Water injection skid process system with adaptive switching and redundancy function and control method

Abstract

The present application relates to the technical field of oilfield water injection, and discloses a water injection skid process system with self-adaptive switching and redundancy function and a control method, which comprises a main process unit composed of a main centrifugal pump and a main plunger pump in series; a self-adaptive switching unit comprising a standby pump group branch, a bypass branch and a control module; and a redundancy design unit adopting double-valve parallel redundancy configuration for key valves and double-path redundancy setting for key sensors. The present application systematically constructs a three-layer structure with main process, self-adaptive switching and redundancy design, and lays a hardware foundation for a high-reliability water injection system. The structure not only realizes the conventional water injection function, but also significantly reduces the single-point failure risk through the introduction of redundancy and self-adaptive mechanism, so that the system can maintain continuous operation in the event of equipment failure or abnormal working conditions, greatly improving the stability and production efficiency of oilfield water injection operation, and solving the problems of poor single-point failure resistance and lack of working condition self-adaptation.

Description

Technical Field

[0001] This invention relates to the field of oilfield water injection technology, specifically to a water injection skid process system and control method with adaptive switching and redundancy functions. Background Technology

[0002] In oilfield development, water injection is a key technology for maintaining formation pressure and enhancing oil recovery. The water injection skid, as the core surface equipment of this process, integrates booster pumps, valves, pipelines, and a monitoring system. It plays a crucial role in pressurizing water that meets quality standards to the required pressure and continuously injecting it into the oil reservoir. Its reliability, operational stability, and adaptability to different stages of water injection directly affect the oilfield's development efficiency and economic benefits.

[0003] Existing water injection skid systems typically employ a single pump unit in series (such as a centrifugal pump + plunger pump) and a single-path valve assembly. In actual operation, cavitation damage to the main pump, bearing overheating, mechanical seal failure, or abnormal opening and closing of critical valves due to scaling or foreign object obstruction can all cause the entire water injection process to be interrupted. The operation and control of traditional water injection skids heavily rely on manual experience and cannot be adjusted in real time according to dynamic water injection needs. They also suffer from poor resistance to single-point failures and a lack of adaptive operation.

[0004] Based on the above situation, there is an urgent need for a water injection skid process system and control method with adaptive switching and redundancy functions to solve the problems of poor resistance to single point of failure and lack of adaptive operation. Summary of the Invention

[0005] The purpose of this invention is to address the problems of existing water injection skid systems, which suffer from poor resistance to single-point failures and lack of adaptive operation when an abnormality occurs.

[0006] The technical solution of the present invention is as follows:

[0007] On the one hand, a water injection skid process system with adaptive switching and redundancy functions is provided, including:

[0008] The main process unit consists of a main centrifugal pump and a main plunger pump connected in series. A valve group, as well as pressure sensors, flow sensors and temperature sensors are arranged along the process to realize conventional water injection pressurization and collect operating parameters.

[0009] The adaptive switching unit includes a backup pump branch, a bypass branch, and a control module. The control module receives measurement data from pressure sensors, flow sensors, and temperature sensors, and automatically activates the backup pump branch and bypass branch after diagnosing equipment failure or process abnormality.

[0010] The redundant design unit employs a dual-valve parallel redundant configuration for at least one valve in the valve group, and implements a dual-path redundant configuration for at least one of the pressure sensor, flow sensor, and temperature sensor, with a spare interface to improve the fault tolerance and reliability of the system.

[0011] Existing water injection skid systems suffer from interruptions in the entire water injection process due to malfunctions. They cannot adjust themselves in real time according to dynamic water injection demands, exhibiting poor resistance to single points of failure and a lack of adaptive operation. This solution systematically constructs a three-layer structure with a main process, adaptive switching, and redundancy design, laying the hardware foundation for a highly reliable water injection system. This structure not only realizes conventional water injection functions but also significantly reduces the risk of single points of failure through the introduction of redundancy and adaptive mechanisms. This allows the system to maintain continuous operation even in the event of equipment failure or abnormal operating conditions, greatly improving the stability and production efficiency of oilfield water injection operations and solving the problems of poor resistance to single points of failure and lack of adaptive operation.

[0012] Furthermore, this solution does not exclusively limit the specific structure of the backup pump group branch. One feasible solution is that the backup pump group branch includes a backup centrifugal pump and its matching valve group redundantly connected in parallel with the main centrifugal pump, and a backup plunger pump and its matching valve group redundantly connected in parallel with the main plunger pump. When this solution is adopted, by setting up backup pump groups and matching valves in parallel for the centrifugal pump and the plunger pump respectively, rapid redundancy switching of pump equipment is realized, ensuring that the system can complete the switching in a short time when the main pump fails, minimizing the water injection interruption time, which is particularly suitable for oilfield water injection scenarios with high requirements for production continuity.

[0013] Furthermore, this solution does not exclusively limit the specific structure of the bypass branch. One feasible solution is that the bypass branch includes a bridging pipe connecting the inlet of the main centrifugal pump and the outlet of the main plunger pump. The bridging pipe is equipped with a control valve, which is used to maintain the flow when the flow rate is lower than a preset flow threshold or when the pump unit fails. When this solution is adopted, the centrifugal pump can be bypassed when the flow rate is lower than the preset flow threshold, avoiding the centrifugal pump from operating in the inefficient zone. It can also temporarily maintain the flow when the main pump fails, enhancing the system's adaptability to complex operating conditions and expanding the system's effective working range.

[0014] On the other hand, a control method for a water injection skid process system with adaptive switching and redundancy functions is provided, including:

[0015] During the operation of the main centrifugal pump, continuous parameter monitoring and diagnosis of equipment faults and process abnormalities are performed.

[0016] Depending on the type of fault or anomaly, it automatically switches to the standby pump set branch, bypass branch, or redundant valves and sensors in the redundant design unit to maintain uninterrupted water injection operation.

[0017] After troubleshooting, the system resumes operation in the main process unit and updates the equipment status information.

[0018] A complete set of intelligent control methods was proposed, realizing closed-loop control from parameter monitoring and fault diagnosis to adaptive switching and system recovery. By combining hardware redundancy with software intelligence, the system not only has fault response capabilities, but also state perception and decision-making capabilities, thus realizing intelligent operation and maintenance of the water injection process.

[0019] Furthermore, to facilitate fault prediction and diagnosis, one feasible solution is: the continuous parameter monitoring and diagnosis of equipment faults and process anomalies includes:

[0020] The operating parameters of the pump group and valve group in the main process unit are collected in real time through pressure sensors, flow sensors, temperature sensors and vibration sensors.

[0021] The control module performs fault prediction and diagnosis based on operating parameters using trend analysis and spectrum analysis algorithms.

[0022] If a fault is predicted or diagnosed, the adaptive switching unit or the backup equipment in the redundancy design unit will be automatically activated according to the preset adaptive switching logic to maintain the continuity of the water injection operation.

[0023] When this solution is adopted, the fusion monitoring of pressure sensors, flow sensors and temperature sensors and intelligent diagnostic algorithms enable high-precision diagnosis of various fault types such as pump overload, valve jamming and parameter mutation, providing a reliable decision basis for system adaptive switching.

[0024] Furthermore, this solution is not limited to the specific steps of the trend analysis algorithm; one feasible solution is that the trend analysis algorithm includes:

[0025] Perform linear fitting on the running parameters within a specified time window and calculate the slope value. k ;

[0026] When the slope value k If the value is negative for several consecutive time windows and its absolute value exceeds the first preset threshold, it is determined that the performance of the pump or pipeline system corresponding to the operating parameter is deteriorating, and an early warning signal is generated.

[0027] When using this scheme, by adjusting the slope of the operating parameters... kThe calculations enable early identification of equipment performance degradation trends, achieving predictive maintenance. This changes the traditional reactive maintenance model, shifting from a passive to a proactive approach, effectively preventing sudden failures and extending equipment lifespan.

[0028] Furthermore, this solution is not limited to the specific steps of the spectrum analysis algorithm; one feasible solution is that the spectrum analysis algorithm includes:

[0029] The vibration spectrum is obtained by performing a fast Fourier transform on the vibration sensor signal.

[0030] The characteristic frequency amplitudes are extracted from the vibration spectrum. These characteristic frequency amplitudes include the 1x rotational frequency amplitude related to pump rotor imbalance, the 2x rotational frequency amplitude related to equipment misalignment, and the fault frequency amplitude related to bearing damage.

[0031] When the growth trend of any characteristic frequency amplitude relative to the baseline value exceeds the second preset threshold, it is determined that the corresponding mechanical fault is developing, and a predictive maintenance alarm is generated.

[0032] This approach allows for the extraction of characteristic frequencies from vibration signals that are associated with mechanical faults such as rotor imbalance, equipment misalignment, and bearing damage, enabling accurate diagnosis and early warning of these faults. This improves the predictive ability for rotating equipment faults and reduces unplanned downtime.

[0033] Furthermore, to facilitate the prediction of cavitation or cavitation in pump units, one feasible solution is as follows: the fault prediction and diagnosis also includes:

[0034] Calculate the fluctuation coefficient of the motor current, where the fluctuation coefficient is the ratio of the standard deviation of the current to the average value;

[0035] When the fluctuation coefficient increases and exceeds the third preset threshold within a series of calculation windows, it is predicted that the pump unit will experience cavitation or cavitation.

[0036] This approach, by calculating the motor current fluctuation coefficient, enables the prediction of cavitation and cavitation phenomena in the pump unit. It provides a simple and effective method for diagnosing mechanical faults using electrical characteristics, enriching the system's fault prediction dimensions and improving the ability to identify potential faults.

[0037] Furthermore, to facilitate intelligent prediction, one feasible solution is that the control module uses a health index model when performing fault prediction. The method for constructing this model includes:

[0038] Multi-dimensional features are extracted from historical normal operation data, including vibration characteristic frequency amplitude, pressure trend slope, and current fluctuation coefficient, and weights are assigned to them.

[0039] The multi-dimensional features are synthesized into a single health index through regression algorithm training.

[0040] In real-time monitoring, the rate of decline of the health index is used to predict the remaining lifespan of the equipment.

[0041] This approach transforms complex equipment status into intuitive quantitative indicators through multi-feature fusion and weight allocation. The model not only assesses current health status but also predicts remaining service life through trends, providing a scientific basis for preventative maintenance and achieving status-based intelligent maintenance. By dynamically allocating weights using historical data, it continuously improves predictive capabilities and effectiveness, realizing intelligent prediction.

[0042] Furthermore, to improve operational response speed, one feasible solution is as follows: when the health index falls below a preset threshold, the control module not only generates an alarm but also automatically generates a maintenance work order and prioritizes maintenance. Simultaneously, it activates the backup pump branch to enter hot standby mode. This solution achieves a complete closed loop from status monitoring and fault warning to maintenance execution. When the health index is below the threshold, the system not only issues an alarm but also automatically triggers the maintenance process, greatly improving operational response speed and automation levels, reducing human intervention, and lowering operational costs.

[0043] Compared with existing technologies, the advantages of this invention are:

[0044] I. A three-layer structure with main process, adaptive switching, and redundancy design was systematically constructed, laying the hardware foundation for a highly reliable water injection system. This structure not only realizes conventional water injection functions, but also significantly reduces the risk of single-point failures through the introduction of redundancy and adaptive mechanisms. This enables the system to maintain continuous operation in the event of equipment failure or abnormal operating conditions, greatly improving the stability and production efficiency of oilfield water injection operations and solving the problems of poor resistance to single-point failures and lack of adaptive operating conditions.

[0045] Second, a complete set of intelligent control methods was proposed, realizing closed-loop control from parameter monitoring and fault diagnosis to adaptive switching and system recovery. It combines hardware redundancy with software intelligence, enabling the system to not only have fault response capabilities, but also state perception and decision-making capabilities, thus realizing intelligent operation and maintenance of the water injection process.

[0046] Third, by integrating pressure, flow and temperature sensors for monitoring and using intelligent diagnostic algorithms, high-precision diagnosis of various fault types such as pump overload, valve jamming, and parameter mutations is achieved, providing a reliable decision-making basis for system adaptive switching. Attached Figure Description

[0047] Figure 1Electrical diagram of a water injection skid process system with adaptive switching and redundancy functions provided in Embodiment 1 of the present invention;

[0048] Figure 2 A device wiring diagram provided for Embodiment 1 of the present invention;

[0049] Figure 3 A flowchart of the control method for a water injection skid process system with adaptive switching and redundancy functions provided in Embodiment 2 of the present invention;

[0050] Figure 4 This is a flowchart of a fault prediction and diagnosis process for a water injection skid process system with adaptive switching and redundancy functions, provided in Embodiment 2 of the present invention.

[0051] Figure 5 A flowchart of the trend analysis algorithm for a water injection skid process system with adaptive switching and redundancy functions provided in Embodiment 2 of the present invention;

[0052] Figure 6 The flowchart of the spectrum analysis algorithm for the water injection skid process system with adaptive switching and redundancy functions provided in Embodiment 2 of the present invention;

[0053] Figure 7 The flowchart illustrates the method for constructing a health index model for a water injection skid process system with adaptive switching and redundancy functions, as provided in Embodiment 2 of the present invention.

[0054] Figure label:

[0055] 100. Main process unit; 200. Adaptive switching unit; 300. Redundancy design unit. Detailed Implementation

[0056] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the term "comprising" or any other variations thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0057] The features and performance of the present invention will be further described in detail below with reference to embodiments.

[0058] Example 1:

[0059] Please refer to Figure 1 A water injection skid process system with adaptive switching and redundancy functions includes:

[0060] The main process unit 100 consists of a main centrifugal pump and a main plunger pump connected in series. A valve group, as well as pressure sensors, flow sensors and temperature sensors are arranged along the process to realize conventional water injection pressurization and collect operating parameters.

[0061] The adaptive switching unit 200 includes a backup pump group branch, a bypass branch, and a control module. The control module receives measurement data from pressure sensors, flow sensors, and temperature sensors, and automatically activates the backup pump group branch and bypass branch after diagnosing equipment failure or process abnormality.

[0062] The redundant design unit 300 employs a dual-valve parallel redundant configuration for at least one valve in the valve group, and implements a dual-path redundant configuration for at least one of the pressure sensor, flow sensor, and temperature sensor, with a spare interface to improve the fault tolerance and reliability of the system.

[0063] Existing water injection skid systems suffer from interruptions in the entire water injection process due to malfunctions. They cannot adjust themselves in real time according to dynamic water injection demands, exhibiting poor resistance to single points of failure and a lack of adaptive operation. This solution systematically constructs a three-layer structure with a main process, adaptive switching, and redundancy design, laying the hardware foundation for a highly reliable water injection system. This structure not only realizes conventional water injection functions but also significantly reduces the risk of single points of failure through the introduction of redundancy and adaptive mechanisms. This allows the system to maintain continuous operation even in the event of equipment failure or abnormal operating conditions, greatly improving the stability and production efficiency of oilfield water injection operations and solving the problems of poor resistance to single points of failure and lack of adaptive operation.

[0064] This solution does not exclusively limit the specific structure of the backup pump group branch. One feasible solution is that the backup pump group branch includes a backup centrifugal pump and its matching valve group redundantly connected in parallel with the main centrifugal pump, and a backup plunger pump and its matching valve group redundantly connected in parallel with the main plunger pump. When this solution is adopted, by setting up backup pump groups and matching valves in parallel for the centrifugal pump and the plunger pump respectively, the rapid redundancy switching of pump equipment is realized, ensuring that the system can complete the switching in a short time when the main pump fails, minimizing the water injection interruption time, which is particularly suitable for oilfield water injection scenarios with high requirements for production continuity.

[0065] This solution does not exclusively limit the specific structure of the bypass branch. One feasible solution is that the bypass branch includes a bridge pipe connecting the inlet of the main centrifugal pump and the outlet of the main plunger pump. The bridge pipe is equipped with a control valve, which is used to maintain the flow when the flow rate is lower than a preset flow threshold or when the pump unit fails. When this solution is adopted, the centrifugal pump can be bypassed when the flow rate is lower than the preset flow threshold, avoiding the centrifugal pump from operating in the inefficient zone. It can also temporarily maintain the flow when the main pump fails, enhancing the system's adaptability to complex operating conditions and expanding the system's effective operating range.

[0066] Reference Figure 2 This embodiment provides a wiring diagram for the equipment and uses a PLC to control each device, ensuring reliability and stability.

[0067] Example 2;

[0068] Please refer to Figure 3 A control method for a water injection skid process system with adaptive switching and redundancy functions, including:

[0069] S100. During the operation of the main centrifugal pump, continuous parameter monitoring and diagnosis of equipment faults and process abnormalities are performed.

[0070] S200: Based on the type of fault or abnormality, it automatically switches to the standby pump set branch, bypass branch, or redundant valves and sensors in the redundant design unit to maintain uninterrupted water injection operation.

[0071] After troubleshooting the S300 system fault, the system resumes operation in the main process unit and updates the equipment status information.

[0072] A complete set of intelligent control methods was proposed, realizing closed-loop control from parameter monitoring and fault diagnosis to adaptive switching and system recovery. By combining hardware redundancy with software intelligence, the system not only has fault response capabilities but also state perception and decision-making capabilities, realizing intelligent operation and maintenance of the water injection process and solving the problem of lack of adaptive operating conditions.

[0073] Reference Figure 4 To facilitate fault prediction and diagnosis, one feasible solution is: the continuous parameter monitoring and diagnosis of equipment faults and process anomalies includes:

[0074] S110. Real-time acquisition of operating parameters of pump group and valve group in the main process unit through pressure sensor, flow sensor, temperature sensor and vibration sensor;

[0075] S120. The control module performs fault prediction and diagnosis based on operating parameters using trend analysis algorithms and spectrum analysis algorithms.

[0076] S130. If a fault is predicted or diagnosed, the adaptive switching unit or the backup equipment in the redundancy design unit is automatically activated according to the preset adaptive switching logic to maintain the continuity of the water injection operation.

[0077] When this solution is adopted, the fusion monitoring of pressure sensors, flow sensors and temperature sensors and intelligent diagnostic algorithms enable high-precision diagnosis of various fault types such as pump overload, valve jamming and parameter mutation, providing a reliable decision basis for system adaptive switching.

[0078] Reference Figure 5 This solution is not limited to the specific steps of the trend analysis algorithm. One feasible solution is that the trend analysis algorithm includes:

[0079] S121. Perform linear fitting on the running parameters within the specified time window and calculate the slope value. k ;

[0080] S122, when the slope value k If the value is negative for several consecutive time windows and its absolute value exceeds the first preset threshold, it is determined that the performance of the pump or pipeline system corresponding to the operating parameter is deteriorating, and an early warning signal is generated.

[0081] When using this scheme, by adjusting the slope of the operating parameters... k The calculations enable early identification of equipment performance degradation trends, achieving predictive maintenance. This changes the traditional reactive maintenance model, shifting from a passive to a proactive approach, effectively preventing sudden failures and extending equipment lifespan.

[0082] Optionally, in this embodiment, the parameter slope is calculated using the following formula. k :

[0083] .

[0084] in i For time series indexing, P i This represents the pressure value at the corresponding moment. n The number of data points in the sliding window.

[0085] When using this scheme, if the slope of the outlet pressure... k A consistently negative value with an absolute value exceeding the set value indicates a steady decline in pump performance, which could indicate impeller wear or channel blockage.

[0086] Reference Figure 6 This solution is not limited to the specific steps of the spectrum analysis algorithm. One feasible solution is that the spectrum analysis algorithm includes:

[0087] S123. Perform a fast Fourier transform on the vibration sensor signal to obtain the vibration spectrum;

[0088] S124. Extract characteristic frequency amplitudes from the vibration spectrum. The characteristic frequency amplitudes include 1 times the rotational frequency amplitude related to pump rotor imbalance, 2 times the rotational frequency amplitude related to equipment misalignment, and fault frequency amplitude related to bearing damage.

[0089] S125. When the growth trend of any characteristic frequency amplitude relative to the baseline value exceeds the second preset threshold, it is determined that the corresponding mechanical fault is developing, and a predictive maintenance alarm is generated.

[0090] This approach allows for the extraction of characteristic frequencies from vibration signals that are associated with mechanical faults such as rotor imbalance, equipment misalignment, and bearing damage, enabling accurate diagnosis and early warning of these faults. This improves the predictive ability for rotating equipment faults and reduces unplanned downtime.

[0091] Specifically, in this embodiment, in feature frequency identification:

[0092] Rotor imbalance: Vibration energy is concentrated at 1 times the rotational frequency.

[0093] Misalignment: Vibrational energy is concentrated at twice the frequency.

[0094] Bearing failure: A specific frequency of bearing failure occurs (determined by the bearing geometry) and is accompanied by high-frequency noise.

[0095] Blade passing frequency: The passing frequency of the blades of a centrifugal pump and its harmonics.

[0096] By establishing a spectral baseline under normal conditions, and during real-time monitoring, if the amplitude of a characteristic frequency shows a monotonically increasing trend, even if its absolute value does not exceed the alarm threshold, it can be predicted that the corresponding fault is developing. For example, if the amplitude of the bearing outer ring fault frequency increases by 20% continuously within a week, it can be predicted that the bearing will fail in a few days.

[0097] Reference Figure 4 To facilitate the prediction of cavitation or cavitation in pump units, one feasible solution is as follows: The fault prediction and diagnosis also includes:

[0098] S140. Calculate the fluctuation coefficient of the motor current, wherein the fluctuation coefficient is the ratio of the standard deviation of the current to the average value;

[0099] S150. When the fluctuation coefficient increases and exceeds the third preset threshold within a series of calculation windows, it is predicted that the pump group will experience cavitation or cavitation.

[0100] This approach, by calculating the motor current fluctuation coefficient, enables the prediction of cavitation and cavitation phenomena in the pump unit. It provides a simple and effective method for diagnosing mechanical faults using electrical characteristics, enriching the system's fault prediction dimensions and improving the ability to identify potential faults.

[0101] Optionally, in this embodiment, the fluctuation coefficient is calculated using the following formula. C :

[0102] ,

[0103] Where N is the total number of current samples collected within the current calculation window. I j For the first j There are 10 current sample values, where I' is the arithmetic mean of all current sample values ​​within the calculation window.

[0104] When using this scheme, the calculated C It is a dimensionless percentage value, suitable for comparing the stability of motors with different power ratings.

[0105] Reference Figure 7 To facilitate intelligent prediction, one feasible solution is that the control module uses a health index model when performing fault prediction. The construction method of this model includes:

[0106] S126. Extract multi-dimensional features from historical normal operation data, including vibration characteristic frequency amplitude, pressure trend slope, and current fluctuation coefficient, and assign weights to them.

[0107] S127. Through regression algorithm training, the multi-dimensional features are synthesized into a single health index.

[0108] In real-time monitoring, the rate of decline of the health index is used to predict the remaining lifespan of the equipment.

[0109] This approach transforms complex equipment status into intuitive quantitative indicators through multi-feature fusion and weight allocation. The model not only assesses current health status but also predicts remaining service life through trends, providing a scientific basis for preventative maintenance and achieving status-based intelligent maintenance. By dynamically allocating weights using historical data, it continuously improves predictive capabilities and effectiveness, realizing intelligent prediction.

[0110] Optionally, in this embodiment, the health index H is calculated using the following formula:

[0111] ,

[0112] in f 1. f 2.f 3. f 4 are the characteristic functions of vibration, pressure, flow rate, and temperature (such as amplitude, trend, fluctuation, etc.). w 1. w 2. w 3. w 4 represents the weights of the corresponding feature functions. b It is obtained through a data-driven approach, using the baseline value as the benchmark.

[0113] When using this approach, the H value is mapped to the remaining lifespan of the equipment. In real-time monitoring, the rate of decrease of the H value is used to predict the time of failure.

[0114] Optionally, in this embodiment, the weight allocation includes:

[0115] Record data for each fault prediction, switching event, and post-event diagnostic results;

[0116] By analyzing this data using machine learning algorithms, the preset thresholds in the trend analysis algorithm and spectrum analysis algorithm, as well as the feature weights in the health index model, are dynamically optimized to improve prediction accuracy.

[0117] To improve operational response speed, one feasible solution is as follows: when the health index falls below a preset threshold, the control module not only generates an alarm but also automatically generates a maintenance work order and prioritizes maintenance. Simultaneously, it activates the backup pump branch to enter hot standby mode. This solution achieves a complete closed loop from status monitoring and fault warning to maintenance execution. When the health index is below the threshold, the system not only issues an alarm but also automatically triggers the maintenance process, significantly improving operational response speed and automation levels, reducing human intervention, and lowering operational costs.

[0118] Specifically, in this embodiment, the sorting maintenance priority includes:

[0119] When multiple fault warnings are detected at the same time, prioritize handling faults that may cause process interruption, and then handle faults that cause efficiency loss.

[0120] Pump group failures have a higher switching priority than valve group failures, and abnormal pressure parameters have a higher switching priority than abnormal temperature parameters.

[0121] The working principle of this embodiment:

[0122] By collecting real-time measurement data from pressure, flow, temperature, and vibration sensors, and employing intelligent algorithms such as trend analysis and spectrum analysis for fault prediction and diagnosis, the control module automatically activates backup pumps, bypass pipelines, or redundant equipment based on preset logic upon detecting performance degradation or fault signs. This seamless switching ensures continuous water injection operations while achieving intelligent closed-loop control throughout the entire process, from condition monitoring and intelligent diagnosis to adaptive maintenance.

[0123] To address the issues of poor resistance to single points of failure and lack of adaptive operation, this solution systematically constructs a three-layer structure with a main process, adaptive switching, and redundancy design, laying the hardware foundation for a highly reliable water injection system. This structure not only realizes conventional water injection functions but also significantly reduces the risk of single points of failure through the introduction of redundancy and adaptive mechanisms. This allows the system to maintain continuous operation even in the event of equipment failure or abnormal operating conditions, greatly improving the stability and production efficiency of oilfield water injection operations and resolving the problems of poor resistance to single points of failure and lack of adaptive operation.

[0124] To achieve intelligent operation and maintenance, this solution proposes a complete set of intelligent control methods, realizing closed-loop control from parameter monitoring and fault diagnosis to adaptive switching and system recovery. It combines hardware redundancy with software intelligence, enabling the system to not only have fault response capabilities, but also state perception and decision-making capabilities, thus realizing intelligent operation and maintenance of the water injection process.

[0125] To facilitate adaptive switching of the system, this solution utilizes a fusion monitoring and intelligent diagnostic algorithm based on pressure, flow, and temperature sensors to achieve high-precision diagnosis of various fault types, such as pump overload, valve jamming, and parameter mutations, providing a reliable decision-making basis for adaptive switching of the system.

[0126] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A control method for a water injection skid process system with adaptive switching and redundancy functions, characterized in that, The water injection skid process system includes: The main process unit consists of a main centrifugal pump and a main plunger pump connected in series. A valve group and pressure, flow and temperature sensors are arranged along the process to realize conventional water injection pressurization and collect operating parameters. The adaptive switching unit includes a backup pump group branch, a bypass branch, and a control module. The control module receives sensor data and automatically activates the backup pump group branch and the bypass branch after diagnosing equipment failure or process abnormality. The redundant design unit adopts a dual-valve parallel redundant configuration for key valves and a dual-path redundant setting for key sensors, and is equipped with a spare interface to improve the fault tolerance and reliability of the system. The backup pump group branch includes a backup centrifugal pump and its matching valve group that are redundantly connected in parallel with the main centrifugal pump, and a backup plunger pump and its matching valve group that are redundantly connected in parallel with the main plunger pump. The bypass branch includes a bridge pipe connecting the inlet of the main centrifugal pump and the outlet of the main plunger pump. The bridge pipe is equipped with a control valve, which is used to maintain the flow in low flow conditions or pump failure. The control method for the water injection skid process system includes: During the operation of the main centrifugal pump, continuous parameter monitoring and diagnosis of equipment faults and process abnormalities are performed. Depending on the type of fault or anomaly, it automatically switches to the standby pump set branch, bypass branch, or redundant valves and sensors in the redundant design unit to maintain uninterrupted water injection operation. After troubleshooting, the system resumes operation in the main process unit and updates the equipment status information. The continuous parameter monitoring and diagnosis of equipment faults and process anomalies includes: The operating parameters of the pump group and valve group in the main process unit are collected in real time through pressure sensors, flow sensors, temperature sensors and vibration sensors. The control module performs fault prediction and diagnosis based on operating parameters using trend analysis and spectrum analysis algorithms. If a fault is predicted or diagnosed, the adaptive switching unit or the backup equipment in the redundancy design unit will be automatically activated according to the preset adaptive switching logic to maintain the continuity of the water injection operation. The trend analysis algorithm includes: Perform linear fitting on the running parameters within a specified time window and calculate the slope value; When the slope value is negative for several consecutive time windows and its absolute value exceeds the first preset threshold, it is determined that the performance of the pump or pipeline system corresponding to the operating parameter is deteriorating, and an early warning signal is generated. The slope k is calculated using the following formula: ， Where i is the time series index, Pi is the pressure value at the corresponding time, and n is the number of data points in the sliding window.

2. The control method for a water injection skid process system with adaptive switching and redundancy functions according to claim 1, characterized in that, The spectrum analysis algorithm includes: The vibration spectrum is obtained by performing a fast Fourier transform on the vibration sensor signal. The characteristic frequency amplitudes are extracted from the vibration spectrum. These characteristic frequency amplitudes include the 1x rotational frequency amplitude related to pump rotor imbalance, the 2x rotational frequency amplitude related to equipment misalignment, and the fault frequency amplitude related to bearing damage. When the growth trend of any characteristic frequency amplitude relative to the baseline value exceeds the second preset threshold, it is determined that the corresponding mechanical fault is developing, and a predictive maintenance alarm is generated.

3. The control method for a water injection skid process system with adaptive switching and redundancy functions according to claim 1, characterized in that, The fault prediction and diagnosis also includes: Calculate the fluctuation coefficient of the motor current, where the fluctuation coefficient is the ratio of the standard deviation of the current to the average value; When the fluctuation coefficient increases and exceeds the third preset threshold within a series of calculation windows, it is predicted that the pump unit will experience cavitation or cavitation.

4. The control method for a water injection skid process system with adaptive switching and redundancy functions according to claim 1, characterized in that, When the control module performs fault prediction, it adopts a health index model, the construction method of which includes: Multi-dimensional features are extracted from historical normal operation data, including vibration characteristic frequency amplitude, pressure trend slope, and current fluctuation coefficient, and weights are assigned to them. The multi-dimensional features are synthesized into a single health index through regression algorithm training. In real-time monitoring, the rate of decline of the health index is used to predict the remaining lifespan of the equipment.

5. The control method for a water injection skid process system with adaptive switching and redundancy functions according to claim 4, characterized in that, When the health index is lower than the preset threshold, the control module not only generates an alarm, but also automatically generates a maintenance work order and sorts the maintenance priorities, while starting the standby pump group branch to enter hot standby state.

Images