An intelligent optimization method and system for oilfield injection-production based on a biological regulation mechanism of artificial bee colony

Abstract

The present invention provides an artificial bee colony intelligent optimization method and system based on a biological regulation mechanism; inspired by the mechanism of regulating physiological indexes in biological systems, on the basis of the leading bees, observing bees, and scout bees in the traditional artificial bee colony algorithm, the optimization method designs a blue light guiding unit, a nectar source regulation unit, and an antenna orientation unit according to the biological regulation law. Among them, the blue light guiding unit acts in the observing bee stage and can enhance the ability of the algorithm to jump out of local optima; the antenna orientation unit acts in the scout bee stage and can, to a certain extent, accelerate the convergence speed of the algorithm; the nectar source regulation unit acts in the leading bee iteration stage and can reduce the execution time of the algorithm. Simulation experiments show that the algorithm of the present invention has higher optimization performance compared with other traditional algorithms, especially in improving the global optimization performance and the algorithm convergence speed, and has improved the operation effect.

Description

Technical Field

[0001] The present invention relates to a bio-intelligent optimization algorithm, in particular to an intelligent optimization method and system for oilfield injection-production based on an artificial bee colony with a biological regulation mechanism. Background Art

[0002] As a newly emerging intelligent optimization algorithm, the artificial bee colony algorithm is a meta-heuristic algorithm based on the bee behavior mechanism proposed by the Dervis Karaboga group in 2005 for optimizing algebraic problems. It can solve multi-variable global optimization problems by simulating the role conversion and division of labor among bees. It is a relatively simple-structured, robust and population-based stochastic optimization algorithm, so it has greater mining potential compared with other algorithms and has certain advantages in solving such problems.

[0003] The artificial bee colony algorithm is usually used to solve problems such as function optimization and scheme optimization. According to the bee population structure in nature, three roles are set: leading bees (employed bees), observing bees (following bees) and scout bees. There are two main behaviors: finding a nectar source and abandoning a nectar source. In the problem to be solved, the nectar source represents the feasible solution of the optimization problem, and its quantity usually corresponds to the number of leading bees. First, the leading bees with memory function will search for food sources in space. After returning to the hive, they will dance the waggle dance in the recruitment area and transmit the nectar source information to the observing bees with a certain probability. The observing bees are the bees waiting near the hive for the leading bees to share the nectar source. They observe the dance of the leading bees and choose the bees they think are satisfactory to follow. The observing bees can accept the information provided by the leading bees and follow the leading bees to collect nectar, or they can not accept the information and turn into new leading bees to search for nectar sources. The best nectar source position is recorded during this process. The leading bees continue to search for new nectar sources near the hive. If a nectar source has not been updated after reaching the specified number of iterations, the leading bees will turn into scout bees, abandon this nectar source and go to search for new nectar sources, starting a new round of iteration until the optimal nectar source is output after meeting the end condition.

[0004] The biggest feature of the artificial bee colony algorithm is the role division of labor, and role conversion can be carried out among various types of bees. Its algorithm steps can be roughly divided into initialization, leading bees recording nectar source information, following bees collecting nectar, and generating scout bees. The traditional artificial bee colony algorithm has the advantages of few control parameters, strong robustness, wide application range, etc., but also has disadvantages such as being prone to premature convergence. There are also relatively large errors when used to solve non-linear multi-parameter problems. At the same time, some improved artificial bee colony algorithms still stay in the stage of being prone to falling into local optima, and the optimization effect needs to be further improved.

[0005] In order to improve the optimization performance of artificial bee colony algorithm, many scholars have carried out a lot of research work, mainly including cross-improvement and self-improvement. Among them, cross-improvement is to make innovative combinations of artificial bee colony algorithm and some other traditional intelligent optimization algorithms to improve the algorithm performance; self-improvement is mainly to improve one or some links in the basic principles of artificial bee colony algorithm in a targeted manner to improve the optimization ability of the algorithm itself.

[0006] In terms of cross-improvement research, in 2017, Liu Gang et al. designed a hybrid discrete artificial bee colony algorithm, and reconnected it with the traditional heuristic algorithm and the greedy random adaptive search algorithm path, successfully reducing the total completion time of the flow workshop scheduling, but the algorithm has limited scope of application. In 2018, Gao Yuxi et al. combined the characteristics of the artificial bee colony algorithm and the particle swarm optimization algorithm to propose a PSO-ABC intelligent hybrid algorithm. The hybrid algorithm has been significantly improved in terms of optimization accuracy and convergence speed, but its performance is poor when dealing with complex multi-peak search algorithms. In 2021, Jiang Wei et al. added the mutation and crossover operators of the differential evolution algorithm to the follow-up bee stage to optimize the home energy management system model. The improved algorithm is easier to jump out of the local optimum, but at the same time, the algorithm operation complexity is greatly increased. In terms of self-improvement research, in 2018, Li Hongjuan et al. proposed an enhanced ABC optimization algorithm to estimate the torque of the asynchronous motor by adjusting the allocation strategy of the new honey source. The improved algorithm can quickly estimate the motor parameters, but the optimization effect is quite different when applied to different asynchronous motors. In 2021, Guan Xuemei and others effectively balanced the local and global search capabilities of the algorithm by adjusting the search radius of bees in the artificial bee colony algorithm, so that the RBF neural network parameters were optimized to a certain extent, but the optimization accuracy still needs to be improved.

[0007] In summary, although a lot of in-depth research has been carried out on the improvement schemes of the traditional artificial bee colony algorithm, most of the research focuses on the cross-improvement algorithm, and some of the improved algorithms are difficult to implement, the running results are unstable, and the algorithm accuracy still needs to be improved. Summary of the invention

[0008] In order to solve the above problems, the present invention proposes an intelligent optimization method for oilfield injection and production based on artificial bee colonies (NEI-ABC) based on the biological regulation mechanism of the human nervous-endocrine-immune system. The intelligent optimization method for oilfield injection and production based on artificial bee colonies based on the biological regulation mechanism is to add a blue light guidance unit, a nectar source control unit and an antenna orientation unit to the traditional artificial bee colony algorithm. The synergistic effect of the three units improves the algorithm in terms of optimization accuracy, convergence speed and ability to jump out of local optimality. The simulation results of the comparison with other artificial bee colony optimization algorithms show that the intelligent optimization method for oilfield injection and production based on artificial bee colonies based on the biological regulation mechanism has better optimization performance than other algorithms.

[0009] An intelligent optimization method for oilfield injection-production based on a bioregulation mechanism artificial bee colony, comprising the following steps:

[0010] Set up a blue light guiding unit for guiding bees to search for nectar sources;

[0011] Set up a nectar source regulation unit for regulating the iteration times of scout bees;

[0012] Set up an antenna orientation unit for scout bees to search for the global optimal position as a new nectar source;

[0013] The optimization method further comprises the following steps:

[0014] S1: Initialize the nectar source and calculate the fitness value of the nectar source;

[0015] S2: The leading bees determine the initial marked nectar source through the blue light guiding unit and search for a new nectar source;

[0016] S3: Obtain the fitness value of the new nectar source, and adopt a greedy selection strategy to retain the nectar source with a better fitness value;

[0017] S4: Calculate the selection probability P of the follower bees to select the leading bees i , and the follower bees follow the nectar source provided by the leading bees according to the selection probability P i ; for the follower bees that do not select to follow the nectar source, they become leading bees and return to step S2;

[0018] S5: Record the current optimal nectar source, add a parameter regulation factor to the nectar source regulation unit to act on the scout bees and define the iteration times n, start iterative calculation, and end until the iteration times n are satisfied, and output the optimal numerical combination; when no optimal solution is found after n iterations of searching reach the threshold, the corresponding leading bees are transformed into scout bees, and the scout bees randomly generate a new nectar source in the search space by using the antenna orientation unit, and return to S3.

[0019] Further, the calculation of the fitness value of the nectar source in S1 specifically includes:

[0020] Let the dimension of the solution space of the problem be D, the number of nectar sources be N, and the iteration times be n; at the nth iteration, the position of the nectar source i can be expressed as where, x id ∈[L d U d ), L d and U d are respectively the upper and lower limits of the solution space, the quality of the nectar source i corresponds to the fitness value fit i of the solution, and the fitness value fit of the nectar source iThe calculation formula is as follows:

[0021]

[0022] where n, N, and D are non-zero natural numbers, i = 1, 2,......, N, d = 1, 2,......, D, and f i is the function value obtained by the objective function at x i .

[0023] Generally, the higher the quality of the nectar source in the artificial bee colony algorithm, the better, that is, the larger the fitness value, the better. For the problem to be optimized, two cases need to be considered: the minimum value problem and the maximum value problem. Therefore, the criteria need to be set according to the specific problem and function.

[0024] Furthermore, in S2, the scout bee determines the initial marked nectar source and searches for new nectar sources through the blue light guiding unit, which specifically includes: The scout bee searches for new nectar sources in the solution space. The formula for generating new nectar sources during the search process is as follows:

[0025]

[0026] In the formula, G b is the guiding function, S represents the strength of adjusting the guiding effect, μ is a fixed constant, and S > 0, μ > 0. k1 and k2 are parameters of the blue light guiding system, k1 > 0, k2 < 0; rand is a function that generates random numbers, and the generated random numbers are uniformly distributed random real numbers greater than or equal to the lower limit value and less than the upper limit value, and e is the natural constant.

[0027] The reason for proposing the blue light guiding unit is that in the traditional artificial bee colony algorithm, the observer bee searches for new nectar sources near the nectar source according to the formula:

[0028] x i = L d + rand(U d - L d );

[0029] According to this search formula, duplicate solutions are likely to be generated, which makes the global optimization ability of the traditional artificial bee colony algorithm weak during the optimization process.

[0030] By introducing the blue light guiding unit into the traditional artificial bee colony algorithm, it can be found through comparison that the iteration number and other parameters are added to the blue light guiding unit. In particular, the generation position of the new nectar source is associated with the iteration number n, rather than being randomly generated in the traditional artificial bee colony algorithm. To a certain extent, it avoids the new nectar source being in the same position as the original nectar source and improves the global search ability of the algorithm.

[0031] Furthermore, in S3, the greedy selection strategy is adopted to retain the nectar source with a better fitness value, which specifically includes:

[0032]

[0033] Among them, j ∈ [1, 2K N] and j ≠ i, is a random number uniformly distributed in [-1, 1];

[0034] The greedy selection strategy means that when the fitness value of the new nectar source x' i is better than the fitness value of the nectar source x i , the new nectar source x' i is used to replace the nectar source x i , otherwise the nectar source x i is retained.

[0035] In S4, calculating the selection probability P of the follower bee to select the leader bee i , the formula is:

[0036]

[0037] Generating a scout bee means that when the leader bee fails to improve the quality of the nectar source x i after trying to search for a new nectar source F times, the leader bee transforms into a scout bee, and the corresponding nectar source x i will be abandoned; where F is the abandonment threshold, which is a preset value.

[0038] Furthermore, in S5, a parameter regulation factor is added to the nectar source regulation unit to act on the scout bee and define the number of iterations n. The specific formula is:

[0039]

[0040] In the formula, R is the concentration change of the regulatory hormone, F is the threshold, is a fixed constant, a1 and a2 are system parameters of the nectar source regulation unit, and a1 > 0, a2 < 0.

[0041] The reason for proposing the nectar source regulation unit is that in the traditional artificial bee colony algorithm, the iteration formula for the nectar source is

[0042]

[0043] According to this formula, the traditional artificial bee colony algorithm has the same number of iterations for any nectar source, increasing the search time. To overcome the above defects, based on the endocrine regulation mechanism, a nectar source regulation unit is added to act on the leader bee, and the iteration time is dynamically adjusted so that the iteration time decreases with the increase of the number of iterations, effectively reducing the time required for algorithm execution.

[0044] Furthermore, in S5, the scout bees use the antenna orientation unit to randomly generate a new nectar source within the search space, which specifically includes: defining the search point of the scout bees to the current global optimal position, and the formula for the antenna orientation unit to generate a new nectar source is:

[0045]

[0046] where g best is the current global optimal nectar source.

[0047] In the traditional artificial bee colony algorithm, when the scout bees generate a new nectar source, they directly use the formula:

[0048] x i = L d + rand(U d - L d ) ;

[0049] to randomly generate a new nectar source within the space. This method has great uncertainty, increases the time complexity of the algorithm operation, and is not conducive to quickly converging to the optimal value. To overcome the above problems, a positioning factor is introduced into the antenna orientation unit based on the immune orientation mechanism, and the search point of the scout bees is defined to the current global optimal position, that is, the generation of the new nectar source is near the current global optimal nectar source, which can increase the local search probability of the scout bees near the optimal point and can accelerate the convergence speed of the algorithm to a certain extent.

[0050] The intelligent optimization method for oilfield injection-production based on the artificial bee colony with biological regulation mechanism is improved in terms of optimization accuracy, convergence speed, and the ability to jump out of local optima. The comparative simulation results on multiple functions with the traditional artificial bee colony algorithm (ABC), the artificial bee colony optimization algorithm (MN-ABC), and the artificial bee colony and particle swarm optimization algorithm (PSO-ABC) show that the intelligent optimization method for oilfield injection-production based on the artificial bee colony with biological regulation mechanism has better optimization performance than other algorithms.

[0051] Preferably, the present invention also provides a system based on the intelligent optimization method for oilfield injection-production based on the artificial bee colony with biological regulation mechanism described above, which is characterized by at least including:

[0052] A blue light guiding unit for guiding the bees to search for nectar sources;

[0053] A nectar source regulation unit for regulating the iteration times of the scout bees;

[0054] An antenna orientation unit for the scout bees to search for the global optimal position as a new nectar source;

[0055] The model optimization module optimizes the artificial bee colony algorithm model by using the blue light guiding unit, the nectar source regulation unit and the antenna orientation unit respectively, and constructs an intelligent optimization model for oilfield injection-production based on the artificial bee colony with a biological regulation mechanism;

[0056] The data acquisition module is used to adopt the sample data to be measured;

[0057] The data prediction module inputs the sample data to be measured into the intelligent optimization model for oilfield injection-production of the artificial bee colony for calculation to obtain an optimization result.

[0058] Meanwhile, the intelligent optimization system for oilfield injection-production based on the artificial bee colony with a biological regulation mechanism further includes:

[0059] A computer-readable storage medium, on which a computer program is stored;

[0060] A non-volatile semiconductor storage element is used to read the data information collected by the data acquisition module;

[0061] The data processing unit calls the computer program through a processing circuit to execute and implement the steps of the intelligent optimization method for oilfield injection-production based on the artificial bee colony with a biological regulation mechanism;

[0062] The data distribution circuit distributes the data to be predicted read from the non-volatile semiconductor storage element to each processing unit, and after being processed by the data processing unit, sends the obtained prediction result to one or more CAN buses to be sent to an external device through a gateway.

[0063] To sum up, in order to solve the problems existing in the traditional artificial bee colony algorithm, such as slow convergence speed and easy to fall into local optimum, inspired by the biological system's regulation mechanism of physiological indexes, the present invention proposes an intelligent optimization method for oilfield injection-production based on the artificial bee colony with a biological regulation mechanism (NEI-ABC); on the basis of the original algorithm's leading bees, observing bees and scouting bees, according to the biological regulation law, a blue light guiding unit, a nectar source regulation unit and an antenna orientation unit are designed. Using the above design idea to improve the honey collection mechanism of observing bees, scouting bees and leading bees, it is closer to the actual honey collection process of the bee colony, can improve the global optimization performance, effectively improve the algorithm convergence speed and improve the operation effect.

[0064] The beneficial effects of the present invention are as follows. The blue light guiding unit is designed based on the action mechanism of the nerve guiding mechanism, and is proposed by simulating the guiding effect of "blue light" on bees when they collect honey. It acts on the observation bee stage and can enhance the ability of the algorithm to jump out of local optima. The antenna orientation unit is designed based on the action mechanism of the immune orientation mechanism, which is derived from the actual function of the antenna in the physiological structure of bees. It acts on the scout bee stage and can, to a certain extent, accelerate the convergence speed of the algorithm. The nectar source regulation unit is proposed according to the endocrine regulation mechanism, combined with the biological background of the change in honey collection efficiency during the working process of bees. It acts on the leading bee iteration stage and can reduce the execution time of the algorithm. Moreover, it has played an active role in solving complex optimization problems such as oilfield injection-production output planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to the provided drawings.

[0066] Figure 1 is a schematic structural diagram of an intelligent optimization method for oilfield injection-production based on a bio-regulation mechanism artificial bee colony according to the present invention;

[0067] Figure 2 is a schematic step diagram of an intelligent optimization method for oilfield injection-production based on a bio-regulation mechanism artificial bee colony according to the present invention;

[0068] Figure 3 is a comparison diagram of the function operation effects of four artificial bee colony algorithms in the specific implementation case of the present invention;

[0069] Figure 4 is a comparison diagram of the oilfield production optimization effects of four artificial bee colony algorithms in the specific implementation case of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0070] The following further describes the specific embodiments of the present invention with reference to the drawings. It should be noted here that the description of these embodiments is for helping to understand the present invention, but does not constitute a limitation to the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0071] As Figure 1 and Figure 2 shown, an intelligent optimization method for oilfield injection-production based on a bio-regulation mechanism artificial bee colony includes the following steps:

[0072] Set up a blue light guiding unit to guide the bees to search for nectar sources;

[0073] Set up a nectar source regulation unit to regulate the number of iterations of scout bees;

[0074] Set up an antenna orientation unit for scout bees to search for the globally optimal position as a new nectar source;

[0075] S1: Randomly initialize the nectar source and calculate the fitness value;

[0076] The S1 calculates the fitness value of the nectar source, specifically including:

[0077] Let the dimension of the solution space of the problem be D, the number of nectar sources be N, and the number of iterations be n; at the nth iteration, the position of nectar source i can be expressed as where, x id ∈[L d U d ), L d and U d are respectively the upper and lower limits of the solution space, the quality of nectar source i corresponds to the fitness value fit i of the solution, and the calculation formula for the fitness value fit i of the nectar source is:

[0078]

[0079] where n, N, and D are non-zero natural numbers, i = 1, 2,......, N, d = 1, 2,......, D, f i is the function value obtained by the objective function on x i .

[0080] S2: The leading bees determine the initial marked nectar source and search for new nectar sources through the blue light guiding unit;

[0081] In the S2, the leading bees determine the initial marked nectar source and search for new nectar sources through the blue light guiding unit, specifically including: The leading bees search for new nectar sources in the solution space, and the formula for generating new nectar sources during the search process is:

[0082]

[0083] In the formula, G b is the guiding function, S represents adjusting the strength of the guiding effect, μ is a fixed constant, and S > 0, μ > 0, k1, k2 are blue light guiding system parameters, k1 > 0, k2 < 0; rand is a function that generates random numbers, and the generated random numbers are uniformly distributed random real numbers greater than or equal to the lower limit value and less than the upper limit value, and e is the natural constant.

[0084] The blue light guiding unit refers to the reference neural guiding mechanism: when a neuron is stimulated, an action potential will move along the neuron and reach the axon terminal, resulting in the depolarization of the neuron. The membrane potential changes, causing voltage-gated ion channels to open, allowing calcium ions to enter, preparing for the release of neurotransmitters. Calcium ions activate an enzyme that separates the vesicles from the synaptic proteins, releasing neurotransmitters into the synaptic cleft. The neurotransmitters in the synaptic cleft bind to the postsynaptic membrane, opening voltage sodium channels, causing excitation or inhibition of the postsynaptic membrane, and thus guiding the cell to make corresponding actions. The nervous system guides the change of the membrane potential of the next neuron's dendritic membrane or cell body membrane by releasing neurotransmitters to act on the receptors, thereby guiding the change of the state of the cell where it is located and realizing the function of information transmission, which is fast and sensitive; its function is determined by the neurotransmitter secretion principle, as shown in the formula:

[0085]

[0086] In the formula, G is the change in the concentration of neurotransmitters, Q is the external stimulus amount, reflecting the strength of the reaction regulation effect, λ is a fixed constant, l1, l2 are system parameters, t is time, and the concentration of neurotransmitters in the cell or tissue fluid is relatively stable before t = 0. When stimulated at t = 0, it shows the change law represented by formula (5).

[0087] The reason for proposing the blue light guiding unit is that in the traditional artificial bee colony algorithm, the formula for the observing bee to search for a new nectar source near the nectar source is:

[0088] x i =L d +rand(U d -L d );

[0089] According to this search formula, duplicate solutions are easily generated, which makes the global optimization ability of the traditional artificial bee colony algorithm weak during the optimization process. By introducing the blue light guiding unit into the traditional artificial bee colony algorithm, it can be found through comparison that the blue light guiding unit adds the number of iterations and other parameters, which to a certain extent avoids the new nectar source and the original nectar source being in the same position and improves the global search ability.

[0090] S3: Obtain the fitness value of the new nectar source, and use the greedy selection strategy to retain the nectar source with a better fitness value;

[0091] The specific steps of using the greedy selection strategy to retain the nectar source with a better fitness value in S3 are as follows:

[0092]

[0093] where j ∈ [1, 2K N] and j ≠ i, is a random number uniformly distributed in [-1, 1];

[0094] The greedy selection strategy means that when the fitness value of the new nectar source x' i is better than that of the nectar source x i , the new nectar source x' i is used to replace the nectar source x i , otherwise the nectar source x i is retained.

[0095] Generally, the higher the quality of the nectar source in the artificial bee colony algorithm, the better, that is, the larger the fitness value, the better. For the problem to be optimized, two cases need to be considered: the minimum value problem and the maximum value problem. Therefore, the standard needs to be set according to the specific problem and function.

[0096] S4: Calculate the selection probability P of the follower bees to select the leading bees i . The follower bees follow the nectar sources provided by the leading bees according to the selection probability P i . For the follower bees that do not select to follow the nectar source, they become leading bees and return to step S2;

[0097] After all the leading bees complete the search process, the leading bees will perform a waggle dance in the recruitment area to share the nectar source with the observer bees. The observer bees will make a choice based on this information. The choice method is roulette wheel based on the fitness value. The selection probability P of the follower bees to select the leading bees calculated in S4 i , and the formula is:

[0098]

[0099] S5: Record the current optimal nectar source, add a parameter regulation factor to the nectar source regulation unit to act on the scout bees and define the iteration number n, start iterative calculation, and end until the iteration number n is satisfied, and output the optimal numerical combination; when no optimal solution is found after n iterations of searching reach the threshold, the corresponding leading bee is transformed into a scout bee. The scout bee randomly generates a new nectar source in the search space using the antenna orientation unit and returns to S3.

[0100] In S5, adding a parameter regulation factor to the nectar source regulation unit to act on the scout bees and define the iteration number n, the specific formula is:

[0101]

[0102] In the formula, R is the concentration change of the regulatory hormone, F is the threshold, is a fixed constant, a1 and a2 are the system parameters of the nectar source regulation unit, and a1>0, a2<0.

[0103] The reason for proposing the nectar source regulation unit is that in the traditional artificial bee colony algorithm, the iterative formula for the leading bee to search for a new nectar source is

[0104]

[0105] According to this formula, in the iterative process of any nectar source by the traditional artificial bee colony algorithm, the specified number of iterations F is the same and cannot be dynamically adjusted for the search process, resulting in an increase in search time. To overcome the above defects, based on the endocrine regulation mechanism, a nectar source regulation unit is added to act on the leading bees, and the threshold is dynamically adjusted, so that the iteration time decreases with the increase of the number of iterations, effectively reducing the time required for algorithm execution.

[0106] The endocrine regulation mechanism: when stimuli such as cold and tension are transmitted to the hypothalamus, the hypothalamus will release a thyrotropin-releasing hormone, and the pituitary gland will receive this hormone and then release thyroid-stimulating hormone. The thyroid-stimulating hormone is transported through the blood and other means to the thyroid gland, and the thyroid gland releases thyroid hormones to regulate metabolism and achieve the purpose of body temperature regulation. The hormones secreted by the endocrine system are transported to various tissues through body fluids and have the characteristics of being extensive and persistent. The law of action change is similar to the concentration change formula of the neurotransmitter.

[0107] Further, in S5, the scout bee uses the antenna orientation unit to randomly generate a new nectar source in the search space, specifically including: defining the search point of the scout bee to the current global optimal position, and the formula for the antenna orientation unit to generate a new nectar source is:

[0108]

[0109] In the formula, g best is the current global optimal position.

[0110] In the traditional artificial bee colony algorithm, when the scout bee generates a new nectar source, it directly randomly generates a new nectar source in the space. This method has great uncertainty, increases the time complexity of the algorithm operation, and is not conducive to quickly converging to the optimal value. To overcome the above problems, an antenna orientation unit is introduced based on the immune orientation mechanism, and the search point of the scout bee is defined to the current global optimal position.

[0111] The immune orientation mechanism: after the white blood cells in the blood detect and identify the antigen, they can accurately phagocytose this antigen and the infected cells without damaging healthy cells; at the same time, white blood cells also have a memory function. When the immune cells identify the antigen, they will record the antigen information and use this information to identify foreign invaders. When encountering the same antigen again, they can quickly mobilize the corresponding antibodies to protect the body from harm. Immune cells identify antigens through antigenic determinants and have the characteristics of orientation and precision. Applying this characteristic to intelligent optimization algorithms can improve the local performance of the algorithms.

[0112] When the emergence of a new nectar source is near the current globally optimal nectar source, the local search probability of scout bees near the optimal point can be increased, which can accelerate the convergence speed of the algorithm to a certain extent.

[0113] The present invention also provides a system for an intelligent optimization method of oilfield injection-production based on the artificial bee colony with a biological regulation mechanism as described above, at least including:

[0114] A blue light guiding unit for guiding bees to search for nectar sources;

[0115] There is also a nectar source regulation unit for regulating the number of iterations of scout bees;

[0116] There is also an antenna orientation unit for scout bees to search for the globally optimal position as a new nectar source;

[0117] And a model optimization module, which respectively uses the blue light guiding unit, the nectar source regulation unit and the antenna orientation unit to optimize the artificial bee colony algorithm model, and constructs an intelligent optimization model for oilfield injection-production based on the artificial bee colony with a biological regulation mechanism;

[0118] And a data acquisition module for using the sample data to be measured;

[0119] And a data prediction module, which inputs the sample data to be measured into the intelligent optimization model for oilfield injection-production of the artificial bee colony for calculation to obtain an optimization result.

[0120] Further, the model optimization module respectively uses the blue light guiding unit, the nectar source regulation unit and the antenna orientation unit to optimize the oilfield injection-production algorithm model of the artificial bee colony, and constructs an intelligent optimization model for oilfield injection-production based on the artificial bee colony with a biological regulation mechanism, specifically including the following steps:

[0121] S1: Initialize the nectar source and calculate the fitness value of the nectar source;

[0122] S2: The leading bees determine the initial marked nectar source and search for a new nectar source through the blue light guiding unit;

[0123] S3: Obtain the fitness value of the new nectar source, and use the greedy selection strategy to retain the nectar source with a better fitness value;

[0124] S4: Calculate the selection probability P of the follower bees to select the leading bees i , and the follower bees follow the nectar source provided by the leading bees according to the selection probability P i ; For the follower bees that do not choose to follow the nectar source, they become leading bees and return to step S2;

[0125] S5: Record the current optimal nectar source, add a parameter regulation factor in the nectar source regulation unit to act on the scout bees and define the iteration number n, start iterative calculation, and end until the iteration number n is satisfied, then output the optimal numerical combination; when no optimal solution is found after n iterations of search reaching the threshold, the corresponding leading bee is transformed into a scout bee, and the scout bee randomly generates a new nectar source within the search space using the antenna orientation unit, and returns to S3.

[0126] The intelligent optimization system for oilfield injection and production based on the artificial bee colony with biological regulation mechanism further includes:

[0127] A computer-readable storage medium, on which a computer program is stored;

[0128] And a non-volatile semiconductor storage element for reading the data information collected by the data acquisition module;

[0129] There is also a data processing unit, which calls the computer program through a processing circuit to execute and implement the steps of the intelligent optimization method for oilfield injection and production based on the artificial bee colony with biological regulation mechanism;

[0130] There is also a data distribution circuit. After the to-be-predicted data read from the non-volatile semiconductor storage element is processed by the data processing unit according to each processing unit, the obtained prediction results are sent to one or more CAN buses to be sent to external devices through a gateway.

[0131] In summary, the blue light guiding unit of the present invention is designed based on the action mechanism of the nerve guiding mechanism, proposed by simulating the guiding effect of "blue light" on bees when they collect nectar, acts in the observation bee stage, and can enhance the ability of the algorithm to jump out of the local optimum; the antenna orientation unit is designed based on the action mechanism of the immune orientation mechanism, derived from the actual function of the antenna in the physiological structure of bees, acts in the scout bee stage, and can accelerate the convergence speed of the algorithm to a certain extent; the nectar source regulation unit is proposed according to the endocrine regulation mechanism, combined with the biological background of the change in nectar collection efficiency during the working process of bees, acts in the leading bee iteration stage, and can reduce the execution time of the algorithm. The corresponding relationships between the newly introduced units and the artificial bee colony algorithm and the biological background are shown in Table 1.

[0132] Table 1 Introduced unit Biological background

[0133]

[0134] The intelligent optimization method for oilfield injection-production based on the artificial bee colony with biological regulation mechanism is improved in terms of optimization accuracy, convergence speed, and the ability to jump out of local optima. The comparative simulation results with the traditional artificial bee colony algorithm (ABC), the artificial bee colony optimization algorithm (MN-ABC), and the particle swarm optimization-based artificial bee colony optimization algorithm (PSO-ABC) on multiple functions show that the intelligent optimization method for oilfield injection-production based on the artificial bee colony with biological regulation mechanism has better optimization performance than other algorithms.

[0135] The specific simulation verification process is as follows:

[0136] In the MATLAB experimental environment, 7 benchmark functions are selected for simulation verification. The known optimal values of the functions are all 0, the dimension is set to D = 30, and x i ∈[-10, 10], and the maximum number of iterations is set to 500 generations. In the particle swarm optimization algorithm (PSO-ABC), the population size is set to 2000, and the individual learning factor and the swarm learning factor are both set to 2. The population sizes of the intelligent optimization method for oilfield injection-production based on the artificial bee colony with biological regulation mechanism (NEI-ABC), the traditional artificial bee colony algorithm (ABC), and the artificial bee colony optimization algorithm (MN-ABC) are all set to 200. The values of relevant variables in the algorithm are shown in Table 2, and the relevant function expressions are shown in Table 3.

[0137] Table 2 Values of relevant variables

[0138]

[0139] Table 3 Relevant benchmark functions

[0140]

[0141] The comparative data of the simulation experiment results after the four algorithms run are shown in Table 4, and the running effects of the 7 functions are as Figure 3 shown.

[0142] Table 4 Comparison of algorithm running effects

[0143]

[0144] Figure 3Figures (a)-(g) are respectively the optimization effect diagrams of the Griewank, Ackley, Rastrigrin, Schewefel, Rosenbrock, Sumsquares, and Dixon-price functions. During the optimization process, the number of iterations of NEI-ABC is reduced by up to 60.2%, 35.8%, 32.6%, 19.6%, 38.6%, 29.6%, and 25.2% respectively compared to the other three algorithms; its optimization accuracy is improved by 98.9%, 81.6%, 54.9%, 86.3%, 95.9%, 77.3%, and 96.4% respectively compared to the other three algorithms. Through the above tests, it can be found that the convergence speed and optimization accuracy of the NEI-ABC algorithm perform relatively the best.

[0145] The injection-production planning in crude oil production, etc. is a typical complex multi-parameter optimization problem. There is a severe non-linearity between the injection water volume and the produced oil output. At present, production decline equations, genetic algorithms, etc. have been applied to crude oil injection-production planning, but there are still problems such as complex implementation, low accuracy, and slow optimization speed. In order to further solve such problems, the oilfield injection-production intelligent optimization system based on the artificial bee colony with biological regulation mechanism of the present invention is applied to the injection-production optimization problem of crude oil production.

[0146] At present, the production of most oilfields shows a downward trend, and problems such as slow convergence speed and easy premature convergence are becoming increasingly prominent when applying traditional injection-production optimization schemes. Therefore, making reasonable optimization of crude oil exploitation work is beneficial to reducing the waste of natural resources and extending the service life of oilfields. During the process of oilfield exploitation, the index to measure its economic benefits cannot be just a single injection water volume or oil production volume. Optimizing with economic benefits as the index is of great significance.

[0147] Furthermore, in the field of oilfield injection-production optimization, a wavelet neural network is used between the injection water volume and the oil production volume to construct an injection-production model function:

[0148]

[0149]

[0150] where, w ij , w jk are the weight parameters of the wavelet neural network, a is the scaling factor, b is the translation factor, I i is the injection water volume of the i-th well, N i is the total number of injection wells, I t is the total daily injection volume, N p is the total number of production wells, Y t is the daily oil production of the oilfield, i is the number of the injection well, j is the number of the production well, and k is the number of neurons in the hidden layer of the wavelet neural network.

[0151] Further, in the field of oilfield injection-production optimization, an optimization equation for the daily net present value of crude oil is established based on the water injection volume and the oil production volume:

[0152]

[0153] where Y j is the daily production of the j-th production well, r o is the crude oil price, r w is the unit water injection cost price, w1 and w2 respectively represent the weight ratios of the water injection volume and the oil production volume, and w1 + w2 = 1, F max is the daily net present value of crude oil. In principle, on the premise of not damaging the formation environment, the oilfield exploitation aims to improve economic benefits, and the larger the optimized production volume, the better.

[0154] According to the injection-production data of four injection wells and their corresponding one production well in a certain oilfield block, as well as the actual conventional adjustment situation of the oilfield, the parameters w1 = 0.6 and w2 = 0.4 are set; in actual situations, this value can be adjusted according to the working conditions, and the optimal injection-production effect can be obtained by optimizing the water injection parameters. The comparison results of the oilfield production optimized by four algorithms are as Figure 4 shown, and the relevant data comparison is shown in Table 5.

[0155] From Figure 4 and Table 5, it can be seen that the ABC algorithm can reach the optimum at the 27th generation, and the predicted value is 113031765.05 RMB; the PSO-ABC algorithm can reach the optimum at the 25th generation, and the predicted value is 113027900.24 RMB; the MN-ABC algorithm can reach the optimum at the 33rd generation, and the predicted value is 113032277.03 RMB; the oilfield injection-production intelligent optimization system based on the artificial bee colony with a biological regulation mechanism of the present invention can reach the optimum at the 24th generation, and the output predicted value is 113032220.44 RMB.

[0156] Table 5 Comparison of Oilfield Daily Production Optimization Data

[0157]

[0158] It can be seen from the result analysis that under the current working conditions, when the water injection volumes of the four wells are 51.7 t / day, 59.4 t / day, 118.8 t / day, and 118.8 t / day respectively, the water injection efficiency is the highest, and the daily net present value is 113,032,220.44 RMB, which can reach the optimum. According to the collected data, the daily net present value under the current working conditions is approximately 104,075,900.1 RMB, and the daily net present value after optimization by the NEI-ABC intelligent algorithm increases by 8.61%. It is very difficult to calculate the theoretical optimum value of oilfield production. In principle, on the premise of not damaging the formation environment and aiming at improving economic benefits, the larger the optimized production volume, the better. Therefore, the intelligent optimization system for oilfield injection-production based on the artificial bee colony with a biological regulation mechanism of the present invention can effectively increase the daily net present value of crude oil production and provide a new way to solve complex multi-parameter problems such as injection-production planning.

[0159] The embodiments of the present invention have been described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, without departing from the principle and spirit of the present invention, various changes, modifications, substitutions, and variations to these embodiments still fall within the protection scope of the present invention.

Claims

1. An intelligent optimization method for oilfield injection-production based on the artificial bee colony with biological regulation mechanism, characterized in that, Including: Establish an optimization equation for the daily net present value of crude oil based on the water injection volume and oil production volume: Among them, Y j is the daily production of the jth production well, r o is the crude oil price, r w is the unit water injection cost price, w1 and w2 respectively represent the weight ratios of water injection volume and oil production volume, and w1 + w2 = 1, F max is the daily net present value of crude oil; Use an intelligent optimization method for oilfield water injection and production based on an artificial bee colony with a biological regulation mechanism to optimize the above water injection parameters to obtain the optimal daily net present value of crude oil; The intelligent optimization method for oilfield assisted production based on an artificial bee colony with a biological regulation mechanism specifically includes the following steps: S1: Initialize the nectar source and calculate the fitness value of the nectar source; S2: The leading bee determines the initial marked nectar source through the blue light guiding unit and searches for a new nectar source; S3: Obtain the fitness value of the new nectar source and retain the nectar source with a better fitness value using a greedy selection strategy; S4: Calculate the selection probability P of the follower bee choosing the leader bee i , and the follower bee follows the nectar source provided by the leader bee according to the selection probability P i ; for the follower bees that do not choose to follow the nectar source, they become leader bees and return to step S2; S5: Record the current optimal nectar source, add a parameter regulation factor to the nectar source regulation unit to act on the scout bee and define the iteration number n, start iterative calculation, and end until the iteration number n is satisfied, and output the optimal numerical combination; when no optimal solution is found after n iterations of search reaching the threshold, the corresponding leading bee is transformed into a scout bee, and the scout bee randomly generates a new nectar source within the search space using the antenna orientation unit, and returns to S3; Among them, in S2, the leading bee determines the initial marked nectar source through the blue light guiding unit and searches for a new nectar source, specifically including: the leading bee searches for a new nectar source in the solution space, and the formula for generating a new nectar source during the search process is: where G b is the guiding function, S represents the strength of adjusting the guiding effect, μ is a fixed constant, and S > 0, μ > 0. k1 and k2 are parameters of the blue light guiding system, k1 > 0, k2 < 0; rand is a function that generates random numbers, and the generated random numbers are uniformly distributed random real numbers greater than or equal to the lower limit value and less than the upper limit value. e is the natural constant; In S5, adding a parameter regulation factor to the nectar source regulation unit to act on the scout bee and define the iteration number n, the specific formula is: In the formula, R is the concentration change amount of the regulatory hormone, F is the threshold, is a fixed constant, a1 and a2 are system parameters of the nectar source regulation unit, and a1 > 0, a2 < 0; In S5, the scout bee randomly generates a new nectar source within the search space using the antenna orientation unit, specifically including: defining the search point of the scout bee to the current global optimal position, and the formula for the antenna orientation unit to generate a new nectar source is: where, g best is the current global optimal position, L d and U d are the upper and lower limits of the solution space respectively.

2. The intelligent optimization method for oilfield injection-production based on the artificial bee colony with a biological regulation mechanism according to claim 1, wherein, S1 calculates the fitness value of the nectar source, specifically including: Let the dimension of the solution space of the problem be D, the number of nectar sources be N, and the number of iterations be n; the position of nectar source i at the nth iteration can be expressed as where x id ∈[L d U d ), the quality of the nectar source i corresponds to the fitness value fit i of the solution, and the calculation formula for the fitness value fit i of the nectar source is as follows: where n, N, and D are non-zero natural numbers, i = 1, 2,......, N, d = 1, 2,......, D, f i is the function value obtained by the objective function at x i above.

3. The intelligent optimization method for oilfield injection and production based on the artificial bee colony with biological regulation mechanism according to claim 1, wherein In S3, using a greedy selection strategy to retain the nectar source with a better fitness value, specifically including: where \(j\in[1, 2KN]\) and \(j\neq i\), is a random number uniformly distributed over \([-1, 1]\); The greedy selection strategy means that when the fitness value of the new nectar source x' i is better than that of the nectar source x i , the new nectar source x' i is used to replace the nectar source x i , otherwise the nectar source x i is retained.

4. The intelligent optimization method for oilfield injection-production based on the artificial bee colony with biological regulation mechanism according to claim 3, characterized in that, In step S4, calculate the probability P that the follower bee selects the leading bee i , and the formula is as follows:

5. The system of the intelligent optimization method for oilfield injection and production based on the artificial bee colony with biological regulation mechanism according to any one of claims 1-4, characterized in that, At least including: A blue light guiding unit for guiding the leading bee to search for the nectar source; A nectar source regulation unit for regulating the iteration number of the scout bee; An antenna orientation unit for the scout bee to search for the global optimal position as a new nectar source; A model optimization module that optimizes the artificial bee colony algorithm model using the blue light guiding unit, the nectar source regulation unit, and the antenna orientation unit respectively, and constructs an intelligent optimization model for oilfield water injection and production based on an artificial bee colony with a biological regulation mechanism; A data acquisition module for using the test sample data; A data prediction module that inputs the test sample data into the intelligent optimization model for oilfield water injection and production of the artificial bee colony for calculation to obtain the optimization result.

6. The intelligent optimization system for oilfield injection and production based on the artificial bee colony with biological regulation mechanism according to claim 5, characterized in that The model optimization module optimizes the artificial bee colony algorithm model using the blue light guiding unit, the nectar source regulation unit, and the antenna orientation unit respectively, and constructs an intelligent optimization model based on an artificial bee colony with a biological regulation mechanism, specifically including the following steps: S1: Initialize the nectar source and calculate the fitness value of the nectar source; S2: The leading bee determines the initial marked nectar source through the blue light guiding unit and searches for a new nectar source; S3: Obtain the fitness value of the new nectar source, and use the greedy selection strategy to retain the nectar source with a better fitness value; S4: Calculate the selection probability P of the follower bee choosing the leader bee i , and the follower bee follows the nectar source provided by the leader bee according to the selection probability P i ; for the follower bees that do not choose to follow the nectar source, they become leader bees and return to step S2; S5: Record the current optimal nectar source, add a parameter regulation factor to the nectar source regulation unit to act on the scout bees and define the iteration number n, start iterative calculation, and end until the iteration number n is satisfied, and output the optimal numerical combination; when the optimal solution is not found after n iterations of search reaching the threshold, the corresponding leading bee is transformed into a scout bee, and the scout bee randomly generates a new nectar source in the search space using the antenna orientation unit, and returns to S3.

7. The intelligent optimization system for oilfield injection and production based on the artificial bee colony with biological regulation mechanism according to claim 6, characterized in that It further includes: A computer-readable storage medium, on which a computer program is stored; A non-volatile semiconductor storage element for reading the data information collected by the data acquisition module; A data processing unit, which calls the computer program through a processing circuit to execute and implement the steps of the intelligent optimization method for oilfield injection-production based on the artificial bee colony with a biological regulation mechanism as described in any one of claims 1-4; A data distribution circuit, after processing the to-be-predicted data read from the non-volatile semiconductor storage element by each processing unit through the data processing unit, sends the obtained prediction result to one or more CAN buses to be sent to an external device through a gateway.

Images