the direction of change in the previous cycle is different from the direction of change in the current cycle, a direction reversal is determined to have occurred. Each time a direction reversal is detected, the cycle counter is incremented by one. If the direction of change remains unchanged for three consecutive cycles, the cycle counter is cleared.

[0047] When the period counter reaches the preset threshold, the periodic detection logic outputs a periodic resonance flag as true. The periodic resonance flag is independent of the instantaneous partial order comparison result and is used to indicate whether there is a continuous structural resonance.

[0048] 2.3) Duration Detection Logic: Used to distinguish between short-term vibration disturbances and long-term resonance; the duration detection logic includes a long-duration counter, the counting period of which is the same as the sampling period; specifically:

[0049] The operating rules of the long-time counter are as follows: when the instantaneous micro-vibration mode flag from the instantaneous partial order comparison logic is abnormal, the long-time counter increments, but does not exceed the preset upper limit value; when the instantaneous micro-vibration mode flag is normal, the long-time counter decrements, but does not go below zero.

[0050] When the value of the long-duration counter reaches the preset high threshold, the duration detection logic outputs a long-duration resonance flag as true. The long-duration resonance flag indicates that the system has been in an abnormal micro-vibration state for a long time, which is inferred to be the continuous existence of structural resonance.

[0051] 2.4) Regional Resonance Statistical Logic: Since structural resonance does not occur in isolation on a single sensor, but simultaneously affects multiple sensors within a region, the regional resonance statistical logic utilizes the spatial adjacency relationship of a matrix to achieve collaborative detection; specifically:

[0052] Each unit sends its own periodic resonance flag and long-term resonance flag to its four neighboring units above, below, left, and right via a dedicated matrix bus. Using the flag from the previous cycle avoids combinational logic loops and ensures that the statistical results of all units are stable within the same clock cycle. Each unit has an internal neighborhood resonance counter to count the number of true resonance flags received in the four neighborhoods. Specifically, the periodic resonance flag and the long-term resonance flag can be processed in a combined manner: as long as any resonance flag of a neighboring unit is true, it is counted as a resonance unit.

[0053] When the value of the neighborhood resonance counter is greater than or equal to the preset threshold, the regional resonance statistical logic outputs the regional resonance flag as true. The regional resonance flag indicates that there is a wide resonance phenomenon in the local area where the current cell is located, and stronger correction measures need to be taken.

[0054] 2.5) Correction Instruction Lookup Table: The correction instruction lookup table is a read-only memory. The address of the read-only memory is encoded by a combination of periodic resonance flags, long-term resonance flags, and regional resonance flags. For example, a three-bit address with periodic flags, long-term flags, and regional flags. Each address unit stores a set of resonance correction instructions. The resonance correction instruction is a multi-bit binary code that contains at least the following fields:

[0055] Data source selection bit: Specifies that the data selector should use the local signal, the neighborhood median signal, the history template signal, or the safe default value;

[0056] Filter mode selection bit: Specifies whether the filter uses the normal parameter set, strong filter parameter set, or super strong filter parameter set;

[0057] Sampling control bit: Specifies whether the sampling controller uses normal sampling, paused sampling, or reduced-frequency sampling;

[0058] Data flag: Indicates whether the output data is "affected by resonance" or "data unavailable";

[0059] In this context, each address unit of the read-only memory stores a complete, mutually exclusive code for the resonance correction instruction. The data source selection bit, filter mode selection bit, sampling control bit, and data tag bit are each encoded independently. However, the correction instruction lookup table ensures that the combination of outputs at the same address is uniquely determined and will not lead to logical contradictions. For example, it will not require both the historical template and the neighborhood median at the same time. The implementation of the correction instruction lookup table is pure combinational logic, and the output depends only on the address input.

[0060] The input to the correction instruction lookup table depends only on the combination state of the resonance flags, and the output is entirely determined by the pre-programmed content, without involving any real-time calculations;

[0061] 3) Execution Layer: The execution layer contains a data selector matrix, a sampling controller matrix, a filter matrix, and a history template register matrix, all of the same size as the main matrix. All matrix units operate in parallel; specifically:

[0062] 3.1) Sampling Controller Matrix: Each unit of the sampling controller matrix receives a sampling control bit from the resonance correction instruction from the logic layer. The sampling controller selects one of three operating modes based on this control bit:

[0063] Normal sampling mode: The data selector is allowed to read new signals from the main process sensors in each sampling period;

[0064] Pause sampling mode: The output of the data selector retains the previous valid value and does not update new data; pause sampling mode is used during periods of intense resonance where it is impossible to obtain a true signal;

[0065] Down-frequency sampling mode: The data selector reads a new signal only once every few sampling cycles, and outputs a hold value for the rest of the cycles; the down-frequency sampling mode is used to reduce the injection rate of erroneous data during resonance, while retaining a certain tracking capability;

[0066] Among them, the sampling control bit and the data source selection bit work independently; when the data source selection bit is specified as the historical template signal, the historical template value remains constant during the resonance period. At this time, the down-frequency sampling will not change the output value, but it can reduce the clock toggle rate of the subsequent filter matrix, thereby reducing dynamic power consumption; in the pure logic implementation, this combination is still retained to ensure the consistency of instruction encoding.

[0067] 3.2) Data Selector Matrix: Each unit of the data selector matrix has multiple data input ports: local main process signal, four neighboring main signals received through the matrix bus, historical template signals from the historical template register matrix, and a safety default value. The data selector selects one of the above inputs to output to the filter matrix of the next stage according to the data source selection bit in the resonance correction instruction.

[0068] The neighborhood median signal is not obtained through numerical calculation, but is selected from the four neighborhood main signals through a hardware comparator tree, which contains only comparators and data selectors.

[0069] 3.3) Historical Template Register Matrix: Each cell of the historical template register matrix is ​​used to store the local main process signal of the last normal cycle before resonance occurs; the update condition of the historical template register matrix is: the instantaneous micro-vibration mode flag is normal and the periodic resonance flag is false. The instantaneous micro-vibration mode flag being normal ensures that the auxiliary environmental sensor partial order relationship is normal in the current cycle; the periodic resonance flag being false further confirms that the system is not in a periodic resonance state, avoiding the mistaken storage of interfered signals during the resonance cycle;

[0070] When both conditions are met, the register is updated to the current main process signal value every cycle; once the periodic resonance flag becomes true, the register stops updating regardless of whether the instantaneous off-order is normal, and retains the last normal value; when the resonance correction instruction requires the use of the history template, the data selector outputs the value stored in the register.

[0071] This mechanism ensures that the control system can still obtain a reasonable reference signal without resonance interference during long-term resonance.

[0072] 3.4) Filter Matrix: Each unit of the filter matrix is ​​a configurable digital filter. The digital filter has three preset parameter sets: normal parameter set (higher cutoff frequency, shorter window), strong filtering parameter set (medium cutoff frequency, medium window), and super-strong filtering parameter set (lowest cutoff frequency, longest window, median filtering can be additionally enabled). The digital filter directly switches to the corresponding parameter set according to the filtering mode selection bit in the resonance correction instruction, without the need for real-time coefficient calculation.

[0073] The filter outputs a purified signal and simultaneously receives a data flag bit from the resonance correction command. The flag bit and the purified signal are packaged together and output to the host system.

[0074] In this embodiment, at least three similar auxiliary environmental sensors are asymmetrically arranged on each main sensor mounting base. Under normal conditions, their output levels have a stable partial order relationship. When structural resonance occurs, this partial order relationship is disrupted, and anomalies can be captured by instantaneous partial order comparison. At the same time, the periodic detection logic identifies repeated oscillations of level changes to distinguish between resonance and random disturbances. The duration detection logic accumulates the duration of abnormal states to filter out brief interferences. The regional resonance statistical logic counts the number of neighborhood resonance markers to confirm the spatial propagation of vibration. After comprehensive judgment, a read-only lookup table outputs a resonance correction instruction containing data source selection, filtering mode, sampling control, and data marking. The execution layer switches to alternative signals such as neighborhood median, historical template, or safe default value accordingly and selects filter parameter groups of different intensities. The entire process relies only on comparators, counters, registers, and read-only memory, completely avoiding numerical calculations.

[0075] Compared with existing technologies, this invention addresses vibration by replacing data sources and switching filtering strategies at the logical level, rather than using numerical compensation; it does not configure a single environmental sensor or globally unified parameters for each main sensor, but instead equips each unit with at least three sensors of the same type and makes independent decisions to achieve positional adaptation; it does not rely on dedicated micro-vibration sensors, but only utilizes the redundant sensors in the sub-matrix; it does not use simple threshold-triggered fixed filtering, but accurately identifies structural resonance through triple verification of periodicity, duration, and spatial range, avoiding malfunctions due to instantaneous disturbances; unlike indiscriminate neighborhood smoothing in image processing, the neighborhood median of this invention is only activated by instructions under specific resonance states; it also differs from sensor fault detection technology, where the partial order comparison is used to determine the structural state of the mounting base rather than the health status of the sensors; and it is even more different from the environment-process correlation matrix scheme, where this invention constructs a physical sensor spatial matrix and outputs signal filtering parameters rather than production process adjustment values. These features collectively achieve computation-free, highly robust, and low-cost adaptive resonance compensation.

[0076] Example 2

[0077] like Figures 2 to 4 As shown, the industrial information acquisition system based on environmental data acquisition and auxiliary data filtering provided in Example 1 includes the following steps during deployment and operation:

[0078] I. Deployment phase, such as Figure 2 As shown, the specific steps include:

[0079] Step A1, Physical installation of the main matrix and sub-matrix: Based on the spatial layout of the monitored object, determine the number of rows A and columns B of the main matrix. Install a main process sensor at the physical location corresponding to each matrix unit. On the mounting base of each main process sensor, arrange at least three auxiliary environmental sensors of the same type in an asymmetrical direction. The specific direction of the asymmetrical arrangement is determined according to the direction of the position relative to the main vibration source or the structural propagation path, such as the front, left side, and top surface.

[0080] Step A2, Calibration of Preset Normal Partial Sequence Code: In the initial state where the system has no vibration or the vibration is negligible, the output levels of the three auxiliary environmental sensors in each sub-matrix are collected. Since the three sensors are installed in different directions, there is a fixed partial sequence relationship between their output levels. This partial sequence relationship is stored in the form of a three-bit binary code in the independent preset partial sequence register of each unit. The preset partial sequence code of each unit can be configured independently to adapt to the directional differences at different positions.

[0081] Step A3, Setting the detection threshold and counting parameters: Set the periodic counter threshold in the periodic detection logic and the long-duration counter high threshold in the duration detection logic. The long-duration counter high threshold can be pre-written into the system register based on field experience, or it can be adjusted during the debugging phase.

[0082] Step A4, Burning the correction instruction lookup table: The contents of the read-only memory are pre-burned. The address of the read-only memory is encoded by a combination of periodic resonance flag, long-term resonance flag and regional resonance flag. Each address unit corresponds to a set of resonance correction instructions. The resonance correction instructions include data source selection bit, filter mode selection bit, sampling control bit and data tag bit. Once the burning content is determined, it is read-only and not written during system operation.

[0083] Step A5, Presetting the filter parameter group: Preset three sets of parameters for each filter: normal parameter group, strong filtering parameter group, and super strong filtering parameter group. These parameter groups are written into the filter configuration register during system initialization and are only switched by instructions during operation without being recalculated.

[0084] Step A6: Initial filling of the historical template register matrix: Under the condition of the system being powered on for the first time and confirming that there is no vibration interference, write the initial signal value of each main process sensor into the corresponding historical template register.

[0085] II. Debugging phase, such as Figure 3 As shown, the specific steps include:

[0086] Step B1, Partial Sequence Relationship Verification: Put the system into normal operation and observe the instantaneous partial sequence comparison logic output of each unit; under normal circumstances, the instantaneous micro-vibration mode flags of all units should be stable to the normal state; if the flag of a certain unit frequently shows slight or severe abnormalities, it indicates that the preset partial sequence code of the unit does not match the actual partial sequence relationship, and the preset partial sequence register of the unit needs to be recalibrated.

[0087] Step B2, Periodic Detection Logic Verification: Manually introduce a brief periodic vibration and observe the output of the periodic detection logic: When the vibration exists, the period counter should increment; when the vibration stops, since the direction of change remains unchanged for more than three consecutive cycles, the period counter should be cleared; verify that the periodic resonance flag is correctly output when the period counter reaches the preset threshold; if the response is too fast or too slow, the period counter threshold can be adjusted.

[0088] Step B3, Duration Detection Logic Verification: Keep the instantaneous micro-vibration mode flag in an abnormal state for a period of time and observe the incrementing behavior of the long-term counter; when the preset high threshold is reached, verify the output of the long-term resonance flag; if a more sensitive or sluggish response is required, the high threshold can be adjusted.

[0089] Step B4, Regional Resonance Statistical Logic Verification: Apply vibration to multiple adjacent units simultaneously, observe whether the neighborhood resonance counter of each unit correctly counts the number of resonance flags in the four neighborhoods, and verify that the regional resonance flag is output when it reaches the threshold. Verify that the regional resonance flag remains false when only a single unit vibrates.

[0090] Step B5, Execution layer response verification: Trigger different combinations of resonance flags respectively, observe whether the data selector, sampling controller and filter of the execution layer switch according to the preset contents of the correction instruction lookup table, and verify whether the data flag bits of the output data are correct;

[0091] Step B6, Historical Template Update Verification: In the non-resonance state, confirm that the historical template register updates with the main signal, then trigger the resonance state, confirm that the historical template register stops updating and retains the last value before resonance, after exiting resonance, confirm that the historical template register resumes updating; after debugging, all thresholds and lookup table contents are locked, and the system enters the formal operation state.

[0092] III. Application stage, such as Figure 4 As shown, the specific steps include:

[0093] Step C1, Data Acquisition and Instantaneous Partial Sequence Comparison: At the beginning of each sampling period, all main process sensors and auxiliary environmental sensors synchronously acquire data. The analog quantity of the auxiliary environmental sensor is converted into a discrete level through a threshold comparator. The instantaneous partial sequence comparison logic of each unit generates a sequential status code based on the current levels of the three auxiliary environmental sensors and compares it with the preset normal partial sequence code to output an instantaneous micro-vibration mode flag.

[0094] Step C2, Parallel Detection of Resonance Features: While obtaining the instantaneous micro-vibration mode markers, three resonance detection logics operate in parallel:

[0095] Step C2.1: The periodic detection logic compares the median level of the three auxiliary environmental sensors in the current cycle with the median level of the previous cycle to determine the direction of change. The direction reversal detector determines whether a reversal has occurred: when a reversal occurs, the cycle counter increments; when the direction of change remains unchanged for more than three consecutive cycles, the cycle counter is reset to zero; when the cycle counter reaches a preset threshold, the periodic resonance flag is output as true, otherwise it is false.

[0096] Step C2.2: The duration detection logic increments or decrements the long-duration counter based on whether the instantaneous micro-vibration mode flag is abnormal; when the long-duration counter reaches the preset high threshold, the long-duration resonance flag is output as true, otherwise it is false.

[0097] Step C2.3: The regional resonance statistical logic reads the resonance flags latched in the previous cycle of the four neighboring cells and counts the number of true flags in the neighborhood. When the number is greater than or equal to a preset threshold, the regional resonance flag is output as true; otherwise, it is output as false.

[0098] Step C3, Resonance State Encoding and Correction Instruction Lookup: Combine the periodic resonance flag, long-term resonance flag, and regional resonance flag into a three-bit address and send it to the correction instruction lookup table. Each bit of the three-bit address represents the true or false state of a flag. The read-only memory outputs the resonance correction instruction corresponding to the address. The instruction includes the data source selection bit, the filter mode selection bit, the sampling control bit, and the data tag bit.

[0099] Step C4, Execution Layer Logic Actions: Based on the resonance correction instruction, each matrix unit in the execution layer simultaneously performs the following actions:

[0100] Step C4.1: The sampling controller selects normal sampling, paused sampling, or down-frequency sampling mode according to the sampling control bit: In paused sampling mode, the output of the data selector holds the previous valid value; in down-frequency sampling mode, the data selector reads a new signal every few cycles; in normal sampling mode, a new signal is read in each cycle.

[0101] Step C4.2: The data selector selects one output from the local signal, the neighborhood median signal, the historical template signal, or the safe default value based on the data source selection bit; the neighborhood median signal is obtained from the four neighborhood main signals through the hardware comparator tree and does not involve numerical calculation.

[0102] Step C4.3: The filter selector switches to the normal parameter group, strong filter parameter group, or super strong filter parameter group according to the filter mode selection bit to filter the signal output by the data selector.

[0103] Step C4.4: The data marker bits are combined with the filtered signal to form the final output of this unit;

[0104] Step C5, Historical Template Update: At the end of each cycle, after the execution layer output stabilizes, each unit checks whether the historical template update conditions are met: the instantaneous micro-vibration mode flag is normal and the periodic resonance flag is false; if the historical template update conditions are met, the current main process signal is written into the historical template register; if the historical template update conditions are not met, the historical template register remains unchanged.

[0105] Step C6, Output and Reporting: Each unit packages the filtered signal together with the data marker bits and sends it to the host system via industrial bus or network;

[0106] Step C7, State Self-Recovery: After the resonance subsides, the periodic counter of the periodic detection logic is cleared to zero because the direction of change remains unchanged for more than three consecutive cycles, and the periodic resonance flag becomes false; the long-duration counter of the duration detection logic begins to decrease periodically until it falls below the threshold, at which point the long-duration resonance flag becomes false; the neighborhood resonance counter in the regional resonance statistics logic is also cleared to zero; finally, the resonance state code is restored to all false, the correction instruction lookup table outputs a normal instruction, and the system automatically recovers to normal operating mode.

[0107] In this embodiment, an auxiliary environmental sensor senses the structural vibration state, the logic layer detects resonance characteristics, a lookup table outputs correction instructions, the execution layer adjusts the data source, sampling method, and filtering intensity, and the output data is simultaneously fed back to the historical template update logic, thus forming a closed loop. The entire closed loop is completed within each sampling period, without relying on external intervention, and all decisions are based on comparators, counters, and read-only memory, without numerical calculations.

[0108] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments for application in other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. An industrial information acquisition system based on environmental data acquisition and auxiliary data filtering, characterized in that, include: Acquisition layer: Contains a main matrix and a set of sub-matrices; Each unit of the main matrix is ​​a main process sensor; each main process sensor is associated with a sub-matrix in the sub-matrix set, and the sub-matrix contains at least three auxiliary environmental sensors of the same type. The logic layer includes instantaneous partial order comparison logic, periodic detection logic, duration detection logic, regional resonance statistics logic, and a correction instruction lookup table. The instantaneous partial order comparison logic generates an instantaneous micro-vibration mode flag based on the output levels of auxiliary environmental sensors within the same sub-matrix. The periodic detection logic outputs a periodic resonance flag based on the number of times the auxiliary environmental sensor levels reverse direction over multiple periods. The duration detection logic outputs a long-term resonance flag based on the number of abnormal consecutive periods indicated by the instantaneous micro-vibration mode flag. The regional resonance statistics logic outputs a regional resonance flag based on the resonance flags of adjacent units. The correction instruction lookup table outputs resonance correction instructions based on the combined state of the periodic resonance flag, long-term resonance flag, and regional resonance flag. Execution layer: contains a data selector matrix, a sampling controller matrix, and a filter matrix with the same size as the main matrix; the data selector matrix, sampling controller matrix, and filter matrix all receive resonance correction instructions, which are used to select the data source, control the sampling mode, and select the filter parameter group, respectively.

2. The industrial information acquisition system based on environmental data acquisition and auxiliary data filtering according to claim 1, characterized in that: The periodic detection logic includes: a comparator, a direction reversal detector, and a period counter; the comparator is used to compare the median level of the auxiliary environmental sensor in the current period with that in the previous period and output the direction of change; the direction reversal detector is used to detect the reversal of the direction of change from the previous period to the current period; the period counter is used to increment the count each time a reversal is detected, and reset the count when the direction of change remains unchanged for more than three consecutive periods; the period counter outputs a periodic resonance flag when it reaches a preset threshold.

3. The industrial information acquisition system based on environmental data acquisition and auxiliary data filtering according to claim 1, characterized in that: The duration detection logic includes a long-time counter; the long-time counter increments when the instantaneous micro-vibration mode flag is abnormal, decrements when the instantaneous micro-vibration mode flag is normal, and outputs the long-time resonance flag when it reaches a preset high threshold.

4. The industrial information acquisition system based on environmental data acquisition and auxiliary data filtering according to claim 1, characterized in that: The regional resonance statistical logic includes: each unit sends its own periodic resonance flag and long-term resonance flag to four neighboring units; each unit contains a neighborhood resonance counter, which is used to count the number of resonance flags received as true in the four neighboring units, and outputs the regional resonance flag when the number is greater than or equal to a preset threshold.

5. The industrial information acquisition system based on environmental data acquisition and auxiliary data filtering according to claim 1, characterized in that: The resonance correction instruction includes sampling control bits; The sampling controller matrix selects a normal sampling mode, a paused sampling mode, or a reduced-frequency sampling mode based on the sampling control bit; in the paused sampling mode, the data selector matrix output retains the previous valid value; in the reduced-frequency sampling mode, the data selector matrix reads a new signal every multiple sampling cycles.

6. The industrial information acquisition system based on environmental data acquisition and auxiliary data filtering according to claim 1, characterized in that: The execution layer also includes a history template register matrix, each cell of which is used to store the local main process signal of the last normal cycle before resonance occurs; when the resonance correction instruction requires the use of a history template, the data selector matrix outputs the corresponding value in the history template register matrix.

7. The industrial information acquisition system based on environmental data acquisition and auxiliary data filtering according to claim 1, characterized in that: The resonance correction instruction includes a data flag bit; the data flag bit is used to indicate that the output data is affected by resonance or the data is unavailable, and is output to the host system together with the purification signal output by the filter matrix.

8. The industrial information acquisition system based on environmental data acquisition and auxiliary data filtering according to claim 1, characterized in that: The correction instruction lookup table is a read-only memory; the address of the read-only memory is formed by a combination of periodic resonance flags, long-term resonance flags and regional resonance flags; each address unit of the read-only memory stores a set of resonance correction instructions.

9. The industrial information acquisition system based on environmental data acquisition and auxiliary data filtering according to claim 1, characterized in that: At least three auxiliary environmental sensors of the same type in the sub-matrix are arranged in an asymmetrical direction on the mounting base of the main process sensor, so that there is a preset partial order relationship between the output levels of the auxiliary environmental sensors under normal conditions. The instantaneous partial order comparison logic compares the current partial order relationship with the preset partial order relationship to generate an instantaneous micro-vibration mode flag.

Images