An oil pressure analysis-based water-mixing energy-saving optimization method

Abstract

The application discloses a water-mixing energy-saving optimization method based on oil pressure analysis, relates to the technical field of water-mixing optimization of oilfield gathering and transportation, and comprises the following steps: collecting real-time parameters and historical data of wells in different metering rooms in a current oilfield gathering and transportation station, including oil pressure, water-mixing flow of each well and total water-mixing flow of the metering room, establishing a water-mixing optimization database and recording the collected parameters of the different metering rooms in the current oilfield gathering and transportation station; obtaining real-time data of each well through the water-mixing optimization database, obtaining an upper limit of oil pressure safety and an oil pressure rising rate threshold based on the change of real-time parameters in a monitoring period and the collected data of different wells in the current oilfield gathering and transportation station in the historical monitoring period. The application can accurately determine the well state based on oil pressure analysis, realize overall collaborative optimization from a single well to a metering room, and consider both safety guarantee and energy saving and consumption reduction in the oilfield gathering and transportation.

Description

Technical Field

[0001] This invention relates to the field of water blending optimization technology in oilfield gathering and transportation, and more specifically, to a water blending energy-saving optimization method based on oil pressure analysis. Background Technology

[0002] As oilfield development enters the mid-to-late production reduction stage, the stable operation and energy conservation of gathering and transportation stations have become the core development goals of the industry. Water blending technology, as a key link in the oilfield gathering and transportation system, can effectively reduce crude oil viscosity, improve fluidity, and avoid the impact of problems such as waxing and blockage on gathering and transportation efficiency by adding hot water or sewage to the oil well produced fluid. Existing water injection optimization technologies mostly adopt fixed water injection flow patterns or simply adjust based on a single well condition parameter. They lack accurate judgment and differentiated optimization for the actual operating conditions of different oil wells. Some oil wells are prone to abnormally high oil pressure due to wax deposition and blockage during production. However, existing technologies lack clear well condition judgment conditions and cannot identify such unbalanced wells in a timely manner, often delaying the timing of optimization and adjustment. This can lead to safety accidents such as equipment damage and pipeline rupture, threatening the stability of the gathering and transportation system. At the same time, existing methods have not established a correlation analysis mechanism between the state of a single well and the overall state of the metering room. The optimization process is mostly limited to local adjustment of a single well, ignoring the synergistic effects of multiple wells in the metering room. Current water injection optimization schemes lack synergistic consideration of energy consumption and safety. Most technologies fail to accurately find the minimum energy consumption critical point under safe operating conditions. The adoption of conservative high water injection flow strategies can easily lead to high operating costs and low practicality. Summary of the Invention

[0003] In view of the problems in related technologies, this invention proposes a water-incorporation energy-saving optimization method based on oil pressure analysis to overcome the above-mentioned technical problems existing in the existing related technologies.

[0004] Therefore, the specific technical solution adopted by the present invention is as follows: A water-incorporation energy-saving optimization method based on oil pressure analysis, comprising the following steps: S1. Collect real-time parameters and historical data of wells in different metering chambers in the current oilfield gathering and transportation station, including oil pressure, water injection flow rate of each well and total water injection flow rate of the metering chamber, establish a water injection optimization database and record the collected parameters of different metering chambers in the current oilfield gathering and transportation station. S2. Obtain real-time data for each well through the water injection optimization database. Based on the changes in real-time parameters within the monitoring period, construct a well status definition model to determine the well status, including unbalanced state, relatively balanced state, and no data state. Obtain the oil pressure safety upper limit and oil pressure rise rate threshold by collecting data from different wells in the current oilfield gathering and transportation station under the historical monitoring period. Optimize the optimization strategy for unbalanced state wells, optimize the water injection volume to optimize unbalanced state wells to a relatively balanced state, and record the labels of different wells. S3. Statistically analyze the well labels of different metering chambers to define the status of the metering chambers. For metering chambers in a relatively balanced state, periodically reduce the total water injection in the metering chamber and simultaneously monitor the well condition and oil pressure trend to adjust the status of individual wells to the optimal balanced state. This allows individual metering chambers to be adjusted to the optimal balanced state. Following the single metering chamber optimization steps, all metering chambers are adjusted to the optimal balanced state in sequence, gradually completing the status adjustment of all metering chambers in the current oilfield gathering and transportation station, obtaining the minimum water injection under safe gathering and transportation to complete the water injection optimization.

[0005] In a preferred embodiment, S1 includes the following steps: S11. Collect real-time parameters of wells under different metering stations in the oilfield gathering and transportation station through sensors, including the real-time oil pressure of each well. Water mixing flow rate and the total water mixing flow rate between metering stations Where i is the well number, j is the metering room number, and t is the parameter acquisition time; S12. The collected parameters are smoothed using an exponentially weighted moving average method to eliminate noise. The specific steps are as follows: ; in, This represents the raw parameters collected from well number i at parameter acquisition time t. The smoothed acquisition parameters of the i-th well under t. This represents the smoothing parameter value at the previous time t-1. This is a smoothing factor, with a value between 0 and 1. S13. Align the data based on the timestamp of the collected parameters, establish a water injection optimization database through MySQL, create archives based on the metering room number, create sub-archives based on the well number in different metering room archives, and record the real-time collected data, the collected data under the historical monitoring cycle, and the water injection adjustment data for each well. The water injection adjustment data includes the water injection adjustment parameters and the adjustment time.

[0006] As a preferred implementation, S21, based on the historical data analysis of each well, the oil pressure rise rate threshold and the oil pressure safety upper limit are obtained, the real-time data of each well is obtained through the water injection optimization database, and the state of each well is defined by the well state definition model in combination with the changes of real-time parameters within the monitoring period, and the label is recorded in the water injection optimization database. S22. For wells in different states, different adjustment and optimization strategies are used to adjust and optimize the well state, and the water injection volume after adjustment and optimization is recorded.

[0007] In a preferred embodiment, S21 includes the following steps: S211. By optimizing the database through water injection, the collected data of different wells in the current oilfield gathering and transportation station under the historical monitoring period are obtained, and the statistical quantile method is used to determine the upper limit of oil pressure for different wells. Configure settings. ,in The function representing the calculation of the higher quantiles. This is the set of historical oil pressure data for well number i under normal operating conditions; The oil pressure rise rate threshold is obtained by analyzing the slope of oil pressure fluctuations within historical normal cycles. ,in These represent the N historical normal monitoring cycles of well number i. The mean value of the linear regression slope of the oil pressure calculated internally and the standard deviation of the corresponding slope; S212. The current monitoring period obtained from the water blending optimization database. Linear regression analysis was performed on the internal oil pressure data sequence to obtain the real-time oil pressure rise rate. : ; in, This represents the slope obtained from the linear regression fit, which is the average rate of increase in oil pressure during this period. The intercept of the regression line. The real-time oil pressure of well number i after smoothing at time t; Based on the real-time oil pressure and oil pressure rise rate during the monitoring period, the state of each well is defined through a well state definition model, which includes the following steps: The definition of a non-equilibrium state is: ; The definition of a relative equilibrium state is: ; The state is defined as "no data" when there is no real-time data for the current well. This refers to the real-time oil pressure data of well i.

[0008] In a preferred embodiment, S221, for a well in an unbalanced state, state optimization is performed by adjusting an optimization strategy. This involves optimizing the water injection rate to bring the well from an unbalanced state to a relatively balanced state. Specifically, this includes the following steps: For wells in an unbalanced state, check the water injection adjustment data in the current well file in the water injection optimization database, determine the adjustment path for wells in an unbalanced state, and if the water injection adjustment time of the current well is within the previous monitoring cycle, mark the well in the current unbalanced state as recently adjusted, and continue monitoring until the end of the next monitoring cycle. When the current well water injection adjustment time During the previous monitoring period, the water injection volume of wells in the current unbalanced state was optimized and adjusted to restore the current well state to a relatively balanced state. The specific steps are as follows: ; in, These represent the adjusted new water flow rate and the original water flow rate, respectively. To determine the water injection flow rate adjustment based on historical water injection flow rate and oil pressure data of current well i, the following steps are taken: Historical operating data of current well i, including water injection flow rate and oil pressure data, are collected. After normalization, the collected historical data is divided into training, validation, and test sets (70% training, 15% validation, and 15% test). A multilayer perceptron is selected as the model for construction, with one neuron in the input layer and one neuron in the output layer. The multilayer perceptron is trained using the training set, and parameters are adjusted using the validation and test sets to obtain the predicted water injection flow rate adjustment output model. The water injection flow rate adjustment and adjustment time are recorded in the corresponding well's sub-file in the water injection optimization database, and the label of the current well is recorded as "adjusted". For wells in a relatively balanced state or those without data, the label for the current well is recorded as "continuous monitoring" in the sub-file of the corresponding well in the water incorporation optimization database.

[0009] In a preferred embodiment, step S3 includes the following steps: S31. Collect tags for sub-files of wells under different metering room files in the water mixing optimization database, determine the metering room status, and mark metering rooms in a relatively balanced state. S32. For metering rooms in a relatively balanced state, the state of each well in the metering room is optimized and adjusted to the best balanced state by adjusting the water injection volume. The best balanced state of the current metering room is obtained, and the total water injection flow rate of the current metering room is recorded as the theoretical minimum total water injection volume. The metering rooms under the jurisdiction of the current oilfield gathering and transportation station are gradually adjusted to the best balanced state to complete the water injection optimization of the oilfield gathering and transportation station.

[0010] In a preferred embodiment, S31 includes the following steps: S311. Sub-file tags of different metering room files in the statistical water-injection optimization database are used to determine the status of the metering room: If all sub-files in the current metrology room file are labeled "continuous monitoring", then the current metrology room status is marked as a relatively balanced state. If the label "adjusted" or "recently adjusted" exists in the sub-files under the current metrology file, then the current metrology state is marked as unbalanced.

[0011] In a preferred embodiment, S32 includes the following steps: S321. For the metering room marked as being in a relatively balanced state, the state of each well in the metering room is optimized and adjusted by adjusting the water injection rate, specifically including the following steps: By step size Reduce the total water injection flow rate in the metering room under the current condition, maintain the new water injection flow rate, and continuously monitor for n cycles. Count the number of wells in the current unbalanced state in the metering room after the monitoring cycle. ; S322, when When =0, repeat step S321 to continue adjusting the total water mixing flow rate in the current metering room; S323, When 0 < ≤ If the wells are in an abnormal, unbalanced state, return to step S221 for state optimization. If all wells return to a relatively balanced state after optimization, update the total water mixing flow rate in the metering room with the new optimized total water mixing flow rate. Simultaneously, repeat step S321 to adjust the total water mixing flow rate in the current metering room until... > If, after optimization, a well cannot restore a relatively balanced state, the total water injection flow rate of the previous cycle is taken as the optimal water injection flow rate for that metering interval. To determine the threshold; S324. Mark the state of the metering room under the optimal water mixing flow rate as the optimal equilibrium state. Repeat steps S321 to S323 to optimize all subordinate metering rooms of the oilfield gathering and transportation station to the optimal equilibrium state, and complete the water mixing optimization of the oilfield gathering and transportation station.

[0012] The beneficial effects of this invention are as follows: 1. This invention sets judgment conditions based on the actual conditions of different wells in the oilfield gathering and transportation station to analyze the state of different wells. For wells in an unbalanced state, it prioritizes optimization to a relatively balanced state to ensure the safety of oilfield gathering and transportation. At the same time, it determines the state of the metering room based on the state of the wells in the metering room. Through trial descent cycles, it comprehensively optimizes the state of the metering room and finally completes the water injection optimization of the oilfield gathering and transportation station. By finding the lowest energy consumption critical point under safe operating conditions, it directly reduces the operating cost of oilfield gathering and transportation. 2. By setting clear well status judgment conditions, such as the upper limit of oil pressure safety and the threshold of oil pressure rise rate, this invention can identify wells in an unbalanced state in a timely manner. These wells often have risks such as wax deposition and blockage. By increasing the water injection flow rate to melt the wax deposits or disperse the blockages, the well can be restored to a relatively balanced and safe operating state, avoiding safety accidents such as equipment damage and pipeline rupture caused by abnormal rise in oil pressure, and ensuring the stable operation of the oilfield gathering and transportation system. 3. This invention determines the state of the metering chamber based on the state of the well, and performs comprehensive optimization through trial descent cycles. The optimization approach, from local to overall, makes the water injection flow rate of the entire metering chamber more reasonable. Single-well optimization ensures overall safety, and the optimal operating point of the entire pipeline system is found by adjusting the total flow rate of the metering chamber, thereby improving the operating efficiency of the metering chamber and ultimately achieving overall optimization of the oilfield gathering and transportation station. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a flowchart of a water-incorporation energy-saving optimization method based on oil pressure analysis according to an embodiment of the present invention. Detailed Implementation

[0015] To further illustrate the various embodiments, the present invention provides accompanying drawings, which are part of the disclosure of the present invention. These drawings are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these drawings, those skilled in the art should be able to understand other possible implementation methods and the advantages of the present invention. The components in the drawings are not drawn to scale, and similar component symbols are generally used to represent similar components.

[0016] According to an embodiment of the present invention, a water-incorporation energy-saving optimization method based on oil pressure analysis is provided.

[0017] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments: Example 1: As Figure 1 As shown, according to an embodiment of the present invention, a water-incorporation energy-saving optimization method based on oil pressure analysis includes the following steps: S1. Collect real-time parameters and historical data of wells in different metering chambers in the current oilfield gathering and transportation station, including oil pressure, water injection flow rate of each well and total water injection flow rate of the metering chamber, establish a water injection optimization database and record the collected parameters of different metering chambers in the current oilfield gathering and transportation station. S11. Collect real-time parameters of wells under different metering stations in the oilfield gathering and transportation station through sensors, including the real-time oil pressure of each well. Water mixing flow rate and the total water mixing flow rate between metering stations Where i is the well number, j is the metering room number, and t is the parameter acquisition time; S12. The collected parameters are smoothed using an exponentially weighted moving average method to eliminate noise. The specific steps are as follows: ; in, This represents the raw parameters collected from well number i at parameter acquisition time t. The smoothed acquisition parameters of the i-th well under t. This represents the smoothing parameter value at the previous time t-1. This is a smoothing factor, with a value between 0 and 1. It should be noted that the smoothing factor can control the weight of current new data and historical data in the weighted average. During the data acquisition process, the sensor will inevitably be affected by various external factors, such as electromagnetic interference and mechanical vibration, which will generate noise. Smoothing by exponential weighted moving average can effectively eliminate the influence of noise. The specific value needs to be set empirically based on the actual well conditions in actual use. The sensors include pressure sensors and electromagnetic flow meters. The smoothed data will be used as real-time parameters and subsequently recorded in the water incorporation optimization database.

[0018] S13. Align the data based on the timestamp of the collected parameters, establish a water injection optimization database through MySQL, create archives based on the metering room number, create sub-archives based on the well number in different metering room archives, and record the real-time collected data, the collected data under the historical monitoring cycle, and the water injection adjustment data for each well. The water injection adjustment data includes the water injection adjustment parameters and the adjustment time.

[0019] S2. Obtain real-time data for each well through the water injection optimization database. Based on the changes in real-time parameters within the monitoring period, construct a well status definition model to determine the well status, including unbalanced state, relatively balanced state, and no data state. Obtain the oil pressure safety upper limit and oil pressure rise rate threshold by collecting data from different wells in the current oilfield gathering and transportation station under the historical monitoring period. Optimize the optimization strategy for unbalanced state wells, optimize the water injection volume to optimize unbalanced state wells to a relatively balanced state, and record the labels of different wells. S21. Based on the historical data analysis of each well, the oil pressure rise rate threshold and oil pressure safety upper limit are obtained. Real-time data of each well is obtained through the water injection optimization database. Combined with the changes in real-time parameters within the monitoring period, the state of each well is defined through the well state definition model, and the label is recorded in the water injection optimization database. S211. By optimizing the database through water injection, the collected data of different wells in the current oilfield gathering and transportation station under the historical monitoring period are obtained, and the statistical quantile method is used to determine the upper limit of oil pressure for different wells. Configure settings. ,in The function representing the calculation of the higher quantiles. This is the set of historical oil pressure data for well number i under normal operating conditions; It should be noted that, The function representing the calculation of high quantiles is mainly used to extract historical oil pressure data from the i-th well under normal operating conditions. In the process, a value that can represent a relatively high level is extracted. You can choose to use a function that calculates the 95th percentile of historical oil pressure data or a function that calculates the mean plus 2 times the standard deviation to serve as the safe upper limit of oil pressure for the current well i. In this way, by analyzing the historical oil pressure of the well during normal, unblocked periods, a pressure value that will not be exceeded most of the time can be found and set as the safety red line.

[0020] The oil pressure rise rate threshold is obtained by analyzing the slope of oil pressure fluctuations within historical normal cycles. ,in These represent the N historical normal monitoring cycles of well number i. The mean value of the linear regression slope of the oil pressure calculated internally and the standard deviation of the corresponding slope; It should be noted that the value of N is usually set to 5, but it can also be adjusted according to the actual situation. The oil pressure rise rate threshold obtained by the historical mean and standard deviation can reflect the recent state of the well and determine the maximum reasonable rate of natural fluctuation of oil pressure over time under normal production conditions.

[0021] S212. The current monitoring period obtained from the water addition optimization database. Linear regression analysis was performed on the internal oil pressure data sequence to obtain the real-time oil pressure rise rate. : ; in, This represents the slope obtained from the linear regression fit, which is the average rate of increase in oil pressure during this period. The intercept of the regression line. The real-time oil pressure of well number i after smoothing at time t; It should be noted that linear regression fits discrete oil pressure data points into a straight line, and its slope directly reflects whether the oil pressure is rising, falling, or remaining stable within that period. The linear regression equation can be solved using the least squares method to obtain the intercept and slope. (Monitoring period) The settings need to be configured according to the monitoring needs of different oilfield gathering and transportation stations. You can consult experts in the relevant fields for configuration and adjustment.

[0022] Based on the real-time oil pressure and oil pressure rise rate during the monitoring period, the state of each well is defined through a well state definition model, which includes the following steps: The definition of a non-equilibrium state is: ; The definition of a relative equilibrium state is: ; The state is defined as "no data" when there is no real-time data for the current well. This refers to the real-time oil pressure data of well i.

[0023] It should be noted that the equilibrium state of each well is defined based on the oil pressure trend analysis of each well, and the equilibrium state of each metering is defined based on the equilibrium state of each well. The unbalanced state is when the rate of increase of wellhead oil pressure exceeds the set value or the current real-time oil pressure exceeds the set safety limit in the most recent monitoring cycle. The relative equilibrium state is when the rate of increase of wellhead oil pressure does not exceed the set value and the current real-time oil pressure does not exceed the set safety limit in the most recent monitoring cycle. The no-data state is when there is no real-time data related to oil pressure, water injection flow rate, etc. for that well.

[0024] S22. For wells in different states, different adjustment and optimization strategies are used to adjust and optimize the well state, and the water injection volume after adjustment and optimization is recorded. S221. For wells in an unbalanced state, state optimization is performed through adjustment and optimization strategies. This involves optimizing the water injection rate to bring the well from an unbalanced state to a relatively balanced state. Specifically, this includes the following steps: For wells in an unbalanced state, check the water injection adjustment data in the current well file in the water injection optimization database, determine the adjustment path for wells in an unbalanced state, and if the water injection adjustment time of the current well is within the previous monitoring cycle, mark the well in the current unbalanced state as recently adjusted, and continue monitoring until the end of the next monitoring cycle. It should be noted that if the current well's water injection adjustment time is within the previous monitoring cycle, it means that the current well has recently undergone water injection adjustment. Since there is a lag in the flow rate adjustment to the oil pressure response, the system will not make new adjustments but will continue to observe the effect of the previous adjustment cycle until the next monitoring cycle when the well status is determined and adjusted.

[0025] When the current well water injection adjustment time During the previous monitoring period, the water injection volume of wells in the current unbalanced state was optimized and adjusted to restore the current well state to a relatively balanced state. The specific steps are as follows: ; in, These represent the adjusted new water flow rate and the original water flow rate, respectively. To determine the water injection flow rate adjustment based on historical water injection flow rate and oil pressure data of current well i, the following steps are taken: Historical operating data of current well i, including water injection flow rate and oil pressure data, are collected. After normalization, the collected historical data is divided into training, validation, and test sets (70% training, 15% validation, and 15% test). A multilayer perceptron is selected as the model for construction, with one neuron in the input layer and one neuron in the output layer. The multilayer perceptron is trained using the training set, and parameters are adjusted using the validation and test sets to obtain the predicted water injection flow rate adjustment output model. The water injection flow rate adjustment and adjustment time are recorded in the corresponding well's sub-file in the water injection optimization database, and the label of the current well is recorded as "adjusted". It should be noted that the input layer has 1 neuron, representing oil pressure, and the output layer has 1 neuron, representing the water mixing flow rate adjustment. During the training of the model to predict the water mixing flow rate adjustment, 1-3 hidden layers can be set, with different combinations of neurons in each hidden layer ranging from 10 to 100. Experiments are used to determine the optimal structure. The ReLU nonlinear activation function is used as the activation function in the hidden layers, mean squared error is used as the loss function, and Adam is chosen as the optimization algorithm. Training data is input into the neural network model in batches, and forward propagation is used to calculate the predicted values. Then, backpropagation is used to calculate the gradient, and the optimization algorithm is used to update the model's weights and biases. Simultaneously, a validation set is used to monitor model performance and prevent overfitting. Test set data is used to evaluate the trained model, and evaluation metrics are calculated. When model performance is unsatisfactory, grid search and random search are used to adjust the learning rate, number of hidden layers, number of neurons, batch size, etc., to obtain the optimal model.

[0026] For wells in a relatively balanced state or those without data, the label for the current well is recorded as "continuous monitoring" in the sub-file of the corresponding well in the water incorporation optimization database.

[0027] It should be noted that for wells in an unbalanced state, the main recommendation is to increase the water injection flow rate to bring them into a safe gathering and transportation state. There are two scenarios: First, if the water injection flow rate of the current well has not been adjusted in the most recent monitoring cycle, then it is recommended to increase the water injection flow rate, for example, "It is recommended to increase the water injection flow rate to 0.5 m³ / h". The specific increase amount is suggested by a neural network model. Second, if the water injection flow rate of the current well has already been adjusted in the most recent monitoring cycle, considering the lag in oil pressure trend changes after the water injection flow rate adjustment, it is recommended to continue monitoring, for example, "It is recommended to continue monitoring; the water injection flow rate was increased 5 hours ago". Labeling wells in different states facilitates subsequent metering-based water injection optimization.

[0028] Example 2: S3. Statistically analyze the well labels of different metering rooms to define the metering room status. For metering rooms in a relatively balanced state, the total water injection in the metering room is reduced periodically, and the well condition and oil pressure trend are monitored simultaneously to adjust the status of individual wells to the optimal balanced state. This allows individual metering rooms to be adjusted to the optimal balanced state. Following the single metering room optimization steps, all metering rooms are adjusted to the optimal balanced state in sequence, gradually completing the status adjustment of all metering rooms in the current oilfield gathering and transportation station, obtaining the minimum water injection under safe gathering and transportation to complete the water injection optimization. S31. Collect tags for sub-files of wells under different metering room files in the water mixing optimization database, determine the metering room status, and mark metering rooms in a relatively balanced state. S311. Sub-file tags of different metering room files in the statistical water-dosing optimization database are used to determine the status of the metering room: If all sub-files in the current metrology room file are labeled "continuous monitoring", then the current metrology room status is marked as a relatively balanced state. If the label "adjusted" or "recently adjusted" exists in the sub-files under the current metrology file, then the current metrology state is marked as unbalanced.

[0029] It should be noted that the identification of the relative equilibrium state of a metering chamber means that if all wells under that metering chamber are in a relatively balanced state, then the metering chamber is considered to be in a relatively balanced state. For metering chambers that are not in a balanced state, it is necessary to further adjust all wells under that metering chamber to a relatively balanced state in order to facilitate the final water injection optimization adjustment of the metering chamber. Starting from a single metering chamber, for wells in a non-balanced state, the water injection flow rate is automatically adjusted according to the optimization suggestions recommended by the system, and the oil pressure trend changes of all wells under that metering chamber are continuously monitored within a monitoring cycle. If there are still wells in a non-balanced state after a monitoring cycle, the well water injection flow rate is automatically adjusted again, and the oil pressure trend changes of all wells under that metering chamber are continuously monitored within the next cycle until all wells under that metering chamber are in a relatively balanced state. This state is mainly achieved by adjusting the opening of the water injection valve of all wells under that metering chamber to a relatively balanced state. However, at this point, the water injection volume in the metering chamber may still be relatively high, and further water injection optimization is needed to adjust the metering chamber to the optimal balanced state.

[0030] S32. For metering rooms in a relatively balanced state, the state of each well in the metering room is optimized and adjusted to the best balanced state by adjusting the water injection volume. The best balanced state of the current metering room is obtained, and the total water injection flow rate of the current metering room is recorded as the theoretical minimum total water injection volume. The metering rooms under the jurisdiction of the current oilfield gathering and transportation station are gradually adjusted to the best balanced state to complete the water injection optimization of the oilfield gathering and transportation station. S321. For the metering room marked as being in a relatively balanced state, the state of each well in the metering room is optimized and adjusted by adjusting the water injection rate, specifically including the following steps: By step size Reduce the total water injection flow rate in the metering chamber under the current condition, maintain the new water injection flow rate, and continuously monitor for n cycles. Count the number of wells in the current unbalanced state in the metering chamber after the monitoring cycle. ; S322, when When =0, repeat step S321 to continue adjusting the total water mixing flow rate in the current metering room; S323, When 0 < ≤ If the wells are in an abnormal, unbalanced state, return to step S221 for state optimization. If all wells return to a relatively balanced state after optimization, update the total water mixing flow rate in the metering room with the new optimized total water mixing flow rate. Simultaneously, repeat step S321 to adjust the total water mixing flow rate in the current metering room until... > If, after optimization, a well cannot restore a relatively balanced state, the total water injection flow rate of the previous cycle is taken as the optimal water injection flow rate for that metering interval. To determine the threshold; S324. Mark the state of the metering room under the optimal water mixing flow rate as the optimal equilibrium state. Repeat steps S321 to S323 to optimize all subordinate metering rooms of the oilfield gathering and transportation station to the optimal equilibrium state, and complete the water mixing optimization of the oilfield gathering and transportation station.

[0031] It should be noted that the step size The value is typically set to 5%, but this can also be adjusted by consulting experts in the field and considering the number of wells in the metering room. n is usually set to 2, but can be adjusted based on the well feedback rate and monitoring cycle duration under actual conditions. The settings need to be based on the number of wells currently in the metering chamber, typically 40% of the total number of wells. These settings can be adjusted according to actual conditions. When a single metering chamber is in a relatively balanced state, the water injection flow rate is gradually reduced according to the system's recommended optimization suggestions. The oil pressure trend of all wells in that metering chamber is continuously monitored over n monitoring cycles. If the oil pressure trend of all wells in that metering chamber is normal over n monitoring cycles, the water injection flow rate will continue to decrease in the next monitoring cycle until a well in that metering chamber begins to show an unbalanced state and cannot be adjusted further. The water injection flow rate of the previous cycle is then the optimal water injection flow rate for that metering chamber. First, each other metering chamber is adjusted to a relatively balanced state, and then the state of each metering chamber is adjusted to the optimal balanced state. This process ultimately completes the water injection optimization of the oilfield gathering and transportation station. The remaining metering chambers are then optimized to the optimal balanced state, meaning that all wells in that metering chamber obtain the lowest possible water injection rate while maintaining safe gathering and transportation.

[0032] In summary, this invention sets judgment conditions based on the actual conditions of different wells in oilfield gathering and transportation stations to analyze the state of different wells. For wells in an unbalanced state, it prioritizes optimization to a relatively balanced state to ensure the safety of oilfield gathering and transportation. At the same time, it determines the state of the metering room based on the state of the wells in the metering room, and comprehensively optimizes the state of the metering room through trial descent cycles. Finally, it completes the water injection optimization of oilfield gathering and transportation stations. By finding the lowest energy consumption critical point under safe operating conditions, it directly reduces the operating cost of oilfield gathering and transportation. By setting clear well status judgment conditions, such as the upper limit of oil pressure safety and the threshold of oil pressure rise rate, wells in an unbalanced state can be identified in a timely manner. These wells often have risks such as wax deposition and blockage. By increasing the water injection flow rate to melt the wax deposits or dissipate the blockages, the well can be restored to a relatively balanced and safe operating state, avoiding safety accidents such as equipment damage and pipeline rupture caused by abnormal oil pressure rise, and ensuring the stable operation of the oilfield gathering and transportation system. By determining the state of the metering room based on the well status, and conducting comprehensive optimization through trial descent cycles, the optimization approach from local to overall can make the water injection flow rate of the entire metering room more reasonable. Single-well optimization ensures overall safety, and by adjusting the total flow rate of the metering room, the optimal operating point of the entire pipeline network system can be found, improving the operating efficiency of the metering room, and thus achieving the overall optimization of the oilfield gathering and transportation station.

[0033] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A water-incorporation energy-saving optimization method based on oil pressure analysis, characterized in that, The method includes the following steps: S1. Collect real-time parameters and historical data of wells in different metering chambers in the current oilfield gathering and transportation station, including oil pressure, water injection flow rate of each well and total water injection flow rate of the metering chamber, establish a water injection optimization database and record the collected parameters of different metering chambers in the current oilfield gathering and transportation station. S2. Obtain real-time data for each well through the water injection optimization database. Based on the changes in real-time parameters within the monitoring period, construct a well status definition model to determine the well status, including unbalanced state, relatively balanced state, and no data state. Obtain the oil pressure safety upper limit and oil pressure rise rate threshold by collecting data from different wells in the current oilfield gathering and transportation station under the historical monitoring period. Optimize the optimization strategy for unbalanced state wells, optimize the water injection volume to optimize unbalanced state wells to a relatively balanced state, and record the labels of different wells. S21. Based on the historical data analysis of each well, the oil pressure rise rate threshold and oil pressure safety upper limit are obtained. Real-time data of each well is obtained through the water injection optimization database. Combined with the changes in real-time parameters within the monitoring period, the state of each well is defined through the well state definition model, and the label is recorded in the water injection optimization database. S211. By optimizing the database through water injection, the collected data of different wells in the current oilfield gathering and transportation station under the historical monitoring period are obtained, and the statistical quantile method is used to determine the upper limit of oil pressure for different wells. Configure settings. ,in The function representing the calculation of the higher quantiles. This is the set of historical oil pressure data for well number i under normal operating conditions; The oil pressure rise rate threshold is obtained by analyzing the slope of oil pressure fluctuations within historical normal cycles. ,in These represent the N historical normal monitoring cycles of well number i. The mean value of the linear regression slope of the oil pressure calculated internally and the standard deviation of the corresponding slope; S212. The current monitoring period obtained from the water blending optimization database. Linear regression analysis was performed on the internal oil pressure data sequence to obtain the real-time oil pressure rise rate. : ; in, This represents the slope obtained from the linear regression fit, which is the average rate of increase in oil pressure during this period. The intercept of the regression line. The real-time oil pressure of well number i after smoothing at time t; Based on the real-time oil pressure and oil pressure rise rate during the monitoring period, the state of each well is defined through a well state definition model, which includes the following steps: The definition of a non-equilibrium state is: ; The definition of a relative equilibrium state is: ; The state is defined as "no data" when there is no real-time data for the current well. This refers to the real-time oil pressure data of well i; S3. Statistically analyze the well labels of different metering chambers to define the status of the metering chambers. For metering chambers in a relatively balanced state, periodically reduce the total water injection in the metering chamber and simultaneously monitor the well condition and oil pressure trend to adjust the status of individual wells to the optimal balanced state. This allows individual metering chambers to be adjusted to the optimal balanced state. Following the single metering chamber optimization steps, all metering chambers are adjusted to the optimal balanced state in sequence, gradually completing the status adjustment of all metering chambers in the current oilfield gathering and transportation station, obtaining the minimum water injection under safe gathering and transportation to complete the water injection optimization.

2. The water-incorporation energy-saving optimization method based on oil pressure analysis according to claim 1, characterized in that, S1 includes the following steps: S11. Collect real-time parameters of wells under different metering stations in the oilfield gathering and transportation station through sensors, including the real-time oil pressure of each well. Water mixing flow rate and the total water mixing flow rate between metering stations Where i is the well number, j is the metering room number, and t is the parameter acquisition time; S12. The collected parameters are smoothed using an exponentially weighted moving average method to eliminate noise. The specific steps are as follows: ; in, This represents the raw parameters collected from well number i at parameter acquisition time t. The smoothed acquisition parameters of the i-th well under t. This represents the smoothing parameter value at the previous time t-1. This is a smoothing factor, with a value between 0 and 1; S13. Align the data based on the timestamp of the collected parameters, establish a water injection optimization database through MySQL, create archives based on the metering room number, create sub-archives based on the well number in different metering room archives, and record the real-time collected data, the collected data under the historical monitoring cycle, and the water injection adjustment data for each well. The water injection adjustment data includes the water injection adjustment parameters and the adjustment time.

3. The water-incorporation energy-saving optimization method based on oil pressure analysis according to claim 2, characterized in that, S2 further includes the following steps: S22. For wells in different states, different adjustment and optimization strategies are used to adjust and optimize the well state, and the water injection volume after adjustment and optimization is recorded.

4. The water-incorporation energy-saving optimization method based on oil pressure analysis according to claim 3, characterized in that, S22 includes the following steps: S221. For wells in an unbalanced state, state optimization is performed through adjustment and optimization strategies. This involves optimizing the water injection rate to bring the well from an unbalanced state to a relatively balanced state. Specifically, this includes the following steps: For wells in an unbalanced state, check the water injection adjustment data in the current well file in the water injection optimization database, determine the adjustment path for wells in an unbalanced state, and if the water injection adjustment time of the current well is within the previous monitoring cycle, mark the well in the current unbalanced state as recently adjusted, and continue monitoring until the end of the next monitoring cycle. If the current well's water injection adjustment time is greater than or equal to the previous monitoring cycle, then the water injection of wells currently in an unbalanced state will be optimized and adjusted to restore the current well state to a relatively balanced state. The specific steps are as follows: ; in, These represent the adjusted new water flow rate and the original water flow rate, respectively. To determine the water injection flow rate adjustment based on historical water injection flow rate and oil pressure data of current well i, the following steps are taken: Historical operating data of current well i, including water injection flow rate and oil pressure data, are collected. After normalization, the collected historical data is divided into training, validation, and test sets (70% training, 15% validation, and 15% test). A multilayer perceptron is selected as the model for construction, with one neuron in the input layer and one neuron in the output layer. The multilayer perceptron is trained using the training set, and parameters are adjusted using the validation and test sets to obtain the predicted water injection flow rate adjustment output model. The water injection flow rate adjustment and adjustment time are recorded in the corresponding well's sub-file in the water injection optimization database, and the label of the current well is recorded as "adjusted". For wells in a relatively balanced state or those without data, the label for the current well is recorded as "continuous monitoring" in the sub-file of the corresponding well in the water incorporation optimization database.

5. The water-incorporation energy-saving optimization method based on oil pressure analysis according to claim 4, characterized in that, S3 includes the following steps: S31. Collect tags for sub-files of wells under different metering room files in the water mixing optimization database, determine the metering room status, and mark metering rooms in a relatively balanced state. S32. For metering rooms in a relatively balanced state, the state of each well in the metering room is optimized and adjusted to the best balanced state by adjusting the water injection volume. The best balanced state of the current metering room is obtained, and the total water injection flow rate of the current metering room is recorded as the theoretical minimum total water injection volume. The metering rooms under the jurisdiction of the current oilfield gathering and transportation station are gradually adjusted to the best balanced state to complete the water injection optimization of the oilfield gathering and transportation station.

6. The water-incorporation energy-saving optimization method based on oil pressure analysis according to claim 5, characterized in that, S31 includes the following steps: S311. Sub-file tags of different metering room files in the statistical water-dosing optimization database are used to determine the status of the metering room: If all sub-files in the current metrology room file are labeled "continuous monitoring", then the current metrology room status is marked as a relatively balanced state. If the sub-files under the current metrology file contain the labels "adjusted" or "recently adjusted", then the current metrology state is marked as unbalanced.

7. The water-incorporation energy-saving optimization method based on oil pressure analysis according to claim 6, characterized in that, S32 includes the following steps: S321. For the metering room marked as being in a relatively balanced state, the state of each well in the metering room is optimized and adjusted by adjusting the water injection rate, specifically including the following steps: By step size Reduce the total water injection flow rate in the metering chamber under the current condition, maintain the new water injection flow rate, and continuously monitor for n cycles. Count the number of wells in the current unbalanced state in the metering chamber after the monitoring cycle. ; S322, when When =0, repeat step S321 to continue adjusting the total water mixing flow rate in the current metering room; S323, When 0 < ≤ If the wells are in an abnormal, unbalanced state, return to step S221 for state optimization. If all wells return to a relatively balanced state after optimization, update the total water mixing flow rate in the metering room with the new optimized total water mixing flow rate. Simultaneously, repeat step S321 to adjust the total water mixing flow rate in the current metering room until... > If, after optimization, a well cannot restore a relatively balanced state, the total water injection flow rate of the previous cycle is taken as the optimal water injection flow rate for that metering interval. To determine the threshold; S324. Mark the state of the metering room under the optimal water mixing flow rate as the optimal equilibrium state. Repeat steps S321 to S323 to optimize all subordinate metering rooms of the oilfield gathering and transportation station to the optimal equilibrium state, and complete the water mixing optimization of the oilfield gathering and transportation station.

Images