Intelligent control system for drilling mud pump

Abstract

The invention discloses an intelligent control system for a drilling mud pump, which relates to the technical field of drilling mud pumps and comprises a sensing and data acquisition layer, a data processing layer and a control layer, comprising pump body state monitoring, shaft and drilling process parameter monitoring and power system monitoring. The data preprocessing layer is responsible for performing cleaning, noise reduction, correction and alignment on original, disordered and unreliable sensor data streams collected on site, and converting the data streams into clean, complete and credible standardized data; the fault diagnosis and predictive maintenance module is used for identifying a current fault, evaluating a performance degradation trend and predicting the residual life of a key part based on the preprocessed data, and comprises valve body fault diagnosis, performance degradation evaluation and life prediction; and the performance and energy efficiency analysis module is used for calculating the operation efficiency and the energy consumption index of the equipment in real time, comparing the operation efficiency and the energy consumption index with the optimal reference and providing quantitative and operable optimization suggestions.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of drilling mud pumps, in particular to an intelligent control system of a drilling mud pump. BACKGROUND

[0002] The drilling mud pump (commonly known as "drilling pump") is the "heart" of the drilling system, and its function is to pump drilling mud to the bottom of the well at high pressure and large capacity to carry cuttings, cool the drill bit, balance the formation pressure and stabilize the well wall. In the whole set of drilling equipment, the mud pump is one of the single machines with the largest energy consumption, and its energy consumption accounts for about 20%-30% of the total energy consumption of the whole drilling process.

[0003] At present, the main problems and limitations of the mud pump control method commonly used in the field are as follows: constant speed pump + throttle regulation This is the most traditional way. The diesel engine or electric motor drives the mud pump at a constant speed, and then the opening of the throttle valve (such as the needle valve on the drilling fluid manifold) is adjusted to change the pipeline resistance, thereby indirectly controlling the displacement and pump pressure. In this way, the output power of the pump is much larger than the actual required power, and a large amount of energy is wasted in the pressure drop of the throttle valve, which is converted into heat energy, and the efficiency is very low. And long-term operation at constant high speed or improper working conditions, especially frequent hydraulic impact, will accelerate the wear, fatigue and damage of the mud pump and its supporting manifold, resulting in frequent maintenance and high cost of spare parts.

[0004] For electric drilling rigs, the front-end rectifier unit of the frequency conversion drive system generally uses traditional diode or phase control thyristor rectification technology. This rectification method not only has large conduction loss itself, but more seriously, it will produce low power factor and high harmonic current, resulting in the apparent power of the grid side being much higher than the active power, making the power supply transformer and cable loss surge, and the overall power utilization efficiency is low. At the same time, the inductors widely used in the system will also become a source of energy consumption that cannot be ignored if they are not designed properly or the magnetic core material is outdated, and their copper loss and iron loss will also increase with the power level.

[0005] Therefore, the present application provides an intelligent control system of a drilling mud pump. SUMMARY

[0006] (I) Technical problems solved In view of the deficiencies of the prior art, the drilling mud pump intelligent control system is provided, real-time volume efficiency and mechanical efficiency are calculated, the unique "energy consumption index" (such as fuel consumption per cubic meter of mud) directly links the equipment efficiency with the operating cost, so that the management personnel can clearly master the pumping cost of each cubic meter of mud. The intelligence of the module is reflected in dynamic benchmark comparison and accurate optimization suggestions. The system can not only find the problem that the current energy consumption is 8% higher than the best level, but also can find the optimal operating parameters under the conditions of meeting the safety and process requirements through analyzing massive historical data, and give specific and operable suggestions such as "reducing the number of strokes to save fuel". This makes every operation decision of the driller have a basis, and directly reduces energy consumption from the source. Finally, the module creates a continuous improvement closed loop: it continuously reduces fuel or power consumption, directly reduces operating costs; by maintaining efficient operation to delay equipment wear and tear and extend equipment life; and by quantitative management and intelligent guidance, the ultimate goal of cost reduction and efficiency improvement and lean operation is achieved under the premise of ensuring drilling safety, which upgrades equipment management from experience dependence to data-driven intelligent new stage, thereby solving the technical problems described in the background art.

[0007] (II) Technical solutions To achieve the above object, the following technical solutions are adopted: the drilling mud pump intelligent control system comprises: The perception and data acquisition layer is responsible for real-time collection of all related data. It includes pump body state monitoring, wellbore and drilling process parameter monitoring and power system monitoring; The data preprocessing layer is responsible for cleaning, denoising, correcting and aligning the original, chaotic and unreliable sensor data stream collected on site, and converting it into clean, complete and reliable standardized data; The fault diagnosis and predictive maintenance module identifies the current fault, evaluates the performance degradation trend, and predicts the remaining life of the key components based on the preprocessed data, including valve fault diagnosis, performance degradation evaluation and life prediction; The performance and energy efficiency analysis module calculates the operating efficiency and energy consumption index of the equipment in real time, compares it with the best benchmark, and provides quantitative and operable optimization suggestions.

[0008] Further, the pump body state monitoring includes: the pump pressure sensor monitors the pressure of the discharge pipeline in real time, the stroke frequency sensor accurately measures the reciprocating frequency of the piston, the lubrication system sensor monitors the oil pressure and temperature, and the vibration and acoustic sensor is installed on the pump body for analyzing early faults of the liquid end.

[0009] Further, the wellbore and drilling process parameter monitoring includes: the flowmeter monitors the actual outlet displacement, compares it with the theoretical displacement to judge the circulation loss or valve leakage; the mud pit volume monitor monitors the total volume change in real time, and the return flow monitor judges whether the wellbore is normally circulating.

[0010] Further, the power system monitoring includes: the engine controller acquires diesel engine speed, torque, fuel consumption, water temperature, oil pressure data; the electric parameter monitoring motor voltage, current, power factor, energy consumption. For the electric drilling rig using variable frequency drive, the system also monitors the efficiency of its rectifier unit and DC bus circuit, including the loss and temperature rise of the rectifier, and the operating state of the key elements of the DC link inductor.

[0011] Further, the system performs signal processing on the real-time pump pressure waveform. First, the time-domain pressure waveform is converted into a frequency spectrum graph by fast Fourier transform to check whether there is an abnormal characteristic frequency. At the same time, the peak value, rising slope and fluctuation period characteristics of the waveform in the time domain are analyzed. These extracted real-time characteristics are automatically compared and matched with various typical fault modes stored in the knowledge base.

[0012] Further, in combination with the vibration signals of each valve box, the system intelligently determines which component of which valve box has a problem, and outputs the diagnosis result.

[0013] Further, the system continuously receives the stroke signal and the flowmeter signal, and calculates the current volumetric efficiency in real time. The system records the daily efficiency value and plots it into a long-term trend curve. Through time series analysis, the system can determine whether the efficiency decline rate is gentle or suddenly accelerated. When the efficiency value is lower than the preset threshold value, or the decline rate exceeds a certain limit value, the system will issue a pump performance degradation warning, prompting attention or preparation for maintenance.

[0014] Further, based on historical maintenance data and physical models, a life consumption model is established for each vulnerable part. The system continuously records the actual working stroke and average working pressure of each component, and according to the life model, the working stroke under different pressures is converted into equivalent damage, and the total designed life of the component is reduced by the equivalent damage life consumed, so that the current remaining life percentage or remaining available stroke can be calculated, and the output result is obtained.

[0015] (Three) beneficial effects The present application provides an intelligent control system for a drilling mud pump, which has the following beneficial effects: 1. By real-time analysis of pump pressure waveform and frequency spectrum, accurate positioning can be achieved at the fault initiation stage. It can not only judge "there is a fault", but also accurately diagnose "No. 3 cylinder discharge valve wear", and give a confidence of 90%. This millisecond-level early warning creates a valuable time window for arranging orderly maintenance, successfully changes unplanned downtime into planned maintenance, and avoids the possibility of serious pump failure or even drilling tool falling into the well caused by fault expansion.

[0016] 2. Through real-time calculation and trend analysis, the system can accurately quantify the health degradation of equipment. It not only provides an instantaneous efficiency value, but more importantly, through long-term trend curves, it clearly reveals the rate and pattern of performance degradation—whether it's smooth natural aging or rapid, abnormal deterioration. This dynamic and continuous assessment method is like establishing a "health record" for the equipment, enabling managers to accurately determine the equipment's "sub-healthy" state. When efficiency falls below a threshold or the rate of decline is abnormal, the system's warnings are no longer simple "fault alarms," ​​but rather "maintenance window reminders" with clear guidance. This allows maintenance teams to plan maintenance in advance at the most appropriate time, avoiding the waste of spare parts caused by premature replacement and completely eliminating secondary damage and unplanned downtime caused by excessive component wear. This data-driven decision-making model significantly improves the overall utilization rate and operational economy of equipment, ultimately achieving lean asset management and reliable assurance for safe production.

[0017] 3. By establishing a precise physical model of component lifespan consumption, comprehensively considering the two core factors of working strokes and load pressure, the scientific quantification of component damage is achieved. It uniformly converts the working history under different pressures into "equivalent damage," thereby dynamically calculating the remaining lifespan with high precision and outputting a clear prediction such as "120,000 strokes remaining." The direct benefit of this capability is the realization of accurate predictive maintenance. Maintenance teams can predict the "end of life" of components weeks or days in advance, much like checking a weather forecast, allowing them to schedule maintenance and spare parts procurement at the most appropriate time. This completely avoids sudden downtime, ensuring the continuity and safety of drilling operations. Simultaneously, it ensures that every component is "used to its fullest potential," avoiding the waste of spare parts and manpower caused by premature replacement, and preventing cascading damage caused by overuse. Ultimately, this achieves a double reduction in spare parts inventory costs and the risk of unexpected downtime, significantly improving the level of lean asset management and operational efficiency.

[0018] 4. By calculating volumetric efficiency and mechanical efficiency in real time, the abstract pump health status is transformed into precise percentages, providing key indicators for preventative maintenance. More importantly, its unique "energy consumption index" (such as oil consumption per cubic meter of mud) directly links equipment efficiency to operating costs, enabling managers to clearly understand the pumping cost of each cubic meter of mud. The module's intelligence is reflected in its dynamic benchmark comparison and precise optimization suggestions. The system can not only detect problems such as "current energy consumption is 8% higher than the optimal level," but also, by analyzing massive amounts of historical data, identify the optimal operating parameters while meeting safety and process requirements, and provide specific and actionable suggestions such as "reducing strokes to save fuel." This ensures that every operational decision made by the driller is based on evidence, directly reducing energy consumption at the source. Ultimately, this module creates a closed loop of continuous improvement: it continuously reduces fuel or electricity consumption, directly cutting operating costs; it slows down equipment wear and extends equipment life by maintaining efficient operation; and through quantitative management and intelligent guidance, it achieves the ultimate goal of cost reduction, efficiency improvement, and lean operation while ensuring drilling safety, elevating equipment management from experience-based to a new stage of data-driven intelligence.

[0019] 5. By introducing advanced rectification and inductance technologies, energy saving is achieved at the source of power conversion. The system not only monitors but also drives the frequency converter to adopt active front-end rectification technology, improving the input power factor to near 1 and significantly suppressing harmonic currents, thereby reducing ineffective power loss on the grid side at its source. Simultaneously, by selecting low-loss magnetic core materials (such as amorphous and nanocrystalline) and optimizing winding processes for high-efficiency inductors, the iron and copper losses of the magnetic components inside the frequency converter system are significantly reduced. This synergistic design of "clean input and high-efficiency conversion," combined with subsequent operational parameter optimization, constructs a full-link energy-saving system from the grid to the drilling mud, further amplifying the energy efficiency advantages of electric drilling rigs. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the intelligent control system for drilling mud pumps of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Please see Figure 1 This invention provides an intelligent control system for drilling mud pumps, comprising: The perception and data acquisition layer is responsible for real-time collection of all relevant data. This includes pump body condition monitoring, wellbore and drilling process parameter monitoring, and power system monitoring.

[0023] Pump body condition monitoring includes: pump pressure sensors monitor the pressure of the discharge line in real time, which is crucial for preventing pump tripping; stroke sensors accurately measure the reciprocating frequency of the piston, which is the core of calculating displacement; lubrication system sensors monitor oil pressure and temperature to ensure normal lubrication of the power end; vibration and acoustic sensors are installed on the pump body to analyze early failures of the fluid end (piston, valve, cylinder sleeve).

[0024] Wellbore and drilling process parameter monitoring includes: flow meters monitor actual outlet displacement, which can be compared with theoretical displacement to determine circulation loss or valve leakage; mud pit volume monitors real-time total volume changes, which are the first line of defense for finding well kicks or losses; return flow monitoring determines whether the wellbore is circulating normally.

[0025] Power system monitoring includes: engine controllers obtain diesel engine speed, torque, fuel consumption, water temperature, oil pressure, and other data. Electrical parameters (for electric drilling rigs) monitor motor voltage, current, power factor, and energy consumption.

[0026] For electric drilling rigs with variable frequency drives, the system also monitors the efficiency of the rectifier unit and DC bus circuit, including the loss and temperature rise of the rectifier, as well as the operating status of key components such as the DC link inductor.

[0027] In particular, for the core variable frequency drive system, the system deeply monitors the operating status of the front-end rectifier unit. Traditional diode rectification or basic thyristor rectification not only produces significant conduction loss, but also injects a large amount of harmonics into the power grid, reducing power grid quality and causing additional energy waste. The system monitors the input current harmonic distortion rate, power factor, and temperature rise of key components (such as IGBTs and diodes) to evaluate their energy efficiency. At the same time, the system also monitors the current and temperature of passive components such as the DC bus support inductor to ensure that they operate in the high-efficiency range and prevent additional energy consumption due to core saturation or excessive copper loss.

[0028] The data preprocessing layer is responsible for cleaning, denoising, correcting, and aligning the raw, chaotic, and unreliable sensor data stream collected on site, transforming it into clean, complete, and reliable standardized data.

[0029] For dynamic signals such as pump pressure and vibration, a digital filter (such as a low-pass filter) is used. The principle is to set a frequency threshold, allowing real pressure fluctuations below this threshold to pass through, while filtering out high-frequency electrical noise or transient interference pulses, resulting in a smooth and true signal curve.

[0030] Statistical threshold method is used for outlier treatment. The system learns from historical normal data and calculates the normal range of each parameter, such as mean ± 3 times standard deviation. When a data point is far beyond this range and cannot be explained by physical principles, the system will consider it as "wild value" and replace it with the previous valid value or interpolation result.

[0031] The system applies a unified and high-precision timestamp to all sensor data. This ensures that the pressure, flow, and vibration data used in analyzing the working conditions at a certain time are strictly corresponding to the same time, avoiding causal misjudgment.

[0032] Different parameters have different sampling frequencies. The system uses resampling techniques to "downsample" high-frequency data, such as calculating the average value per second, or "upsample" low-frequency data by interpolation to estimate intermediate values, so that all data can be matched and calculated on a unified time grid.

[0033] The fault diagnosis and predictive maintenance module, based on preprocessed data, identifies current faults, assesses performance degradation trends, and predicts the remaining life of key components, including valve body fault diagnosis, performance degradation assessment, and life prediction.

[0034] Valve body fault diagnosis specifically includes: A healthy mud pump has a stable and regular pump pressure waveform. When the valve or piston wears out, the spring breaks, or the pump pressure waveform and spectrum change slightly but can be identified.

[0035] The system performs signal processing on the real-time pump pressure waveform. First, the time-domain pressure waveform is converted to a frequency-domain spectrum graph through Fast Fourier Transform (FFT) to check for abnormal characteristic frequencies. At the same time, the waveform's peak value, rising slope, fluctuation period, and other characteristics are analyzed in the time domain. For example, suction valve leakage will cause slow pressure rise, and discharge valve damage will cause pressure peak drop. These extracted real-time features are automatically compared and matched with various typical fault patterns stored in the knowledge base, such as "discharge valve wear waveform" and "piston leak frequency spectrum."

[0036] Combined with the vibration signals of each valve box (which valve box has abnormal vibration), the system intelligently determines which component of which valve box has a problem and outputs the diagnosis result, such as "No. 3 cylinder discharge valve is severely worn, confidence 90%".

[0037] By real-time analysis of pump pressure waveform and spectrum, accurate positioning can be achieved in the early stage of failure. It can not only judge "fault", but also accurately diagnose "No. 3 cylinder discharge valve wear", and give a 90% confidence level. This millisecond-level early warning creates a valuable time window for arranging orderly maintenance, successfully turning unplanned downtime into planned maintenance, and avoiding the possibility of serious pump failure or even drill pipe falling into the well accident.

[0038] Performance degradation evaluation is specifically: The volumetric efficiency of the pump = actual displacement / theoretical displacement. The efficiency of a new pump is close to 100%. With the wear of internal seals (piston, cylinder sleeve), internal leakage will increase, resulting in a decrease in actual displacement and a decrease in volumetric efficiency.

[0039] The system continuously receives stroke signals (calculates theoretical displacement) and flowmeter signals (actual displacement), and calculates the current volumetric efficiency in real time. The system records the efficiency value of each day and plots a long-term trend curve. Through time series analysis, the system can determine whether the efficiency decline rate is flat (normal wear) or suddenly accelerated (abnormal failure precursor). When the efficiency value is lower than the preset threshold (such as 85%), or the decline rate exceeds a certain limit, the system will issue a "pump performance degradation" warning, prompting attention or preparation for maintenance.

[0040] Through real-time calculation and trend analysis, the health decay of the equipment can be accurately quantified. It not only provides an instantaneous efficiency value, but more importantly, through a long-term trend curve, it clearly reveals the rate and mode of performance degradation - is it a smooth natural aging, or is it a sharp abnormal degradation. This dynamic and continuous evaluation method is like establishing a "health record" for the equipment, allowing management personnel to accurately determine the "sub-health" state of the equipment. When the efficiency is below the threshold or the decline rate is abnormal, the system's warning is no longer a simple "fault alarm", but a "maintenance window reminder" with clear guidance. This allows the maintenance team to plan maintenance in advance at the most appropriate time, avoiding both the waste of spare parts caused by premature replacement and the secondary damage and unplanned downtime caused by excessive wear of parts. This data-based decision-making mode significantly improves the overall utilization rate and operational economy of the equipment, ultimately achieving lean management of assets and reliable protection of safety production.

[0041] Life prediction is specifically: The life of consumable parts (such as pistons and cylinder sleeves) is mainly affected by two factors: cumulative working time (strokes) and working intensity (load pressure).

[0042] A life consumption model is established for each wear part based on historical maintenance data and physical models. For example, the model stipulates that the life of a piston is 1 million strokes at rated pressure; when the pressure exceeds the rated value by 10%, the life is halved. The system continuously records the actual working strokes and average working pressure of each component. According to the life model, the working strokes at different pressures are converted into "equivalent damage". The total designed life of the component is reduced by the "equivalent damage" life that has been consumed, and the current remaining life percentage or remaining available strokes can be calculated. The output result is as follows: "The current piston remaining life is expected to be 15 days or 1.2 million strokes".

[0043] By establishing an accurate life consumption physical model, the two core factors of working strokes and load pressure are comprehensively considered, and the scientific quantification of component damage is realized. It unifies the working history at different pressures into "equivalent damage", so as to dynamically calculate the high-precision remaining life, and output the explicit prediction such as "remaining 1.2 million strokes". The direct benefit brought by this ability is to realize accurate predictive maintenance. The maintenance team can predict the "end of life" of the component weeks or days in advance, just like checking the weather forecast, so as to arrange maintenance and spare parts procurement at the most appropriate time. This completely avoids sudden shutdown, ensures the continuity and safety of drilling operations. At the same time, it ensures that each component is "used to the full", avoiding the waste of spare parts and manpower caused by premature replacement, and preventing chain damage caused by overuse, ultimately realizing the double reduction of spare parts inventory cost and unexpected downtime risk, significantly improving the lean level of asset management and operational efficiency.

[0044] The performance and energy efficiency analysis module calculates the running efficiency and energy consumption index of the device in real time, and compares it with the best benchmark to provide quantitative and operable optimization suggestions.

[0045] The system obtains the pump stroke and outlet flow meter readings in real time, and calculates the theoretical displacement of the pump under no leakage condition through a fixed formula according to the cylinder diameter of the pump and the real-time stroke.

[0046] Volumetric efficiency = (actual displacement measured by flow meter / calculated theoretical displacement) * 100% This percentage directly reflects the health status of the pump hydraulic end. The higher the percentage, the better the sealing between the piston, valve and cylinder, and the less the internal leakage. The efficiency of a new pump is usually above 95%, and a decrease in efficiency is a direct signal of internal wear.

[0047] For diesel engines, the real-time fuel consumption rate and speed are obtained through the engine controller to calculate the power input to the pump; for electric motors, the voltage, current and power factor are obtained through the electric parameter sensor to calculate the input electric power. The actual displacement and pump pressure are used to calculate the water power actually transmitted to the mud by the pump.

[0048] Mechanical Efficiency = (Output Water Power / Input Shaft Power) * 100% This percentage reflects the transmission efficiency of the pump power end (gear, bearing, crosshead, etc.) and mechanical friction loss. A drop in mechanical efficiency may indicate poor lubrication, bearing wear, or alignment problems.

[0049] These two efficiency values are displayed on the dashboard along with their historical trend curves.

[0050] Energy Consumption Index = Total Energy Consumption per Unit of Time / Actual Displacement per Unit of Time For diesel power: Energy Consumption Index = Fuel Consumption per Hour (L) / Displacement per Hour (m3) = L / m3 For electric power: Energy Consumption Index = Power Consumption per Hour (Degree) / Displacement per Hour (m3) = Degree / m3 The historical best value is a dynamic benchmark library that the system records and learns from the pump's best energy consumption index under different conditions (different displacement, pressure) when it is in good health.

[0051] Theoretical / manufacturer's data is the performance curve provided by the reference equipment manufacturer.

[0052] The system automatically compares the real-time calculated energy consumption index with the best value under the current similar condition in the benchmark library, outputs an intuitive comparison result, such as: "the current energy consumption is 8% higher than the best level", or through a traffic light system (green = good, yellow = attention, red = poor) for early warning.

[0053] The system analyzes a large amount of historical operation data to find the internal relationship between energy consumption index and controllable parameters such as pump strokes and pressure under different well depths, different formations, and different pump pressures.

[0054] The system identifies those parameter combinations that achieve the lowest energy consumption under the premise of meeting the drilling process requirements (such as displacement not lower than a certain lower limit). Based on the learned pattern, the system will perform simulation calculations in the background for the current downhole working conditions and drilling requirements.

[0055] It answers the question: "Under the premise of ensuring minimum displacement and required pump pressure, how will the energy consumption index change if I adjust the stroke slightly (for example, from 120 SPM to 115 SPM)?" It is important to note that the system will not make suggestions that may affect drilling safety. All suggestions must meet the pre-set safety constraints (such as displacement cannot be lower than XX, pump pressure must be higher than YY), and ultimately, it will generate a clear, quantitative, and actionable suggestion that is directly pushed to the driller.

[0056] Example: "Optimization suggestion: calculated that reducing the ROP from 120 SPM to 118 SPM will result in a 0.5% decrease in volume efficiency (within safety range), but the energy index can be reduced by 3%, saving about 4 liters of fuel per hour. Do you want to implement it?"

[0057] In addition, for electric drilling rigs, the system can also make hardware energy efficiency upgrade suggestions at the power electronics level based on the assessment of the rectifier unit efficiency.

[0058] Rectifier system energy efficiency upgrade suggestion: the current monitoring shows that the power factor of the rectifier unit is only 0.75, and the harmonic distortion rate is as high as 40%. It is recommended to upgrade the existing diode rectifier cabinet to an active front-end rectifier cabinet. After the upgrade, the power factor can be stabilized at 0.98 or above, and the harmonic distortion rate can be less than 5%, which is expected to reduce the total loss on the grid side by about 8%-15% and avoid possible grid penalties.

[0059] Inductor energy efficiency optimization suggestion: the system identifies that the DC bus reactor has about 3% power loss under the current load through model calculation and temperature rise monitoring. It is recommended to choose a new generation of high-efficiency inductor with amorphous alloy magnetic core and litz wire winding when updating the equipment, which is expected to reduce this part of the loss by more than 50% and improve the overall efficiency of the system.

[0060] These suggestions, together with operation parameter optimization, constitute the multi-level and three-dimensional energy-saving strategy of the system, which fully explores the energy-saving potential from hardware foundation to operation control, and maximizes energy efficiency improvement and carbon emission reduction.

[0061] By calculating the volume efficiency and mechanical efficiency in real time, the abstract pump health status is converted into a precise percentage, providing a key indicator for preventive maintenance. More importantly, its unique "energy consumption index" (such as fuel consumption per cubic meter of mud) directly links device efficiency with operating costs, enabling managers to clearly understand the pumping cost of each cubic meter of mud. The intelligence of the module is reflected in its dynamic benchmarking and precise optimization suggestions. The system not only finds out that "the current energy consumption is 8% higher than the best level", but also analyzes massive historical data to find the optimal operating parameters under the premise of meeting safety and process requirements, and gives specific and actionable suggestions such as "reduce the ROP to save fuel". This enables the driller to make every operational decision with solid evidence, directly reducing energy consumption from the source. Ultimately, the module creates a continuous improvement loop: it continuously reduces fuel or power consumption, directly reducing operating costs; by maintaining high efficiency, it delays equipment wear and tear, extending equipment life; and by quantifying management and intelligent guidance, it achieves the ultimate goal of cost reduction and efficiency improvement and lean operation under the premise of ensuring drilling safety, upgrading equipment management from experience-based to data-driven and intelligent new stage.

[0062] The above-described embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the above-described embodiments can be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art can be aware that units and algorithm steps of the examples described in connection with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether the functions are performed by hardware or software depends on the specific application and design constraints of the technical solutions.

[0063] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place or can be distributed on multiple network units. Part or all of the units can be selected to achieve the purpose of the embodiments according to actual needs.

[0064] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. An intelligent control system for a drilling mud pump, characterized by: Comprise: A perception and data acquisition layer responsible for real-time collection of all relevant data; Including pump body state monitoring, wellbore and drilling process parameter monitoring and power system monitoring; The data preprocessing layer is responsible for cleaning, denoising, correcting and aligning the original, chaotic and unreliable sensor data stream collected on site, transforming it into clean, complete and reliable standardized data; Fault diagnosis and predictive maintenance module, based on preprocessed data, identify current faults, assess performance degradation trends, and predict the remaining life of key components, including valve body fault diagnosis, performance degradation assessment and life prediction; Performance and energy efficiency analysis module, real-time calculation of device running efficiency and energy consumption index, and comparison with the best benchmark, providing quantitative and operable optimization suggestions.

2. The intelligent control system of the drilling mud pump according to claim 1, characterized in that: Pump body state monitoring includes: pump pressure sensor real-time monitoring of discharge pipeline pressure, stroke sensor accurate measurement of piston reciprocating frequency, lubrication system sensor monitoring engine oil pressure and temperature, vibration and acoustic sensors installed on the pump body for analyzing early faults of the liquid end.

3. The intelligent control system of the drilling mud pump according to claim 1, characterized in that: Wellbore and drilling process parameter monitoring includes: flowmeter monitoring actual outlet displacement, compared with theoretical displacement to judge circulation loss or valve leakage; mud pit volume monitor real-time monitoring of total volume change, return flow monitoring to judge whether the wellbore is normally circulating.

4. The intelligent control system of the drilling mud pump according to claim 1, characterized in that: Power system monitoring includes: engine controller acquiring diesel engine speed, torque, fuel consumption, water temperature, oil pressure data; electrical parameter monitoring motor voltage, current, power factor, energy consumption; For electric drilling rigs using variable frequency drive, the system also monitors the efficiency of the rectifier unit and DC bus circuit, including the loss and temperature rise of the rectifier, as well as the operating state of the key components of the DC link inductor.

5. The intelligent control system of the drilling mud pump according to claim 1, characterized in that: The system performs signal processing on the real-time pump pressure waveform. First, the time-domain pressure waveform is converted to a frequency spectrum graph by fast Fourier transform to check if there are any abnormal characteristic frequencies; at the same time, the peak value, rising slope and fluctuation period characteristics of the waveform in the time domain are analyzed; Automatically compare and match these extracted real-time features with various typical fault patterns stored in the knowledge base.

6. The intelligent control system of the drilling mud pump according to claim 1, characterized in that: Combined with the vibration signals of each valve box, the system intelligently determines which component of which valve box is problematic and outputs the diagnosis result.

7. The intelligent control system of the drilling mud pump according to claim 1, characterized in that: The system continuously receives the stroke signal and the flow meter signal, and calculates the current volumetric efficiency in real time; the system records the daily efficiency value and draws a long-term trend curve; through time series analysis, the system determines whether the efficiency decline rate is gentle or suddenly accelerated; when the efficiency value is lower than the preset threshold, or the decline rate exceeds a certain limit, the system will issue a pump performance degradation warning, prompting attention or preparation for maintenance.

8. The intelligent control system of the drilling mud pump according to claim 1, characterized in that: Based on historical maintenance data and physical models, a life consumption model is established for each wear part; the system continuously records the actual working stroke and average working pressure of each component, and according to the life model, the working stroke under different pressures is converted into equivalent damage, and the total designed life of the component is subtracted from the consumed equivalent damage life, so as to calculate the current remaining life percentage or remaining available stroke, and output the result.

Images