Working condition analysis method and device of fracturing pump, computer equipment and storage medium

Abstract

The invention relates to the technical field of fracturing, and discloses a working condition analysis method and device of a fracturing pump, computer equipment and a storage medium, and the method comprises the following steps: obtaining operation data of a power end and a fluid end in the fracturing pump; corresponding key parameters are calculated according to the operation data, and a current working condition curve of the fracturing pump is generated based on the key parameters; the overload condition of the fracturing pump is analyzed through the current working condition curve, and an analysis result is obtained. The problems that in the prior art, the working condition of the fracturing pump cannot be comprehensively monitored in real time, and overload cannot be accurately judged are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fracturing, in particular to a working condition analysis method and device of a fracturing pump, a computer device and a storage medium. BACKGROUND

[0002] Fracturing technology is widely used in oil and gas well stimulation and water injection well stimulation, and is one of the effective measures for improving oil and gas well recovery rate at present. As the core equipment of fracturing operation, the reliability of fracturing pump is directly related to the efficiency and safety of the whole fracturing operation. In actual application, the monitoring of the working condition of the existing fracturing pump operation only relies on single data and human monitoring, lacks comprehensive collection and analysis of multi-parameter operation data, and cannot accurately judge the overload condition of the fracturing pump by calculating key parameters and generating a current working condition curve. This leads to the working condition of the fracturing pump operation often exceeding the recommended operation curve, and long-time construction easily causes the key components of the power end to be damaged in advance, thereby reducing the service life and operation reliability of the fracturing pump. SUMMARY

[0003] Therefore, the embodiments of the present application provide a working condition analysis method and device of a fracturing pump, a computer device and a storage medium to solve the problem that the working condition of the fracturing pump cannot be monitored in real time and comprehensively and the overload cannot be accurately judged in the prior art.

[0004] In a first aspect, the embodiments of the present application provide a working condition analysis method of a fracturing pump, and the method comprises: obtaining operation data of a power end and a fluid end in the fracturing pump; calculating corresponding key parameters according to the operation data, and generating a current working condition curve of the fracturing pump based on the key parameters; analyzing the overload condition of the fracturing pump by using the current working condition curve to obtain an analysis result.

[0005] Further, the calculating corresponding key parameters according to the operation data comprises: extracting flow data, pressure data, rotating speed data and temperature data in the operation data; calculating connecting rod load according to the pressure data, calculating instantaneous power according to the flow data, calculating instantaneous torque of a crankshaft according to the rotating speed data, and obtaining lubricating oil return temperature according to the temperature data; taking the connecting rod load, the instantaneous power, the instantaneous torque of the crankshaft and the lubricating oil return temperature as the key parameters.

[0006] Further, the generating a current working condition curve of the fracturing pump based on the key parameters comprises: The key parameters are fused based on a mapping relationship between preset mechanical parameters and preset distribution weights to obtain a fusion parameter representing a running state of the fracturing pump. The fusion parameters are arranged in a time sequence to obtain a parameter sequence. A multi-dimensional parameter space is constructed based on the parameter sequence, and a current working condition curve of the fracturing pump is constructed in the multi-dimensional parameter space.

[0007] Further, the method further comprises: monitoring a change trend of the lubricating oil return temperature; generating a temperature abnormality signal when the change trend of the lubricating oil return temperature has an abnormality; adjusting a preset distribution weight corresponding to the lubricating oil return temperature based on the temperature abnormality signal.

[0008] Further, the current working condition curve is used to analyze an overload condition of the fracturing pump to obtain an analysis result, which comprises: obtaining a target working condition curve of the fracturing pump, wherein the target working condition curve is a working condition curve generated according to optimal parameters of the fracturing pump; comparing the current working condition curve with the target working condition curve to obtain a comparison result; if the comparison result is that the current working condition curve matches the target working condition curve, it is determined that the analysis result is that the fracturing pump has no overload condition, or if the comparison result is that the current working condition curve does not match the target working condition curve, it is determined that the analysis result is that the fracturing pump has an overload condition.

[0009] Further, the method further comprises: inputting the running data and the key parameters into a pre-trained analysis model; extracting features of the running data and the key parameters through the pre-trained analysis model to obtain a feature vector; performing working condition recognition on the fracturing pump based on the feature vector to obtain a recognition result; calibrating the analysis result using the recognition result to obtain a calibrated analysis result.

[0010] Further, after the current working condition curve is used to analyze the overload condition of the fracturing pump to obtain an analysis result, the method further comprises: triggering an alarm mechanism to generate corresponding overload alarm information when the analysis result is that the fracturing pump has an overload condition; determining an overload type of the fracturing pump according to the running data and the key parameters, wherein the overload type comprises at least one of power overload, torque overload, and load overload; Obtain the target handling strategy corresponding to the overload type, and send the overload alarm information and the target handling strategy to the monitoring terminal.

[0011] Secondly, embodiments of the present invention provide a fracturing pump operating condition analysis device, the device comprising: The acquisition module is used to acquire operating data from the power end and hydraulic end of the fracturing pump; The calculation module is used to calculate the corresponding key parameters based on the operating data, and generate the current operating condition curve of the fracturing pump based on the key parameters; The analysis module is used to analyze the overload condition of the fracturing pump using the current operating condition curve and obtain the analysis results.

[0012] Thirdly, embodiments of the present invention provide a computer device, including: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the method described in the first aspect or any corresponding embodiment thereof.

[0013] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing computer instructions that cause a computer to perform the method described in the first aspect or any of its corresponding embodiments.

[0014] The method provided in this application has the following beneficial effects: The method provided in this application acquires multi-parameter operating data from the power and hydraulic ends of the fracturing pump, enabling comprehensive perception and data foundation construction of the fracturing pump's operating status. Furthermore, by calculating key parameters and generating a current operating condition curve, the collected data is transformed into a visual indicator that intuitively reflects the equipment's load status, providing a dynamic and quantitative analytical basis for overload judgment. Finally, overload analysis is performed using the current operating condition curve, achieving real-time and accurate assessment of the fracturing pump's operating status and fault early warning. This effectively overcomes the limitations of traditional single-parameter monitoring, significantly improves the reliability and safety of fracturing pump operations, and extends the service life of key components. Attached Figure Description

[0015] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0016] Figure 1This is a flowchart illustrating the operating condition analysis method for a fracturing pump according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a sensor deployed on a fracturing pump according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the recommended plunger horsepower curve according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the recommended operating curve according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of a fracturing pump alarm system according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the working process of a fracturing pump alarm system according to an embodiment of the present invention; Figure 7 This is a structural block diagram of a fracturing pump operating condition analysis device according to an embodiment of the present invention; Figure 8 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] According to embodiments of the present invention, a method, apparatus, computer equipment, and storage medium for analyzing the operating conditions of fracturing pumps are provided. It should be noted that the steps shown in the flowcharts in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowcharts, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0019] This embodiment provides a method for analyzing the operating conditions of a fracturing pump. Figure 1 This is a flowchart of a fracturing pump operating condition analysis method according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps: Step S101: Obtain the operating data of the power end and hydraulic end of the fracturing pump.

[0020] In the embodiments of this application, such as Figure 2As shown, multiple sensors are deployed at key locations on the fracturing pump body to directly measure physical signals: speed and temperature sensors are installed at the power end (i.e., the transmission part of the fracturing pump, typically including components such as crankshaft and connecting rod); pressure and flow sensors are installed at the hydraulic end (i.e., the fluid working part of the fracturing pump, typically including components such as plunger and cylinder). Subsequently, the data acquisition module synchronously collects and formats the measured operating data (including input speed, lubricating oil return temperature, discharge pressure, and displacement) from all the above sensors in real time, forming a time-series dataset, laying a solid data foundation for parameter calculation and curve generation.

[0021] Among them, the speed sensor is used to measure the input speed of the power input shaft, which is crucial for calculating the crankshaft torque; the temperature sensor is installed on the return oil line of the lubrication system to measure the return oil temperature, which can serve as an auxiliary basis for judging whether the system generates abnormal heat due to overload; the pressure sensor is used to measure the discharge pressure at the pump outlet; and the flow sensor is used to measure the pump's displacement (i.e., flow rate). The two data points output by the pressure sensor and the flow sensor are the core of calculating the pump's instantaneous power and connecting rod load.

[0022] Step S102: Calculate the corresponding key parameters based on the operating data, and generate the current operating condition curve of the fracturing pump based on the key parameters.

[0023] In this embodiment, firstly, flow rate data, pressure data, rotational speed data, and temperature data are extracted from the operating data. Then, calculations are performed based on physical principles and a preset algorithm: the connecting rod load is calculated based on the pressure data (i.e., discharge pressure) and the known plunger cross-sectional area; this parameter directly reflects the force exerted by the hydraulic end on the power end. The instantaneous power of the fracturing pump is calculated based on the flow rate data (i.e., displacement) and pressure data, characterizing the instantaneous output capability of the entire machine. The instantaneous crankshaft torque is calculated based on the rotational speed data (i.e., input rotational speed) and the calculated power or torque characteristics, used to assess the torsional load of the core transmission components. Simultaneously, the temperature data (i.e., lubricating oil return temperature) is either directly read or smoothed and filtered before being used as the lubricating oil return temperature parameter to assist in determining the system's thermal load status. These calculated connecting rod loads, instantaneous power, crankshaft instantaneous torque, and lubricating oil return temperature are collectively determined as key parameters characterizing the health status and load level of the fracturing pump. Then, based on the mapping relationship between preset key parameters and preset weights (for example, under normal operating conditions, instantaneous power and connecting rod load may be assigned higher weights), these parameters are weighted and fused to obtain a fused parameter that can comprehensively characterize the operating state of the fracturing pump. As time progresses, the continuously generated fused parameters are arranged in a time series to form a parameter sequence. Finally, based on this parameter sequence, points are plotted and fitted in a multi-dimensional parameter space composed of dimensions such as time, load, power, torque, and temperature to dynamically and comprehensively display the motion trajectory of the fracturing pump's real-time operating points, i.e., the current operating condition curve of the fracturing pump.

[0024] It should be noted that the preset weight allocation can be dynamically adjusted. For example, when an abnormal upward trend in the lubricating oil return temperature is detected, the weight of the temperature parameter is automatically increased, making the fused parameters and the subsequently generated operating condition curve more sensitive to abnormal heat load.

[0025] Step S103: Analyze the overload condition of the fracturing pump using the current operating condition curve to obtain the analysis results.

[0026] In this embodiment, firstly, a target operating condition curve customized for this type of fracturing pump is obtained from pre-stored data. This curve is generated based on the optimal performance parameters and safe operating boundaries of the fracturing pump, representing the safe operating range under ideal conditions. Then, the current operating condition curve, reflecting the equipment's operating status, is precisely compared with the target operating condition curve as a benchmark within the same multi-dimensional parameter space. The comparison process includes checking whether the current operating condition curve exceeds the envelope area defined by the target operating condition curve, or determining whether the trajectory deviation between the two in key parameter dimensions exceeds a preset tolerance threshold, thereby obtaining the comparison result. Finally, a logical judgment is made based on the comparison result: if the comparison result shows that the current operating condition curve matches the target operating condition curve (i.e., the current curve is within the safe envelope of the target curve), the analysis result is determined to be that the fracturing pump is not overloaded; if the comparison result shows that the two do not match (i.e., the current curve partially or completely exceeds the safe boundary of the target curve), the analysis result is determined to be that the fracturing pump is overloaded. This analysis process realizes the transformation from data to decision, providing a direct basis for subsequent alarms and control.

[0027] Matching decisions do not have to be simple binary judgments. Dynamic thresholds or fuzzy logic can be introduced. For example, when the current operating condition curve hovers near the boundary of the target curve or slightly exceeds it briefly, it can be judged as a critical state and a prompt message can be output, thereby providing a more refined classification and warning of overload risk.

[0028] In this embodiment of the application, the calculation of corresponding key parameters based on the running data includes the following steps A1-A3: Step A1: Extract the flow rate data, pressure data, speed data, and temperature data from the operating data.

[0029] Specifically, the system uses its built-in data processing logic (which can be used as part of the data analysis and processing module) to identify and separate flow data (i.e., time-series data representing the fracturing pump displacement measured by the flow sensor at the hydraulic end), pressure data (i.e., time-series data representing the fracturing pump discharge pressure measured by the pressure sensor at the hydraulic end), rotational speed data (i.e., time-series data representing the input shaft rotational speed measured by the rotational speed sensor at the power end), and temperature data (i.e., time-series data representing the lubricating oil return temperature measured by the temperature sensor at the power end) from the real-time collected and stored operating data (which includes synchronous or asynchronous readings from multiple sensors at the power and hydraulic ends).

[0030] In addition, this data extraction process includes necessary data cleaning and synchronization alignment operations. For example, interpolation or resampling is performed on the timing differences caused by different sensors due to sampling frequency or transmission delay to ensure that the extracted flow rate data, pressure data, speed data and temperature data are strictly aligned on the timestamp.

[0031] Step A2: Calculate the connecting rod load based on the pressure data, calculate the instantaneous power based on the flow rate data, calculate the crankshaft instantaneous torque based on the rotational speed data, and obtain the lubricating oil return temperature based on the temperature data.

[0032] Specifically, the conversion from raw data to mechanical parameters is based on physical principles and preset algorithms. It transforms extracted basic sensor data into characteristic parameters that directly reflect the stress and energy state of key components in the fracturing pump. Simultaneously, the system can incorporate parameter verification logic, such as checking whether the calculated connecting rod load or crankshaft instantaneous torque is within physically possible ranges, to initially ensure the validity of the data. The conversion process includes, but is not limited to: The connecting rod load is calculated based on the pressure data by multiplying the pressure data (i.e., discharge pressure) measured by the hydraulic end pressure sensor with the plunger cross-sectional area stored in the system, which is determined by the fracturing pump model. This calculation directly reflects the magnitude of the force acting on the power end connecting rod.

[0033] Calculating instantaneous power based on flow data involves combining the flow data (i.e. displacement) measured by the hydraulic end flow sensor with the aforementioned pressure data, and then calculating the real-time output power of the fracturing pump according to the power calculation formula (e.g., power = pressure × flow).

[0034] Calculating the instantaneous torque of the crankshaft based on the rotational speed data usually requires combining the calculated instantaneous power with the rotational speed data (input rotational speed) measured by the speed sensor at the power end. The relationship between torque, power, and rotational speed (e.g., torque = power / (2π × rotational speed / 60)) is used to solve the problem, thereby assessing the torsional load on the crankshaft.

[0035] The lubricating oil return temperature is obtained from the temperature data. The calculation mainly refers to the necessary signal processing of the temperature data (raw value of lubricating oil return temperature) directly measured by the temperature sensor at the power end, such as digital filtering to eliminate noise interference, or calculating the average value in the short term to obtain a stable reading, and finally outputting the lubricating oil return temperature parameter value for auxiliary analysis.

[0036] Step A3 uses connecting rod load, instantaneous power, crankshaft instantaneous torque, and lubricating oil return temperature as key parameters.

[0037] Specifically, the four parameters output in real time by the calculation module—connecting rod load (characterizing the force acting on the connecting rod at the hydraulic end), instantaneous power (characterizing the real-time output capacity of the entire machine), crankshaft instantaneous torque (characterizing the torsional load of the core transmission components at the power end), and processed lubricating oil return temperature (characterizing the thermal load state of the system)—are logically or structurally aggregated and bound together. By defining these four parameters as key parameters, a multi-dimensional evaluation system is established. This system simultaneously covers multiple key aspects of the fracturing pump, including mechanical load (reflected by connecting rod load and crankshaft instantaneous torque), energy load (reflected by instantaneous power), and thermal load (reflected by lubricating oil return temperature). This provides a structured input data foundation for subsequently generating current operating condition curves that comprehensively reflect the overall operating conditions of the equipment. Furthermore, a data object or array with a specific structure is created within the program. This object stores these four parameters and their corresponding timestamps in a preset format, ensuring that they are transmitted to the curve generation module as a whole.

[0038] By extracting flow rate, pressure, speed, and temperature data from operational data, precise screening and classification of multi-source heterogeneous sensor data were achieved, providing structured input for subsequent calculations. By calculating connecting rod load from pressure data, instantaneous power from flow rate data, crankshaft instantaneous torque from speed data, and lubricating oil return temperature from temperature data, the raw data were transformed into physical parameters reflecting the core mechanical state and thermal load of the fracturing pump, enhancing the physical significance and engineering applicability of the operating condition analysis. By using these parameters as key parameters, a multi-dimensional evaluation system covering mechanical load, energy load, and thermal load was constructed, laying a complete data foundation for generating comprehensive operating condition curves.

[0039] In this embodiment of the application, generating the current operating condition curve of the fracturing pump based on key parameters includes the following steps: Step B1: Based on the mapping relationship between preset mechanical parameters and preset weight allocation, the key parameters are fused to obtain fused parameters that characterize the operating status of the fracturing pump.

[0040] Specifically, the system has pre-stored mapping relationships that clearly define the preset weights assigned to each key parameter (for example, under standard operating conditions, instantaneous power and connecting rod load may be assigned higher weights to directly reflect mechanical and energy loads, while lubricating oil return temperature may be assigned a relatively lower auxiliary weight under normal conditions). Based on this mapping relationship, the values ​​of the key parameters acquired in real time are multiplied by their corresponding preset weights, and then all weighted parameter values ​​are summed (or other specific fusion algorithms, such as weighted averages, are used) to finally calculate a single, dimensionless or dimensionless value, i.e., the fused parameter.

[0041] In addition, the preset weight values ​​can be set and optimized based on the design characteristics and historical operating data of different fracturing pump models, through prior AI model training or expert experience, to ensure that the fusion results can most effectively represent the operating status of the specific pump type.

[0042] Step B2: Arrange the fusion parameters according to the time series to obtain the parameter sequence.

[0043] Specifically, the purpose of constructing dynamic operating condition curves for data structuring is to organize and arrange multiple fusion parameters calculated at discrete time points (each fusion parameter representing the comprehensive operating state of the fracturing pump at a specific moment) according to their chronological order of generation. This can be achieved by maintaining a data cache or database table with timestamps. Whenever a new fusion parameter is calculated, it is associated with the timestamp corresponding to that parameter (e.g., the moment of data acquisition or calculation completion), and these time-stamped fusion parameters are sequentially stored in a linear data structure (such as an array, list, or queue) according to the order of their timestamps. Through this process, the fusion parameters, which originally represented instantaneous states, are organized into a continuous, time-ordered data set, i.e., a parameter sequence.

[0044] This arrangement process can also include checking and processing the continuity of data. For example, when data is missing at a certain point in time due to data transmission failure, an interpolation algorithm (such as linear interpolation) can be used to estimate the missing value based on the fusion parameters of the previous and next times, thus maintaining the continuity of the parameter sequence on the time axis.

[0045] Step B3: Construct a multidimensional parameter space based on the parameter sequence, and then construct the current operating condition curve of the fracturing pump in the multidimensional parameter space.

[0046] Specifically, the role of operational condition visualization and quantitative analysis is to transform a time-sequential sequence of parameters into a graphical representation that dynamically displays the evolution of the comprehensive operating status of the fracturing pump. The implementation is as follows: First, a multi-dimensional parameter space is constructed. This space is an abstract mathematical structure, whose coordinate axes include at least a time axis and at least one dimension axis that can characterize the magnitude of the fused parameter value or the comprehensive state level it represents (for example, in a two-dimensional space, the horizontal axis represents time, and the vertical axis represents the value of the fused parameter; or a higher-dimensional space may also include the decomposition dimensions of key parameters). Then, each data point in the parameter sequence (i.e., each timestamp and its corresponding fused parameter value) is used as a set of coordinates and mapped or plotted into this pre-defined multi-dimensional parameter space. Finally, a specific algorithm (such as linear interpolation or curve fitting) is used to connect the discrete data points in an orderly manner, forming a continuous and smooth trajectory line. This trajectory line, generated in the multi-dimensional parameter space, that shows the path of change in the comprehensive operating status of the fracturing pump within a specific time period, is defined as the current operational condition curve of the fracturing pump. The construction process can be dynamic and rolling, retaining only the parameter sequence for the most recent period (e.g., the past hour) to generate and update the current operating condition curve.

[0047] By fusing key parameters based on a preset mapping relationship, multi-parameter weighted integration was achieved, generating fused parameters that can comprehensively characterize the operating status of the fracturing pump, thus improving the comprehensiveness and accuracy of the status assessment. By arranging the fused parameters according to a time series, a parameter sequence reflecting the dynamic changes in equipment status was formed, providing structured data support for time series analysis. By constructing a multi-dimensional parameter space based on the parameter sequence and generating the current operating condition curve, abstract data was transformed into an intuitive and traceable graphical trajectory, realizing the dynamic visualization and precise characterization of the fracturing pump's operating status.

[0048] In this embodiment of the application, the method further includes the following steps C1-C3: Step C1: Monitor the changing trend of lubricating oil return temperature.

[0049] Specifically, the intelligent assisted diagnosis and dynamic weight adjustment pre-monitoring function continuously observes the lubricating oil return temperature, a key parameter, to capture its dynamic behavior characteristics. This is achieved as follows: periodically extracting a data window of the most recent specified time length (e.g., data from the past 30 minutes) from the lubricating oil return temperature data sequence, and applying a trend analysis algorithm to the temperature data within this window: determining whether the temperature is stable, rising, or falling by calculating the linear regression slope of the temperature data within that time period; or identifying anomalies by comparing the deviation of the current temperature value from the normal operating temperature range established based on historical data.

[0050] Step C2: When there is an abnormal trend in the change of lubricating oil return temperature, a temperature abnormality signal is generated.

[0051] Specifically, the trend of lubricating oil return temperature (e.g., a positive slope indicating a sustained rapid temperature rise, or a marker indicating that the temperature has exceeded the normal fluctuation range) is compared with preset abnormal condition judgment criteria. Judgment criteria may include: determining whether the calculated temperature rise slope exceeds a preset slope threshold; determining whether the continuous temperature rise exceeds a preset time length; or determining whether the temperature value exceeds a dynamic early warning boundary established based on historical data within a short period. When at least one preset abnormal condition judgment criterion is met, a temperature abnormality signal is generated. This signal is a logical signal or message transmitted within the system, clearly indicating the occurrence of an abnormal thermal state event in the lubrication system and activating the weight adjustment mechanism.

[0052] Step C3: Adjust the preset weighting corresponding to the lubricating oil return temperature based on the temperature anomaly signal.

[0053] Specifically, based on preset rules (e.g., defining a lookup table for different weight adjustment ranges corresponding to different anomaly levels), the preset weight allocation set stored in the system for parameter fusion is modified, that is, the preset weight allocation corresponding to the lubricating oil return temperature is adjusted: when an abnormal temperature trend is confirmed, the weight value is increased, for example, from 0.1 in the normal state to 0.3 or higher. Furthermore, the adjustment mechanism can be reversible; that is, if the temperature trend is continuously monitored and subsequent monitoring shows that the lubricating oil return temperature trend returns to normal, the corresponding preset weight allocation will be gradually or directly restored to the default value.

[0054] By monitoring the changing trend of lubricating oil return temperature, continuous tracking of thermal load status is achieved, enhancing the foresight of fault early warning; by generating temperature anomaly signals when trends are abnormal, potential overheating risks are promptly transformed into manageable logical events, improving the response capability to latent faults; by adjusting the preset allocation weights corresponding to lubricating oil return temperature based on temperature anomaly signals, adaptive optimization of the evaluation model is achieved, making the fused parameters and operating condition curves more sensitive to thermal load anomalies and improving the judgment accuracy under complex operating conditions.

[0055] In this embodiment of the application, the overload condition of the fracturing pump is analyzed using the current operating condition curve to obtain the analysis results, including the following steps D1-D3: Step D1: Obtain the target operating condition curve of the fracturing pump, wherein the target operating condition curve is the operating condition curve generated based on the optimal parameters of the fracturing pump.

[0056] Specifically, by accessing a database or configuration file pre-stored with reference data for various fracturing pump models, the target operating condition curve is obtained based on the specific model of the currently monitored fracturing pump. The target operating condition curve (e.g.) Figure 3The recommended plunger horsepower curve shown, or Figure 4 The recommended operating curve shown is not generated in real time, but is pre-calculated and stored based on the optimal parameters determined during the design phase of this type of fracturing pump (i.e., multiple recommended optimal operating points are pre-configured under the premise of ensuring safety, reliability and long service life, usually taking into account factors such as structural strength, efficiency and thermal balance). The target operating curve defines the ideal envelope region or boundary trajectory of the pump for safe and efficient operation in a multi-dimensional parameter space, and serves as the benchmark for judging whether the current operating state is overloaded.

[0057] Step D2: Compare the current operating condition curve with the target operating condition curve to obtain the comparison result.

[0058] Specifically, within the same multidimensional parameter space, the current operating condition curve is spatially compared with a target operating condition curve serving as a benchmark. The comparison process may include checking whether the current operating condition curve lies entirely within the safe operating envelope defined by the target operating condition curve, or whether its trajectory remains consistent with the target curve. The comparison results are output using a specific algorithm (e.g., calculating the Euclidean distance between the two curves at different parameter points, or determining whether a point on the current curve exceeds the boundary function of the target curve). The comparison results can be Boolean values ​​(true / false) indicating whether an error has occurred; percentages indicating the degree of compliance; or structured data objects containing detailed information such as the location and magnitude of the error.

[0059] Step D3: If the comparison result shows that the current operating condition curve matches the target operating condition curve, then the analysis result is determined to be that the fracturing pump is not overloaded; or, if the comparison result shows that the current operating condition curve does not match the target operating condition curve, then the analysis result is determined to be that the fracturing pump is overloaded.

[0060] Specifically, if the comparison result shows that the current operating condition curve matches the target operating condition curve, meaning the current operating condition curve is entirely within the safety envelope specified by the target operating condition curve and no out-of-bounds alarm conditions are triggered, then the analysis result is determined to be that the fracturing pump is not overloaded, representing a positive signal that the equipment is in a safe operating state. Alternatively, if the comparison result shows that the current operating condition curve does not match the target operating condition curve, meaning that part or all of the trajectory of the current operating condition curve exceeds the safety boundary of the target operating condition curve, then the analysis result is determined to be that the fracturing pump is overloaded, representing an alarm signal that the equipment faces risks and requires intervention.

[0061] By acquiring the target operating condition curve of the fracturing pump, an ideal operating benchmark is provided for overload judgment. By comparing the current operating condition curve with the target operating condition curve, a precise spatial match between the real-time operating status and the safety boundary is achieved, giving the overload judgment a clear geometric and logical basis. By determining the analysis results based on the comparison results, the complex curve differences are transformed into a clear conclusion of whether or not overload exists, outputting a judgment result that directly supports decision-making, thereby improving the reliability and operability of overload analysis.

[0062] In this embodiment of the application, the method further includes: Step S201: Input the running data and key parameters into the pre-trained analysis model.

[0063] In this embodiment, a pre-trained analysis model is invoked (this model is an intelligent analysis module obtained by collecting historical operating data and corresponding operating condition labels of various fracturing pump models and training them with an AI model). Then, two data streams (operating data and key parameters) are combined and input into the model: the first data stream is the operating data without deep calculations (i.e., the raw sensor reading sequence of flow rate, pressure, speed, and temperature); the second data stream is the key parameters derived through physical calculations (i.e., connecting rod load, instantaneous power, crankshaft instantaneous torque, and lubricating oil return temperature). This method of inputting both raw data and derived parameters allows the pre-trained analysis model to perform comprehensive analysis based on both low-level signal characteristics and high-level physical semantics, providing a calibration basis for inconsistent analysis results.

[0064] Step S202: Feature vectors are obtained by extracting features from the running data and key parameters through a pre-trained analysis model.

[0065] In this embodiment, when operational data (raw flow rate, pressure, speed, and temperature data) and key parameters (calculated connecting rod load, instantaneous power, crankshaft instantaneous torque, and lubricating oil return temperature) are input into the pre-trained analysis model, the multi-layer neural network structure within the model (such as convolutional layers for capturing local temporal patterns and fully connected layers for synthesizing global features) is activated. The model automatically and non-linearly processes and transforms these input data in layers, filtering out redundant information while amplifying subtle features strongly correlated with the equipment's health status, resulting in feature vectors. These feature vectors integrate the most discriminative information from the input data, providing machine-readable input for intelligent operating condition identification.

[0066] Step S203: Based on the feature vector, the operating condition of the fracturing pump is identified to obtain the identification result.

[0067] In this embodiment, the subsequent classification layer (usually a classifier such as a Softmax layer or a support vector machine) of the pre-trained analysis model receives the feature vector and, based on the classification boundary learned during the training phase and existing in the high-dimensional feature space, determines the comprehensive operating mode represented by the vector. This determination process is called operating condition identification, and its output is a clear classification conclusion, i.e., the identification result. This identification result is usually a specific state label, such as normal operating condition, early overload warning, power overload, abnormal mechanical load, etc. It originates from the model's learning of various faults and normal patterns in massive historical data, providing another data-driven operating condition judgment independent of curve comparison methods. In addition, the identification result can be accompanied by a confidence probability to indicate the degree of certainty with which the model makes this judgment; when the confidence is low, the result can be used to indicate the need for calibration operations, or combined with other parameters for comprehensive judgment.

[0068] Step S204: Use the identification results to calibrate the analysis results to obtain calibrated analysis results.

[0069] In this embodiment, the identification result (i.e., the operating condition independently determined by the pre-trained analysis model) and the analysis result (i.e., the conclusion of whether or not there is overload obtained by comparing the current operating condition curve with the target operating condition curve) are subjected to consistency verification and decision fusion. Based on calibration rules (e.g., when the two conclusions are consistent, the original analysis result is directly adopted; when the two conflict, if the confidence level of the identification result is higher than a set threshold, the identification result is used to overwrite the original analysis result), the analysis result after intelligent verification and calibration is output.

[0070] By inputting operational data and key parameters into a pre-trained analysis model, the underlying signals and high-level physical parameters are integrated, providing multi-dimensional data input for intelligent diagnosis. Feature vectors are obtained through feature extraction from the model, uncovering deep, fault-related abstract features in the data and improving the discriminative power of condition identification. Identification results are obtained by performing operating condition identification based on feature vectors, realizing intelligent classification based on a data-driven model and providing auxiliary judgment criteria independent of curve comparison. By calibrating the analysis results using the identification results, cross-validation and decision optimization of multiple method conclusions are achieved, effectively reducing the risk of false alarms and false negatives and improving the robustness of the overall system.

[0071] In this embodiment of the application, after analyzing the overload condition of the fracturing pump using the current operating condition curve and obtaining the analysis results, the method further includes: Step S301: When the analysis results indicate that the fracturing pump is overloaded, the alarm mechanism is triggered to generate corresponding overload alarm information.

[0072] In this embodiment, the analysis results are continuously monitored. When the state of the analysis results changes from no overload to overload of the fracturing pump, this state change serves as a trigger condition to activate the alarm mechanism. Once activated, the alarm mechanism executes its logic to generate overload alarm information containing necessary basic information. The overload alarm information includes at least the alarm type (e.g., overload), trigger time, and the equipment identifier of the fracturing pump, providing an alarm event-driven signal for further determining the response strategy.

[0073] Step S302: Determine the overload type of the fracturing pump based on the operating data and key parameters. The overload type includes at least one of power overload, torque overload, and load overload.

[0074] In this embodiment, based on operational data (especially pressure, flow rate, and speed) and key parameters (instantaneous power, crankshaft instantaneous torque, and connecting rod load), these are compared with safety thresholds for different components: Power overload is determined by whether the instantaneous power exceeds its rated power threshold; torque overload is determined by whether the crankshaft instantaneous torque exceeds its allowable torque threshold; and load overload is determined by whether the connecting rod load exceeds its design load threshold. Parallel logic is used to determine the overload type of the fracturing pump, refining the generalized alarm into fault types targeting specific subsystems or components.

[0075] Step S303: Obtain the target handling strategy corresponding to the overload type, and send the overload alarm information and the target handling strategy to the monitoring terminal.

[0076] In this embodiment, the strategy library pre-stores corresponding response plans for different overload types (such as power overload, torque overload, and load overload), i.e., target handling strategies (for example, for power overload, the strategy could be to reduce displacement or construction pressure; for torque overload, the strategy could be to check and adjust the transmission system). The target handling strategy matching the overload type is indexed from this strategy library. Subsequently, the overload alarm information (including time, device ID, alarm level, etc.) and the target handling strategy are encapsulated and sent to the monitoring terminal (such as a customer's mobile APP or computer terminal) via a communication interface (such as a wired network or wireless module).

[0077] By triggering an alarm mechanism to generate overload alarm information when the analysis result indicates overload, immediate warning of risk status is achieved, ensuring timely reporting of abnormal operating conditions. By determining the overload type based on operating data and key parameters, the generalized overload alarm is refined into specific types such as power overload, torque overload, and load overload, enabling the location of the root cause of the fault. By obtaining the target handling strategy corresponding to the overload type and sending it along with the alarm information to the monitoring terminal, a complete information chain is provided to the operators, realizing the integrated push of alarms and handling suggestions, significantly improving the efficiency and accuracy of operation and maintenance response.

[0078] This embodiment provides a fracturing pump alarm system. Figure 5 This is a schematic diagram of the fracturing pump alarm system according to an embodiment of the present invention, as shown below. Figure 5 As shown, the system includes: a sensor array, a data acquisition module, and a data analysis and diagnostic module. The sensor array includes: a flow sensor, a pressure sensor, a speed sensor, and a temperature sensor. The data analysis and diagnostic module includes: a data storage unit, a data analysis and processing unit, and a data fault alarm unit. The sensor array is used to collect operating data from the power and hydraulic ends of the fracturing pump: flow sensor measures flow data, pressure sensor measures pressure data, speed sensor measures speed data, and temperature sensor measures temperature data. The data acquisition module is used to acquire the operational data collected by the sensor array and transmit the operational data to the data analysis and diagnostic module.

[0079] The data analysis and diagnostic module is used to process and analyze operational data: a data storage unit stores operational data; a data analysis and processing unit calculates corresponding key parameters based on the operational data, generates the current operating condition curve of the fracturing pump based on the key parameters, and analyzes the overload situation of the fracturing pump using the current operating condition curve to obtain analysis results; a data fault alarm unit triggers an alarm mechanism to generate corresponding overload alarm information when the analysis results indicate that the fracturing pump is overloaded, determines the overload type of the fracturing pump based on the operational data and key parameters, obtains the target handling strategy corresponding to the overload type, and sends the overload alarm information and target handling strategy to the monitoring terminal.

[0080] As an example, such as Figure 6As shown, the workflow of the fracturing pump alarm system includes: First, the pressure sensor collects pressure data. The data analysis and processing unit determines the connecting rod load based on this pressure data and judges whether the connecting rod load is normal. If the connecting rod load is abnormal, the data fault alarm unit outputs a connecting rod overload alarm message and triggers a pressure reduction strategy. If the connecting rod load is normal, the speed sensor collects speed data. The data analysis and processing unit calculates the crankshaft instantaneous torque based on this speed data and judges whether the torque is normal. If the torque is abnormal, the data fault alarm unit outputs a torque overload alarm message and triggers a speed reduction or displacement reduction strategy. If the torque is normal, the flow sensor collects flow data. The data analysis and processing unit calculates the instantaneous power based on this flow data and judges whether the power is normal. If the power is abnormal, the data fault alarm unit outputs a power overload alarm message and triggers a speed reduction, displacement reduction, or pressure reduction strategy. If the power is normal, it is also necessary to judge whether the tank temperature alarm is triggered. If the tank temperature alarm is not triggered, the data analysis and processing unit compares the generated real-time operating condition curve with the baseline operating condition curve, and then plots the real-time operating condition curve to complete the monitoring and analysis of the fracturing pump operating condition.

[0081] This embodiment also provides a fracturing pump operating condition analysis device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0082] This embodiment provides a device for analyzing the operating conditions of a fracturing pump, such as... Figure 7 As shown, it includes: The acquisition module 71 is used to acquire the operating data of the power end and hydraulic end of the fracturing pump; The calculation module 72 is used to calculate the corresponding key parameters based on the operating data, and generate the current operating condition curve of the fracturing pump based on the key parameters; Analysis module 73 is used to analyze the overload condition of the fracturing pump using the current operating condition curve and obtain the analysis results.

[0083] In this embodiment of the application, the calculation module 72 includes: an extraction submodule and a fusion submodule; The extraction submodule is used to extract flow rate data, pressure data, speed data, and temperature data from the operating data; calculate the connecting rod load based on the pressure data, calculate the instantaneous power based on the flow rate data, calculate the crankshaft instantaneous torque based on the speed data, and obtain the lubricating oil return temperature based on the temperature data; and use the connecting rod load, instantaneous power, crankshaft instantaneous torque, and lubricating oil return temperature as key parameters.

[0084] The fusion submodule is used to fuse key parameters based on the mapping relationship between preset mechanical parameters and preset weight allocation to obtain fused parameters that characterize the operating status of the fracturing pump; arrange the fused parameters according to the time series to obtain the parameter sequence; construct a multi-dimensional parameter space based on the parameter sequence, and construct the current operating condition curve of the fracturing pump in the multi-dimensional parameter space.

[0085] In this embodiment of the application, the calculation module 72 further includes: an adjustment submodule, used to monitor the changing trend of the lubricating oil return temperature; when there is an abnormality in the changing trend of the lubricating oil return temperature, generate a temperature abnormality signal; and adjust the preset allocation weight corresponding to the lubricating oil return temperature based on the temperature abnormality signal.

[0086] In this embodiment of the application, the analysis module 73 is specifically used to obtain the target operating condition curve of the fracturing pump, wherein the target operating condition curve is the operating condition curve generated based on the optimal parameters of the fracturing pump; the current operating condition curve is compared with the target operating condition curve to obtain the comparison result; if the comparison result shows that the current operating condition curve matches the target operating condition curve, the analysis result is determined to be that the fracturing pump is not overloaded, or if the comparison result shows that the current operating condition curve does not match the target operating condition curve, the analysis result is determined to be that the fracturing pump is overloaded.

[0087] In this embodiment of the application, the device further includes: an input module, used to input operating data and key parameters into a pre-trained analysis model; to extract features from the operating data and key parameters through the pre-trained analysis model to obtain feature vectors; to identify the operating conditions of the fracturing pump based on the feature vectors to obtain identification results; and to calibrate the analysis results using the identification results to obtain calibrated analysis results.

[0088] In this embodiment of the application, the device further includes: a triggering module, used to trigger an alarm mechanism to generate corresponding overload alarm information when the analysis result indicates that the fracturing pump is overloaded; to determine the overload type of the fracturing pump based on operating data and key parameters, wherein the overload type includes at least one of power overload, torque overload, and load overload; to obtain the target handling strategy corresponding to the overload type, and to send the overload alarm information and the target handling strategy to the monitoring terminal.

[0089] Please see Figure 8 , Figure 8 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 8As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system).

[0090] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0091] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0092] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device as shown by a landing page for an app. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, which can be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0093] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0094] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0095] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0096] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method of analyzing the operating conditions of a fracturing pump, characterized in that, The method comprises: acquiring operation data of a power end and a fluid end of a fracturing pump; calculating corresponding key parameters according to the operation data, and generating a current working condition curve of the fracturing pump based on the key parameters; analyzing overload conditions of the fracturing pump by using the current working condition curve to obtain an analysis result.

2. The method of claim 1, wherein, The calculation of the corresponding key parameters according to the operation data comprises: extracting flow data, pressure data, rotating speed data and temperature data from the operation data; calculating connecting rod load according to the pressure data, calculating instantaneous power according to the flow data, calculating instantaneous torque of a crankshaft according to the rotating speed data, and acquiring lubricating oil return temperature according to the temperature data; taking the connecting rod load, the instantaneous power, the instantaneous torque of the crankshaft and the lubricating oil return temperature as the key parameters.

3. The method of claim 1, wherein, The generation of the current working condition curve of the fracturing pump based on the key parameters comprises: fusing the key parameters based on a mapping relationship between preset mechanical parameters and preset distribution weights to obtain fused parameters representing a running state of the fracturing pump; arranging the fused parameters in a time sequence to obtain a parameter sequence; constructing a multi-dimensional parameter space based on the parameter sequence, and constructing the current working condition curve of the fracturing pump in the multi-dimensional parameter space.

4. The method of claim 2, wherein, The method further comprises: monitoring a change trend of the lubricating oil return temperature; generating a temperature abnormality signal when the change trend of the lubricating oil return temperature has an abnormal condition; adjusting a preset distribution weight corresponding to the lubricating oil return temperature based on the temperature abnormality signal.

5. The method of claim 1, wherein, The analysis of the overload conditions of the fracturing pump by using the current working condition curve to obtain an analysis result comprises: acquiring a target working condition curve of the fracturing pump, wherein the target working condition curve is a working condition curve generated according to optimal parameters of the fracturing pump; comparing the current working condition curve with the target working condition curve to obtain a comparison result; if the comparison result is that the current working condition curve matches the target working condition curve, determining that the analysis result is that the fracturing pump has no overload condition, or if the comparison result is that the current working condition curve does not match the target working condition curve, determining that the analysis result is that the fracturing pump has an overload condition.

6. The method of claim 1, wherein, The method further comprises: inputting the operation data and the key parameters into a pre-trained analysis model; extracting features of the operation data and the key parameters by using the pre-trained analysis model to obtain a feature vector; performing working condition identification on the fracturing pump based on the feature vector to obtain an identification result; calibrating the analysis result by using the identification result to obtain a calibrated analysis result.

7. The method of claim 1, wherein, After analyzing the overload conditions of the fracturing pump by using the current working condition curve to obtain an analysis result, the method further comprises: triggering an alarm mechanism to generate corresponding overload alarm information when the analysis result is that the fracturing pump has an overload condition; determining an overload type of the fracturing pump according to the operation data and the key parameters, wherein the overload type comprises at least one of power overload, torque overload and load overload. The overload type corresponding target treatment strategy is acquired, and the overload alarm information and the target treatment strategy are sent to a monitoring terminal.

8. A working condition analysis device for a fracturing pump, characterized in that, The device comprises: An acquisition module is configured to acquire operation data of a power end and a fluid end of the fracturing pump; A calculation module is configured to calculate corresponding key parameters according to the operation data, and generate a current working condition curve of the fracturing pump based on the key parameters; An analysis module is configured to analyze an overload condition of the fracturing pump by using the current working condition curve, and obtain an analysis result.

9. A computer device, comprising: Comprise: A memory and a processor, which are in communication connection with each other, and the memory stores computer instructions; the processor executes the computer instructions to perform the method in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer instructions, and the computer instructions are used to make the computer execute the method in any one of claims 1 to 7.

Images