An automated state detection method and system for smart buckets

Abstract

This invention discloses an automated status detection method and system for smart smart buckets, belonging to the field of big data analysis technology. The invention constructs an analysis set by extracting historical transport pressure data and timestamps from the target smart smart bucket, and combines this with current transport data to filter high-frequency pressure values ​​and weighted fusion to obtain a detection threshold. This effectively filters out invalid pressure fluctuations caused by transport dynamic interference, providing a precise data benchmark for subsequent analysis. A neural network model is used to extract time-series features to generate pressure change trends, enabling batch analysis of smart smart buckets with the same attributes. This achieves accurate pressure trend prediction and unified batch detection, improving detection efficiency and consistency. Simultaneously, the longest historical transport time is combined with the maximum pressure value predicted by the neural network model for judgment and early warning, solving the problems of pressure data disturbance, inconsistent standards, and lag in traditional detection, ensuring the real-time and reliable status detection of smart smart buckets during transport.

Description

Technical Field

[0001] This invention relates to the field of big data analytics, specifically to an automated status detection method and system for a smart chip bucket. Background Technology

[0002] With its built-in chip enabling pressure sensing, vibration sensing, positioning, and data recording, the Smart Tank allows for visualized display of remaining liquid levels, automatic alerts to suppliers when levels are low, precise tracking of transport routes and severe collisions with timely alarms, and support for recycling, reuse, and historical loading record traceability. Combined with ton containers, it facilitates easy repackaging. It significantly improves the safety of liquid transportation, supply chain response efficiency, and resource recycling rate, solving the inefficiencies and delays of traditional containers in areas such as status monitoring, information traceability, and supply-demand coordination. It provides crucial support for the intelligent management of liquid materials and possesses significant practical value and industry-revolutionary significance.

[0003] During the service life of the smart bucket, its transportation process often faces complex dynamic disturbances, such as road bumps, sudden stops and starts of the transport vehicle, and centrifugal forces during turning, which cause the liquid inside the bucket to slosh violently. During this sloshing, the kinetic energy of the liquid is transferred to the space above the liquid surface, resulting in uneven and unsteady pressure fluctuations in that area. These fluctuations are not caused by changes in the actual state of the liquid inside the bucket, but rather originate from the dynamic disturbances during transportation. These unsteady pressure fluctuations directly interfere with the pressure detection accuracy of the smart bucket's built-in chip, causing the pressure data collected by the chip to contain a large amount of invalid interference information. This, in turn, affects the analysis and judgment of the true state of the liquid inside the bucket, restricting the reliability of the smart bucket's core functions such as pressure sensing and state monitoring in transportation scenarios. Therefore, there is an urgent need for an automated state detection method and system for smart buckets. Summary of the Invention

[0004] The purpose of this invention is to provide an automated status detection method and system for smart buckets, in order to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an automated state detection method applied to a smart battery, the automated state detection method comprising the following steps:

[0006] Step S1: Select any smart chip bucket as the research object, denoted as the target smart chip bucket, obtain the bucket internal pressure data and the timestamp data corresponding to the bucket internal pressure data collection generated during multiple historical transportation processes of the target smart chip bucket, and construct the transportation bucket internal pressure analysis set of the target smart chip bucket in the transportation state.

[0007] Step S1-1: Extract the historical transportation records of the target smart chip bucket, determine the start and end timestamps of each historical transportation process, calculate the time interval between the start and end timestamps, and obtain the transportation duration of the target smart chip bucket.

[0008] Step S1-2: For each historical transportation process, based on the start and end timestamps, extract the barrel pressure data and the corresponding timestamp data from the storage unit of the target smart barrel, archive and store them to build a transportation barrel pressure analysis set. The barrel pressure data is represented as the pressure value of the gas space above the liquid surface of the liquid inside the smart barrel.

[0009] By selecting a target smart bucket, the start and end timestamps of its historical multiple transportation processes are extracted to determine the transportation duration. Based on these start and end timestamps, the corresponding bucket pressure data and timestamp data are extracted from the storage unit and archived to construct a transportation bucket pressure analysis set. This provides targeted and complete historical data support for subsequent pressure time series analysis, threshold calculation, etc., ensuring that the data accurately corresponds to the transportation process.

[0010] Step S2: Real-time acquisition of the internal pressure data and timestamp data of the target smart chip barrel during the current transportation process; construct a pressure time series curve of the target smart chip barrel based on the timestamp data and internal pressure data; traverse and read the internal pressure data and timestamp data of the transportation barrel internal pressure analysis set to construct a reference pressure time series curve.

[0011] Step S2-1: During the current transportation of the target smart bucket, the built-in smart chip collects the bucket pressure data in real time. The smart chip has a built-in pressure sensor. The collection timestamp data corresponding to the pressure data in each bucket is recorded synchronously. The real-time collected bucket pressure data and the corresponding timestamp data are associated in the collection order. The timestamp data is selected as the horizontal axis to establish the x-axis, and the bucket pressure data is selected as the vertical axis to establish the y-axis. The pressure time series curve of the target smart bucket is generated based on the x-axis and y-axis.

[0012] Step S2-2: Traverse the historical transportation pressure data and the corresponding collection timestamp data of each transportation process stored in the transportation barrel pressure analysis set. Arrange the extracted barrel pressure data continuously according to the collection timestamp data in chronological order. Select the arranged transportation timestamp data as the horizontal axis to establish the x-axis and select the arranged barrel pressure data as the vertical axis to establish the y-axis. Construct a reference pressure time series curve of the target smart barrel based on the x-axis and y-axis.

[0013] By using a smart chip to collect real-time data on the pressure inside the target smart bucket during transportation, along with corresponding timestamp data, and constructing a pressure time-series curve, and by traversing historical data from the transportation bucket pressure analysis set and arranging them by time to construct a reference pressure time-series curve, we can provide intuitive charts showing how the pressure inside the bucket changes over time during current and historical transportation processes. This lays a visual data foundation for the subsequent extraction of pressure detection thresholds.

[0014] Step S3: Analyze and filter the pressure time series curve of the smart chip bucket to obtain the first pressure detection threshold; analyze and filter the reference pressure time series curve to obtain the second pressure detection threshold; use weighted fusion calculation to combine the first pressure detection threshold and the second pressure detection threshold to obtain the pressure detection threshold of the target smart chip bucket;

[0015] Step S3-1: Based on the pressure data values ​​inside the target smart chip bucket in the pressure time series curve, calculate the frequency of each pressure data value inside the bucket, and select the pressure data value inside the bucket with the highest frequency as the first pressure detection threshold of the target smart chip bucket.

[0016] Step S3-2: Based on the pressure data values ​​inside the bucket in the reference pressure time series curve of the target smart bucket, calculate the frequency of occurrence of each pressure data value inside the bucket, and select the pressure data value inside the bucket with the highest frequency of occurrence as the second pressure detection threshold of the target smart bucket.

[0017] Step S3-3: Set the weight coefficient of the first pressure detection threshold as α and the weight coefficient of the second pressure detection threshold as β, where the sum of α and β is 1; multiply the first pressure detection threshold by α to obtain the first weighted value; multiply the second pressure detection threshold by β to obtain the second weighted value; calculate the sum of the first weighted value and the second weighted value to obtain the pressure detection threshold of the target smart chip bucket.

[0018] The formula for calculating the pressure detection threshold of the target smart chip bucket is as follows:

[0019] ;

[0020] In the formula, P max P1 represents the first pressure detection threshold; P2 represents the second pressure detection threshold.

[0021] First, calculate the average value of all pressure data values ​​inside the target smart tank in the time-series curve graph, and record it as the current average pressure. Then, calculate the absolute difference between the current average pressure and the first pressure detection threshold, and record it as the first difference.

[0022] Simultaneously, the average value of all pressure data in the tank in the reference pressure time series curve is calculated and recorded as the historical pressure average value. The absolute difference between the historical pressure average value and the second pressure detection threshold is calculated and recorded as the second difference value.

[0023] The first difference and the second difference are added together to obtain the total difference. Finally, the first difference is divided by the total difference to obtain the weight coefficient α of the first pressure detection threshold. The second difference is divided by the total difference to obtain the weight coefficient β of the second pressure detection threshold, and α+β=1 is satisfied.

[0024] By calculating the frequency of pressure data values ​​in each tank from the pressure time-series curves of the target smart tank and the reference pressure time-series curves, and selecting the highest frequency value as the first pressure detection threshold and the second pressure detection threshold, and then obtaining the pressure detection threshold through weighted fusion, a clear judgment benchmark is provided for subsequent monitoring and screening of tank pressure data.

[0025] Step S4: Monitor the internal pressure data of the target smart chip bucket according to the pressure time series curve of the pressure detection threshold, filter out the internal pressure data of the target smart chip bucket that meet the monitoring conditions, and the timestamp data corresponding to the internal pressure data, and construct the internal pressure data prediction and analysis set of the target smart chip bucket, denoted as the internal pressure prediction set.

[0026] Step S4-1: Traverse all the pressure data inside the target smart chip bucket and the timestamp data corresponding to the pressure data inside the bucket contained in the pressure time series curve. Sequentially determine the relationship between each pressure data value inside the bucket and the pressure detection threshold. Select the pressure data inside the bucket whose pressure value does not exceed the pressure detection threshold according to the relationship.

[0027] Step S4-2: For the internal pressure data of the target smart chip bucket that is greater than the pressure detection threshold, replace the internal pressure data of the corresponding timestamp with the pressure detection threshold to obtain the alternative internal pressure data of the target smart chip bucket; archive and store the selected internal pressure data and the alternative internal pressure data to construct the internal pressure data prediction and analysis set of the target smart chip bucket, which is denoted as the internal pressure prediction set.

[0028] By monitoring the pressure time-series curve of the target smart bucket based on the pressure detection threshold, the data within the bucket that does not exceed the threshold is filtered out, and the data that exceeds the threshold is replaced with the pressure detection threshold to obtain alternative data. This alternative data is then archived to construct a bucket pressure prediction set, providing standardized pressure data and corresponding timestamp data for subsequent neural network model processing, ensuring the rationality and usability of the prediction analysis data.

[0029] Step S5: Use a neural network model to process the predicted pressure set inside the target smart chip bucket to obtain the trend of pressure change data inside the target smart chip bucket during the current transportation process.

[0030] The internal pressure prediction set of the target smart chip bucket, arranged chronologically by timestamp, is used as input to a neural network model. The neural network model comprises an input layer, a hidden layer, and an output layer. The input layer performs format conversion on the internal pressure data and corresponding timestamp data. The hidden layer performs feature extraction and temporal correlation operations on the converted data. The output layer outputs the predicted internal pressure values ​​for subsequent preset time nodes during the current transportation process. Based on the sequence of predicted internal pressure values ​​for subsequent time nodes arranged chronologically, the changing trend of the target smart chip bucket's internal pressure data during the current transportation process is predicted.

[0031] The neural network model used in step S5 is LSTM, which can specifically process the bucket pressure prediction set data arranged in chronological order of timestamps. It captures the long-term and short-term dependencies of pressure data over time by using the temporal correlation operation of the hidden layer, accurately extracts temporal features, and then outputs the bucket pressure prediction value at the subsequent preset time node.

[0032] By inputting the barrel pressure data and corresponding timestamp data arranged sequentially in the barrel pressure prediction set into the neural network model, and performing input layer format conversion, hidden layer feature extraction and temporal correlation operation, and outputting the predicted values ​​of subsequent time nodes in the output layer, the trend of the barrel pressure data change of the target smart barrel during current transportation is obtained, providing a trend basis for pressure change for subsequent automated status detection.

[0033] Step S6: Simultaneously use steps S1 to S5 to process the smart barrels of the same attribute during the current transportation process, obtain the trend of pressure data change of each smart barrel during the transportation of the same batch, and perform automated status detection on the smart barrels of the same attribute.

[0034] Step S6-1: Select smart buckets with the same attributes as the target smart bucket during the current transportation process to form a set of smart buckets in the same batch. The smart buckets with the same attributes are smart buckets with the same type and specifications of liquid and the same bucket model. For each smart bucket in the set of smart buckets in the same batch, execute steps S1 to S5 respectively to obtain the trend of the change of the bucket pressure data in the bucket during the current transportation process.

[0035] Step S6-2: Analyze and extract the longest historical transportation time of the target smart chip bucket from step S1-1; analyze and calculate the transportation time already transported in the current transportation process to obtain the maximum remaining transportation time of the target smart chip bucket; input the maximum remaining transportation time into the neural network model of step S5 to obtain the predicted value of the bucket pressure when the maximum remaining transportation time ends, and record it as the maximum pressure value at the end.

[0036] Step S6-3: Detect the internal pressure threshold of the smart bucket according to the preset internal pressure threshold. Mark the smart bucket with the maximum cutoff pressure value exceeding the internal pressure threshold as abnormal and issue a warning signal; mark the smart bucket with the maximum cutoff pressure value not exceeding the internal pressure threshold as normal, and complete the automatic status detection of the smart bucket.

[0037] By selecting smart buckets with the same attributes to form a batch, and performing steps S1 to S5 on each smart bucket to obtain the pressure change trend, the maximum remaining transportation time is obtained by combining the longest historical transportation time with the current transportation time, and the maximum remaining transportation time is input into the neural network model to obtain the cutoff maximum pressure value. Then, it is compared with the preset bucket pressure threshold to mark the status, thus realizing batch automated status detection of smart buckets in the same batch. This ensures the consistency and pertinence of status judgment, and can identify abnormal statuses and issue early warnings in a timely manner.

[0038] Furthermore, an automated status detection system for a smart bucket includes a historical data acquisition module, a time series diagram construction module, a detection threshold calculation module, a predictive data filtering module, a pressure trend prediction module, and a batch status detection module.

[0039] The historical data acquisition module is used to acquire the internal pressure data and corresponding timestamp data of the target smart chip bucket during historical transportation and construct a transportation internal pressure analysis set; the time series graph construction module is used to construct a pressure time series curve graph of the target smart chip bucket during current transportation and historical transportation; the detection threshold calculation module is used to calculate the pressure detection threshold of the target smart chip bucket; the prediction data filtering module is used to construct a predicted internal pressure set of the target smart chip bucket; the pressure trend prediction module is used to obtain the changing trend of the internal pressure data of the target smart chip bucket during current transportation; the batch status detection module is used to perform automated status detection on the same batch of smart chip buckets;

[0040] The output of the historical data acquisition module is electrically connected to the input of the time series diagram construction module; the output of the time series diagram construction module is electrically connected to the input of the detection threshold calculation module; the output of the detection threshold calculation module is electrically connected to the input of the prediction data filtering module; the output of the prediction data filtering module is electrically connected to the input of the pressure trend prediction module; and the output of the pressure trend prediction module is electrically connected to the input of the batch state detection module.

[0041] The historical data acquisition module includes a historical transportation record extraction unit and a historical pressure data archiving unit. The historical transportation record extraction unit is used to extract the historical transportation records of the target smart chip bucket and determine the start and end timestamps and transportation duration of each transportation. The historical pressure data archiving unit is used to extract the pressure data inside the bucket and the corresponding timestamp data based on the start and end timestamps of the historical transportation and archive and store them.

[0042] The time series diagram construction module includes a current pressure time series diagram construction unit and a reference pressure time series diagram construction unit; the current pressure time series diagram construction unit is used to construct a pressure time series curve based on the current transported barrel pressure data and corresponding timestamp data; the reference pressure time series diagram construction unit is used to construct a reference pressure time series curve based on the historical transported barrel pressure data and corresponding timestamp data.

[0043] The detection threshold calculation module includes a basic threshold extraction unit and a threshold weighted fusion unit; the basic threshold extraction unit is used to extract the most frequently occurring pressure value from the current and reference pressure time series curves; the threshold weighted fusion unit is used to obtain the pressure detection threshold of the target smart chip bucket through weighted fusion calculation;

[0044] The predictive data filtering module includes a compliance pressure filtering unit and an over-threshold data processing unit. The compliance pressure filtering unit is used to filter out the bucket pressure data and corresponding timestamp data that do not exceed the pressure detection threshold. The over-threshold data processing unit is used to replace the bucket pressure data that exceeds the threshold with the pressure detection threshold and archive it together with the compliance data.

[0045] The pressure trend prediction module includes a neural network data processing unit and a pressure trend generation unit. The neural network data processing unit is used to input the data in the barrel pressure prediction set into the neural network model for format conversion, feature extraction, and time series correlation calculation. The pressure trend generation unit is used to generate the barrel pressure data change trend based on the pressure prediction values ​​of subsequent time nodes output by the neural network model.

[0046] The batch status detection module includes a cutoff pressure prediction unit and a status threshold judgment unit; the cutoff pressure prediction unit is used to predict the pressure value inside the barrel at the end of the transportation based on the longest historical transportation time and the current transportation time; the status threshold judgment unit is used to compare the cutoff pressure prediction value with the preset barrel pressure threshold and mark the status of the smart barrel.

[0047] Compared with the prior art, the beneficial effects of the present invention are:

[0048] 1. This invention constructs an analysis set by extracting the internal pressure data and timestamps of the target smart chip bucket during historical transportation, and generates a pressure time series curve and a reference curve by combining the current transportation data. High-frequency pressure values ​​are screened and weighted to obtain the detection threshold, effectively filtering out invalid pressure fluctuations caused by transportation dynamic interference, providing accurate data benchmarks for subsequent analysis, and solving the problem of pressure data being affected by disturbances in traditional detection.

[0049] 2. This invention filters compliant pressure data based on detection thresholds, replaces data exceeding the threshold to construct a prediction set, and uses a neural network model to extract time-series features to generate pressure change trends. The above process is then executed on batches of smart chip buckets with the same attributes, achieving accurate prediction of pressure trends and unified batch detection, improving detection efficiency and consistency, and overcoming the shortcomings of inconsistent batch detection standards in traditional methods.

[0050] 3. This invention combines the longest historical transportation time with the current transportation time to obtain the maximum remaining transportation time, uses a neural network model to predict the maximum pressure value before the deadline, compares it with a preset threshold to complete the status judgment and issue an early warning, forming a complete process from data collection and processing to status warning, ensuring the real-time and reliable status detection of the smart bucket during transportation, and solving the problem of lagging detection in traditional methods. Attached Figure Description

[0051] Figure 1 This is a flowchart illustrating an automated status detection method for a smart bucket according to the present invention.

[0052] Figure 2 This is a schematic diagram of an automated status detection system for a smart battery according to the present invention. Detailed Implementation

[0053] 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.

[0054] Example 1: As Figure 1 As shown, the present invention provides a technical solution, an automated state detection method applied to a smart bucket, the automated state detection method comprising the following steps:

[0055] Step S1: Select any smart chip bucket as the research object, denoted as the target smart chip bucket, obtain the bucket internal pressure data and the timestamp data corresponding to the bucket internal pressure data collection generated during multiple historical transportation processes of the target smart chip bucket, and construct the transportation bucket internal pressure analysis set of the target smart chip bucket in the transportation state.

[0056] Step S1-1: Extract the historical transportation records of the target smart chip bucket, determine the start and end timestamps of each historical transportation process, calculate the time interval between the start and end timestamps, and obtain the transportation duration of the target smart chip bucket.

[0057] Step S1-2: For each historical transportation process, based on the start and end timestamps, extract the barrel pressure data and the corresponding timestamp data from the storage unit of the target smart barrel, archive and store them to build a transportation barrel pressure analysis set. The barrel pressure data is represented as the pressure value of the gas space above the liquid surface of the liquid inside the smart barrel.

[0058] In practical implementation, after selecting the target smart bucket, it is necessary to extract the complete historical transportation records from its storage unit. Each transportation cycle is accurately defined by the start and end timestamps to ensure that the extracted bucket pressure data and corresponding timestamps strictly fall within each transportation cycle. This ensures that the pressure analysis set inside the transportation bucket can truly reflect the pressure characteristics of different transportation processes and provide a reliable historical benchmark for subsequent analysis.

[0059] Step S2: Real-time acquisition of the internal pressure data and timestamp data of the target smart chip barrel during the current transportation process; construct a pressure time series curve of the target smart chip barrel based on the timestamp data and internal pressure data; traverse and read the internal pressure data and timestamp data of the transportation barrel internal pressure analysis set to construct a reference pressure time series curve.

[0060] Step S2-1: During the current transportation of the target smart bucket, the built-in smart chip collects the bucket pressure data in real time. The smart chip has a built-in pressure sensor. The collection timestamp data corresponding to the pressure data in each bucket is recorded synchronously. The real-time collected bucket pressure data and the corresponding timestamp data are associated in the collection order. The timestamp data is selected as the horizontal axis to establish the x-axis, and the bucket pressure data is selected as the vertical axis to establish the y-axis. The pressure time series curve of the target smart bucket is generated based on the x-axis and y-axis.

[0061] Step S2-2: Traverse the historical transportation pressure data and the corresponding collection timestamp data of each transportation process stored in the transportation barrel pressure analysis set. Arrange the extracted barrel pressure data continuously according to the collection timestamp data in chronological order. Select the arranged transportation timestamp data as the horizontal axis to establish the x-axis and select the arranged barrel pressure data as the vertical axis to establish the y-axis. Construct a reference pressure time series curve of the target smart barrel based on the x-axis and y-axis.

[0062] In practical implementation, the construction of the current pressure time series curve needs to rely on the real-time acquisition function of the smart chip to ensure the synchronization of pressure data and timestamps, and the acquisition frequency should be adapted to the dynamic changes of the transportation scenario; the reference pressure time series curve needs to arrange the historical transportation data in chronological order to eliminate the interference of time intervals between different transportation cycles, so that the time axis scales of the two curves are consistent, making it easy to intuitively compare the pressure change patterns.

[0063] Step S3: Analyze and filter the pressure time series curve of the smart chip bucket to obtain the first pressure detection threshold; analyze and filter the reference pressure time series curve to obtain the second pressure detection threshold; use weighted fusion calculation to combine the first pressure detection threshold and the second pressure detection threshold to obtain the pressure detection threshold of the target smart chip bucket;

[0064] Step S3-1: Based on the pressure data values ​​inside the target smart chip bucket in the pressure time series curve, calculate the frequency of each pressure data value inside the bucket, and select the pressure data value inside the bucket with the highest frequency as the first pressure detection threshold of the target smart chip bucket.

[0065] Step S3-2: Based on the pressure data values ​​inside the bucket in the reference pressure time series curve of the target smart bucket, calculate the frequency of occurrence of each pressure data value inside the bucket, and select the pressure data value inside the bucket with the highest frequency of occurrence as the second pressure detection threshold of the target smart bucket.

[0066] Step S3-3: Set the weight coefficient of the first pressure detection threshold as α and the weight coefficient of the second pressure detection threshold as β, where the sum of α and β is 1; multiply the first pressure detection threshold by α to obtain the first weighted value; multiply the second pressure detection threshold by β to obtain the second weighted value; calculate the sum of the first weighted value and the second weighted value to obtain the pressure detection threshold of the target smart chip bucket.

[0067] In practical implementation, the extraction of the first and second pressure detection thresholds must be based on a sufficient amount of pressure data. Frequency statistics should cover the complete cycle of the pressure time series curve to avoid high-frequency values ​​deviating from the actual stable pressure due to insufficient data. During weighted fusion, α can focus on the real-time characteristics of the current transportation, while β can focus on the universality of historical data. Through a reasonable ratio of the two, the final pressure detection threshold can reflect the current state and also draw on historical experience.

[0068] Step S4: Monitor the internal pressure data of the target smart chip bucket according to the pressure time series curve of the pressure detection threshold, filter out the internal pressure data of the target smart chip bucket that meet the monitoring conditions, and the timestamp data corresponding to the internal pressure data, and construct the internal pressure data prediction and analysis set of the target smart chip bucket, denoted as the internal pressure prediction set.

[0069] Step S4-1: Traverse all the pressure data inside the target smart chip bucket and the timestamp data corresponding to the pressure data inside the bucket contained in the pressure time series curve. Sequentially determine the relationship between each pressure data value inside the bucket and the pressure detection threshold. Select the pressure data inside the bucket whose pressure value does not exceed the pressure detection threshold according to the relationship.

[0070] Step S4-2: For the internal pressure data of the target smart chip bucket that is greater than the pressure detection threshold, replace the internal pressure data of the corresponding timestamp with the pressure detection threshold to obtain the alternative internal pressure data of the target smart chip bucket; archive and store the selected internal pressure data and the alternative internal pressure data to construct the internal pressure data prediction and analysis set of the target smart chip bucket, which is denoted as the internal pressure prediction set.

[0071] In practice, screening compliant stress data requires judging the data in the stress time series curve one by one to ensure that no stress value at any time point is missed. When processing data exceeding the threshold, the original data of the corresponding timestamp must be replaced with the stress detection threshold to maintain the continuity of the time series. This ensures that the constructed bucket stress prediction set can filter out abnormal interference while retaining the true time series correlation, providing high-quality data for subsequent predictions.

[0072] Step S5: Use a neural network model to process the predicted pressure set inside the target smart chip bucket to obtain the trend of pressure change data inside the target smart chip bucket during the current transportation process.

[0073] The internal pressure prediction set of the target smart chip bucket, arranged chronologically by timestamp, is used as input to a neural network model. The neural network model comprises an input layer, a hidden layer, and an output layer. The input layer performs format conversion on the internal pressure data and corresponding timestamp data. The hidden layer performs feature extraction and temporal correlation operations on the converted data. The output layer outputs the predicted internal pressure values ​​for subsequent preset time nodes during the current transportation process. Based on the sequence of predicted internal pressure values ​​for subsequent time nodes arranged chronologically, the changing trend of the target smart chip bucket's internal pressure data during the current transportation process is predicted.

[0074] In practical implementation, the inputs of the neural network model must be strictly arranged in timestamp order, and the format conversion of the input layer should be adapted to the model's requirements for data dimensions; the feature extraction of the hidden layer should focus on capturing the trend features and time-series correlation patterns of pressure changes over time, and the number of preset time nodes in the output layer should be set in combination with the accuracy requirements of transportation monitoring so that the generated pressure change trend can accurately reflect the subsequent pressure development trend.

[0075] Step S6: Simultaneously use steps S1 to S5 to process the smart barrels of the same attribute during the current transportation process, obtain the trend of pressure data change of each smart barrel during the transportation of the same batch, and perform automated status detection on the smart barrels of the same attribute.

[0076] Step S6-1: Select smart buckets with the same attributes as the target smart bucket during the current transportation process to form a set of smart buckets in the same batch. The smart buckets with the same attributes are smart buckets with the same type and specifications of liquid and the same bucket model. For each smart bucket in the set of smart buckets in the same batch, execute steps S1 to S5 respectively to obtain the trend of the change of the bucket pressure data in the bucket during the current transportation process.

[0077] Step S6-2: Analyze and extract the longest historical transportation time of the target smart chip bucket from step S1-1; analyze and calculate the transportation time already transported in the current transportation process to obtain the maximum remaining transportation time of the target smart chip bucket; input the maximum remaining transportation time into the neural network model of step S5 to obtain the predicted value of the bucket pressure when the maximum remaining transportation time ends, and record it as the maximum pressure value at the end.

[0078] Step S6-3: Detect the internal pressure threshold of the smart bucket according to the preset internal pressure threshold. Mark the smart bucket with the maximum cutoff pressure value exceeding the internal pressure threshold as abnormal and issue a warning signal; mark the smart bucket with the maximum cutoff pressure value not exceeding the internal pressure threshold as normal, and complete the automatic status detection of the smart bucket.

[0079] In practice, the selection of smart tanks in the same batch must strictly adhere to the consistency standards of the type, specifications, and tank model of the liquid being loaded to ensure comparability among individuals within the batch. When calculating the maximum remaining transport time, the time units of the longest historical transport time and the current transport time must be accurately matched. The neural network model's prediction of the maximum cutoff pressure value must be based on the pressure change trend of each smart tank. When comparing thresholds, a unified preset tank pressure standard must be used to ensure the fairness and accuracy of status detection.

[0080] Example 2, as Figure 2 As shown, the present invention provides an automated status detection system for a smart bucket, which includes a historical data acquisition module, a time series diagram construction module, a detection threshold calculation module, a predictive data filtering module, a pressure trend prediction module, and a batch status detection module.

[0081] The historical data acquisition module is used to acquire the internal pressure data and corresponding timestamp data of the target smart chip bucket during historical transportation and construct a transportation internal pressure analysis set; the time series graph construction module is used to construct a pressure time series curve graph of the target smart chip bucket during current transportation and historical transportation; the detection threshold calculation module is used to calculate the pressure detection threshold of the target smart chip bucket; the prediction data filtering module is used to construct a predicted internal pressure set of the target smart chip bucket; the pressure trend prediction module is used to obtain the changing trend of the internal pressure data of the target smart chip bucket during current transportation; the batch status detection module is used to perform automated status detection on the same batch of smart chip buckets;

[0082] The output of the historical data acquisition module is electrically connected to the input of the time series diagram construction module; the output of the time series diagram construction module is electrically connected to the input of the detection threshold calculation module; the output of the detection threshold calculation module is electrically connected to the input of the prediction data filtering module; the output of the prediction data filtering module is electrically connected to the input of the pressure trend prediction module; and the output of the pressure trend prediction module is electrically connected to the input of the batch state detection module.

[0083] The historical data acquisition module includes a historical transportation record extraction unit and a historical pressure data archiving unit. The historical transportation record extraction unit is used to extract the historical transportation records of the target smart chip bucket and determine the start and end timestamps and transportation duration of each transportation. The historical pressure data archiving unit is used to extract the pressure data inside the bucket and the corresponding timestamp data based on the start and end timestamps of the historical transportation and archive and store them.

[0084] The time series diagram construction module includes a current pressure time series diagram construction unit and a reference pressure time series diagram construction unit; the current pressure time series diagram construction unit is used to construct a pressure time series curve based on the current transported barrel pressure data and corresponding timestamp data; the reference pressure time series diagram construction unit is used to construct a reference pressure time series curve based on the historical transported barrel pressure data and corresponding timestamp data.

[0085] The detection threshold calculation module includes a basic threshold extraction unit and a threshold weighted fusion unit; the basic threshold extraction unit is used to extract the most frequently occurring pressure value from the current and reference pressure time series curves; the threshold weighted fusion unit is used to obtain the pressure detection threshold of the target smart chip bucket through weighted fusion calculation;

[0086] The predictive data filtering module includes a compliance pressure filtering unit and an over-threshold data processing unit. The compliance pressure filtering unit is used to filter out the bucket pressure data and corresponding timestamp data that do not exceed the pressure detection threshold. The over-threshold data processing unit is used to replace the bucket pressure data that exceeds the threshold with the pressure detection threshold and archive it together with the compliance data.

[0087] The pressure trend prediction module includes a neural network data processing unit and a pressure trend generation unit. The neural network data processing unit is used to input the data in the barrel pressure prediction set into the neural network model for format conversion, feature extraction, and time series correlation calculation. The pressure trend generation unit is used to generate the barrel pressure data change trend based on the pressure prediction values ​​of subsequent time nodes output by the neural network model.

[0088] The batch status detection module includes a cutoff pressure prediction unit and a status threshold judgment unit; the cutoff pressure prediction unit is used to predict the pressure value inside the barrel at the end of the transportation based on the longest historical transportation time and the current transportation time; the status threshold judgment unit is used to compare the cutoff pressure prediction value with the preset barrel pressure threshold and mark the status of the smart barrel.

[0089] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of equivalents of the claims be included within the present invention.

Claims

1. An automated state detection method for a smart battery, characterized in that: The automated state detection method includes the following steps: Step S1: Select any smart chip bucket as the research object, denoted as the target smart chip bucket, obtain the bucket internal pressure data and the timestamp data corresponding to the bucket internal pressure data collection generated during multiple historical transportation processes of the target smart chip bucket, and construct the transportation bucket internal pressure analysis set of the target smart chip bucket in the transportation state. Step S2: Real-time acquisition of the internal pressure data and timestamp data of the target smart chip barrel during the current transportation process; construct a pressure time series curve of the target smart chip barrel based on the timestamp data and internal pressure data; traverse and read the internal pressure data and timestamp data of the transportation barrel internal pressure analysis set to construct a reference pressure time series curve. Step S3: Analyze the pressure data values ​​inside the target smart chip bucket based on the pressure time-series curves of the target and reference pressure time-series curves, calculate the frequency of each pressure data value, and select the pressure data value with the highest frequency in the pressure time-series curve as the first pressure detection threshold of the target smart chip bucket; select the pressure data value with the highest frequency in the reference pressure time-series curve as the second pressure detection threshold of the target smart chip bucket; use weighted fusion calculation to combine the first and second pressure detection thresholds to obtain the pressure detection threshold of the target smart chip bucket. Step S4: Monitor the internal pressure data of the target smart chip bucket according to the pressure time series curve of the pressure detection threshold, filter out the internal pressure data of the target smart chip bucket that meet the monitoring conditions, and the timestamp data corresponding to the internal pressure data, and construct the internal pressure data prediction and analysis set of the target smart chip bucket, denoted as the internal pressure prediction set. Step S5: Use a neural network model to process the predicted pressure set inside the target smart chip bucket to obtain the trend of pressure change data inside the target smart chip bucket during the current transportation process. Step S6: Simultaneously use steps S1 to S5 to process the smart barrels of the same attribute during the current transportation process, obtain the trend of pressure data change of each smart barrel during the transportation of the same batch, and perform automated status detection on the smart barrels of the same attribute.

2. The automated state detection method for a smart battery according to claim 1, characterized in that: The specific steps of step S1 are as follows: Step S1-1: Extract the historical transportation records of the target smart chip bucket, determine the start and end timestamps of each historical transportation process, calculate the time interval between the start and end timestamps, and obtain the transportation duration of the target smart chip bucket. Step S1-2: For each historical transportation process, based on the start and end timestamps, extract the internal pressure data and the corresponding timestamp data of the internal pressure data collection from the storage unit of the target smart chip bucket, archive and store them to construct a transportation bucket internal pressure analysis set. The internal pressure data represents the pressure value of the gas space above the liquid surface inside the smart chip bucket.

3. The automated state detection method for a smart battery according to claim 2, characterized in that: The specific steps of step S2 are as follows: Step S2-1: During the current transportation of the target smart bucket, the built-in smart chip collects the bucket pressure data in real time. The smart chip has a built-in pressure sensor. The collection timestamp data corresponding to the pressure data in each bucket is recorded synchronously. The real-time collected bucket pressure data and the corresponding timestamp data are associated in the collection order. The timestamp data is selected as the horizontal axis to establish the x-axis, and the bucket pressure data is selected as the vertical axis to establish the y-axis. The pressure time series curve of the target smart bucket is generated based on the x-axis and y-axis. Step S2-2: Traverse the historical transport pressure data and corresponding collection timestamp data of each transport process stored in the transport barrel pressure analysis set. Arrange the extracted barrel pressure data continuously according to the collection timestamp data in chronological order. Select the arranged transport timestamp data as the horizontal axis to establish the x-axis, and select the arranged barrel pressure data as the vertical axis to establish the y-axis. Based on the x-axis and y-axis, construct a reference pressure time series curve of the target smart barrel.

4. The automated state detection method for a smart battery according to claim 3, characterized in that: In step S3, the weighting coefficient of the first pressure detection threshold is set as α, and the weighting coefficient of the second pressure detection threshold is set as β, where the sum of α and β is 1; the first pressure detection threshold is multiplied by α to obtain the first weighted value; the second pressure detection threshold is multiplied by β to obtain the second weighted value; the sum of the first weighted value and the second weighted value is calculated to obtain the pressure detection threshold of the target smart chip bucket.

5. The automated state detection method for a smart battery according to claim 4, characterized in that: The specific steps of step S4 are as follows: Step S4-1: Traverse all the pressure data inside the target smart chip bucket and the timestamp data corresponding to the pressure data inside the bucket contained in the pressure time series curve. Sequentially determine the relationship between each pressure data value inside the bucket and the pressure detection threshold. Select the pressure data inside the bucket whose pressure value does not exceed the pressure detection threshold according to the relationship. Step S4-2: For the internal pressure data of the target smart chip bucket that is greater than the pressure detection threshold, replace the internal pressure data of the corresponding timestamp with the pressure detection threshold to obtain the alternative internal pressure data of the target smart chip bucket; archive and store the selected internal pressure data and the alternative internal pressure data to construct the internal pressure data prediction and analysis set of the target smart chip bucket, denoted as the internal pressure prediction set.

6. The automated state detection method for a smart battery according to claim 5, characterized in that: In step S5, the bucket pressure prediction set of the target smart bucket, arranged in chronological order, is used as input to the neural network model. The neural network model includes an input layer, a hidden layer, and an output layer. The input layer is used to convert the format of the bucket pressure data and the corresponding timestamp data. The hidden layer is used to extract features and perform temporal correlation operations on the transformed data; the output layer is used to output the predicted barrel pressure values ​​for subsequent preset number of time nodes in the current transportation process; based on the sequence of barrel pressure prediction values ​​for subsequent time nodes in chronological order, the trend of barrel pressure data change of the target smart barrel in the current transportation process is predicted.

7. The automated state detection method for a smart battery according to claim 6, characterized in that: The specific steps of step S6 are as follows: Step S6-1: Select smart buckets with the same attributes as the target smart bucket during the current transportation process to form a set of smart buckets in the same batch. The smart buckets with the same attributes are smart buckets with the same type and specifications of liquid and the same bucket model. For each smart bucket in the set of smart buckets in the same batch, execute steps S1 to S5 respectively to obtain the trend of the change of the bucket pressure data in the bucket during the current transportation process. Step S6-2: Analyze and extract the longest historical transportation time of the target smart chip bucket from step S1-1; Analyze and calculate the transport time already taken in the current transport process to obtain the maximum remaining transport time of the target smart chip bucket; input the maximum remaining transport time into the neural network model in step S5 to obtain the predicted pressure value inside the bucket when the maximum remaining transport time ends, and record it as the maximum pressure value at the end; Step S6-3: Detect the internal pressure threshold of the smart bucket according to the preset internal pressure threshold, mark the smart bucket with the maximum cutoff pressure value exceeding the internal pressure threshold as an abnormal state, and issue a warning signal; The smart bucket is marked as normal if the maximum cutoff pressure value does not exceed the internal pressure threshold, thus completing the automated status detection of the smart bucket.

8. An automated state detection system for a smart battery, comprising the automated state detection method for a smart battery as described in any one of claims 1-7, characterized in that: The automated condition detection system includes a historical data acquisition module, a time series diagram construction module, a detection threshold calculation module, a predictive data filtering module, a pressure trend prediction module, and a batch condition detection module. The historical data acquisition module is used to acquire the internal pressure data and corresponding timestamp data of the target smart chip bucket during historical transportation and construct a transportation internal pressure analysis set; the time series graph construction module is used to construct the pressure time series curve graph of the target smart chip bucket during current transportation and historical transportation; the detection threshold calculation module is used to calculate the pressure detection threshold of the target smart chip bucket; the prediction data filtering module is used to construct the internal pressure prediction set of the target smart chip bucket; the pressure trend prediction module is used to obtain the changing trend of the internal pressure data of the target smart chip bucket during current transportation; and the batch status detection module is used to perform automated status detection on the same batch of smart chip buckets.

9. An automated status detection system for a smart battery according to claim 8, characterized in that: The historical data acquisition module includes a historical transportation record extraction unit and a historical pressure data archiving unit. The historical transportation record extraction unit is used to extract the historical transportation records of the target smart chip bucket and determine the start and end timestamps and transportation duration of each transportation. The historical pressure data archiving unit is used to extract the pressure data inside the bucket and the corresponding timestamp data based on the start and end timestamps of the historical transportation and archive and store them. The time series diagram construction module includes a current pressure time series diagram construction unit and a reference pressure time series diagram construction unit; the current pressure time series diagram construction unit is used to construct a pressure time series curve based on the current transported barrel pressure data and corresponding timestamp data; the reference pressure time series diagram construction unit is used to construct a reference pressure time series curve based on the historical transported barrel pressure data and corresponding timestamp data. The detection threshold calculation module includes a basic threshold extraction unit and a threshold weighted fusion unit; the basic threshold extraction unit is used to extract the most frequently occurring pressure value from the current and reference pressure time series curves; the threshold weighted fusion unit is used to obtain the pressure detection threshold of the target smart bucket through weighted fusion calculation.

10. An automated status detection system for a smart battery according to claim 8, characterized in that: The predictive data filtering module includes a compliance pressure filtering unit and an over-threshold data processing unit. The compliance pressure filtering unit is used to filter out the bucket pressure data and corresponding timestamp data that do not exceed the pressure detection threshold. The over-threshold data processing unit is used to replace the bucket pressure data that exceeds the threshold with the pressure detection threshold and archive it together with the compliance data. The pressure trend prediction module includes a neural network data processing unit and a pressure trend generation unit. The neural network data processing unit is used to input the data in the pressure prediction set in the bucket into the neural network model for format conversion, feature extraction, and time series correlation calculation. The pressure trend generation unit is used to generate the pressure data change trend in the bucket based on the pressure prediction values ​​of subsequent time nodes output by the neural network model. The batch status detection module includes a cutoff pressure prediction unit and a status threshold judgment unit; the cutoff pressure prediction unit is used to predict the pressure value inside the barrel at the end of the transportation based on the longest historical transportation time and the current transportation time; the status threshold judgment unit is used to compare the cutoff pressure prediction value with the preset barrel pressure threshold and mark the status of the smart barrel.

Images