Concrete production quality traceability system based on digital twinning

Abstract

The application relates to the technical field of concrete production quality monitoring, in particular to a concrete production quality traceability system based on digital twinning, which comprises the following steps: collecting mixer operation signals and batching data, establishing a reference by using the no-load operation signals, generating equipment state characteristics, and simultaneously identifying the distribution of attached objects and shedding events; a power decoupling module compensates and differentially adjusts the mixing power according to the equipment state characteristics to obtain decoupled power; an abnormality attribution determination module determines the theoretical water consumption based on the decoupled power, combines the change trend between consecutive batches and the equipment state mutation points, and determines the abnormality attribution type; and a traceability analysis module performs traceability accordingly. By stripping the interference of the attached objects on the power signal, the application realizes accurate determination of whether the quality abnormality is caused by material factors, performs traceability only when the material factor is confirmed, and avoids false traceability caused by equipment state interference.

Description

Technical Field

[0001] This invention relates to the field of concrete production quality monitoring technology, and in particular to a concrete production quality traceability system based on digital twins. Background Technology

[0002] During concrete production, deposits gradually accumulate inside the mixer, increasing mixing resistance and causing current signal drift, which affects the judgment of the actual mixing state of the material. Current quality traceability systems mostly rely directly on mixing current or power signals for quality analysis, failing to distinguish whether signal changes originate from fluctuations in material moisture content or changes in the state of deposits on the equipment. This easily leads to misjudging equipment factors as material factors, resulting in incorrect traceability directions and reducing the accuracy and reliability of quality traceability.

[0003] Chinese Patent Publication No. CN121577862A discloses a concrete quality assessment system and method, belonging to the field of concrete quality testing technology. The concrete quality assessment system includes a communication terminal and a data acquisition system installed on the mixer truck, as well as a data cloud platform for the mixing plant. The data acquisition system includes a real-time torque value acquisition device, a temperature and humidity sensor, a speed sensor, and a vibration sensor, which can collect concrete state and working condition data. The cloud stores real-time data, basic information of the transportation task, and historical data samples with quality results. The concrete quality assessment method includes: acquiring real-time data through the data acquisition system and acquiring basic information and historical samples through the cloud; preprocessing the acquired data; matching historical samples to calculate dynamic benchmark values ​​and calculating the deviation and rate of change of real-time torque; triggering an early warning when the threshold is exceeded. This invention realizes real-time monitoring of the entire concrete transportation process, improves the accuracy of assessment, and effectively reduces potential engineering quality hazards and economic losses.

[0004] Therefore, the existing technology has the following problems: The system only triggers warnings by exceeding the threshold for torque deviation and rate of change, without identifying and compensating for interference from the mixer's attachments. It cannot distinguish whether the signal change originates from fluctuations in the material's moisture content or changes in the state of the equipment's attachments, which can easily lead to misjudging equipment factors as material factors and causing incorrect tracing directions. Summary of the Invention

[0005] To address this, the present invention provides a concrete production quality traceability system based on digital twins, which overcomes the problem that existing technologies only trigger warnings by exceeding thresholds for torque deviation and rate of change, without addressing the identification and compensation of interference from mixer attachments. This makes it impossible to distinguish whether signal changes originate from fluctuations in material moisture content or changes in the state of equipment attachments, and easily leads to misjudging equipment factors as material factors, resulting in incorrect traceability directions.

[0006] To achieve the above objectives, the present invention provides a concrete production quality traceability system based on digital twins, comprising: The data acquisition module is used to collect the operating signals of the mixer and the batching data of the current batch. The equipment status monitoring module is used to monitor the equipment status, including: The reference establishment unit is used to establish an operating reference based on the operating signals of the mixer when it is unloaded. The status feature generation unit is used to generate equipment status features characterizing the degree of internal deposits of the mixer based on the deviation between the no-load operation signal of the mixer when it is currently no-load and the operation reference. The distribution analysis unit is used to identify the uniformity of the distribution of the deposits inside the mixer based on the fluctuation pattern of the no-load operation signal, and to generate the distribution characteristics of the deposits based on the uniformity of the distribution. The detachment event identification unit is used to identify attachment detachment events based on the fluctuation pattern of the no-load operation signal. When a detachment event is identified, the detachment time is marked as a sudden change point in the equipment state. The power decoupling module is used to determine the real-time mixing power based on the operating signal of the mixer under load, compensate the real-time mixing power according to the equipment status characteristics, and make differentiated adjustments to the compensation according to the distribution characteristics of the attached material to obtain the decoupled power. An anomaly attribution determination module is used to determine the theoretical water consumption based on the decoupling power, and to determine the anomaly attribution type based on the changing trend characteristics of the theoretical water consumption and the decoupling power in consecutive batches, combined with the abrupt change point of the equipment state. The source tracing analysis module performs source tracing analysis based on the anomaly attribution type.

[0007] Furthermore, the benchmark establishment unit includes: The no-load gap identification subunit is used to collect no-load operation signals during the no-load operation period from the time the mixer finishes unloading one batch of concrete to the time it is ready to feed the next batch. The benchmark calculation subunit is used to perform statistical averaging on multiple consecutive no-load operation signals to obtain the operation benchmark.

[0008] Furthermore, the state feature generation unit includes: The state deviation calculation subunit is used to perform difference calculation between the current no-load operation signal of the mixer and the operation reference to obtain the state deviation value; The cumulative feature generation subunit is used to perform feature decomposition on the distribution pattern of the state deviation value in the time domain, extract the peak feature and the continuous feature in the distribution pattern, use the peak feature as a characterization of the local accumulation degree of the attachment, use the continuous feature as a characterization of the coverage of the attachment, and combine the peak feature and the continuous feature to generate the equipment state feature value at the current moment. The feature change rate calculation subunit is used to track the rate of change of the equipment status feature value in multiple consecutive no-load operation cycles and generate the deposit accumulation rate feature. The accumulation type identification subunit is used to classify the current accumulation state of the attachment into a gradual accumulation type or a step accumulation type based on the accumulation rate characteristics of the attachment.

[0009] Furthermore, the distribution analysis unit includes: The fluctuation pattern extraction subunit is used to extract the fluctuation amplitude sequence and fluctuation frequency sequence of the no-load operation signal in the time domain. The uniformity assessment subunit is used to determine whether the distribution of the deposits inside the mixer is uniform or non-uniform based on the degree of dispersion of the fluctuation amplitude sequence and the fluctuation frequency sequence, and when it is determined to be non-uniform, to determine the deposit aggregation area based on the peak position of the fluctuation amplitude sequence. A distribution feature generation subunit is used to generate attachment distribution features based on the determination result of the uniformity evaluation subunit.

[0010] Furthermore, the detachment event identification unit includes: The mutation detection subunit is used to perform time-domain mutation detection on the no-load operation signal. When the amplitude of the no-load operation signal decreases within a preset time window and the decrease is greater than the preset detachment threshold, it is determined that the attached material has detached. The mutation marker subunit is used to record the moment when the attachment detachment event occurs as the equipment state mutation point and mark the spatial location corresponding to the equipment state mutation point.

[0011] Furthermore, the power decoupling module includes: The real-time power determination unit is used to determine the real-time mixing power based on the torque current signal and speed signal of the mixer when it is running under load. The compensation amount determination unit is used to determine the basic power compensation amount based on the equipment state characteristic value; The differential adjustment unit is used to determine the corresponding compensation adjustment method based on the distribution pattern characterized by the distribution characteristics of the attached material; The decoupling execution unit is used to compensate the real-time stirring power based on the compensation adjustment method and the basic power compensation amount, thereby removing the power interference components caused by uneven distribution or accumulation of deposits to obtain decoupling power.

[0012] Furthermore, the anomaly attribution determination module includes: The theoretical water consumption determination unit is used to determine the theoretical water consumption based on the material moisture content corresponding to the decoupling power and the aggregate weighing value of the current mixing batch. The water volume comparison unit is used to compare the theoretical water volume with the actual water volume in the ingredient data and generate a water volume comparison result. The trend analysis unit is used to track the change trajectory of the decoupling power in multiple consecutive production batches and extract the trend characteristics of the change trajectory, which include continuous change characteristics or intermittent change characteristics. The mutation matching unit is used to match the time of the anomaly occurrence with the time axis of the device state mutation point to determine whether the time of the anomaly occurrence corresponds to the time of the attachment falling off. The attribution determination unit is used to determine the abnormal attribution type as material factor abnormality or non-material factor abnormality based on the water volume comparison results, trend characteristics and mutation matching results.

[0013] Furthermore, the trend analysis unit includes: The change trajectory tracking subunit is used to acquire the numerical sequence of the decoupling power in multiple consecutive production batches as the change trajectory. The continuous feature extraction subunit is used to extract continuous change features when the change trajectory changes continuously in the same direction within a preset number of batches of time windows and the change amplitude at the beginning and end of the time window is greater than a preset amplitude threshold. An intermittent feature extraction subunit is used to extract intermittent change features when the change trajectory undergoes at least one directional reversal within a preset number of batches of time windows and the change amplitude before and after the reversal is greater than a preset fluctuation threshold. The feature output subunit is used to take the extracted continuous change feature or the intermittent change feature as the trend feature.

[0014] Furthermore, the attribution determination unit includes: The first determination subunit is used to determine the anomaly attribution type as non-material factor anomaly when the water volume comparison result is consistent with the theoretical water volume and the actual water volume. The second determination subunit is used to determine the anomaly attribution type as attachment detachment when the water volume comparison result is inconsistent with the theoretical water volume and the time of the anomaly occurs matches the abrupt change point of the equipment state. The third determination subunit is used to determine the anomaly attribution type as the actual fluctuation of material moisture content when the water volume comparison result is inconsistent with the theoretical water volume and the time of the anomaly does not match the sudden change point of the equipment state.

[0015] Furthermore, the source tracing analysis module includes: The material traceability execution unit is used to trace back the material dispensing sequence record of the batching system according to the timestamp of the abnormality when the abnormality is attributed to the actual fluctuation of the material moisture content, determine the target silo that was in the dispensing state at the time of the abnormality, match the target silo to the corresponding raw material supply batch according to the raw material entry registration record, and output the raw material batch traceability result. The equipment maintenance prompt unit is used to output corresponding equipment maintenance prompts or cleaning prompts when the abnormality attribution type is non-material factors or attachment detachment.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: This embodiment accurately locates the idle period from the time the mixer unloads material to the time before the next feeding cycle through the idle gap identification subunit. During this period, pure equipment operation signals unaffected by materials are collected, and the reference calculation subunit performs statistical averaging processing on the signals of multiple consecutive idle periods in a sliding window manner. This realizes the dynamic updating and adaptive adjustment of the operating reference, enabling the operating reference to follow the slow changes in equipment characteristics during long-term operation of the mixer. This provides an accurate and reliable reference for the subsequent generation of state characteristics and avoids the systematic deviation introduced by the fixed reference due to equipment aging or wear.

[0017] Furthermore, this embodiment quantifies the no-load signal deviation into a representation of the degree of attachment by calculating the state deviation, overcoming the deficiency of existing technologies that cannot directly perceive the state of attachments inside the mixer; it overcomes the deficiency of a single amplitude feature being unable to distinguish between local accumulation and coverage by extracting peak and continuous dual features through time-domain waveform decomposition; it achieves an upgrade from static evaluation to dynamic accumulation trend perception by tracking the rate of change of feature values ​​in continuous periods; and it overcomes the deficiency of a unified compensation strategy being unable to adapt to different accumulation modes by identifying two accumulation types, gradual and step, enabling the compensation strategy to adaptively match the actual growth mode of the attachments.

[0018] Furthermore, this embodiment extracts the fluctuation amplitude sequence and fluctuation frequency sequence of the no-load operation signal to transform the distribution state of the attachment into quantifiable features, overcoming the defect of not being able to perceive the spatial distribution of the attachment; it determines uniformity and non-uniformity by comprehensively considering the amplitude dispersion and frequency concentration, and locates the clustering area when it is non-uniform, providing a spatial basis for differentiated compensation; by outputting structured features containing distribution type and clustering area, the power decoupling module can apply differentiated compensation to different areas, avoiding the accuracy loss of uniform compensation.

[0019] Furthermore, this embodiment converts torque current and rotational speed into real-time stirring power through a real-time power determination unit, providing the original data foundation for power decoupling; a compensation amount determination unit establishes a mapping relationship between equipment state characteristic values ​​and basic power compensation amounts, ensuring that the compensation amount accurately corresponds to the degree of deposit accumulation; a differential adjustment unit adopts uniform compensation across the entire domain or differential angle compensation based on the orientation information of the aggregation area, depending on whether the deposit distribution is uniform or non-uniform, overcoming the problem in the prior art where a uniform compensation amount is used regardless of the deposit distribution, leading to undercompensation in aggregation areas and overcompensation in non-aggregate areas, so that each angle position in the decoupling power can accurately reflect the actual material stirring resistance under the state of the deposit at that position; and a decoupling execution unit deducts the corresponding compensation amount point by point according to the angle position, achieving accurate removal of deposit interference, providing decoupling power data that accurately reflects the true state of the material for subsequent moisture content inversion and anomaly attribution determination, avoiding misjudgment of anomaly attribution due to inaccurate compensation.

[0020] Furthermore, this embodiment converts decoupling power into theoretical water consumption through a theoretical water consumption determination unit, providing a quantitative basis for attribution determination; it achieves self-verification of moisture content inversion results through a water consumption comparison unit, overcoming the defect of being susceptible to signal interference when relying solely on decoupling power; it extracts continuous or intermittent features through a trend analysis unit, enabling the differentiation between gradual changes in equipment status and material fluctuations; it provides spatiotemporal evidence for attachment detachment through a mutation matching unit; and it performs a joint determination of triple evidence through an attribution determination unit, only determining that the material moisture content is fluctuating and triggering tracing when the water consumption is inconsistent and there is no mutation matching, thus overcoming the problem of mis-tracing caused by equipment interference and material fluctuations in existing technologies.

[0021] Furthermore, in this embodiment, when the material traceability execution unit confirms the actual fluctuation of the material's moisture content, it traces back the material release sequence record based on the abnormal timestamp, matches the target silo, and associates it with the raw material supply batch, achieving precise positioning of abnormal raw materials and providing a clear basis for defining responsibility for quality issues. The equipment maintenance prompt unit outputs equipment maintenance or cleaning prompts when there are non-material factors or detached attachments, enabling rapid handling of non-material factors. The entire traceability analysis module only initiates raw material traceability when the abnormality is confirmed to originate from material factors, overcoming the problem of ineffective traceability caused by performing raw material traceability for all abnormalities in existing technologies. This makes the traceability process more targeted and efficient. Simultaneously, the digital twin visualization interface enhances the visualization and operability of the traceability results through annotation and display. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the concrete production quality traceability system based on digital twins in this embodiment; Figure 2This is a schematic diagram of the equipment status monitoring module in the concrete production quality traceability system based on digital twins in this embodiment; Figure 3 This is a schematic diagram of the power decoupling module in the concrete production quality traceability system based on digital twins in this embodiment. Detailed Implementation

[0023] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0024] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0025] Please see Figure 1 As shown, it is a schematic diagram of the concrete production quality traceability system based on digital twin in this embodiment.

[0026] This embodiment provides a concrete production quality traceability system based on digital twins, including: The data acquisition module is used to collect the operating signals of the mixer and the batching data of the current batch. The equipment status monitoring module is used to monitor the equipment status, including: The reference establishment unit is used to establish an operating reference based on the operating signals of the mixer when it is unloaded. The status feature generation unit is used to generate equipment status features characterizing the degree of internal deposits of the mixer based on the deviation between the no-load operation signal of the mixer when it is currently no-load and the operation reference. The distribution analysis unit is used to identify the uniformity of the distribution of the deposits inside the mixer based on the fluctuation pattern of the no-load operation signal, and to generate the distribution characteristics of the deposits based on the uniformity of the distribution. The detachment event identification unit is used to identify attachment detachment events based on the fluctuation pattern of the no-load operation signal. When a detachment event is identified, the detachment time is marked as a sudden change point in the equipment state. The power decoupling module is used to determine the real-time mixing power based on the operating signal of the mixer under load, compensate the real-time mixing power according to the equipment status characteristics, and make differentiated adjustments to the compensation according to the distribution characteristics of the attached material to obtain the decoupled power. An anomaly attribution determination module is used to determine the theoretical water consumption based on the decoupling power, and to determine the anomaly attribution type based on the changing trend characteristics of the theoretical water consumption and the decoupling power in consecutive batches, combined with the abrupt change point of the equipment state. The source tracing analysis module performs source tracing analysis based on the anomaly attribution type.

[0027] In this embodiment, the data acquisition module includes an operation signal acquisition unit and a batching data acquisition unit. The operation signal acquisition unit is used to acquire the operation signals of the mixer during operation in real time. Specifically, the operation signals include no-load operation signals and loaded operation signals. During the no-load operation period from the time the mixer finishes unloading one batch of concrete to the time it feeds the next batch, the operation signal acquisition unit acquires the torque current signal and winding temperature signal of the mixing motor under no-load conditions at a millisecond sampling frequency, as the no-load operation signal; during the loaded operation of the mixer, the operation signal acquisition unit acquires the torque current signal and speed signal of the mixing motor, as the loaded operation signal. The batching data acquisition unit is used to acquire the batching data of the current batch of concrete from the batching plant control system. Specifically, the batching data includes the aggregate weighing value, cement weighing value, actual water consumption, and admixture weighing value corresponding to the current batch of concrete production. The batching data is stored in association with the current production batch number and production timestamp.

[0028] Please see Figure 2 As shown, it is a schematic diagram of the equipment status monitoring module in the concrete production quality traceability system based on digital twins in this embodiment.

[0029] Specifically, the benchmark establishment unit includes: The no-load gap identification subunit is used to collect no-load operation signals during the no-load operation period from the time the mixer finishes unloading one batch of concrete to the time it is ready to feed the next batch. The benchmark calculation subunit is used to perform statistical averaging on multiple consecutive no-load operation signals to obtain the operation benchmark.

[0030] In this embodiment, the no-load gap identification subunit is used to identify the no-load gap during the operation of the mixer. The mixer's production operation is cyclical, with each cycle consisting of a feeding stage, a mixing stage, and a discharging stage. After the discharging stage is completed and before the feeding stage of the next cycle begins, there is no material in the mixer, and it is in a no-load operating state; this period is the no-load operating segment. The no-load gap identification subunit identifies the start and end times of this no-load operating segment by monitoring the opening and closing signals of the mixer's discharge gate and the operating current signal, and collects the torque current signal and winding temperature signal of the mixing motor during this period.

[0031] The reference calculation subunit is used to perform statistical averaging on multiple consecutive no-load operating signals to obtain the operating reference. Specifically, the reference calculation subunit uses a sliding window method to statistically average the torque current signal and winding temperature signal of multiple consecutive no-load operating periods to calculate the no-load torque current reference value and the no-load temperature reference value, which constitute the operating reference. After each no-load operating period is completed, the reference calculation subunit adds the signal data of that period to the sliding window and removes the oldest signal data in the window.

[0032] This embodiment accurately locates the idle period from the time the mixer unloads material to the time before the next feeding cycle by using the idle gap identification subunit. During this period, pure equipment operation signals unaffected by materials are collected, and the reference calculation subunit performs statistical averaging of signals from multiple consecutive idle periods in a sliding window manner. This achieves dynamic updating and adaptive adjustment of the operating reference, enabling the operating reference to follow the slow changes in equipment characteristics during long-term operation of the mixer. This provides an accurate and reliable reference for the generation of subsequent state characteristics and avoids the systematic deviations introduced by fixed references due to equipment aging or wear.

[0033] Specifically, the state feature generation unit includes: The state deviation calculation subunit is used to perform difference calculation between the current no-load operation signal of the mixer and the operation reference to obtain the state deviation value; The cumulative feature generation subunit is used to perform feature decomposition on the distribution pattern of the state deviation value in the time domain, extract the peak feature and the continuous feature in the distribution pattern, use the peak feature as a characterization of the local accumulation degree of the attachment, use the continuous feature as a characterization of the coverage of the attachment, and combine the peak feature and the continuous feature to generate the equipment state feature value at the current moment. The feature change rate calculation subunit is used to track the rate of change of the equipment status feature value in multiple consecutive no-load operation cycles and generate the deposit accumulation rate feature. The accumulation type identification subunit is used to classify the current accumulation state of the attachment into a gradual accumulation type or a step accumulation type based on the accumulation rate characteristics of the attachment.

[0034] In this embodiment, the state deviation calculation subunit performs a step-by-step difference calculation between the current no-load operating signal and the corresponding signal type in the operating reference to obtain the state deviation value. This state deviation value reflects the degree of deviation of the mixer's no-load operating characteristics from the reference value at the current moment. The greater the deviation, the higher the degree of accumulation of deposits inside the mixer.

[0035] The cumulative feature generation subunit is used to perform feature decomposition on the distribution of the state deviation value in the time domain, extracting peak features and persistence features. Specifically, the cumulative feature generation subunit takes the waveform of the state deviation value changing with time during a single no-load operation period as the distribution pattern in the time domain, extracts the maximum amplitude value from this waveform as the peak feature, which characterizes the severity of the accumulation of deposits in a local area of ​​the mixer; and extracts the continuous duration of the deviation value being greater than a preset baseline from this waveform as the persistence feature, which characterizes the coverage of deposits inside the mixer. The preset baseline is determined in the following way: during the initial operation phase of the system, deviation value data of multiple consecutive no-load operation periods are collected and their average value is calculated. This average value is multiplied by a coefficient pre-calibrated according to the mixer model and on-site working conditions to obtain the initial preset baseline; during subsequent production operation, new deviation value data of no-load operation periods are continuously collected in a sliding window manner and the average value is updated, so that the preset baseline is dynamically adjusted with the slow changes in equipment characteristics during long-term operation of the mixer.

[0036] The characteristic change rate calculation subunit is used to track the rate of change of the equipment status characteristic value over multiple consecutive no-load operation cycles, and generate the deposit accumulation rate characteristic. Specifically, the characteristic change rate calculation subunit records the equipment status characteristic value corresponding to each no-load operation period, forming a sequence of the characteristic value changing over time, and determines the deposit accumulation rate characteristic based on the change of the current characteristic value relative to the historical characteristic value and the corresponding time span.

[0037] The accumulation type identification subunit is used to classify the current accumulation state of the deposits into a gradual accumulation type or a step accumulation type based on the deposit accumulation rate characteristics. Specifically, when the deposit accumulation rate characteristic is continuously positive and lower than or equal to a preset rate threshold, it is marked as a gradual accumulation type, which corresponds to the slow growth state of deposits during long-term operation of the mixer; when the deposit accumulation rate characteristic is continuously greater than the preset rate threshold within a preset time window, it is marked as a step accumulation type, which corresponds to the abnormal state of rapid aggravation of deposits during mixing. The preset rate threshold is determined through online calibration: during the initial stage of system operation or the first operation after equipment maintenance, the deposit accumulation rate characteristic values ​​of multiple consecutive production batches are collected, and the maximum value is used as the initial rate threshold; in subsequent operation, the above collection and calibration process is repeated every preset number of production batches to update the preset rate threshold, so that the preset rate threshold always matches the normal accumulation rate under the current operating conditions. The preset time window is determined according to an integer multiple of the single batch production cycle of the mixer, and its duration is set to the total duration corresponding to an integer number of production cycles.

[0038] This embodiment quantifies the no-load signal deviation into a representation of the degree of attachment by calculating the state deviation, overcoming the deficiency of existing technologies that cannot directly perceive the state of attachments inside the mixer; it overcomes the deficiency of a single amplitude feature in distinguishing between local accumulation and coverage by extracting peak and continuous dual features through time-domain waveform decomposition; it achieves an upgrade from static evaluation to dynamic accumulation trend perception by tracking the rate of change of feature values ​​in continuous periods; and it overcomes the deficiency of a unified compensation strategy in adapting to different accumulation modes by identifying two accumulation types, gradual and step, enabling the compensation strategy to adaptively match the actual growth mode of attachments.

[0039] Specifically, the distribution analysis unit includes: The fluctuation pattern extraction subunit is used to extract the fluctuation amplitude sequence and fluctuation frequency sequence of the no-load operation signal in the time domain. The uniformity assessment subunit is used to determine whether the distribution of the deposits inside the mixer is uniform or non-uniform based on the degree of dispersion of the fluctuation amplitude sequence and the fluctuation frequency sequence, and when it is determined to be non-uniform, to determine the deposit aggregation area based on the peak position of the fluctuation amplitude sequence. A distribution feature generation subunit is used to generate attachment distribution features based on the determination result of the uniformity evaluation subunit.

[0040] In this embodiment, the distribution analysis unit is used to analyze the uniformity of the distribution of adhering materials inside the mixer based on the fluctuation pattern of the no-load operation signal. The fluctuation pattern extraction subunit performs fluctuation pattern analysis on the operation signal of the same no-load operation period used by the state feature generation unit, and extracts the fluctuation amplitude sequence and fluctuation frequency sequence of the no-load operation signal in the time domain. Specifically, when the mixer is running without load, the drive motor needs to overcome the mechanical resistance of the rotating mixing shaft. When there are no adhering materials inside the mixer or the adhering materials are evenly distributed, the fluctuation amplitude of the no-load operation signal remains basically consistent at each angle of rotation of the mixing shaft; when the adhering materials are unevenly distributed inside the mixer, that is, different angle positions of the mixing shaft correspond to different adhering amounts, the mechanical resistance increases when the mixing shaft rotates to the angle corresponding to the area where the adhering materials are concentrated, and the amplitude of the no-load operation signal increases significantly at that angle position. The fluctuation pattern extraction subunit analyzes the real-time waveform of the no-load operation signal within a single no-load operation period, extracts the amplitude change sequence within each rotation of the mixing shaft in the waveform as the fluctuation amplitude sequence; and extracts the repetition frequency characteristics of each amplitude fluctuation in the waveform as the fluctuation frequency sequence.

[0041] The uniformity assessment subunit determines whether the distribution of the deposits inside the mixer is uniform or non-uniform based on the dispersion of the amplitude values ​​in the fluctuation amplitude sequence from their average value and the concentration of the frequency components in the fluctuation frequency sequence. Specifically, when the dispersion of each amplitude value in the fluctuation amplitude sequence is lower than a preset dispersion threshold and the proportion of the dominant frequency component in the fluctuation frequency sequence is higher than a preset concentration threshold, it is determined to be a uniform distribution; when the dispersion of each amplitude value in the fluctuation amplitude sequence is higher than the preset dispersion threshold or the proportion of the dominant frequency component in the fluctuation frequency sequence is lower than the preset concentration threshold, it is determined to be a non-uniform distribution.

[0042] In this embodiment, the preset discrete threshold is determined as follows: The no-load operating signal of the mixer in its factory-delivered no-load state or after thorough cleaning is obtained as a reference signal. The degree of dispersion of each amplitude in the fluctuation amplitude sequence corresponding to the reference signal from its average value is calculated, and this degree of dispersion is used as the reference discrete value. Historical no-load operating signals of the same model mixer under normal operating conditions are statistically analyzed to obtain the upper limit of normal fluctuation of amplitude dispersion, which is used as the redundancy margin of the reference discrete value. The sum of the reference discrete value and the redundancy margin is used as the preset discrete threshold, so that the preset discrete threshold is located in the transition range between the normal fluctuation range of the reference signal and the abnormal fluctuation range corresponding to the uneven distribution of adhering materials, thus balancing judgment sensitivity and false judgment suppression capability. The upper limit of normal fluctuation is the boundary value of the statistical distribution range of amplitude dispersion in the historical no-load operating signals of the same model mixer under normal operating conditions.

[0043] In this embodiment, the preset concentration threshold is determined as follows: The no-load operating signal of the mixer in its factory-delivered no-load state or after thorough cleaning is obtained as a reference signal. The proportion of the main frequency component in the fluctuation frequency sequence corresponding to the reference signal is calculated, and this proportion is used as the reference concentration value. Historical no-load operating signals of the same model mixer under standard operating conditions are statistically analyzed to obtain the normal fluctuation lower limit of the main frequency component proportion, which is used as the redundancy margin of the reference concentration value. The preset concentration threshold is obtained by subtracting the redundancy margin from the reference concentration value, ensuring that the preset concentration threshold is located in the transition range between the normal concentration of the reference signal and the dispersion corresponding to the uneven distribution of the deposits, thus balancing judgment sensitivity and false judgment suppression capability. The normal fluctuation lower limit is the boundary value of the statistical distribution range of the main frequency component proportion in the historical no-load operating signals of the same model mixer under standard operating conditions. The distribution feature generation subunit generates deposit distribution features based on the judgment result of the uniformity evaluation subunit. The distribution characteristics of the attachments include at least the distribution type information, and when it is determined to be a non-uniform distribution, it also includes the location information of the attachment aggregation area.

[0044] This embodiment extracts the amplitude and frequency sequences of the no-load operation signal to transform the distribution state of the attached objects into quantifiable features, overcoming the deficiency of not being able to perceive the spatial distribution of the attached objects. By comprehensively considering the amplitude dispersion and frequency concentration, it determines whether the distribution is uniform or non-uniform, and locates the clustering area when it is non-uniform, providing a spatial basis for differentiated compensation. By outputting structured features that include the distribution type and clustering area, the power decoupling module can apply differentiated compensation to different areas, avoiding the accuracy loss of uniform compensation.

[0045] Specifically, the detachment event identification unit includes: The mutation detection subunit is used to perform time-domain mutation detection on the no-load operation signal. When the amplitude of the no-load operation signal decreases within a preset time window and the decrease is greater than the preset detachment threshold, it is determined that the attached material has detached. The mutation marker subunit is used to record the moment when the attachment detachment event occurs as the equipment state mutation point and mark the spatial location corresponding to the equipment state mutation point.

[0046] In this embodiment, the detachment event identification unit is used to identify events of sudden detachment of deposits inside the mixer and record relevant spatiotemporal information. During long-term operation of the mixer, deposits gradually thicken on the surfaces of the mixing arms, liners, and drum walls. When the accumulated deposits reach a certain thickness or are subjected to external factors such as material impact or vibration, some deposits may suddenly detach. After the deposits detach, the mechanical resistance of the mixer during no-load operation decreases, which is reflected in the no-load operation signal as a sudden drop in signal amplitude. This embodiment utilizes this physical phenomenon to identify detachment events by detecting abrupt changes in amplitude in the no-load operation signal.

[0047] The mutation detection subunit performs time-domain mutation detection on the operating signals of the same idle operating period used by the state feature generation unit. Specifically, the mutation detection subunit traverses the amplitude sequence of the idle operating signal within a single idle operating period using a sliding window, and calculates the change in amplitude at the current sampling point relative to the amplitude at the previous moment. When a negative change is detected and the decrease is greater than a preset detachment threshold, it is determined that an attachment detachment event has occurred at the current moment. The preset detachment threshold is determined as follows: The no-load operating signal of the mixer in its factory-delivered state or after thorough cleaning is used as a reference signal. The amplitude fluctuation range of this reference signal is calculated, and the upper limit of this fluctuation range is used as the reference fluctuation value. The amplitude decrease corresponding to multiple small detachment events of the same model mixer under normal operating conditions is statistically analyzed, and the minimum amplitude decrease obtained from the statistics is used as the detachment judgment benchmark. The sum of the reference fluctuation value and the detachment judgment benchmark is used as the preset detachment threshold, ensuring that the preset detachment threshold is located in the transition range between normal signal fluctuations and amplitude decreases corresponding to the detachment of attachments. This avoids misjudging signal noise or small fluctuations during normal operation as detachment events, while ensuring that real detachment events can be effectively detected. After a detachment event is determined to have occurred, the mutation detection subunit generates a detachment event trigger signal and transmits this signal to the mutation marking subunit.

[0048] After receiving the detachment event trigger signal, the mutation marking subunit records the current moment as the equipment state mutation point and marks the corresponding spatial location of the mutation point in the digital twin model. Specifically, the mixer's no-load operation signal corresponds to the stirring shaft rotating to different spatial angles at different times, and the signal amplitude at each sampling moment corresponds to the spatial position of the stirring shaft at that moment. When the detachment event is triggered, the mutation marking subunit determines the spatial angle position of the stirring shaft when the attachment detachment occurs based on the correspondence between the current moment and the stirring shaft rotation angle, and marks the spatial position in the digital twin model. The marked equipment state mutation point and its spatial location information are output to the anomaly attribution determination module, which is used to determine whether the time of the anomaly occurrence corresponds to the time of attachment detachment in the subsequent quality anomaly attribution determination. That is, when the anomaly attribution determination module detects a quality anomaly and the time of the anomaly occurrence matches the marked equipment state mutation point, the anomaly can be preferentially attributed to attachment detachment.

[0049] Please see Figure 3 As shown, Figure 3 This is a schematic diagram of the power decoupling module in the concrete production quality traceability system based on digital twins in this embodiment.

[0050] Specifically, the power decoupling module includes: The real-time power determination unit is used to determine the real-time mixing power based on the torque current signal and speed signal of the mixer when it is running under load. The compensation amount determination unit is used to determine the basic power compensation amount based on the equipment state characteristic value; The differential adjustment unit is used to determine the corresponding compensation adjustment method based on the distribution pattern characterized by the distribution characteristics of the attached material; The decoupling execution unit is used to compensate the real-time stirring power based on the compensation adjustment method and the basic power compensation amount, thereby removing the power interference components caused by uneven distribution or accumulation of deposits to obtain decoupling power.

[0051] In this embodiment, the power decoupling module is used to separate the interference component of the adhering material on the stirring power from the real-time stirring power, restoring the power component corresponding to the actual stirring state of the material. The real-time power determination unit determines the real-time stirring power based on the torque current signal and speed signal of the mixer under load, according to the motor power calculation principle. Specifically, during the operation of the mixer under load, the electromagnetic torque output by the drive motor overcomes various resistances inside the mixer, including material stirring resistance, mechanical friction resistance of the stirring shaft, and additional resistance caused by the adhering material. The torque current signal reflects the magnitude of the electromagnetic torque output by the motor, and the speed signal reflects the current rotational speed of the stirring shaft. Multiplying the two and dividing by the transmission efficiency coefficient yields the real-time stirring power. This real-time stirring power includes three components: material stirring, mechanical friction, and adhering material resistance.

[0052] The compensation amount determination unit determines the base power compensation amount based on the equipment state characteristic value. The equipment state characteristic value is generated by the state characteristic generation unit, and its magnitude represents the degree of accumulation of deposits inside the mixer. The higher the degree of deposit accumulation, the greater the additional resistance the mixer needs to overcome during no-load operation; correspondingly, the interference of deposits on real-time mixing power is more significant during load operation. The compensation amount determination unit has a pre-set mapping relationship between the equipment state characteristic value and the base power compensation amount. This mapping relationship is pre-established based on the correspondence between the degree of deposits and the power increment under normal load operation of the mixer. Specifically: in a clean state with no deposits inside the mixer, no-load power and load power under different operating conditions are collected as reference power values; after the mixer has been running for a period of time and deposits have accumulated to different degrees, no-load power and load power under the same operating conditions are collected again. The increment of the load power under the deposit state relative to the load power under the clean state is correlated with the equipment state characteristic value under the deposit state to establish a characteristic value-compensation amount mapping relationship. The compensation amount determination unit determines the corresponding basic power compensation amount based on the device state characteristic value at the current moment through the mapping relationship.

[0053] The differential adjustment unit determines the corresponding compensation adjustment method based on the distribution pattern characterized by the attachment distribution features. The attachment distribution features are generated by the distribution analysis unit and include information on whether the distribution is uniform or non-uniform, as well as the location information of the attachment aggregation area when it is determined to be non-uniform. When the attachments are uniformly distributed inside the mixer, the additional resistance generated by the attachments on the rotation of the stirring arms at various angles is basically the same. The differential adjustment unit uniformly distributes the basic power compensation amount across the entire range, without distinguishing angles or regions; that is, it adopts a uniform compensation method across the entire range, allocating and deducting the basic power compensation amount according to the original amplitude ratio of the real-time stirring power, so that each angular position receives the same compensation intensity.

[0054] When deposits are unevenly distributed inside the mixer, the additional resistance at the angular position corresponding to the deposit accumulation area is much greater than that in the non-accumulation area. The differential adjustment unit establishes a compensation intensity distribution curve that varies with the rotation angle of the stirring shaft based on the orientation information of the deposit accumulation area. This curve assigns a higher compensation intensity within the angular range corresponding to the deposit accumulation area and a lower compensation intensity within the angular range corresponding to the non-accumulation area. A smooth transition compensation intensity is used between the accumulation and non-accumulation areas to avoid abrupt changes in compensation intensity that introduce new signal distortion. Specifically, when the stirring shaft rotates to the angular range corresponding to the accumulation area, the basic power compensation is increased and used as the actual compensation within that angular range; when the stirring shaft rotates to the angular range corresponding to the non-accumulation area, the basic power compensation is decreased and used as the actual compensation within that angular range. Through this angle-differentiated compensation, the power components corresponding to each angular position in the decoupled power can accurately reflect the actual material stirring resistance under the deposit state at that position, avoiding the problem of undercompensation in the accumulation area and overcompensation in the non-accumulation area caused by uniform compensation.

[0055] Based on the aforementioned differentially adjusted compensation intensity distribution curve, the decoupling execution unit performs point-by-point compensation processing on the real-time stirring power according to the angular position. The actual compensation amount corresponding to the angular position is subtracted from the real-time stirring power at each angular position to obtain the decoupling power component corresponding to each angular position, which is then combined to form a complete decoupling power waveform.

[0056] This embodiment converts torque current and rotational speed into real-time stirring power through a real-time power determination unit, providing the original data foundation for power decoupling. A compensation amount determination unit establishes a mapping relationship between equipment state characteristic values ​​and basic power compensation amounts, ensuring a precise correspondence between the compensation amount and the degree of deposit accumulation. A differentiated adjustment unit, based on the uniform or non-uniform distribution of deposits, employs either uniform compensation across the entire domain or differentiated angle compensation based on the orientation information of the aggregation area. This overcomes the problem in existing technologies where a uniform compensation amount is used regardless of deposit distribution, leading to undercompensation in aggregation areas and overcompensation in non-aggregate areas. This ensures that each angle position in the decoupling power accurately reflects the actual material stirring resistance under the deposit state at that location. The decoupling execution unit deducts the corresponding compensation amount point-by-point according to the angle position, achieving precise removal of deposit interference. This provides accurate decoupling power data reflecting the true state of the material for subsequent moisture content inversion and anomaly attribution determination, avoiding misjudgments of anomaly attribution due to inaccurate compensation.

[0057] Specifically, the anomaly attribution determination module includes: The theoretical water consumption determination unit is used to determine the theoretical water consumption based on the material moisture content corresponding to the decoupling power and the aggregate weighing value of the current mixing batch. The water volume comparison unit is used to compare the theoretical water volume with the actual water volume in the ingredient data and generate a water volume comparison result. The trend analysis unit is used to track the change trajectory of the decoupling power in multiple consecutive production batches and extract the trend characteristics of the change trajectory, which include continuous change characteristics or intermittent change characteristics. The mutation matching unit is used to match the time of the anomaly occurrence with the time axis of the device state mutation point to determine whether the time of the anomaly occurrence corresponds to the time of the attachment falling off. The attribution determination unit is used to determine the abnormal attribution type as material factor abnormality or non-material factor abnormality based on the water volume comparison results, trend characteristics and mutation matching results.

[0058] Specifically, the trend analysis unit includes: The change trajectory tracking subunit is used to acquire the numerical sequence of the decoupling power in multiple consecutive production batches as the change trajectory. The continuous feature extraction subunit is used to extract continuous change features when the change trajectory changes continuously in the same direction within a preset number of batches of time windows and the change amplitude at the beginning and end of the time window is greater than a preset amplitude threshold. An intermittent feature extraction subunit is used to extract intermittent change features when the change trajectory undergoes at least one directional reversal within a preset number of batches of time windows and the change amplitude before and after the reversal is greater than a preset fluctuation threshold. The feature output subunit is used to take the extracted continuous change feature or the intermittent change feature as the trend feature.

[0059] Specifically, the attribution determination unit includes: The first determination subunit is used to determine the anomaly attribution type as non-material factor anomaly when the water volume comparison result is consistent with the theoretical water volume and the actual water volume. The second determination subunit is used to determine the anomaly attribution type as attachment detachment when the water volume comparison result is inconsistent with the theoretical water volume and the time of the anomaly occurs matches the abrupt change point of the equipment state. The third determination subunit is used to determine the anomaly attribution type as the actual fluctuation of material moisture content when the water volume comparison result is inconsistent with the theoretical water volume and the time of the anomaly does not match the sudden change point of the equipment state.

[0060] In this embodiment, the anomaly attribution determination module is used to comprehensively determine the attribution type of quality anomalies based on decoupling power, batching data, and abrupt changes in equipment status, in order to determine whether to trigger material layer tracing. The theoretical water consumption determination unit determines the theoretical water consumption based on the material moisture content corresponding to the decoupling power and the aggregate weighing value of the current mixing batch. Specifically, the decoupling power reflects the power consumption under actual material mixing conditions, and its magnitude is directly related to the fluidity of the material during the mixing process. When the material moisture content is high, the material fluidity is enhanced, the mixing resistance is reduced, and the decoupling power is correspondingly reduced; when the material moisture content is low, the material fluidity is weakened, the mixing resistance is increased, and the decoupling power is correspondingly increased.

[0061] The theoretical water consumption determination unit pre-sets a correlation between decoupling power and moisture content. This correlation is established in advance through laboratory calibration: under laboratory conditions, using the same model mixer and the same mixing process as on-site, multiple batches of concrete with different moisture contents and the same mix proportions are prepared. The decoupling power values ​​during the mixing process of each batch of concrete are recorded, and a decoupling power-moisture content calibration curve is established. After obtaining the decoupling power of the current batch, the moisture content of the current material is determined by looking up the calibration curve. Then, based on the aggregate weighing value of the current batch, the theoretical water consumption required for the material to reach the target slump under this moisture content is calculated.

[0062] The water comparison unit compares the theoretical water consumption with the actual water consumption in the batching data to generate a water comparison result. When the theoretical water consumption matches the actual water consumption, it indicates that the currently inverted water content is consistent with the water content corresponding to the actual water consumption in the batching data, and the quality anomaly is highly likely not caused by fluctuations in the material's moisture content. When the theoretical water consumption does not match the actual water consumption, it indicates that the actual water consumption deviates from the water consumption that should be based on the true moisture content of the material, and the water comparison unit marks this deviation as a water comparison anomaly. The matching of the theoretical and actual water consumption means that the deviation between them is within a preset deviation tolerance range; the deviation tolerance range is predetermined based on the metering accuracy of the batching system and the normal fluctuation range of the production conditions.

[0063] The trend analysis unit is used to track the change trajectory of the decoupling power in multiple consecutive production batches and extract the trend features of the change trajectory. The trend features include continuous change features or intermittent change features. Specifically, the change trajectory tracking subunit acquires the decoupling power values ​​of multiple consecutive production batches in a sliding window manner, forming a numerical sequence of decoupling power changing over time, as the change trajectory. The continuous feature extraction subunit determines the change trajectory: when the sequence continuously changes in the same direction within a time window of a preset number of batches, that is, the difference sign between adjacent batches is consistent, and the change amplitude of the decoupling power between the start and end times within the time window is greater than a preset amplitude threshold, it is extracted as a continuous change feature. This feature corresponds to the situation of gradual changes in equipment status, such as gradual changes in transmission system efficiency. The intermittent feature extraction subunit determines the change trajectory: when the sequence undergoes at least one directional reversal within a preset number of batches' time window (i.e., the sign of the difference between adjacent batches changes from positive to negative or from negative to positive), and the magnitude of the change before and after the reversal is greater than a preset fluctuation threshold, it is extracted as an intermittent change feature. This feature corresponds to the situation of material batch differences or changes in the distribution state of adhering substances. The feature output subunit outputs the extracted continuous or intermittent change features as the trend features to the attribution determination unit.

[0064] The mutation matching unit is used to match the time of an anomaly occurrence with the time of a change in the equipment state, determining whether the time of the anomaly occurrence corresponds to the time of attachment detachment. The time of a change in the equipment state is marked and recorded by the detachment event identification unit. When the attribution determination unit detects a quality anomaly, the mutation matching unit obtains the specific timestamp of the anomaly occurrence and compares it with the timestamp recorded at the time of the change in the equipment state, calculating the time difference. When the time difference is less than a preset time matching tolerance, it is determined that the time of the anomaly occurs corresponds to the time of attachment detachment; otherwise, it is determined that there is no match. The time matching tolerance is determined based on the duration of a single batch production cycle of the mixer. The matching result of the mutation matching unit is output to the attribution determination unit.

[0065] The attribution determination unit, based on the water volume comparison result, the trend characteristics, and the mutation matching result, determines the current quality anomaly attribution type using three determination sub-units. The first determination sub-unit determines the anomaly attribution type as non-material factor anomaly when the water volume comparison result shows that the theoretical water volume matches the actual water volume. Since the matching water volume indicates that the inverted water content is consistent with the actual water volume, the change in decoupling power is not due to a true fluctuation in the material's moisture content, but rather to changes in equipment status or signal interference; therefore, there is no need to trigger material layer tracing. The second determination sub-unit determines the anomaly attribution type as attachment detachment when the water volume comparison result shows that the theoretical water volume does not match the actual water volume, and the time of the anomaly matches the mutation point in the equipment status. Although the discrepancy in water volume at this time superficially points to material factors, the mutation matching results indicate that the time of the anomaly occurs exactly corresponds to the time of attachment detachment. After the attachment detaches, the no-load operation signal changes abruptly, which directly affects the accuracy of decoupling power, causing deviation in the moisture content inversion, and thus leading to the discrepancy in water volume comparison. Therefore, it should be attributed primarily to the attachment detachment and should not trigger material layer tracing.

[0066] The third determination subunit is used to determine the anomaly attribution type as true fluctuation in material moisture content when the water volume comparison result shows that the theoretical water volume does not match the actual water volume, and the time of the anomaly does not match the abrupt change point of the equipment state. In this case, the water volume comparison anomaly indicates that the actual water volume does not match the water volume required for the true moisture content of the material, and the abrupt change matching mismatch excludes the interference of attached material detachment. At the same time, the trend analysis has excluded the gradual change factor of the equipment state. Therefore, the anomaly is determined to be caused by the fluctuation of material moisture content, triggering the traceability analysis module to perform material layer traceability.

[0067] This embodiment converts decoupling power into theoretical water consumption through a theoretical water consumption determination unit, providing a quantitative basis for attribution determination; it achieves self-verification of moisture content inversion results through a water consumption comparison unit, overcoming the defect of being susceptible to signal interference when relying solely on decoupling power; it extracts continuous or intermittent features through a trend analysis unit, enabling the differentiation between gradual changes in equipment status and material fluctuations; it provides spatiotemporal evidence for attachment detachment through a mutation matching unit; and it performs a joint determination of triple evidence through an attribution determination unit, only determining that the material moisture content fluctuates and triggering source tracing when the water consumption is inconsistent and there is no mutation matching, thus overcoming the problem of misattribution caused by equipment interference and material fluctuations in existing technologies.

[0068] Specifically, the source tracing analysis module includes: The material traceability execution unit is used to trace back the material dispensing sequence record of the batching system according to the timestamp of the abnormality when the abnormality is attributed to the actual fluctuation of the material moisture content, determine the target silo that was in the dispensing state at the time of the abnormality, match the target silo to the corresponding raw material supply batch according to the raw material entry registration record, and output the raw material batch traceability result. The equipment maintenance prompt unit is used to output corresponding equipment maintenance prompts or cleaning prompts when the abnormality attribution type is non-material factors or attachment detachment.

[0069] In this embodiment, the source tracing analysis module is used to perform corresponding processing operations based on the attribution type output by the anomaly attribution judgment module. Raw material source tracing is only initiated when the anomaly is confirmed to originate from material factors; otherwise, only equipment maintenance prompts are output. The material source tracing execution unit is triggered when the anomaly attribution type is "true fluctuation in material moisture content," and is used to locate the source of the abnormal raw material. "True fluctuation in material moisture content" means that the moisture content of aggregates in a certain silo or a certain supply batch used in the current batch of concrete has undergone a real change. This change causes a mismatch between the actual moisture content of the material and the expected water consumption in the batching data, thereby triggering the source tracing requirement.

[0070] Specifically, the material traceability execution unit obtains the production batch number and timestamp at the time of the anomaly, and traces back the material dispensing sequence record of the batching system based on the timestamp. The material dispensing sequence record details the start time, end time, and flow rate of each aggregate bin in each production batch. Based on the time of the anomaly, the material traceability execution unit searches for all aggregate bins that are in the dispensing state at that time, and determines the target bin. After determining the target bin, the material traceability execution unit associates the batch of materials dispensed by the target bin at that time with the raw material supply batch according to the raw material entry registration record, tracing back to the specific raw material supplier, raw material origin, and entry date. Finally, it generates a raw material batch traceability result containing raw material supplier information, raw material origin, entry date, and the original inspection report number of the batch of raw materials, and marks and displays it through a digital twin visualization interface, providing a clear basis for the determination of responsibility for quality problems and subsequent handling.

[0071] The equipment maintenance prompt unit is triggered when the anomaly attribution type is either non-material factor or attached material detachment. When the anomaly attribution type is non-material factor, it indicates that the quality anomaly is not caused by the material itself, but by a change in the state of the mixer equipment. In this case, there is no need to trace the raw material source, and the equipment maintenance prompt unit outputs a corresponding equipment maintenance prompt, such as prompting to check or calibrate the transmission system. When the anomaly attribution type is attached material detachment, it indicates that the quality anomaly is caused by a sudden drop in the no-load signal due to the sudden detachment of attached material. The equipment maintenance prompt unit outputs a mixer cleaning prompt, prompting to clean the inside of the mixer to eliminate the interference of the detached attached material on the subsequent production process.

[0072] This embodiment achieves precise location of abnormal raw materials by using the material traceability execution unit to backtrack the material release sequence record based on the abnormal timestamp, match the target silo, and associate it with the raw material supply batch when confirming the true fluctuation of the material moisture content. This provides a clear basis for defining responsibility for quality issues. The equipment maintenance prompt unit outputs equipment maintenance or cleaning prompts when there are non-material factors or attachments falling off, enabling rapid handling of non-material factors. The entire traceability analysis module only initiates raw material traceability when the abnormality is confirmed to originate from material factors. This overcomes the problem of ineffective traceability caused by performing raw material traceability for all abnormalities in the existing technology, making the traceability process more targeted and efficient. At the same time, the digital twin visualization interface is used for annotation and display, which improves the visualization and operability of the traceability results.

[0073] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A concrete production quality traceability system based on digital twins, characterized in that, include: The data acquisition module is used to collect the operating signals of the mixer and the batching data of the current batch. The equipment status monitoring module is used to monitor the equipment status, including: The reference establishment unit is used to establish an operating reference based on the operating signals of the mixer when it is unloaded. The status feature generation unit is used to generate equipment status features characterizing the degree of internal deposits of the mixer based on the deviation between the no-load operation signal of the mixer when it is currently no-load and the operation reference. The distribution analysis unit is used to identify the uniformity of the distribution of the deposits inside the mixer based on the fluctuation pattern of the no-load operation signal, and to generate the distribution characteristics of the deposits based on the uniformity of the distribution. The detachment event identification unit is used to identify attachment detachment events based on the fluctuation pattern of the no-load operation signal. When a detachment event is identified, the detachment time is marked as a sudden change point in the equipment state. The power decoupling module is used to determine the real-time mixing power based on the operating signal of the mixer under load, compensate the real-time mixing power according to the equipment status characteristics, and make differentiated adjustments to the compensation according to the distribution characteristics of the attached material to obtain the decoupled power. An anomaly attribution determination module is used to determine the theoretical water consumption based on the decoupling power, and to determine the anomaly attribution type based on the changing trend characteristics of the theoretical water consumption and the decoupling power in consecutive batches, combined with the abrupt change point of the equipment state. The source tracing analysis module performs source tracing analysis based on the anomaly attribution type.

2. The concrete production quality traceability system based on digital twins according to claim 1, characterized in that, The benchmark establishment unit includes: The no-load gap identification subunit is used to collect no-load operation signals during the no-load operation period from the time the mixer finishes unloading one batch of concrete to the time it is ready to feed the next batch. The benchmark calculation subunit is used to perform statistical averaging on multiple consecutive no-load operation signals to obtain the operation benchmark.

3. The concrete production quality traceability system based on digital twins according to claim 1, characterized in that, The state feature generation unit includes: The state deviation calculation subunit is used to perform difference calculation between the current no-load operation signal of the mixer and the operation reference to obtain the state deviation value; The cumulative feature generation subunit is used to perform feature decomposition on the distribution pattern of the state deviation value in the time domain, extract the peak feature and the continuous feature in the distribution pattern, use the peak feature as a characterization of the local accumulation degree of the attachment, use the continuous feature as a characterization of the coverage of the attachment, and combine the peak feature and the continuous feature to generate the equipment state feature value at the current moment. The feature change rate calculation subunit is used to track the rate of change of the equipment status feature value in multiple consecutive no-load operation cycles and generate the deposit accumulation rate feature. The accumulation type identification subunit is used to classify the current accumulation state of the attachment into a gradual accumulation type or a step accumulation type based on the accumulation rate characteristics of the attachment.

4. The concrete production quality traceability system based on digital twins according to claim 1, characterized in that, The distribution analysis unit includes: The fluctuation pattern extraction subunit is used to extract the fluctuation amplitude sequence and fluctuation frequency sequence of the no-load operation signal in the time domain. The uniformity assessment subunit is used to determine whether the distribution of the deposits inside the mixer is uniform or non-uniform based on the degree of dispersion of the fluctuation amplitude sequence and the fluctuation frequency sequence, and when it is determined to be non-uniform, to determine the deposit aggregation area based on the peak position of the fluctuation amplitude sequence. A distribution feature generation subunit is used to generate attachment distribution features based on the determination result of the uniformity evaluation subunit.

5. The concrete production quality traceability system based on digital twins according to claim 1, characterized in that, The detachment event identification unit includes: The mutation detection subunit is used to perform time-domain mutation detection on the no-load operation signal. When the amplitude of the no-load operation signal decreases within a preset time window and the decrease is greater than the preset detachment threshold, it is determined that the attached material has detached. The mutation marker subunit is used to record the moment when the attachment detachment event occurs as the equipment state mutation point and mark the spatial location corresponding to the equipment state mutation point.

6. The concrete production quality traceability system based on digital twins according to claim 1, characterized in that, The power decoupling module includes: The real-time power determination unit is used to determine the real-time mixing power based on the torque current signal and speed signal of the mixer when it is running under load. The compensation amount determination unit is used to determine the basic power compensation amount based on the equipment state characteristic value; The differential adjustment unit is used to determine the corresponding compensation adjustment method based on the distribution pattern characterized by the distribution characteristics of the attached material; The decoupling execution unit is used to compensate the real-time stirring power based on the compensation adjustment method and the basic power compensation amount, thereby removing the power interference components caused by uneven distribution or accumulation of deposits to obtain decoupling power.

7. The concrete production quality traceability system based on digital twins according to claim 1, characterized in that, The anomaly attribution determination module includes: The theoretical water consumption determination unit is used to determine the theoretical water consumption based on the material moisture content corresponding to the decoupling power and the aggregate weighing value of the current mixing batch. The water volume comparison unit is used to compare the theoretical water volume with the actual water volume in the ingredient data and generate a water volume comparison result. The trend analysis unit is used to track the change trajectory of the decoupling power in multiple consecutive production batches and extract the trend characteristics of the change trajectory, which include continuous change characteristics or intermittent change characteristics. The mutation matching unit is used to match the time of the anomaly occurrence with the time axis of the device state mutation point to determine whether the time of the anomaly occurrence corresponds to the time of the attachment falling off. The attribution determination unit is used to determine the abnormal attribution type as material factor abnormality or non-material factor abnormality based on the water volume comparison results, trend characteristics and mutation matching results.

8. The concrete production quality traceability system based on digital twins according to claim 7, characterized in that, The trend analysis unit includes: The change trajectory tracking subunit is used to acquire the numerical sequence of the decoupling power in multiple consecutive production batches as the change trajectory. The continuous feature extraction subunit is used to extract continuous change features when the change trajectory changes continuously in the same direction within a preset number of batches of time windows and the change amplitude at the beginning and end of the time window is greater than a preset amplitude threshold. An intermittent feature extraction subunit is used to extract intermittent change features when the change trajectory undergoes at least one directional reversal within a preset number of batches of time windows and the change amplitude before and after the reversal is greater than a preset fluctuation threshold. The feature output subunit is used to take the extracted continuous change feature or the intermittent change feature as the trend feature.

9. The concrete production quality traceability system based on digital twins according to claim 7, characterized in that, The attribution determination unit includes: The first determination subunit is used to determine the anomaly attribution type as non-material factor anomaly when the water volume comparison result is consistent with the theoretical water volume and the actual water volume. The second determination subunit is used to determine the anomaly attribution type as attachment detachment when the water volume comparison result is inconsistent with the theoretical water volume and the time of the anomaly occurs matches the abrupt change point of the equipment state. The third determination subunit is used to determine the anomaly attribution type as the actual fluctuation of material moisture content when the water volume comparison result is inconsistent with the theoretical water volume and the time of the anomaly does not match the sudden change point of the equipment state.

10. The concrete production quality traceability system based on digital twins according to claim 9, characterized in that, The source tracing analysis module includes: The material traceability execution unit is used to trace back the material dispensing sequence record of the batching system according to the timestamp of the abnormality when the abnormality is attributed to the actual fluctuation of the material moisture content, determine the target silo that was in the dispensing state at the time of the abnormality, match the target silo to the corresponding raw material supply batch according to the raw material entry registration record, and output the raw material batch traceability result. The equipment maintenance prompt unit is used to output corresponding equipment maintenance prompts or cleaning prompts when the abnormality attribution type is non-material factors or attachment detachment.

Images