A cooling system pressure regulation method and device based on multi-source data fusion

Abstract

The application discloses a cooling system pressure regulation method and device based on multi-source data fusion, wherein the method comprises the following steps: collecting multi-source data from a cooling system by a sensor and performing fusion processing to generate a standardized sensor data set; calculating a comprehensive system pressure value by using a data fusion technology according to the standardized sensor data set to obtain a fusion pressure data set; determining a system pressure state by threshold comparison based on the fusion pressure data set; obtaining channel flow data and adjusting the valve opening degree based on the system pressure state to obtain an optimized regulation result; transmitting the optimized regulation result to a pressure maintaining system and controlling the valve flow by using a pressure stabilizing adjustment algorithm to determine a pressure balance index. Therefore, the cooling system pressure regulation method based on multi-source data fusion can dynamically respond to pressure changes, stably maintain the system pressure and effectively prevent the gas blocking phenomenon.

Description

Technical Field

[0001] This invention relates to the field of engine cooling system control technology, and in particular to a cooling system pressure regulation method and device based on multi-source data fusion. Background Technology

[0002] As a key subsystem for the operation of internal combustion engines, the pressure stability of the engine cooling system directly affects the engine's thermal management efficiency and service life. In heavy-duty applications such as commercial vehicles, the cooling system faces more complex pressure regulation challenges: on the one hand, it needs to cope with the intense heat exchange process under high engine load conditions, and on the other hand, it needs to handle dynamic pressure fluctuations caused by complex road conditions.

[0003] Traditional cooling systems generally employ a passive regulation combination of a mechanical pressure relief valve and an expansion tank. This design has three inherent drawbacks: First, there is a lag in response. The mechanical pressure relief valve requires more than 50 milliseconds to fully open from pressure exceeding the limit, while the pressure fluctuations generated by modern commercial vehicles under rapid acceleration or long downhill conditions can reach the millisecond level. Second, the regulation precision is insufficient. A single mechanical valve cannot achieve gradient pressure regulation, which can easily lead to over- or under-pressure relief in the system. Finally, there is a disconnect between monitoring and control. Existing systems typically use pressure monitoring only for alarm functions and lack a real-time linkage mechanism with the regulation actions.

[0004] A more prominent technical challenge lies in the gas-liquid separation process. When the coolant temperature exceeds 75 degrees Celsius, the rate of bubble formation within the system accelerates significantly. If these bubbles are not separated in time, they will create air resistance in critical areas such as the cylinder head water jacket, not only reducing cooling efficiency but also causing localized overheating and even cylinder block cracking. Existing degassing systems employ a single-channel air guide design, which cannot achieve efficient gas separation and recycling. In long downhill conditions where the hydraulic retarder operates continuously, the rapid formation of a large number of bubbles often leads to a sudden drop in system pressure, causing a chain reaction of failures such as negative pressure at the water pump inlet and obstructed coolant circulation. Although some improvement solutions attempt to introduce electronic pressure sensors, they only achieve electronic monitoring functions and fail to establish a closed-loop control system from pressure sensing to valve regulation. Especially under complex operating conditions with multiple coupled parameters, they lack the ability to collaboratively analyze parameters such as pressure, temperature, and flow rate. Summary of the Invention

[0005] This invention provides a cooling system pressure regulation method and device based on multi-source data fusion, which can dynamically respond to pressure changes, stably maintain system pressure, and effectively eliminate air lock phenomenon.

[0006] To achieve the above objectives, in a first aspect, the present invention provides a cooling system pressure regulation method based on multi-source data fusion, comprising: collecting multi-source data from the cooling system through sensors and performing fusion processing to generate a standardized sensor dataset; calculating a comprehensive system pressure value using data fusion technology based on the standardized sensor dataset to obtain a fused pressure dataset; determining the system pressure state based on the fused pressure dataset through threshold comparison; acquiring channel flow data based on the system pressure state and adjusting valve openings to obtain an optimized regulation result; transmitting the optimized regulation result to a pressure-maintaining system and controlling valve flow using a pressure-stabilizing regulation algorithm to determine a pressure balance index; monitoring the system pressure based on the pressure balance index and triggering a pressure relief channel when the pressure exceeds the limit to restore it to a stable range, thereby obtaining the final regulation result.

[0007] Secondly, this invention provides a cooling system pressure regulation device based on multi-source data fusion. Based on the aforementioned cooling system pressure regulation method based on multi-source data fusion, the cooling system pressure regulation device based on multi-source data fusion includes: a generation module, a first obtaining module, a first determining module, a second obtaining module, a second determining module, and a third obtaining module. The generation module is used to collect multi-source data from the cooling system through sensors and perform fusion processing to generate a standardized sensor dataset. The first obtaining module is used to calculate the comprehensive system pressure value based on the standardized sensor dataset using data fusion technology to obtain a fused pressure dataset. The first determining module is used to determine the system pressure state based on the fused pressure dataset through threshold comparison. The second obtaining module is used to acquire channel flow data based on the system pressure state and adjust the valve opening to obtain an optimized regulation result. The second determining module is used to transmit the optimized regulation result to the pressure holding system and use a pressure stabilization algorithm to control the valve flow, determining a pressure balance index. The third obtaining module is used to monitor the system pressure based on the pressure balance index and trigger a pressure relief channel when the pressure exceeds the limit, restoring it to a stable range, thus obtaining the final regulation result.

[0008] Thirdly, the present invention provides an electronic device, comprising: at least one processor; and

[0009] A memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the cooling system pressure regulation method based on multi-source data fusion as described above.

[0010] Fourthly, the present invention provides a computer-readable storage medium including a computer program and instructions, which, when the computer program or the instructions are executed on a computer, cause the computer to perform the cooling system pressure regulation method based on multi-source data fusion as described above.

[0011] Compared with the prior art, the cooling system pressure regulation method and device based on multi-source data fusion according to the present invention generates a standardized dataset by multi-source data fusion, adjusts the valve opening by combining real-time pressure status, and realizes closed-loop control based on pressure balance index. It can dynamically respond to pressure changes, stably maintain system pressure, and effectively eliminate air resistance. Attached Figure Description

[0012] Figure 1 This is a flowchart illustrating a cooling system pressure control method based on multi-source data fusion according to Embodiment 1 of the present invention.

[0013] Figure 2 This is a schematic diagram of a cooling system pressure regulation device based on multi-source data fusion in Embodiment 2 of the present invention;

[0014] Figure 3 This is a schematic diagram of the structure of an electronic device according to Embodiment 3 of the present invention. Detailed Implementation

[0015] The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and not intended to limit the scope of the invention. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the embodiments of the present invention, and not all structures.

[0016] To facilitate understanding, the main implementation concepts of the various embodiments of the present invention will be briefly described first.

[0017] In traditional engine cooling system pressure regulation, the passive adjustment combination of a mechanical pressure relief valve and an expansion tank is insufficient to handle high-dynamic pressure fluctuations and multi-parameter coupled conditions. The opening response time of the mechanical pressure relief valve differs by orders of magnitude from millisecond-level pressure fluctuations, resulting in significant lag in pressure regulation. The discrete action characteristics of a single mechanical valve cannot achieve pressure gradient regulation, easily leading to over- or under-pressure relief. The lack of real-time data linkage between the pressure monitoring unit and the control actuator makes it impossible to construct a closed-loop control system. The single-channel air guide design in the gas-liquid separation stage cannot effectively handle the rapidly generated bubble clusters under high-temperature conditions in continuous operation of the hydraulic retarder, leading to aggravated air resistance effects and causing negative pressure at the water pump inlet and obstruction of coolant circulation.

[0018] For example, in long downhill driving conditions for commercial vehicles, the engine remains under high load. Once the coolant temperature exceeds the critical threshold, the bubble formation rate increases exponentially. At this point, traditional systems, lacking multi-source data fusion capabilities, cannot simultaneously analyze the dynamic coupling relationship between pressure, temperature, and flow parameters, leading to a mismatch between the gas-liquid separation channel's guiding efficiency and the system pressure demand. Bubbles form localized air resistance at the cylinder head water jacket, causing a sudden drop in coolant circulation flow. Before the mechanical valves respond, the system pressure has already deviated from the safe threshold, triggering a chain reaction of thermal management failures.

[0019] If the above problems are not addressed, system pressure fluctuations will exceed the allowable range of material fatigue strength, accelerating the aging and failure of cylinder block sealing components; localized overheating caused by vapor lock may lead to cylinder head deformation or even cracking; obstructed coolant circulation will reduce radiator heat exchange efficiency, causing engine overheating and protective shutdown. Under dynamic operating conditions, lag and inaccuracy in pressure regulation will significantly shorten the service life of critical components, increasing system maintenance costs and safety risks.

[0020] To address the aforementioned challenges, this invention first investigates the difficulties of pressure regulation lag and multi-parameter coupling, discovering that the discrete action mode of traditional mechanical valves cannot match millisecond-level dynamic fluctuations. To resolve this, this invention attempts to introduce a multi-source data fusion mechanism into the pressure sensing stage, constructing a real-time data closed loop by simultaneously collecting pressure, temperature, and flow parameters. Further analysis reveals that simply increasing the number of sensors exacerbates data heterogeneity; therefore, a standardized processing procedure is proposed to eliminate differences in multi-source data formats. To solve the problem of lag in regulation execution, this invention explores a pressure state grading determination and valve pre-adjustment strategy, where continuous calculation of fused pressure values ​​enables pre-regulation. Addressing insufficient gas-liquid separation efficiency, this invention innovatively dynamically correlates the pressure relief channel triggering conditions with pressure balance indicators, initiating multi-stage pressure relief in advance during the bubble formation acceleration phase.

[0021] Example 1, Figure 1 This is a flowchart illustrating a cooling system pressure control method based on multi-source data fusion according to Embodiment 1 of the present invention, as shown below. Figure 1 As shown, Embodiment 1 provides a cooling system pressure control method based on multi-source data fusion, including:

[0022] Step S100: Collect multi-source data from the cooling system through sensors and perform fusion processing to generate a standardized sensor dataset;

[0023] Step S200: Calculate the integrated system pressure value using data fusion technology based on the standardized sensor dataset to obtain the fused pressure dataset;

[0024] Step S300: Determine the system pressure state based on the fused pressure dataset by threshold comparison;

[0025] Step S400: Based on the system pressure status, obtain the channel flow data and adjust the valve opening to obtain the optimized control result;

[0026] Step S500: The optimized control result is transmitted to the pressure holding system and the valve flow is controlled by the pressure stabilization and regulation algorithm to determine the pressure balance index;

[0027] Step S600: Monitor the system pressure according to the pressure balance index and trigger the pressure relief channel when the pressure exceeds the limit to restore it to the stable range, and obtain the final control result;

[0028] Multi-source data refers to heterogeneous data collected from sensors of different types and locations within the cooling system. This can be achieved using a combination of pressure, temperature, and flow sensors. Multi-source data fusion improves the comprehensiveness and accuracy of pressure monitoring. Standardized sensor datasets are collections of multi-source data that have undergone format unification and dimension normalization. This can be achieved using Z-score normalization or minimum-maximum normalization methods, eliminating the interference of different sensor data dimensions on subsequent analysis. The comprehensive system pressure value is a quantitative indicator reflecting the overall system pressure level. This can be achieved by using a weighted average method or Kalman filtering algorithm to fuse multi-source pressure data, addressing the issue of single sensors being susceptible to local pressure fluctuations. Threshold comparison involves matching real-time pressure data with a preset safety range. This can be achieved using piecewise linear discrimination or dynamic threshold adjustment algorithms to quickly identify abnormal system pressure states. Channel flow data refers to the real-time flow velocity information of fluids in each branch of the cooling system. This can be collected using ultrasonic or electromagnetic flow meters, providing flow reference parameters for valve opening adjustment. Pressure stabilization algorithms refer to control strategies that maintain dynamic pressure balance in a system. Specifically, they can be implemented using PID control or fuzzy control algorithms, suppressing pressure fluctuations through closed-loop feedback to regulate valve flow. Pressure relief channels, on the other hand, are dedicated pipeline structures used to release excessive pressure. These can be implemented using bypass pipelines controlled by solenoid valves, preventing system pressure runaway through rapid pressure relief.

[0029] In practical applications, this invention collects multi-source data from the cooling system using sensors and performs fusion processing to generate a standardized sensor dataset. This step involves using multiple sensors to simultaneously collect parameters such as pressure, temperature, and flow rate, and preprocessing and standardizing the format of the collected raw data for subsequent analysis. Based on the standardized sensor dataset, data fusion technology is used to calculate the comprehensive system pressure value, resulting in a fused pressure dataset. This step generates a more accurate and stable system pressure estimate by weighted fusion and time-series analysis of the multi-source data. The system pressure state is determined based on the fused pressure dataset through threshold comparison. By setting multi-level pressure thresholds, the fused pressure data is classified to identify whether the system is in a normal, alert, or dangerous state. Based on the system pressure state, channel flow rate data is acquired and valve openings are adjusted to obtain optimized control results. Based on the pressure state judgment results, the system further collects flow rate data and adjusts the openings of relevant valves according to a preset pressure-flow mapping relationship to achieve proactive regulation of system pressure. The optimized control results are transmitted to the pressure-maintaining system, and a pressure-stabilizing regulation algorithm is used to control valve flow rate, determining the pressure balance index. This step transmits the aforementioned control commands to the execution unit and continuously adjusts the valve flow rate through a closed-loop control algorithm until the system pressure reaches a stable state, thus obtaining a quantified pressure balance index. Based on this pressure balance index, the system pressure is monitored, and a pressure relief channel is triggered when the pressure exceeds the limit, restoring the system to a stable range, thus obtaining the final control result. The system continuously monitors the deviation between the actual pressure and the balance index. When the deviation exceeds a set threshold, the pressure relief mechanism is automatically activated, and the system pressure is dynamically adjusted to return to a stable range.

[0030] In other words, this pressure regulation method based on multi-source data fusion achieves accurate sensing and rapid adjustment of cooling system pressure through real-time data acquisition, fusion analysis, and closed-loop control, effectively solving the problems of slow response and insufficient accuracy of traditional mechanical pressure regulation systems.

[0031] As a preferred embodiment, the solution of the present invention is implemented as follows:

[0032] First, multiple sensors, including pressure, temperature, and flow sensors, are deployed in the cooling system. These sensors are installed at key locations such as the water pump outlet, radiator inlet and outlet, and cylinder block water jacket. The sensor acquisition frequency is set to 100Hz to capture millisecond-level pressure fluctuations.

[0033] The collected raw data is low-pass filtered to remove high-frequency noise, and then converted into a uniform format through data standardization. The standardized dataset includes fields such as timestamps, pressure values, temperature values, and flow rates.

[0034] Next, the standardized dataset is fused using the Kalman filter algorithm. This algorithm comprehensively considers the measurement errors of each sensor and the dynamic characteristics of the system to obtain the optimal estimated system pressure value. The fused pressure data is then arranged into a time series at 10ms intervals.

[0035] The system has three preset pressure thresholds: normal operating range, warning range, and danger range. The current pressure status of the system is determined by comparing the fused pressure data with these thresholds.

[0036] Based on the pressure status, the system further acquires flow data for the corresponding channels. For example, when the pressure is within the warning range, the flow rate in the pressure relief channel will be the primary focus. The system has a built-in pressure-flow characteristic curve to guide valve opening adjustments. The valves are driven by a stepper motor, enabling opening adjustment with an accuracy of 0.1%.

[0037] The optimized control commands are transmitted to the pressure-holding actuator via the CAN bus. The actuator uses a PID control algorithm to continuously adjust the valve opening based on real-time pressure feedback until the system pressure stabilizes near the target value. The average pressure after stabilization and the fluctuation range together constitute the pressure balance index.

[0038] The system continuously monitors the deviation between the actual pressure and the balance index. When the deviation exceeds a preset threshold (e.g., ±5%), a multi-stage pressure relief mechanism is triggered. The pressure relief channels adopt a parallel design, which can selectively open according to the degree of pressure exceeding the limit, achieving precise pressure regulation.

[0039] Based on the above analysis, this invention achieves high-precision and rapid-response control of cooling system pressure. Multi-source data fusion technology improves the accuracy of pressure sensing and overcomes the vulnerability of single sensors to interference. Real-time data analysis and closed-loop control mechanisms shorten the pressure regulation response time to milliseconds, effectively addressing severe pressure fluctuations under dynamic operating conditions. Multi-stage pressure relief design enhances the system's adaptability to different levels of pressure anomalies, avoiding over- or under-pressure relief. Furthermore, by establishing a correlation analysis model for parameters such as pressure, temperature, and flow rate, the system can initiate preventative pressure regulation before bubble formation intensifies, effectively suppressing the vapor lock effect and ensuring stable coolant circulation. These improvements collectively enhance the pressure stability and thermal management efficiency of the cooling system, extend the service life of key components, and reduce the risk of system failure.

[0040] In this embodiment, step S100 includes: Step S101, acquiring raw data streams of pressure and temperature indicators through a sensor array and performing signal preprocessing to filter noise interference, obtaining a pre-cleaned data set. Step S102, integrating the pressure and temperature indicators using a multi-source fusion method based on the pre-cleaned data set to obtain an initial fused dataset. Step S103, determining whether the pressure and temperature indicator values ​​in the pre-cleaned data set exceed a preset threshold range; if they do, performing data calibration on the initial fused dataset to generate a calibrated dataset; if they do not exceed the threshold, using the initial fused dataset as the calibrated dataset. Step S104, standardizing the calibrated dataset, converting it to a unified format, and generating the standardized sensor dataset.

[0041] Specifically, the sensor array consists of pressure and temperature sensors. The pressure sensor is located at the intersection of the main circulation pipe and branch pipes of the cooling system, while the temperature sensor is installed at the cylinder head water jacket outlet and in the hydraulic retarder cooling circuit. Signal preprocessing uses a second-order Butterworth filter to remove 50Hz power frequency noise from the raw data stream, with the sampling frequency set to 1kHz. The multi-source fusion method employs a Kalman filter algorithm to synchronize and align the pressure and temperature indices in time and space, with preset threshold ranges set at 0.2-1.5MPa for pressure and 20-120℃ for temperature. Data calibration corrects out-of-limit data points using linear interpolation, and standardization converts pressure data to MPa units and temperature data to degrees Celsius units. Timestamps are standardized to Unix time format.

[0042] When the pressure sensor array acquires pressure fluctuation data in the main circulation pipeline, it simultaneously obtains cylinder head water jacket temperature gradient change data. During signal preprocessing, digital filtering eliminates pulse noise caused by engine vibration. When the instantaneous pressure value exceeds 1.5 MPa or the temperature reaches 120°C, a data calibration program is triggered to perform a weighted average of data from three adjacent sampling points. During standardization, pressure data retains two decimal places, and temperature data retains one decimal place, ensuring consistent data dimensions during subsequent weighting. The calibrated dataset forms a continuous pressure-temperature correlation curve over time, providing time-aligned foundational data for calculating the comprehensive pressure value.

[0043] In this embodiment, step S200 includes: Step S201, acquiring pressure and temperature data from the standardized sensor dataset, and obtaining a preliminary comprehensive pressure value using a weighted processing method. Step S202, considering the influence of environmental variables on the preliminary comprehensive pressure value, adjusting it using a smoothing filter if the fluctuation range exceeds a preset threshold, and determining a smoothed pressure reference value. Step S203, based on the smoothed pressure reference value, performing deviation analysis and correction using historical pressure data sequences to generate the fused pressure dataset over a time series.

[0044] The weighted processing method assigns weight coefficients based on the difference in measurement accuracy between pressure and temperature sensors, with the weight of the more accurate sensor increasing to 0.6-0.8. The smoothing filter tool uses a moving average algorithm with a window length of 3-5 sampling periods. When the ambient temperature change rate exceeds 2℃ / s, the window is automatically shortened to 2 periods. Deviation analysis compares the difference between the current pressure reference value and the historical average pressure under the same working conditions. If the deviation exceeds ±5%, the correction coefficient is updated. The correction coefficient is calculated by fitting the most recent 10 sets of historical data using the least squares method.

[0045] Specifically, pressure and temperature data are weighted with 0.7 and 0.3 respectively, and then superimposed to form a preliminary comprehensive pressure value. When the variation of this value between adjacent sampling points exceeds a preset ±10% threshold, moving average filtering is activated to suppress abrupt pressure spikes within ±3%. During the correction phase, stored historical pressure sequences are retrieved and matched against the benchmark pressure curve corresponding to the current engine speed and load rate. When the real-time pressure value deviates from the benchmark value by ±5% for three consecutive sampling periods, a correction factor is generated to adjust the current pressure reference value. After these three stages of weighting, filtering, and correction, the temporal continuity of the final output fused pressure dataset is improved by 40%, and the data standard deviation is reduced to below 35% of the original data, providing a stable data foundation for subsequent pressure status determination.

[0046] In this embodiment, step S300 includes: Step S301, analyzing the fused stress dataset using a threshold comparison algorithm; if the dataset value exceeds a preset threshold, it is marked as a potential abnormal state; otherwise, it is marked as a normal state, thus obtaining a preliminary system stress state. Step S302, acquiring state analysis data based on the preliminary system stress state, performing in-depth checks on abnormal states, and determining the specific classification of abnormal states. Step S303, comprehensively determining the final system stress state based on the preliminary system stress state and the specific classification of abnormal states.

[0047] When the pressure value in the fused pressure dataset exceeds the dynamically adjusted threshold, a potential abnormal state is marked. The state analysis data further extracts pressure fluctuation characteristics. For example, during the continuous operation of the hydraulic retarder, if a sudden pressure drop is accompanied by a rapid temperature rise, it is classified as a pneumatic resistance anomaly; if the pressure continues to rise and the temperature remains stable, it is classified as an overpressure anomaly. Through specific classification, the system invokes pre-stored control strategies. For example, a pneumatic resistance anomaly triggers a multi-stage pressure relief channel, while an overpressure anomaly prioritizes adjusting the opening of the main circulation valve. Thus, the determination of the system pressure state is upgraded from a single threshold judgment to multi-dimensional analysis, avoiding control failures caused by confusion of anomaly types, while shortening the anomaly response time to the millisecond level, ensuring that pressure fluctuations are effectively suppressed before critical components are damaged.

[0048] As a preferred embodiment, the solution of the present invention is implemented as follows:

[0049] The system analyzes the fused pressure dataset using a threshold comparison algorithm. If a dataset value exceeds a preset threshold, it is marked as a potential abnormal state; otherwise, it is marked as a normal state, thus obtaining a preliminary system pressure status. Specifically, multiple pressure threshold ranges can be set, such as 0-2 MPa for the normal range, 2-3 MPa for the slightly abnormal range, 3-4 MPa for the moderately abnormal range, and above 4 MPa for the severely abnormal range. The system compares the real-time pressure values ​​with these ranges and provides corresponding status labels.

[0050] Based on the preliminary system pressure status, status analysis data is obtained, and abnormal states are thoroughly examined to determine their specific classification. For example, auxiliary parameters such as temperature and flow rate can be used to determine whether the abnormal pressure is caused by coolant boiling, pipe blockage, or pump failure. Furthermore, machine learning algorithms such as decision trees or support vector machines can be used to train a classification model based on historical fault data to achieve automatic classification of abnormal states.

[0051] Based on the preliminary system stress state and the specific classification of abnormal states, the final system stress state is comprehensively determined. Therefore, the preliminary state and the results of in-depth analysis can be weighted and integrated to obtain a more accurate system stress state assessment. For example, four levels can be set: normal, slightly abnormal, moderately abnormal, and severely abnormal, and dynamically adjusted according to factors such as the specific type and duration of the abnormality.

[0052] Based on the above analysis, it is evident that this invention, by employing multi-level threshold comparison and deep anomaly analysis, enables the system to promptly detect potential pressure anomalies and accurately determine their specific type and severity. This method significantly improves the sensitivity and accuracy of pressure monitoring, providing a reliable basis for subsequent control decisions. Furthermore, by introducing intelligent algorithms such as machine learning, the system possesses adaptive learning capabilities, continuously optimizing the accuracy of anomaly identification and effectively reducing false alarms and missed alarms.

[0053] In this embodiment, step S400 includes: step S401, acquiring flow data through the channel and combining it with the system pressure status, comparing it using a pre-established mapping table to determine whether the flow data matches the pressure status, and obtaining a preliminary matching conclusion; step S402, if the preliminary matching conclusion shows inconsistency, calibrating the flow data to determine the calibrated flow information; if they match, using the flow data as the calibrated flow information; step S403, based on the calibrated flow information and the system pressure status, querying the valve control strategy table to generate a valve opening adjustment command, and sending it to the actuator to adjust the valve, thereby obtaining the optimized control result.

[0054] The pre-established mapping table contains the correspondence between pressure status and flow rate data. For example, when the pressure status is abnormal, the corresponding flow rate threshold range is 30-50 liters per minute, and when the pressure status is normal, the corresponding flow rate threshold range is 20-40 liters per minute. The error range of the calibrated flow rate information is controlled within ±2%, and the calibration method includes dynamically correcting the zero-point drift of the flow sensor based on the pressure status. The valve control strategy table adopts a two-dimensional matrix structure, where one dimension is the pressure status classification and the other dimension is the flow deviation level. Each intersection corresponds to a specific valve opening adjustment range. For example, when the pressure status is abnormal and the flow deviation level is level one, the corresponding valve opening is reduced by 15%.

[0055] Specifically, flow data is acquired through a dedicated channel independent of the pressure sensor to avoid signal interference. When the system pressure is deemed abnormal, the mapping table automatically switches to high-pressure mode, at which point the upper limit of the flow threshold is increased to 60 liters per minute. During calibration, if the flow data deviates from the preset correspondence with the pressure status by more than 5%, a sensor zero-point reset operation is triggered to eliminate accumulated errors caused by long-term operation. When a valve opening adjustment command is generated, the feedback signal of the actuator is simultaneously verified. If the actual opening deviates from the command by more than 3%, a secondary calibration process is initiated. Through a dual verification mechanism of pressure status and flow data, millisecond-level synchronization between valve action and system pressure changes is achieved, ensuring that the control results meet the dynamic requirements of the engine under high-load conditions.

[0056] In practical applications, the embodiments of the present invention are as follows:

[0057] In step S401, flow data from each channel of the cooling system is acquired using flow sensors and compared with the system pressure status using a pre-established mapping table. The mapping table contains standard flow ranges corresponding to different pressure states. By looking up the table, it is determined whether the actual flow data falls within the standard range for the corresponding pressure state, thus obtaining a preliminary matching conclusion.

[0058] In step S402, if the preliminary matching results show inconsistencies, the flow data is calibrated. The calibration methods include: using a Kalman filter algorithm to filter noise from the original flow data; combining temperature sensor data to perform temperature compensation on the flow; and performing trend analysis based on historical data to eliminate abnormal fluctuations. After calibration, more accurate flow information is obtained. If the preliminary matching results are consistent, the original flow data is directly used as the calibrated flow information.

[0059] In step S403, based on the calibrated flow information and system pressure status, a preset valve control strategy table is queried. This strategy table contains the optimal valve opening configuration under different pressure-flow combinations. Specific valve opening adjustment commands are generated based on the table lookup results, such as "increase valve opening by 10% for valve 1, decrease valve opening by 5% for valve 2," etc. The commands are then sent to the actuator via the CAN bus, which controls the valve motor to adjust the opening, ultimately obtaining the optimized control result.

[0060] This invention improves data reliability by timely detecting and calibrating anomalies through matching analysis of flow rate and pressure status. The use of a preset strategy table makes valve adjustment more targeted and predictive, avoiding pressure fluctuations that may result from blind adjustments. Simultaneously, the close integration of flow control and pressure status forms a closed-loop feedback mechanism, making system pressure regulation more sensitive and precise. Compared to traditional single-parameter control, this method better handles pressure fluctuations under complex operating conditions, improving the stability and reliability of the cooling system.

[0061] In this embodiment, step S500 includes: Step S501, transmitting the optimized control result to the pressure-maintaining system via a transmission system and performing data verification to determine whether the data integrity meets a preset standard, thereby obtaining a transmission verification result. Step S502, if the transmission verification result shows that it does not meet the standard, data repair is performed to determine the repaired control data; if it meets the standard, the received optimized control result is used as the repaired control data. Step S503, inputting the repaired control data into the pressure-stabilizing algorithm to simulate and calculate the system pressure state after valve flow adjustment, and outputting a quantitative value for measuring the system stability as the pressure balance index.

[0062] The data verification process compares the transmitted optimization and control results item by item through a preset integrity verification protocol, including data packet length, check code, and key parameter range; the data repair uses redundant coding technology to fill in missing fields, and interpolates and replaces outliers by combining the statistical distribution characteristics of historical control data; the pressure stabilization algorithm constructs a pressure-flow dynamic equation based on a fluid dynamics model, predicts the system pressure change trend after valve adjustment through iterative calculation, and finally outputs a pressure balance index as a quantitative benchmark for system stability.

[0063] Specifically, when the optimized control results are transmitted to the pressure-maintaining system, a data integrity check is first performed. By comparing the data packet header information with preset standards, it identifies whether there are missing fields or incorrect checksums. If the check fails, backup information from the redundant data storage module is retrieved and combined with the valid portion of the currently transmitted data to repair the data, ensuring the integrity and logical consistency of the control data. After the repaired data is input into the pressure-stabilizing algorithm, the algorithm establishes a differential equation model of pressure fluctuations based on the geometric parameters of the cooling system pipelines, fluid viscosity, and valve characteristic parameters, simulating the pressure distribution under different valve openings. Through multiple iterative calculations, the algorithm outputs a pressure balance index, which includes quantitative parameters such as pressure fluctuation amplitude, stabilization time, and probability of deviation from the threshold, used to guide the triggering conditions and adjustment range of subsequent pressure relief channels. This process effectively solves the control failure problem caused by abnormal data transmission in traditional systems by dynamically verifying the reliability of transmitted data and generating control benchmarks based on accurate mathematical models.

[0064] In practical applications, the specific implementation methods of the present invention are as follows:

[0065] The optimized control results are transmitted to the pressure-maintaining system via a transmission system and undergo data verification. During data verification, the system checks whether the data integrity meets preset standards, obtaining the transmission verification result. If the transmission verification result shows that the data does not meet the preset standards, data repair is performed. The data repair process uses a redundancy check algorithm to reconstruct missing or erroneous data segments and determine the repaired control data. If the transmission verification result meets the preset standards, the received optimized control results are directly used as the repaired control data. The repaired control data is input into the pressure stabilization algorithm. The pressure stabilization algorithm uses a PID control model to simulate and calculate the system pressure state after valve flow adjustment based on the repaired control data. The algorithm outputs a quantified value to measure the system stability as a pressure balance index.

[0066] Specifically, the pressure stabilization algorithm first calculates the deviation between the current system pressure and the target pressure. Then, based on the deviation value, the rate of change of the deviation, and the cumulative deviation, it calculates the proportional, derivative, and integral terms, respectively. The weighted sum of these three terms yields the control variable, which is used to adjust the valve opening. Finally, the algorithm outputs a quantitative value between 0 and 100 as a pressure balance index, based on how close the adjusted system pressure is to the target pressure. Here, 0 indicates that the system pressure is completely out of control, and 100 indicates that an ideal equilibrium state has been reached.

[0067] Based on the above analysis, this invention achieves closed-loop control throughout the entire process, from data transmission to pressure regulation. The data verification and repair mechanism improves the system's anti-interference capability and avoids malfunctions caused by data transmission errors. The introduction of the pressure regulation algorithm enables the system to quickly respond to pressure changes based on real-time data and accurately adjust valve flow. The output of the pressure balance index provides a quantitative basis for system status assessment and subsequent control, effectively improving the accuracy and stability of cooling system pressure regulation.

[0068] In this embodiment, step S600 includes: Step S601, acquiring system pressure data through real-time monitoring, determining whether the system pressure data is complete, and obtaining a processed pressure dataset. Step S602, comparing and analyzing the processed pressure dataset with the pressure balance index; if the current pressure state deviates from the pressure balance index beyond a preset tolerance, a pressure relief channel is triggered. Step S603, acquiring the system pressure state after the pressure relief operation, and dynamically adjusting the system parameters using a pressure stabilization algorithm to bring the system pressure closer to the pressure balance index. Step S604, monitoring the adjusted system pressure in real-time; when it stabilizes within the range corresponding to the pressure balance index, generating and outputting the final control result containing the pressure relief record, adjustment parameters, and the final stable pressure value.

[0069] Specifically, when acquiring system pressure data in real time, a multi-channel parallel acquisition mode is used to ensure data integrity. If data loss exceeds a set proportion, a redundant sensor retransmission mechanism is activated. During comparative analysis with pressure balance indicators, preset tolerances are dynamically adjusted according to the current system operating conditions. For example, when the hydraulic retarder is working, the tolerance range is reduced to 70% of the standard value. When using a pressure stabilization algorithm for dynamic adjustment, the algorithm input parameters include the duration of the pressure relief channel opening, the current coolant temperature, and the historical pressure fluctuation frequency. When generating the final control result, the pressure relief record includes the trigger time, pressure relief amount, and channel number, and the adjustment parameters cover the valve opening gradient and the number of iterations of the pressure stabilization algorithm.

[0070] Specifically, system pressure data is collected in real time through a distributed sensor network, and data integrity verification is based on a dual standard of timestamp continuity and numerical reasonableness. When a pressure deviation from the balance index is detected, the pressure relief channel selects the corresponding physical path to activate according to the direction of deviation; for example, the top pressure relief valve is activated when the high pressure exceeds the limit, and the bottom return pump is activated when the low pressure exceeds the limit. The pressure stabilization algorithm adopts a feedforward-feedback composite control mode during the dynamic adjustment phase. Feedforward control predicts the adjustment range based on the pressure relief amount, while feedback control corrects the adjustment speed based on the real-time pressure difference. During the final control result generation process, the system automatically associates the current pressure relief event with historical control records to form a comprehensive report including time series comparison charts and parameter change curves. Pressure stability determination adopts a sliding window mechanism; when the pressure fluctuation amplitude is less than 20% of the balance index threshold within three consecutive monitoring cycles, the system is determined to have returned to a stable state.

[0071] In practical applications, the present invention is specifically implemented as follows:

[0072] Step S601 involves acquiring system pressure data through real-time monitoring, determining the completeness of the system pressure data, and obtaining a processed pressure dataset. Specifically, a distributed pressure sensor network is used to collect pressure data from key nodes of the cooling system in real time, with a sampling frequency of 100Hz. The collected raw data undergoes an integrity check, and outliers and missing values ​​are removed to form the processed pressure dataset.

[0073] Step S602: For the processed pressure dataset, a comparative analysis is performed based on the pressure balance index. If the current pressure state deviates from the pressure balance index beyond a preset tolerance, the pressure relief channel is triggered. Specifically, the processed pressure dataset is compared with the preset pressure balance index, and the deviation value is calculated. If the deviation value exceeds a preset tolerance range of ±5%, the pressure relief channel is triggered. The pressure relief channel includes a main pressure relief valve and an auxiliary pressure relief valve; one or more pressure relief valves are selected to be opened based on the magnitude of the deviation value.

[0074] Step S603: Obtain the system pressure status after the pressure relief operation and dynamically adjust the system parameters using a pressure stabilization algorithm to bring the system pressure closer to the pressure balance target. Specifically, after the pressure relief operation is completed, system pressure data is collected again and input into a pre-trained neural network model. Based on the current pressure status and the target pressure balance target, the model outputs adjustment suggestions for key parameters such as pump speed and throttle valve opening. The actuator adjusts relevant components in real time according to the adjustment suggestions, gradually bringing the system pressure closer to the pressure balance target.

[0075] Step S604: Monitor the adjusted system pressure in real time. When it stabilizes within the range corresponding to the pressure balance index, generate and output the final control result, which includes pressure relief records, adjustment parameters, and the final stable pressure value. Specifically, continuously monitor the adjusted system pressure. When the pressure value remains stable within the target range for more than 30 seconds, it is determined to have reached a stable state. Subsequently, a control report is generated, recording the opening time and flow rate of each valve during the pressure relief process, the adjustment process of system parameters, and the pressure value at the final stabilization point. This report is output as the final control result to the central control unit for subsequent analysis and optimization.

[0076] Through the above technical solution, this invention achieves precise control of cooling system pressure. Real-time monitoring and data analysis promptly identify abnormal pressure states and trigger targeted pressure relief operations. Combined with intelligent algorithms, system parameters are dynamically adjusted to ensure rapid pressure recovery and stabilization within the target range. This closed-loop control mechanism significantly improves the system's response speed and adjustment accuracy to pressure fluctuations, effectively preventing over- or under-pressure relief. Simultaneously, the complete recording of the control process provides data support for system optimization, contributing to further improvements in the cooling system's operating efficiency and reliability.

[0077] Example 2, Figure 2 This is a schematic diagram of a cooling system pressure regulation device based on multi-source data fusion according to Embodiment 2 of the present invention, as shown below. Figure 2 As shown, Embodiment 2 provides a cooling system pressure control device based on multi-source data fusion, which is based on the cooling system pressure control method based on multi-source data fusion described in Embodiment 1. The cooling system pressure control device based on multi-source data fusion includes: a generation module 201, a first obtaining module 202, a first determining module 203, a second obtaining module 204, a second determining module 205, and a third obtaining module 206. The generation module 201 is used to collect multi-source data from the cooling system through sensors and perform fusion processing to generate a standardized sensor dataset. The first obtaining module 202 is used to calculate the comprehensive system pressure value based on the standardized sensor dataset using data fusion technology to obtain a fused pressure dataset. The first determining module 203 is used to determine the system pressure state based on the fused pressure dataset through threshold comparison. The second obtaining module 204 is used to acquire channel flow data based on the system pressure state and adjust the valve opening to obtain an optimized control result. The second determining module 205 is used to transmit the optimized control result to the pressure holding system and use a pressure stabilization algorithm to control the valve flow and determine the pressure balance index. The third module 206 is used to monitor the system pressure according to the pressure balance index and trigger the pressure relief channel when the pressure exceeds the limit to restore it to the stable range, thereby obtaining the final control result.

[0078] The various variations and specific examples of the cooling system pressure regulation method based on multi-source data fusion provided in Embodiment 1 are also applicable to the cooling system pressure regulation device based on multi-source data fusion provided in this embodiment. Through the foregoing detailed description of a cooling system pressure regulation method based on multi-source data fusion, those skilled in the art can clearly understand the implementation method of the cooling system pressure regulation device based on multi-source data fusion in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0079] Example 3, Figure 3 This is a schematic diagram of the structure of an electronic device according to Embodiment 3 of the present invention, as shown below. Figure 3 As shown, Embodiment 3 also provides an electronic device 300, which may include a processor 301 and a memory 302.

[0080] Memory 302 is used to store programs. Memory 302 may include volatile memory, such as random-access memory (RAM), such as static random-access memory (SRAM), double data rate synchronous dynamic random-access memory (DDR SDRAM), etc.; memory may also include non-volatile memory, such as flash memory. Memory 302 is used to store computer programs (such as application programs, functional modules, etc. that implement the above methods), computer instructions, etc. The computer programs, computer instructions, etc., can be partitioned and stored in one or more memories 302. Furthermore, the computer programs, computer instructions, data, etc., can be accessed by processor 301.

[0081] The aforementioned computer programs and instructions can be stored in one or more partitions of memory 302. Furthermore, the aforementioned computer programs and instructions can be invoked by processor 301.

[0082] The processor 301 is configured to execute the computer program stored in the memory 302 to implement the various steps of the methods described in the above embodiments. For details, please refer to the relevant descriptions in the preceding method embodiments.

[0083] The processor 301 and the memory 302 can be independent structures or integrated structures. When the processor 301 and the memory 302 are independent structures, the memory 302 and the processor 301 can be coupled together via bus 303.

[0084] The electronic device in this embodiment can execute the technical solution in the above method. Its specific implementation process and technical principle are the same, and will not be repeated here.

[0085] Example 4: Example 4 also provides a computer-readable storage medium including a computer program and instructions, which, when executed on a computer, cause the computer to perform the cooling system pressure regulation method based on multi-source data fusion according to any embodiment of the present invention.

[0086] Computer-readable storage media include various media that can store program code, such as USB flash drives, external hard drives, ROM, RAM, magnetic disks, or optical disks.

[0087] This embodiment also provides a computer program product, which includes: a computer program stored in a readable storage medium, at least one processor of an electronic device can read the computer program from the readable storage medium, and the at least one processor executes the computer program to cause the electronic device to perform the solution provided in any of the above embodiments.

[0088] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this invention disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this invention can be achieved, and this is not limited herein.

[0089] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A method for regulating cooling system pressure based on multi-source data fusion, characterized in that, include: A standardized sensor dataset is generated by collecting multi-source data from the cooling system through sensors and performing fusion processing. The integrated system pressure value is calculated using data fusion technology based on the standardized sensor dataset to obtain the fused pressure dataset; The system pressure status is determined by threshold comparison based on the fused pressure dataset. Based on the system pressure status, channel flow data is obtained and valve opening is adjusted to obtain optimized control results; The optimized control results are transmitted to the pressure holding system, and a pressure stabilization algorithm is used to control the valve flow to determine the pressure balance index. The system pressure is monitored based on the pressure balance index, and the pressure relief channel is triggered when the pressure exceeds the limit to restore it to a stable range, thus obtaining the final control result.

2. The cooling system pressure control method based on multi-source data fusion as described in claim 1, characterized in that, The process of collecting multi-source data from the cooling system via sensors and fusing it to generate a standardized sensor dataset includes: The raw data streams of pressure and temperature indicators are acquired by the sensor array and the signals are preprocessed to filter out noise interference, resulting in a pre-cleaned data set. Based on the initially cleaned dataset, the pressure and temperature indices are integrated using a multi-source fusion method to obtain an initial fused dataset. Determine whether the pressure and temperature index values ​​in the initially cleaned dataset exceed a preset threshold range. If they do, perform data calibration on the initial fused dataset to generate a calibrated dataset. If they do not exceed the threshold, use the initial fused dataset as the calibrated dataset. The calibrated dataset is standardized and converted into a uniform format to generate the standardized sensor dataset.

3. The cooling system pressure control method based on multi-source data fusion as described in claim 1, characterized in that, The process of calculating the integrated system pressure value using data fusion technology based on a standardized sensor dataset to obtain the fused pressure dataset includes: Pressure and temperature data are obtained from the standardized sensor dataset, and a weighted processing method is used to obtain a preliminary comprehensive pressure value. If the fluctuation range of the preliminary comprehensive pressure value exceeds a preset threshold, it is adjusted using a smoothing filter tool to determine the smoothed pressure reference value, taking into account the influence of environmental variables. Based on the smoothed pressure reference value, deviation analysis and correction are performed using historical pressure data sequences to generate the fused pressure dataset in the time series.

4. The cooling system pressure control method based on multi-source data fusion as described in claim 1, characterized in that, The method of determining the system stress state based on threshold comparison using a fused stress dataset includes: The fused stress dataset is analyzed using a threshold comparison algorithm. If the dataset value exceeds a preset threshold, it is marked as a potential abnormal state; otherwise, it is marked as a normal state, thus obtaining the preliminary system stress state. Based on the preliminary system pressure status, obtain status analysis data, conduct in-depth inspection of abnormal states, and determine the specific classification of abnormal states. Based on the specific classification of the preliminary system pressure status and abnormal status, the final system pressure status is determined comprehensively.

5. The cooling system pressure control method based on multi-source data fusion as described in claim 1, characterized in that, The process of acquiring channel flow data based on system pressure status and adjusting valve opening to obtain optimized control results includes: Traffic data is acquired through the channel and combined with the system pressure status. A pre-established mapping table is used for comparison to determine whether the traffic data matches the pressure status and to obtain a preliminary matching conclusion. If the preliminary matching results show inconsistency, the traffic data is calibrated to determine the calibrated traffic information; if they match, the traffic data is used as the calibrated traffic information. Based on the calibrated flow information and the system pressure status, the valve control strategy table is queried to generate a valve opening adjustment command, which is then sent to the actuator to adjust the valve, thereby obtaining the optimized control result.

6. The cooling system pressure control method based on multi-source data fusion as described in claim 1, characterized in that, The process of transmitting optimized control results to the pressure-maintaining system and using a pressure-stabilizing regulation algorithm to control valve flow, and determining pressure balance indicators, includes: The optimized control results are transmitted to the pressure holding system through the transmission system and the data is verified to determine whether the data integrity meets the preset standard, and the transmission verification result is obtained. If the transmission verification result does not meet the standard, data repair is performed to determine the repaired control data; if it meets the standard, the received optimized control result is used as the repaired control data. The repaired control data is input into the pressure stabilization algorithm to simulate and calculate the pressure state of the system after the valve flow is adjusted, and outputs a quantitative value to measure the stability of the system as the pressure balance index.

7. The cooling system pressure control method based on multi-source data fusion as described in claim 1, characterized in that, The process of monitoring system pressure based on pressure balance indicators and triggering a pressure relief channel when pressure exceeds limits to restore it to a stable range, resulting in the final control outcome, includes: By acquiring system pressure data through real-time monitoring, determining whether the system pressure data is complete, and obtaining a processed pressure dataset; For the sorted pressure dataset, a comparative analysis is performed in conjunction with the pressure balance index. If the current pressure state deviates from the pressure balance index by more than a preset tolerance, the pressure relief channel is triggered. The system pressure status after the pressure relief operation is obtained, and the system parameters are dynamically adjusted using a pressure stabilization and regulation algorithm to make the system pressure approach the pressure balance index. The system pressure is monitored in real time after adjustment. When the pressure stabilizes within the range corresponding to the pressure balance index, the final control result, which includes pressure relief records, adjustment parameters, and the final stable pressure value, is generated and output.

8. A cooling system pressure regulation device based on multi-source data fusion, based on the cooling system pressure regulation method based on multi-source data fusion as described in any one of claims 1-7, characterized in that, The cooling system pressure regulation device based on multi-source data fusion includes: The generation module is used to collect multi-source data from the cooling system through sensors and perform fusion processing to generate a standardized sensor dataset. The first obtaining module is used to calculate the integrated system pressure value based on the standardized sensor dataset using data fusion technology, and obtain the fused pressure dataset. The first determining module is used to determine the system pressure state based on the fused pressure dataset by threshold comparison; The second module is used to acquire channel flow data based on the system pressure state and adjust the valve opening to obtain optimized control results. The second determining module is used to transmit the optimized control results to the pressure-maintaining system and use a pressure-stabilizing regulation algorithm to control the valve flow rate, thereby determining the pressure balance index; and The third module is used to monitor the system pressure according to the pressure balance index and trigger the pressure relief channel when the pressure exceeds the limit to restore it to the stable range, thereby obtaining the final control result.

9. An electronic device, characterized in that, include: At least one processor; as well as A memory that is communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform a cooling system pressure regulation method based on multi-source data fusion as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, It includes computer programs and instructions, which, when run on a computer, cause the computer to perform a cooling system pressure regulation method based on multi-source data fusion as described in any one of claims 1-7.

Images