Intelligent completion single-core cable layered power distribution and bit error rate joint optimization method

Abstract

The application discloses a layered power distribution and bit error rate joint optimization method for intelligent completion single-core cable, which comprises the following steps: S1, system initialization and channel estimation; S2, real-time signal processing based on a DFE-MLSE joint equalizer; S3, bit error rate optimization based on a soft threshold decision; S4, equalizer coefficient adaptive update based on SPSA; and S5, layered power dynamic distribution optimization; wherein, the steps S1 to S5 are cyclically executed in the system running process, and continuous joint optimization of communication quality and power efficiency is realized. The application forms an adaptive optimization method of a DFE-MLSE joint equalizer + soft threshold decision; direct current power carrier communication adopts a Manchester coding mode, and a gradient projection power distribution algorithm with the minimum weighted bit error rate as the target and the total power and layered power constraints is formed. Through the equalizer and the power distribution, the optimization of communication reliability and power is realized in a strong interference deep well communication environment.

Description

Technical Field

[0001] This invention relates to the field of oil and gas well completion technology, and in particular to a method for joint optimization of layered power allocation and bit error rate of intelligent well completion single-core cables. Background Technology

[0002] Existing oil reservoirs are complex and diverse, with severe interlayer heterogeneity. In order to achieve rational and reliable development of oil fields, intelligent injection and production technology, by configuring a downhole permanent reservoir dynamic monitoring system, can monitor the production and water injection information of each reservoir in real time, which helps staff to adjust production plans in a timely manner and achieve scientific development of oil fields.

[0003] In intelligent stratified injection and production technology, single-core cables are often used for data transmission between wells and the surface. During intelligent well completion and stratified injection and production, the signal from a single-core cable experiences varying degrees of amplitude attenuation and phase shift under conditions of 150℃ high temperature, 60MPa pressure, and cable tension. This manifests as waveform distortion of the analog signal at the receiving end. When the distortion exceeds the fault tolerance range of the decoding circuit, incorrect decoding occurs, leading to an increased bit error rate. Improving the noise immunity of the cable communication system and optimizing power distribution for each layer are crucial for the stability and reliability of oilfield production. Summary of the Invention

[0004] The purpose of this application is to provide a method for joint optimization of layered power allocation and bit error rate of intelligent well completion single-core cables, so as to solve the problems mentioned in the background art.

[0005] The present invention solves its problems through the following technical solution: A joint optimization method for layered power allocation and bit error rate of intelligent well completion single-core cables includes the following steps: S1: System initialization and channel estimation; S2: Real-time signal processing based on DFE-MLSE joint equalizer; S3: Bit error rate optimization based on soft threshold decision; S4: SPSA-based adaptive update of equalizer coefficients; S5: Hierarchical power dynamic allocation optimization; Steps S1 to S5 are executed cyclically during system operation to achieve continuous joint optimization of communication quality and power efficiency.

[0006] Optionally, step S1 includes: After initializing the equalizer coefficients, the actual transmission signal sequence is sent through a single-core cable. Analyze the equalizer structure and output: (1) Where K1 is the order of the initial feedforward filter; K2 is the order of the initial feedback filter; r[n] is the input signal that enters the equalizer feedforward network through a single-core cable; f is the feedforward filter coefficient, f is a K1+1 order vector, and f=[f -K1 ,...,f0] T b represents the feedback filter coefficients, b is a K2-order vector, and b = [b1, ..., b2]. K2 ] T .

[0007] Optionally, step S2 includes: By adjusting the feedforward filter coefficient f and the feedback filter coefficient b, the overall system bit error rate can be minimized. (2) (3) Where N is the preferred training length of the training sequence; This is the actual transmitted signal sequence; Output symbol for MLSE decision; For indicator functions; Let be the cost function.

[0008] Optionally, step S3 includes: S31, MLSE output After soft thresholding, soft decision symbols are generated. Find the optimal feedforward filter coefficients f and feedback filter coefficients b to minimize the overall bit error rate of the soft-decision system. (4) (5) (6) in, x is the input value; T m Let m be the threshold parameter; , Soft T (x) represents the output value; S32. Smooth approximation of soft decision bit error rate estimation and adaptive trigger coefficient update; Bit error rate (BER) estimation is performed using the BER observation window length W. (7) By using the sigmoid function to perform a smooth approximation of the non-differentiable indicator function, a continuously optimizable cost function is constructed: (8) (9) Where β is the disturbance amplitude attenuation coefficient; Here, Δ is the distance metric function; Δ is the decision threshold. Triggering condition for adaptive update of equalizer coefficients: If Keep the coefficient; if Trigger coefficient update.

[0009] Optionally, step S4 includes: S41. Optimize the DFE-MLSE equalizer coefficients using a simultaneous perturbation random approximation algorithm: The cost function is enhanced, and its total cost function is: (10) Assuming the current iteration is the m-th iteration, the feedforward filter coefficients and the feedback filter coefficients are fm and fm respectively. m and b m Perform SPSA gradient estimation using the following steps: Generate random perturbation vectors: (11) (12) Determine the amplitude of the disturbance: ; γ is the initial perturbation amplitude; γ is the bit error rate amplification factor. Calculate the cost function under positive and negative perturbations: (13) (14) Estimate gradient change: (15) (16) in, This means taking the reciprocal of each element. ,have ; Design an adaptive step size for the feedforward filter coefficients f and the feedback filter coefficients b. and Perform adaptive step size parameter updates: (17) (18) Feedforward filter step size: (19) Feedback filter step size: (20) Adaptive update of soft threshold: (twenty one) in, This is the initial value of the feedforward filter step size; This is the initial value for the feedback filter step size; This represents the feedforward filter step size attenuation coefficient. This refers to the step size attenuation coefficient of the feedback filter. This refers to the sensitivity of the feedforward portion; For the sensitivity of the feedback section; The expected bit error rate performance metric for the system; The step size of the feedforward filter is a constant that is divided by zero. The step size of the feedback filter is a constant that prevents division by zero; both are typically taken as values ​​of... ; This is the step size coefficient for adaptive adjustment of the soft threshold, used to control the threshold. Adjustment rate as bit error rate deviation changes; S42. Equalizer stability design based on energy constraints and minimum phase constraints; Apply energy constraints to the feedforward filter to prevent amplifier saturation: If the limit is exceeded, normalize; This is the maximum energy constraint for the feedforward filter; Applying a minimum phase constraint to the feedback filter ensures The root lies inside the unit circle; among them, for Transform the complex variable, For feedback filters Domain transfer function.

[0010] Optionally, the specific model for hierarchical power dynamic allocation optimization in step S5 is as follows: (twenty two)

[0011] in, For layer index; For the first Layer-allocated transmit power; For the first The priority weight of the layer satisfies ; The complementary error function is defined as follows: ; For the first Layer channel system integrated parameters, ,in For the first Layer channel power gain, To transmit the amplitude scaling factor, Information transmission rate; For the first Layer Manchester coding inter-symbol interference factor, satisfying ; This represents the total usable power of a single-core cable. The first Minimum and maximum permissible power of the layer; For the first Layer in power The bit error rate is below; For the first The layer requires a high bit error rate.

[0012] In summary, the technical effects and advantages of this invention are as follows: This invention establishes a joint optimization method for layered power allocation and bit error rate of intelligent well completion single-core cables; to mitigate inter-symbol interference in downhole 7-layer injection-production data transmission, an adaptive optimization method combining a DFE-MLSE joint equalizer and soft threshold decision is developed; DC power line carrier communication employs Manchester coding, forming a gradient projection power allocation algorithm that integrates total power and layered power constraints with the goal of minimizing weighted bit error rate. Through equalizer and power allocation, communication reliability and power are optimized in a highly interfering deep well communication environment. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. 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 method for joint optimization of layered power allocation and bit error rate of intelligent well completion single-core cable according to an embodiment of the present invention; Figure 2 This is a diagram of a DFE-MLSE equalizer combined with a soft threshold decision structure provided in an embodiment of the present invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0016] The following is an overview of the subject matter described in detail herein. This overview is not intended to limit the scope of the claims.

[0017] To address the technical challenges of high bit error rate and uneven power distribution in single-core cable data transmission within intelligent well completion systems, this invention proposes a joint solution integrating equalization and power optimization. The method includes: establishing a joint equalizer structure based on feedforward filters, feedback filters, and maximum likelihood sequence estimation, and introducing a soft threshold decision mechanism to reduce error propagation; using the soft decision bit error rate as the optimization objective, adaptively updating the filter coefficients using a simultaneous perturbation stochastic approximation algorithm, and designing independent dual-step update strategies for the feedforward and feedback filters to ensure stability; simultaneously, establishing an optimization model that minimizes the bit error rate and integrates total power and layered power constraints, dynamically allocating the transmission power of each layer through gradient projection. The system also includes energy constraint and minimum phase constraint processing for the filter coefficients, as well as collaborative adaptive adjustment of the soft decision threshold. This invention improves the reliability of data transmission for real-time monitoring of layered production profiles and precise flow control in intelligent well completion communication by reducing the bit error rate and optimizing power distribution, thus supporting production optimization.

[0018] This embodiment proposes a joint optimization method for layered power allocation and bit error rate of intelligent well completion single-core cables, including the following steps: 1. For example Figure 1 As shown in the overall scheme, after initializing the equalizer coefficients according to the following steps, the actual transmission signal sequence is sent through a single-core cable. Analyze the equalizer structure and output: (1-1) Algorithm structure parameter initialization: The preferred training length N; For a typical cable channel in an intelligent well completion system, initialize the order K1 of the feedforward filter; Initialize the order K2 of the feedback filter to determine the number of states of the maximum likelihood sequence estimation (MLSE) detector; Viterbi backtracking depth L covers the effective memory length of the channel; The bit error rate observation window length W or finite-length sliding window W is the same concept as the bit error rate observation window W, both referring to the window used for estimating bit error rate. The sliding window length is adjusted to prevent excessively large windows from causing large estimation errors and excessively long windows from reducing the algorithm's tracking speed for channel changes; (1-2) Adaptive update trigger parameter initialization The bit error rate (BER) is updated to a threshold T. When the BER estimated by the sliding window exceeds T, the equalizer coefficients are adaptively updated. The expected bit error rate performance index of the system ; (1-3) Parameters of the Simultaneous Perturbation Stochastic Approximation (SPSA) algorithm: Initial step size: Initial value of the feedforward filter step size Initial value of feedback filter step size ; Step size attenuation coefficient: Step size attenuation coefficient of feedforward filter Feedback filter step size attenuation coefficient ; BER sensitivity coefficient: feedforward sensitivity ; Feedback section sensitivity ; Initial disturbance amplitude , used for SPSA gradient estimation of random perturbations; Disturbance amplitude attenuation coefficient: ; BER magnification factor: To enhance the algorithm's ability to escape local optima, the perturbation amplitude can be appropriately increased. (1-4) System constraints and regularization parameters: Maximum energy constraint of feedforward filter: To prevent amplifier overload and numerical instability; Regularization coefficient: ,in These are the regularization coefficients for the feedforward filter, the feedback filter, and the soft threshold regularization coefficient, respectively.

[0019] 2. A joint equalizer based on the equivalent channel model, decision feedback equalizer-maximum likelihood sequence estimation (DFE-MLSE), and Viterbi sequence detection, by adjusting the feedforward filter coefficients. and feedback filter coefficients To minimize the overall system bit error rate (BER):

[0020] : Feedforward filter FFF coefficients ( (order) vector to suppress forward inter-symbol interference; : Feedback filter FBF coefficients ( (Order) vector, using the decided symbols to eliminate backward inter-symbol interference; r: The signal passing through the single-core cable enters the equalizer feedforward network as the input signal; The actual transmitted signal sequence; : MLSE decision output symbol; It is an indicator function: ; It is the cost function.

[0021] 3. Optimization of MLSE detection bit error rate based on soft threshold decision, MLSE output After soft thresholding, soft decision symbols are generated. That is, Find the optimal feedforward filter coefficients f and feedback filter coefficients. To minimize the overall bit error rate (BER) of the soft-decision system:

[0022] in: Soft threshold function: : Input value; Threshold parameter for the mth iteration; : Output value.

[0023] 4. Smooth approximation soft decision bit error rate estimation and adaptive trigger coefficient update; (4-1) Using a finite-length sliding window That is, statistics of the most recent Perform bit error rate estimation:

[0024] (4-2) By using the Sigmoid function to make a smooth approximation of the non-differentiable indicator function, a continuously optimizable cost function is constructed:

[0025] in, It is a distance metric function; (4-3) The system uses a distance-based soft-decision error assessment to evaluate real-time communication quality, i.e., quantizing decision symbols. With test sequence Differences between them: ,in This is the threshold for the judgment.

[0026] (4-4) When the estimated bit error rate exceeds the dynamic threshold, the equalizer coefficients are automatically updated. The triggering condition is: if Keep the coefficient; if Trigger coefficient update.

[0027] 5. Optimize the DFE-MLSE equalizer coefficients using the Simultaneous Perturbation Stochastic Approximation (SPSA) algorithm, following these steps: (5-1) The cost function is enhanced to improve the stability of the optimization process and the performance of the final solution. The total cost function is: (5-2) Assume the current is the th In the next iteration, the filter coefficients are... and Perform SPSA gradient estimation using the following steps: Generate random perturbation vectors:

[0028]

[0029] Each perturbation element and The elements are given equal probability of +1 or -1, representing feedforward and feedback perturbations respectively, and each element independently follows a symmetric Bernoulli distribution. Determine the amplitude of the disturbance ; Calculate the cost function under positive and negative perturbations:

[0030]

[0031] Estimate gradient change:

[0032]

[0033] in, This means taking the reciprocal of each element. ,have .

[0034] The filter coefficients are updated based on the gradient estimation, becoming the feedforward filter coefficients. and feedback filter coefficients Design adaptive step size and Perform adaptive step size parameter updates:

[0035]

[0036] Feedforward filter step size: Feedback filter step size: (5-3) Adaptive update of soft threshold:

[0037] in, This is the initial value of the feedforward filter step size; This is the initial value for the feedback filter step size; This represents the feedforward filter step size attenuation coefficient. This refers to the step size attenuation coefficient of the feedback filter. This refers to the sensitivity of the feedforward portion; For the sensitivity of the feedback section; The expected bit error rate performance metric for the system; The step size of the feedforward filter is a constant that is divided by zero. The step size of the feedback filter is a constant that prevents division by zero; both are typically taken as values ​​of... ; This is the step size coefficient for adaptive adjustment of the soft threshold, used to control the threshold. Adjustment rate as bit error rate deviation changes.

[0038] 6. Equalizer stability design based on energy constraints and minimum phase constraints (6-1) Apply energy constraints to the feedforward filter to prevent amplifier saturation: If the limit is exceeded, normalization is performed.

[0039] (6-2) Apply a minimum phase constraint to the feedback filter to ensure The root lies within the unit circle, ensuring system stability; where, for Transform the complex variable, For feedback filters Domain transfer function; normalization is used to ensure that the power of the feedforward filter is within a reasonable range, and the feedback filter meets the minimum phase condition through root position testing and correction, thereby reducing the risk of divergence and oscillation.

[0040] 7. Layer power optimization allocation for iterative intelligent well completion layered injection and production based on gradient descent and feasible region projection: The DC power carrier communication for intelligent well completion layered injection and production uses Manchester encoding for data transmission. With the goal of minimizing the weighted bit error rate, it integrates the comprehensive constraints of total power, single-layer power range and bit error rate to optimize the allocation of power for 7 layers. The power adjustment direction is calculated by gradient descent method to maximize power efficiency and optimize bit error rate reliability.

[0041] (7-1) Initialize the iteration counter ; (7-2) Initialize based on average power distribution (7-3) Iteration steps (for the ) (next iteration): (7-3-1) Calculate the objective function with respect to each In the present gradient at:

[0042] in, ;in It is the bit rate in bps; : No. Layer channel power gain; : No. Layer sends amplitude scaling factor; : No. Manchester ISI factor; (7-3-2) Differentiate it: in, (7-3-3) Gradient direction using weighted normalization: in It is the learning rate step size of the algorithm; (7-3-4) Calculate candidate power points: (7-3-5) Project it onto the feasible region superior, feasible region Power vector satisfying inequality constraints ; (7-3-6) The projection operation first projects each ;Detect total power constraint If the conditions are not met, then all will be reduced proportionally. Until the total power constraint is met, finally check the BER constraint: ; The specific model for hierarchical power dynamic allocation optimization is as follows: in, For layer index; For the first Layer-allocated transmit power; For the first The priority weight of the layer satisfies ; The complementary error function is defined as follows: ; For the first Layer channel system integrated parameters, ,in For the first Layer channel power gain, To transmit the amplitude scaling factor, Information transmission rate; For the first Layer Manchester coding inter-symbol interference factor, satisfying ; This represents the total usable power of a single-core cable. The first Minimum and maximum permissible power of the layer; For the first Layer in power The bit error rate is below; For the first The layer requires a high bit error rate.

[0043] (7-3-7) Calculate the change in the objective function value or the norm of the power vector between two consecutive iterations. If it is less than the preset threshold, the algorithm is considered to have converged and the iteration is stopped; otherwise, continue iterating from (7-1).

[0044] Finally, it should be noted that: the accompanying drawings of the embodiments disclosed in this invention only involve structures related to the embodiments disclosed in this invention, and other structures can refer to general designs. In the absence of conflict, the same embodiment and different embodiments of this invention can be combined with each other.

[0045] In conclusion, the above description is only a preferred embodiment of the present invention and is 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 method for joint optimization of layered power allocation and bit error rate in intelligent well completion single-core cables, characterized in that, Includes the following steps: S1: System initialization and channel estimation; S2: Real-time signal processing based on DFE-MLSE joint equalizer; S3: Bit error rate optimization based on soft threshold decision; S4: SPSA-based adaptive update of equalizer coefficients; S5: Hierarchical power dynamic allocation optimization; Steps S1 to S5 are executed cyclically during system operation to achieve continuous joint optimization of communication quality and power efficiency.

2. The method for joint optimization of layered power allocation and bit error rate of intelligent well completion single-core cable according to claim 1, characterized in that, Step S1 includes: After initializing the equalizer coefficients, the actual transmission signal sequence is sent through a single-core cable. Analyze the equalizer structure and output: （1） Where K1 is the order of the initial feedforward filter; K2 is the order of the initial feedback filter; r[n] is the input signal that enters the equalizer feedforward network through a single-core cable; f is the feedforward filter coefficient, f is a K1+1 order vector, and f=[f -K1 ,...,f0] T b represents the feedback filter coefficients, b is a K2-order vector, and b = [b1, ..., b2]. K2 ] T .

3. The method for joint optimization of layered power allocation and bit error rate of intelligent well completion single-core cable according to claim 1, characterized in that, Step S2 includes: By adjusting the feedforward filter coefficient f and the feedback filter coefficient b, the overall system bit error rate can be minimized. （2） （3） Where N is the preferred training length of the training sequence; This is the actual transmitted signal sequence; Output symbol for MLSE decision; For indicator functions; Let be the cost function.

4. The method for joint optimization of layered power allocation and bit error rate of intelligent well completion single-core cable according to claim 1, characterized in that, Step S3 includes: S31, MLSE output After soft thresholding, soft decision symbols are generated. Find the optimal feedforward filter coefficients f and feedback filter coefficients b to minimize the overall bit error rate of the soft-decision system. （4） （5） （6） in, x is the input value; T m Let m be the threshold parameter; , Soft T (x) represents the output value; S32. Smooth approximation of soft decision bit error rate estimation and adaptive trigger coefficient update; Bit error rate (BER) estimation is performed using the BER observation window length W. （7） By using the sigmoid function to perform a smooth approximation of the non-differentiable indicator function, a continuously optimizable cost function is constructed: （8） （9） Where β is the disturbance amplitude attenuation coefficient; Here, Δ is the distance metric function; Δ is the decision threshold. Triggering condition for adaptive update of equalizer coefficients: If Keep the coefficient; if Trigger coefficient update.

5. The method for joint optimization of layered power allocation and bit error rate of intelligent well completion single-core cable according to claim 1, characterized in that, Step S4 includes: S41. Optimize the DFE-MLSE equalizer coefficients using a simultaneous perturbation random approximation algorithm: The cost function is enhanced, and its total cost function is: （10） Assuming the current iteration is the m-th iteration, the feedforward filter coefficients and the feedback filter coefficients are fm and fm respectively. m and b m Perform SPSA gradient estimation using the following steps: Generate random perturbation vectors: （11） （12） Determine the amplitude of the disturbance: ; γ is the initial perturbation amplitude; γ is the bit error rate amplification factor. Calculate the cost function under positive and negative perturbations: （13） （14） Estimate gradient change: （15） （16） in, This means taking the reciprocal of each element. ,have ; Design an adaptive step size for the feedforward filter coefficients f and the feedback filter coefficients b. and Perform adaptive step size parameter updates: （17） （18） Feedforward filter step size: (19) Feedback filter step size: (20) Adaptive update of soft threshold: （21） in, This is the initial value of the feedforward filter step size; This is the initial value for the feedback filter step size; This represents the feedforward filter step size attenuation coefficient. This refers to the attenuation coefficient of the feedback filter step size. This refers to the sensitivity of the feedforward portion; For the sensitivity of the feedback section; The expected bit error rate performance metric for the system; The step size of the feedforward filter is a constant that is divided into zero. The step size of the feedback filter is a constant that prevents division by zero; both are typically taken as values ​​of... ; This is the step size coefficient for adaptive adjustment of the soft threshold, used to control the threshold. Adjustment rate as bit error rate deviation changes; S42. Equalizer stability design based on energy constraints and minimum phase constraints; Apply energy constraints to the feedforward filter to prevent amplifier saturation: If the limit is exceeded, normalize; This is the maximum energy constraint for the feedforward filter; Applying a minimum phase constraint to the feedback filter ensures The root lies inside the unit circle; in, for Transform the complex variable, For feedback filters Domain transfer function.

6. The method for joint optimization of layered power allocation and bit error rate of intelligent well completion single-core cable according to claim 1, characterized in that, The specific model for hierarchical power dynamic allocation optimization in step S5 is as follows: （22） in, For layer index; For the first Layer-allocated transmit power; For the first The priority weight of the layer satisfies ; The complementary error function is defined as follows: ; For the first Layer channel system integrated parameters, ,in For the first Layer channel power gain, To transmit the amplitude scaling factor, Information transmission rate; For the first Layer Manchester coding inter-symbol interference factor, satisfying ; This represents the total usable power of a single-core cable. The first Minimum and maximum permissible power of the layer; For the first Layer in power The bit error rate is below; For the first The layer requires a high bit error rate.

Images