A method for automatically controlling equipment in a hydraulic station

Abstract

The present invention discloses a method for automatic control of equipment in a hydraulic station, which specifically relates to the technical field of automatic control of equipment. By collecting the operating data of the main hydraulic pump and performing fault prediction, the potential faults of the main equipment are discovered in time, and the preloading process of the standby hydraulic pump is started according to the time when the fault occurs. By obtaining the oil viscosity abnormality information and the load incremental stability information during the preloading process of the standby hydraulic pump, and calculating the oil viscosity abnormality coefficient and the load incremental stability coefficient based on the information, the stability of the standby pump during the switching process is effectively evaluated, the working status of the standby pump is comprehensively analyzed, the potential loading hidden dangers are identified in advance, and corresponding control is performed through the output of the risk model to avoid equipment failure or system instability due to improper switching. According to the comparison between the standby switching hidden danger risk index and the preset threshold, the switching risk of the standby hydraulic pump is dynamically judged to ensure the smooth operation and seamless switching of the hydraulic system.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment automation control, and more particularly to an equipment automation control method in a hydraulic station. Background Art

[0002] Hydraulic systems are widely used in various industrial equipment, such as machinery manufacturing, aerospace, automotive production, and metallurgy, as a means of efficient energy transmission and control. The hydraulic station is the core component of the hydraulic system, primarily responsible for transmitting power to the actuators, thereby completing various mechanical operations. During normal operation, the various devices within the hydraulic station (such as hydraulic pumps, valves, and actuators) must ensure system stability, reliability, and efficient response. However, in practical hydraulic system applications, ensuring seamless and rapid handover between devices when the primary device fails or load fluctuations require switching to a backup device remains a technical challenge. Traditional hydraulic station control systems often use a hard handover strategy, directly unloading the primary device and switching to the backup device. These backup devices are often not preheated or operate at low load, leading to startup delays and uneven oil temperatures. This switching method cannot quickly restore efficient system response and, in severe cases, can even cause hydraulic shock or equipment failure, disrupting production or operations, impacting productivity and equipment life. Summary of the Invention

[0003] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an automatic control method for equipment in a hydraulic station to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A method for automatic control of equipment in a hydraulic station, comprising the following steps:

[0006] Step S1, collecting operating data of the main hydraulic pump, and predicting the failure time of the main hydraulic pump based on the operating data of the main hydraulic pump, and starting the preloading process of the backup hydraulic pump according to the failure time;

[0007] Step S2, obtaining oil viscosity abnormality information and load increasing stability information of the standby hydraulic pump during the preloading process;

[0008] Step S3: constructing a standby switching hidden danger risk model based on the abnormal oil viscosity information and load increasing stability information during the standby hydraulic pump preloading process, outputting a standby switching hidden danger risk index, and evaluating potential loading hidden dangers during the standby hydraulic pump preloading process;

[0009] Step S4: comparing the standby switching hidden danger risk index with a preset standby switching hidden danger risk index threshold, and performing classification control on the hydraulic pumps.

[0010] In a preferred embodiment, high-precision sensors are installed to collect real-time operating data of the main hydraulic pump, including temperature, pressure, flow, vibration, and oil viscosity;

[0011] Data preprocessing is performed on the collected main hydraulic pump operation data, including missing value processing and data standardization;

[0012] The fault occurrence time is used as the target variable, the operating data of the main hydraulic pump is used as the feature variable, and a feature vector is constructed based on the operating data of the main hydraulic pump;

[0013] Build a failure occurrence time prediction model: ,in is the output of the failure occurrence time prediction model, is the weight vector The transpose of is the characteristic variable, is the bias term;

[0014] Construct the objective function;

[0015] Use the historical operating data of the main hydraulic pump to train the fault occurrence time prediction model and minimize the objective function to find the optimal weight vector and bias ;

[0016] Input the real-time collected main hydraulic pump operation data into the trained fault occurrence time prediction model to predict the fault occurrence time of the main hydraulic pump;

[0017] The starting conditions for starting the preloading process of the standby hydraulic pump according to the fault occurrence time are: ,in is the current time, is the time when the fault occurred, The lead time required to preload the standby hydraulic pump, If and When the values ​​are equal, the preloading process of the standby hydraulic pump is started.

[0018] In a preferred embodiment, by acquiring the oil viscosity abnormality information during the preloading process of the standby hydraulic pump, analyzing the oil viscosity during the preloading process of the standby hydraulic pump, and calculating the oil viscosity abnormality coefficient, the degree of oil viscosity abnormality during the preloading process of the standby hydraulic pump is measured;

[0019] The logic for obtaining the oil viscosity abnormality coefficient is as follows:

[0020] The oil viscosity of the standby hydraulic pump during preloading is collected by a high-precision sensor to construct an oil viscosity time series: ,in represents the oil viscosity collected at time t, t={1,2,...,T}, where T is a positive integer;

[0021] Cluster analysis is performed on the oil viscosity time series, and the oil viscosity anomaly coefficient is calculated as follows:

[0022] Step S21, using the silhouette coefficient method to determine the initial number of cluster centers SL, randomly selecting the oil viscosity at SL time points from the oil viscosity time series as the initial cluster centers;

[0023] Step S22, calculating the Euclidean distance between the oil viscosity at each time point in the oil viscosity time series and the cluster center, and assigning the oil viscosity at each time point to the cluster center with the closest Euclidean distance;

[0024] Step S23, calculating the average value of the oil viscosity in each cluster center and using it as the new cluster center coordinates;

[0025] Step S24, repeating steps S22 and S23 to iteratively update the cluster center until the cluster center no longer changes;

[0026] Calculate the deviation of the oil viscosity at each time point from the center of its cluster: ,in represents the deviation of the oil viscosity from the center of its cluster, Indicates the oil viscosity value at time point t, represents the oil viscosity value of the kth cluster center;

[0027] Calculate the oil viscosity abnormal coefficient: ,in is the abnormal coefficient of oil viscosity, is the deviation between the oil viscosity at time point t and its cluster center, Time point The oil viscosity value, is the time point contained in the k-th cluster center.

[0028] In a preferred embodiment, the load increasing stability information of the standby hydraulic pump during the preloading process is obtained, the load increasing stability of the standby hydraulic pump during the preloading process is analyzed, and the load increasing stability coefficient is calculated to measure the load increasing stability of the standby hydraulic pump during the preloading process;

[0029] The logic for obtaining the load increasing stability coefficient is as follows:

[0030] Gradually apply system load to the standby hydraulic pump until it reaches the system load of the main hydraulic pump, and record the system load values ​​at different times And the response time of the standby hydraulic pump to system load changes at different times ;

[0031] Calculate the load change rate: ,in represents the load change rate at time t, is the system load value at time t, is the system load value at time t-1, is the sampling time interval;

[0032] Calculate the response lag time: ,in For the response delay time, is the response time of the standby hydraulic pump to system load changes, is the expected response time;

[0033] Calculate the load increment stability value at different times: ,in Indicates the load increasing stable value at time t, Transfer function for the system The phase angle, ,in is the static gain of the system, is the time constant of the system, is the imaginary unit, is the angular frequency;

[0034] Calculate the load increasing stability factor: ,in is the load increasing stability factor, is the load increasing stable value at time r, is the time window size.

[0035] In a preferred embodiment, a standby switching hidden danger risk model is constructed based on the oil viscosity abnormality coefficient and the load increasing stability coefficient, and a standby switching hidden danger risk index is output. The model is based on the following formula , where They represent the preset proportional coefficients of the oil viscosity abnormality coefficient and the load increasing stability coefficient, respectively, and Both are greater than 0.

[0036] In a preferred embodiment, in step S4, the standby switching hidden danger risk index is compared with a preset standby switching hidden danger risk index threshold, and the hydraulic pump is controlled by classification, specifically as follows:

[0037] If the standby switching hidden danger risk index is greater than the standby switching hidden danger risk index threshold, the current standby hydraulic pump is marked as a switching risk device, and early warning maintenance is performed on the marked switching risk device;

[0038] If the standby switching hidden danger risk index is less than or equal to the standby switching hidden danger risk index threshold, the standby hydraulic pumps that are not marked as switching risk equipment will be sorted from small to large according to the standby switching hidden danger risk index, and a standby equipment switching sorting table will be generated, and the standby equipment ranked first will be switched.

[0039] The technical effects and advantages of the present invention are as follows:

[0040] 1. The present invention collects the operating data of the main hydraulic pump and performs fault prediction, so as to timely discover the potential faults of the main equipment and start the preloading process of the standby hydraulic pump according to the time of the fault occurrence. By obtaining the oil viscosity abnormality information and load incremental stability information of the standby hydraulic pump in real time during the preloading process, and calculating the oil viscosity abnormality coefficient and load incremental stability coefficient based on this, the stability of the standby pump during the switching process is effectively evaluated, and the working status of the standby pump is comprehensively analyzed. When the oil viscosity is abnormal or the load incremental is unstable, potential loading hazards can be identified in advance, and corresponding control can be performed through the output of the risk model to avoid equipment failure or system instability due to improper switching. According to the comparison between the standby switching hazard risk index and the preset threshold, the switching risk of the standby hydraulic pump is dynamically judged. When the risk is high, early warning and maintenance are automatically performed to ensure that the performance of the standby equipment meets the switching requirements. When the risk is low, the standby hydraulic pump with the lowest risk is preferentially selected for switching to ensure smooth operation and seamless switching of the hydraulic system. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings;

[0042] Figure 1 Flowchart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0044] Example: Figure 1 The present invention provides a method for automatically controlling equipment in a hydraulic station, comprising the following steps:

[0045] Step S1, collecting operating data of the main hydraulic pump, and predicting the failure time of the main hydraulic pump based on the operating data of the main hydraulic pump, and starting the preloading process of the backup hydraulic pump according to the failure time;

[0046] Step S2, obtaining oil viscosity abnormality information and load increasing stability information of the standby hydraulic pump during the preloading process;

[0047] Step S3: constructing a standby switching hidden danger risk model based on the abnormal oil viscosity information and load increasing stability information during the standby hydraulic pump preloading process, outputting a standby switching hidden danger risk index, and evaluating potential loading hidden dangers during the standby hydraulic pump preloading process;

[0048] Step S4, comparing the standby switching hidden danger risk index with a preset standby switching hidden danger risk index threshold, and performing classification control on the hydraulic pump;

[0049] Step S1, collecting operating data of the main hydraulic pump, and predicting the failure time of the main hydraulic pump based on the operating data of the main hydraulic pump, and starting the preloading process of the backup hydraulic pump according to the failure time;

[0050] By installing high-precision sensors, the operating data of the main hydraulic pump is collected in real time, including temperature, pressure, flow, vibration, oil viscosity and other parameters;

[0051] It should be noted that the high-precision sensors for collecting operating data of the main hydraulic pump include temperature sensors (such as thermocouples and RTDs, which can be installed at the outlet of the hydraulic pump to monitor the oil temperature of the hydraulic pump), pressure sensors (such as strain gauge pressure sensors and piezoresistive pressure sensors, which can be installed at the outlet of the hydraulic pump to monitor the output pressure in real time), flow sensors (such as vortex flowmeters, electromagnetic flowmeters, and positive displacement flow sensors, which can be installed at the outlet of the hydraulic pump to monitor the output flow of the pump), vibration sensors (such as acceleration sensors, which can be installed on the surface of the hydraulic pump body to monitor the surface vibration signal of the hydraulic pump), and oil viscosity sensors (such as rotational viscosity sensors and oscillating viscosity sensors, which can be installed in the hydraulic oil tank to monitor the viscosity of the oil);

[0052] Data preprocessing of the collected main hydraulic pump operating data includes missing value processing (if there is missing data, it can be filled using mean filling or other interpolation methods) and data standardization (due to different sensor data dimensions, data standardization is usually required, for example, using Z-score standardization);

[0053] The fault occurrence time is used as the target variable, the operating data of the main hydraulic pump is used as the feature variable, and the feature vector is constructed based on the operating data of the main hydraulic pump: ,in represents the eigenvector at time t, They represent the temperature, pressure, flow rate, vibration, and oil viscosity at time t respectively;

[0054] Build a failure occurrence time prediction model: ,in is the output of the fault occurrence time prediction model (the output of the fault occurrence time prediction model is the fault occurrence time), is the weight vector The transpose of is the characteristic variable, is the bias term;

[0055] Construct the objective function: ,in is a regularization term used to prevent overfitting, is the prediction error of each sample, , is the total number of samples, is a hyperparameter used to control the trade-off between error and model complexity;

[0056] Use the historical operating data of the main hydraulic pump to train the fault occurrence time prediction model and minimize the objective function to find the optimal weight vector and bias ;

[0057] Input the real-time collected main hydraulic pump operation data into the trained fault occurrence time prediction model to predict the fault occurrence time of the main hydraulic pump;

[0058] The starting conditions for starting the preloading process of the standby hydraulic pump according to the fault occurrence time are: ,in is the current time, is the time when the fault occurred, The lead time required to preload the standby hydraulic pump, If and When the values ​​are equal, the preloading process of the standby hydraulic pump is started;

[0059] It should be noted that the above formulas are all dimensionless and numerical calculations. Common dimensionless methods include Min-Max normalization and Z-Score normalization, which will not be described here.

[0060] Step S2, obtaining oil viscosity abnormality information and load increasing stability information of the standby hydraulic pump during the preloading process;

[0061] By acquiring the abnormal oil viscosity information during the preloading process of the standby hydraulic pump, analyzing the oil viscosity during the preloading process of the standby hydraulic pump, and calculating the abnormal oil viscosity coefficient, the abnormal degree of the oil viscosity during the preloading process of the standby hydraulic pump is measured;

[0062] The logic for obtaining the oil viscosity abnormality coefficient is as follows:

[0063] The oil viscosity of the standby hydraulic pump during preloading is collected by a high-precision sensor to construct an oil viscosity time series: ,in represents the oil viscosity collected at time t, t={1,2,...,T}, where T is a positive integer;

[0064] Cluster analysis is performed on the oil viscosity time series, and the oil viscosity anomaly coefficient is calculated as follows:

[0065] Step S21, using the silhouette coefficient method to determine the initial number of cluster centers SL, randomly selecting the oil viscosity at SL time points from the oil viscosity time series as the initial cluster centers;

[0066] It should be noted that the silhouette coefficient method is a commonly used clustering effect evaluation method. Calculating the silhouette coefficient can help select the optimal number of initial clusters. The silhouette coefficient is used to measure the closeness of each point to its similar points and the degree of separation from the nearest cluster. The calculation formula of the silhouette coefficient is: ,in represents the silhouette coefficient, The average distance between a sample point and all other sample points in the same class represents the cohesion of the sample point. The separation degree of the sample point is represented by the average distance between the sample point and all the nearest sample points in different classes. By calculating the silhouette coefficient under different cluster numbers, the number of clusters that maximizes the silhouette coefficient is selected as the initial number of cluster centers SL.

[0067] Step S22, calculating the Euclidean distance between the oil viscosity at each time point in the oil viscosity time series and the cluster center, and assigning the oil viscosity at each time point to the cluster center with the closest Euclidean distance;

[0068] In an alternative example, the Euclidean distance is calculated: for each time point t, the Euclidean distance between the oil viscosity value and all cluster centers is calculated. , : ,in is the oil viscosity value at time point t, is the oil viscosity value of the kth cluster center;

[0069] Step S23, calculating the average value of the oil viscosity in each cluster center and using it as the new cluster center coordinates;

[0070] Step S24, repeating steps S22 and S23 to iteratively update the cluster center until the cluster center no longer changes;

[0071] Calculate the deviation of the oil viscosity at each time point from the center of its cluster: ,in represents the deviation of the oil viscosity from the center of its cluster, Indicates the oil viscosity value at time point t, represents the oil viscosity value of the kth cluster center;

[0072] Calculate the oil viscosity abnormal coefficient: ,in is the abnormal coefficient of oil viscosity, is the deviation between the oil viscosity at time point t and its cluster center, Time point The oil viscosity value, is the time point contained in the k-th cluster center;

[0073] It should be noted that the above formulas are all dimensionless and numerical calculations. Common dimensionless methods include Min-Max normalization and Z-Score normalization, which will not be described here.

[0074] The oil viscosity anomaly coefficient in this invention is a key indicator used to measure the degree of oil viscosity anomaly in the standby hydraulic pump during the preloading process. Oil viscosity is the resistance coefficient to the flow of hydraulic oil in a hydraulic system, directly affecting the hydraulic pump's operating efficiency and system stability. During hydraulic system operation, oil viscosity can be affected by a variety of factors, such as temperature fluctuations, oil aging, and load fluctuations, which in turn affect the performance of the standby hydraulic pump. By calculating the oil viscosity anomaly coefficient, potential hidden dangers can be promptly identified and instability in the standby pump caused by abnormal oil viscosity can be prevented, thereby achieving more accurate and efficient fault warning and risk control. A large oil viscosity anomaly coefficient indicates that the oil viscosity of the standby hydraulic pump has significantly deviated from the normal range during the preloading process. This may be caused by factors such as low oil temperature, oil aging, or contaminants entering the system. When the oil viscosity increases abnormally, the hydraulic pump's operating flow rate is affected, pressure fluctuations increase, and this can lead to increased wear within the pump body. In severe cases, it can even cause the hydraulic pump to become stuck or unable to start. Therefore, a large oil viscosity anomaly coefficient is a warning sign of potential backup pump system failure, reminding the system to pay close attention to oil temperature and viscosity during backup switching to prevent unstable startup or inability of the backup hydraulic pump to effectively carry the load due to excessive oil viscosity. In contrast, a small oil viscosity anomaly coefficient indicates that the oil viscosity of the backup hydraulic pump fluctuates minimally during the preloading process, remaining within the normal operating range. This generally indicates that the hydraulic oil is in ideal operating condition, and the backup pump can maintain stability and efficiency when taking over the load. A small anomaly coefficient indicates that the hydraulic system's temperature control, filtration, and lubrication are effective, the oil is not contaminated or excessively aged, and the hydraulic pump's workload can be smoothly transferred from the primary pump to the backup pump, reducing response delays during system switching and improving the efficiency and safety of the switching process. Assessing potential loading hazards during the backup hydraulic pump preloading process based on the oil viscosity anomaly coefficient provides real-time, accurate information on oil viscosity deviations, helping the hydraulic system monitor the operating status of the backup pump. By promptly detecting abnormal changes in oil viscosity, potential operational risks facing the hydraulic pump can be identified in advance, such as excessively high oil viscosity leading to pump overload or startup difficulties, or excessively low oil viscosity leading to insufficient lubrication and increased wear. This allows the system to take preventative measures before problems occur, preventing the backup pump from being unable to effectively switch or experiencing long-term unstable operation, thereby improving the reliability and efficiency of the entire hydraulic system.

[0075] It should be noted that high-precision sensors monitor the operating status of the backup hydraulic pump in real time, capturing oil viscosity data at every moment. This data is then constructed into a time series representing the oil viscosity value collected at time t. This data not only provides a basis for subsequent fault prediction and performance evaluation, but also provides real-time feedback on abnormalities during the backup hydraulic pump's preloading process. Next, cluster analysis is used to identify outliers in the oil viscosity data. Cluster analysis is an unsupervised learning method that groups oil viscosity values ​​in a time series and determines whether abnormal fluctuations exist by calculating the eigenvalues ​​of each cluster group. In cluster analysis, the silhouette coefficient method is first used to determine the number of cluster centers. The silhouette coefficient method is a metric used to evaluate clustering effectiveness. It determines the optimal number of clusters by calculating the similarity between each data point and its cluster center. This method not only maximizes cluster compactness and separation but also automatically selects an appropriate number of clusters, ensuring the accuracy of the cluster analysis. After determining the number of clusters, initial cluster centers are randomly selected from the oil viscosity time series. The Euclidean distance between the oil viscosity at each time point and all cluster centers is calculated. The oil viscosity at each time point is assigned to the closest cluster center based on its distance from the cluster center. This process helps group the data into groups with similar characteristics. As clustering progresses, the average oil viscosity within each cluster center is calculated and used as the new cluster center coordinates. This process is repeated until the cluster centers stabilize, meaning their positions no longer change. Using this cluster analysis, the deviation between the oil viscosity at each time point and its corresponding cluster center is calculated. The magnitude of the deviation directly reflects the degree of oil viscosity anomaly at that time point. Generally speaking, a large deviation in oil viscosity at a given time point indicates a significant deviation from normal values, potentially indicating an abnormality in the hydraulic system, such as excessively high or low oil viscosity. The oil viscosity anomaly coefficient is then calculated.

[0076] By obtaining the load increasing stability information of the standby hydraulic pump during the preloading process, analyzing the load increasing stability of the standby hydraulic pump during the preloading process, and calculating the load increasing stability coefficient, the load increasing stability of the standby hydraulic pump during the preloading process is measured;

[0077] The logic for obtaining the load increasing stability coefficient is as follows:

[0078] Gradually apply system load to the standby hydraulic pump until it reaches the system load of the main hydraulic pump, and record the system load values ​​at different times And the response time of the standby hydraulic pump to system load changes at different times ;

[0079] Calculate the load change rate: ,in represents the load change rate at time t, is the system load value at time t, is the system load value at time t-1, is the sampling time interval;

[0080] Calculate the response lag time: ,in For the response delay time, is the response time of the standby hydraulic pump to system load changes, is the expected response time;

[0081] Calculate the load increment stability value at different times: ,in Indicates the load increasing stable value at time t, Transfer function for the system The phase angle, ,in is the static gain of the system, is the time constant of the system, is the imaginary unit, is the angular frequency;

[0082] It should be noted that The static gain of the system represents the proportional relationship between the system input and output. The time constant of the system is a parameter that measures the response speed of the system (set according to the time situation). is the imaginary unit ( ), The angular frequency represents the frequency of load change;

[0083] Calculate the load increasing stability factor: ,in is the load increasing stability factor, is the load increasing stable value at time r, is the time window size;

[0084] It should be noted that the above formulas are all dimensionless and numerical calculations. Common dimensionless methods include Min-Max normalization and Z-Score normalization, which will not be described here.

[0085] The load-increment stability coefficient used in this invention measures the load-increment stability during the standby hydraulic pump's preloading process. The load-increment stability coefficient reflects the hydraulic pump's response stability as the load gradually increases, and is directly related to the hydraulic system's operational stability and reliability under load fluctuations. A larger load-increment stability coefficient generally indicates that the system maintains a stable response during the gradual load increase, that the hydraulic pump's performance is not excessively affected by load fluctuations, and that the system can smoothly carry the expected load, thus avoiding overload, vibration, or other instability. A smaller load-increment stability coefficient, on the other hand, may indicate a delayed or unstable system response during the load-increment process, potentially leading to pump overheating, pressure fluctuations, or equipment failure. In extreme cases, it could even cause system crash or equipment damage. Assessing potential loading hazards during the standby hydraulic pump's preloading process based on the load-increment stability coefficient has significant benefits. First, the load-increment stability coefficient provides a quantitative basis for the hydraulic system's dynamic stability. During the initial startup of the standby hydraulic pump, the system's load-increment process is typically slow. If the standby hydraulic pump cannot stably adapt to load changes during this process, it may be unable to respond to a primary pump failure. At this time, the evaluation based on the load increase stability coefficient can effectively monitor the stability of the system during the load increase process, provide timely early warning for the system, and improve the reliability and response speed of the standby pump.

[0086] It should be noted that a gradually increasing load is applied to the backup hydraulic pump until it reaches the predetermined load level of the main hydraulic pump. The load increase is gradual to observe the backup hydraulic pump's response under different load levels. At each load stage, the system load value and the backup hydraulic pump's response time to the load change are recorded, and the load change rate at each time point is calculated. The load change rate reflects the speed of the system load change. Load changes are typically gradual, but the speed of change has a significant impact on system stability. If the load change rate is too rapid, the system may not respond in time, leading to instability or overload. Therefore, by calculating the load change rate, we can identify load change trends that may lead to system instability. The response lag time refers to the delay in the backup hydraulic pump's response to the load change. Due to the inertia of the hydraulic pump and the characteristics of the hydraulic fluid, the backup pump's response will not be instantaneous. Therefore, it is necessary to calculate the time required for the backup hydraulic pump to reach the expected response after the load change. To further analyze the system's load-increase stability, the phase angle of the system's transfer function is used to calculate the load-increase stability value. The transfer function phase angle reflects the system's dynamic response to load changes. By calculating the load increment stability values ​​at different time points, the system's instantaneous response to load changes can be determined. Finally, the load increment stability coefficient is a quantitative evaluation of the overall load increment stability performance within a time window. This coefficient evaluates the standby hydraulic pump's adaptability to load increments by calculating the load increment stability values ​​at various time points within that time window.

[0087] Step S3: constructing a standby switching hidden danger risk model based on the abnormal oil viscosity information and load increasing stability information during the standby hydraulic pump preloading process, outputting a standby switching hidden danger risk index, and evaluating potential loading hidden dangers during the standby hydraulic pump preloading process;

[0088] Construct a standby switching hidden danger risk model based on the oil viscosity abnormality coefficient and load increasing stability coefficient, and output the standby switching hidden danger risk index The model is based on the following formula , where They represent the preset proportional coefficients of the oil viscosity abnormality coefficient and the load increasing stability coefficient, respectively, and All greater than 0;

[0089] It should be noted that the above formulas are all dimensionless and numerical calculations. Common dimensionless methods include Min-Max normalization and Z-Score normalization, which will not be described here. Set it according to the actual situation. For example, adopt the expert empowerment method, that is, invite experts in related fields to determine the preset proportion coefficients of various indicators through professional opinion surveys and comprehensive evaluations, for example, It can be 0.5, 0.5;

[0090] From the above calculation expression, it can be seen that the larger the oil viscosity abnormality coefficient and the smaller the load incremental stability coefficient, the greater the standby switching hidden danger risk index, indicating that the oil viscosity abnormality and load incremental instability are more serious, and the preloading process of the standby hydraulic pump is at a higher risk, which may lead to instability or even failure during the switching process. Conversely, the smaller the oil viscosity abnormality coefficient and the larger the load incremental stability coefficient, the smaller the standby switching hidden danger risk index, indicating that the preloading process of the standby hydraulic pump is relatively stable and can smoothly take over the main pump load. The hidden dangers in the preloading process are relatively small, and the switching operation can continue;

[0091] In step S4, the standby switching hidden danger risk index is compared with a preset standby switching hidden danger risk index threshold, and the hydraulic pump is controlled by classification, as follows:

[0092] If the standby switching hidden danger risk index is greater than the standby switching hidden danger risk index threshold, it means that there is a high risk in the preloading process of the standby hydraulic pump. The current standby hydraulic pump is marked as a switching risk device, and early warning maintenance is performed on the marked switching risk device.

[0093] If the standby switching hidden danger risk index is less than or equal to the standby switching hidden danger risk index threshold, it means that the preloading process of the standby hydraulic pump is relatively stable. The standby hydraulic pumps that are not marked as switching risk equipment will be sorted from small to large according to the standby switching hidden danger risk index, and a standby equipment switching sorting table will be generated. The standby equipment ranked first will be switched.

[0094] The present invention collects the operating data of the main hydraulic pump and performs fault prediction, so as to timely discover the potential faults of the main equipment and start the preloading process of the standby hydraulic pump according to the time when the fault occurs. By obtaining the oil viscosity abnormality information and load incremental stability information during the preloading process of the standby hydraulic pump in real time, and calculating the oil viscosity abnormality coefficient and load incremental stability coefficient based on this, the stability of the standby pump during the switching process is effectively evaluated, and the working status of the standby pump is comprehensively analyzed. When the oil viscosity is abnormal or the load incremental is unstable, the potential loading hidden dangers can be identified in advance, and corresponding control can be performed through the output of the risk model to avoid equipment failure or system instability due to improper switching. According to the comparison between the standby switching hidden danger risk index and the preset threshold, the switching risk of the standby hydraulic pump is dynamically judged. When the risk is high, early warning and maintenance are automatically performed to ensure that the performance of the standby equipment meets the switching requirements. When the risk is low, the standby hydraulic pump with the lowest risk is preferentially selected for switching to ensure smooth operation and seamless switching of the hydraulic system.

[0095] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0096] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0097] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for automatic control of equipment in a hydraulic station, characterized by: The steps include: Step S1, collecting operating data of the main hydraulic pump, and predicting the failure time of the main hydraulic pump based on the operating data of the main hydraulic pump, and starting the preloading process of the backup hydraulic pump according to the failure time; Step S2, obtaining oil viscosity abnormality information and load increasing stability information of the standby hydraulic pump during the preloading process; Step S3: constructing a standby switching hidden danger risk model based on the abnormal oil viscosity information and load increasing stability information during the standby hydraulic pump preloading process, outputting a standby switching hidden danger risk index, and evaluating potential loading hidden dangers during the standby hydraulic pump preloading process; Step S4, comparing the standby switching hidden danger risk index with a preset standby switching hidden danger risk index threshold, and performing classification control on the hydraulic pump; By acquiring the abnormal oil viscosity information during the preloading process of the standby hydraulic pump, analyzing the oil viscosity during the preloading process of the standby hydraulic pump, and calculating the abnormal oil viscosity coefficient, the abnormal degree of the oil viscosity during the preloading process of the standby hydraulic pump is measured; The logic for obtaining the oil viscosity abnormality coefficient is as follows: The oil viscosity of the standby hydraulic pump during preloading is collected by a high-precision sensor to construct an oil viscosity time series: ,in represents the oil viscosity collected at time t, t={1,2,...,T}, where T is a positive integer; Cluster analysis is performed on the oil viscosity time series, and the oil viscosity anomaly coefficient is calculated as follows: Step S21, using the silhouette coefficient method to determine the initial number of cluster centers SL, randomly selecting the oil viscosity at SL time points from the oil viscosity time series as the initial cluster centers; Step S22, calculating the Euclidean distance between the oil viscosity at each time point in the oil viscosity time series and the cluster center, and assigning the oil viscosity at each time point to the cluster center with the closest Euclidean distance; Step S23, calculating the average value of the oil viscosity in each cluster center and using it as the new cluster center coordinates; Step S24, repeating steps S22 and S23 to iteratively update the cluster center until the cluster center no longer changes; Calculate the deviation of the oil viscosity at each time point from the center of its cluster: ,in represents the deviation of the oil viscosity from the center of its cluster, Indicates the oil viscosity value at time point t, represents the oil viscosity value of the kth cluster center; Calculate the oil viscosity abnormal coefficient: ,in is the abnormal coefficient of oil viscosity, is the deviation between the oil viscosity at time point t and its cluster center, Time point The oil viscosity value, is the time point contained in the k-th cluster center; By obtaining the load increasing stability information of the standby hydraulic pump during the preloading process, analyzing the load increasing stability of the standby hydraulic pump during the preloading process, and calculating the load increasing stability coefficient, the load increasing stability of the standby hydraulic pump during the preloading process is measured; The logic for obtaining the load increasing stability coefficient is as follows: Gradually apply system load to the standby hydraulic pump until it reaches the system load of the main hydraulic pump, and record the system load values ​​at different times And the response time of the standby hydraulic pump to system load changes at different times ; Calculate the load change rate: ,in represents the load change rate at time t, is the system load value at time t, is the system load value at time t-1, is the sampling time interval; Calculate the response lag time: ,in For the response delay time, is the response time of the standby hydraulic pump to system load changes, is the expected response time; Calculate the load increment stability value at different times: ,in Indicates the load increasing stable value at time t, Transfer function for the system The phase angle, ,in is the static gain of the system, is the time constant of the system, is the imaginary unit, is the angular frequency; Calculate the load increasing stability factor: ,in is the load increasing stability factor, is the load increasing stable value at time r, is the time window size; Construct a standby switching hidden danger risk model based on the oil viscosity abnormality coefficient and load increasing stability coefficient, and output the standby switching hidden danger risk index The model is based on the following formula , where They represent the preset proportional coefficients of the oil viscosity abnormality coefficient and the load increasing stability coefficient, respectively, and Both are greater than 0.

2. The method for automatic control of equipment in a hydraulic station according to claim 1, characterized in that: By installing high-precision sensors, the operating data of the main hydraulic pump is collected in real time, including temperature, pressure, flow, vibration, and oil viscosity; Data preprocessing is performed on the collected main hydraulic pump operation data, including missing value processing and data standardization; The fault occurrence time is used as the target variable, the operating data of the main hydraulic pump is used as the feature variable, and a feature vector is constructed based on the operating data of the main hydraulic pump; Build a failure occurrence time prediction model: ,in is the output of the failure occurrence time prediction model, is the weight vector The transpose of is the characteristic variable, is the bias term; Construct the objective function; Use the historical operating data of the main hydraulic pump to train the fault occurrence time prediction model and minimize the objective function to find the optimal weight vector and bias ; Input the real-time collected main hydraulic pump operation data into the trained fault occurrence time prediction model to predict the fault occurrence time of the main hydraulic pump; The starting conditions for starting the preloading process of the standby hydraulic pump according to the fault occurrence time are: ,in is the current time, is the time when the fault occurred, The lead time required to preload the standby hydraulic pump, If and When the values ​​are equal, the preloading process of the standby hydraulic pump is started.

3. The method for automatic control of equipment in a hydraulic station according to claim 1, characterized in that: In step S4, the standby switching hidden danger risk index is compared with a preset standby switching hidden danger risk index threshold, and the hydraulic pump is controlled by classification, as follows: If the standby switching hidden danger risk index is greater than the standby switching hidden danger risk index threshold, the current standby hydraulic pump is marked as a switching risk device, and early warning maintenance is performed on the marked switching risk device; If the standby switching hidden danger risk index is less than or equal to the standby switching hidden danger risk index threshold, the standby hydraulic pumps that are not marked as switching risk equipment will be sorted from small to large according to the standby switching hidden danger risk index, and a standby equipment switching sorting table will be generated, and the standby equipment ranked first will be switched.

Images