Remote control dispatching system suitable for multiple scenes of oil field

Abstract

The invention discloses a remote control scheduling system suitable for multiple scenes of an oil field, and relates to the technical field of control scheduling, the system comprises an instruction generation unit, a decision control unit, an adjustment control unit and an acquisition sending unit, the acquisition sending unit periodically acquires equipment operation parameters and transmits the equipment operation parameters to the decision control unit; the decision control unit adopts a particle swarm algorithm to complete initial task allocation, performs secondary allocation in combination with equipment operation parameters, predicts predicted productivity of equipment within a production time limit based on an efficiency attenuation model, compares an initial task load and judges whether adjustment is needed or not; marking equipment which needs to be adjusted and of which the theoretical target power exceeds a rated upper limit as incremental equipment, and calculating to-be-allocated increments; a to-be-distributed increment is optimized and split to non-increment equipment allowed by a load through a dynamic coupling adaptation algorithm, and an instruction generation unit generates a corresponding instruction and transmits the instruction to an adjustment control unit, so that dynamic and efficient scheduling under multiple scenes of an oil field is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of control scheduling technology, in particular to a remote control scheduling system suitable for multiple scenes of oilfields. BACKGROUND

[0002] In the actual production process of oilfields, the traditional production task control scheduling mode has the following problems: on the one hand, the initial task allocation is mostly dependent on manual experience or simple equal division, which is difficult to achieve the coordinated optimization of global production efficiency and total energy consumption, and is prone to cause unbalanced equipment load, overloading of some equipment or idle capacity, in addition, in the face of task increment, there is a lack of scientific secondary allocation mechanism, which is difficult to quickly allocate the incremental task to the adaptive equipment, and it is difficult to meet the dynamic and efficient scheduling requirements in different production scenes of oilfields, which ultimately affects the overall production benefit; on the other hand, the characteristics of the decrease of equipment operating efficiency with the increase of cumulative running time are not considered, which cannot accurately predict the actual capacity of the equipment within the entire production time limit, often leading to the failure of some equipment to complete the initial task on time, or over-consumption of energy to complete the task, therefore, the existing control scheduling mode has been difficult to meet the comprehensive requirements of efficiency, energy consumption control and dynamic adaptability of oilfield production, and there is an urgent need for a remote control scheduling system which can realize scientific task allocation, accurate capacity prediction and adapt to multiple scenes. SUMMARY

[0003] The present application aims to provide a remote control scheduling system suitable for multiple scenes of oilfields, which allocates tasks through multi-objective optimization and dynamic coupling adaptation algorithm, and predicts capacity through efficiency decay model, to achieve the purpose of dynamic control scheduling in multiple scenes of oilfields.

[0004] The technical solution to achieve the purpose of the present application is as follows:

[0005] A remote control scheduling system suitable for multiple scenes of oilfields, comprising a control end and an oilfield controlled end, the control end comprising an instruction generation unit and a decision control unit, and the oilfield controlled end comprising an adjustment control unit and a collection and sending unit:

[0006] The control end allocates the initial task quantity according to the production task quantity and production time limit input by the user, generates and sends the initial allocation instruction, receives the equipment operating parameters, obtains the equipment operating efficiency through the efficiency decay model which fuses the time decay term and the load decay term, calculates the predicted capacity within the production time limit, compares the initial task quantity to locate the equipment to be adjusted, selects and marks the incremental equipment based on the difference between the theoretical target power and the upper limit of the rated power of the equipment to be adjusted, combines the total amount of the incremental task to be allocated which cannot be completed by the incremental equipment within the production time limit, and completes the secondary allocation of the increment based on the power redundancy, capacity redundancy coefficient and unit energy consumption coefficient of the non-incremental equipment through the dynamic coupling adaptation algorithm, generates the theoretical target power and sends the secondary allocation instruction.

[0007] The oilfield controlled end receives the initial allocation instruction and the secondary allocation instruction in turn and controls the corresponding equipment to operate at corresponding power, collects and processes the equipment operation parameters, and feeds back to the control end.

[0008] Further, the efficiency decay model is used to predict the expected production capacity of the equipment within the production time limit, including the following steps:

[0009] The abnormal values in the equipment operation parameters are removed by the Lyapunov criterion, and the mean values of the parameters in the first three sampling periods are used to fill the vacancies after removing the abnormal values.

[0010] Considering the long-term operation of the equipment, the efficiency decay model including time decay term and load decay term is constructed according to the mechanical wear and load fatigue of the equipment under different cumulative running time, which takes the cumulative running time of the equipment as the variable, reflects the ratio relationship between the current running efficiency and the efficiency in the new state. In order to make the model fit the actual operation law of the equipment, the actual efficiency data of the equipment under different cumulative running time is collected, and the least square method is used to fit the efficiency decay coefficient in the model. By minimizing the sum of squares of the error between the actual efficiency and the calculated efficiency of the model, the decay coefficient suitable for the actual state of the equipment is determined, and finally the model reflecting the efficiency decay law of the equipment is formed.

[0011] The power utilization rate of the equipment is calculated, that is, the current actual running power of the equipment is divided by the rated running power when the production reference capacity is used, and then the current efficiency calculated by the efficiency decay model is multiplied by the reference capacity to obtain the actual production capacity of the equipment in the current sampling period.

[0012] The number of sampling periods contained in the remaining production time of the equipment is determined, the actual production capacity of the equipment in each remaining sampling period is predicted by combining the efficiency decay model, the cumulative production capacity in the remaining production time is obtained, and then the remaining cumulative production capacity is added to the completed production capacity of the equipment, and finally the expected production capacity of the equipment within the entire production time limit is obtained.

[0013] Further, when the least square method is used to fit the efficiency decay coefficient, the actual efficiency data of the equipment under different cumulative running time is collected, which is substituted into the efficiency decay model formula, and the sum of squares of the error between the actual efficiency and the calculated efficiency of the model is minimized to generate the efficiency decay coefficient suitable for the actual running state of the equipment.

[0014] Further, the decision control unit calculates the theoretical target power of the initial task amount and locates the incremental equipment, including the following steps:

[0015] According to the initial task amount of the device, the production time limit, the efficiency decay model and the benchmark productivity, a calculation formula of the theoretical target power is derived through an equation relationship that the initial task amount needs to meet the actual productivity of each sampling period within the production time limit, and the theoretical target power is the minimum running power required for the device to just complete the initial task amount;

[0016] The derived theoretical target power is compared with the upper limit of the rated power of the device. If the theoretical target power is less than or equal to the upper limit of the rated power, a control instruction containing the theoretical target power is generated and transmitted to the adjustment control unit through a wireless network to adjust the running power of the device to the theoretical target power. If the theoretical target power exceeds the upper limit of the rated power, it is determined that the device cannot complete the initial task by adjusting the power, and it is marked as an incremental device, and the upper limit of the rated power of the device is recorded.

[0017] Further, when calculating the to-be-allocated increment, the maximum predicted productivity of the incremental device running at the upper limit of the rated power is calculated first, and the maximum productivity that the device can complete within the entire production time limit by running at the upper limit of the rated power is predicted in combination with the upper limit of the rated power of the incremental device, the number of sampling periods contained in the remaining production time, the benchmark productivity and the current efficiency. Then, the initial task amount of the incremental device is subtracted from the maximum predicted productivity to obtain a difference value, which is the to-be-allocated increment of the device, which needs to be split to other non-incremental devices for completion.

[0018] Further, the initial task amount is calculated based on the particle swarm algorithm, including the following steps:

[0019] The total amount of the current period production task of the oil field, the production time limit, and the benchmark productivity, energy consumption and upper limit of rated power of each device in the previous sampling period of each device are obtained;

[0020] The initial task amount of each device is set as a decision variable, and the initial task amounts of all devices constitute a decision variable set. Three types of constraint conditions are set. The first is the total amount constraint, the sum of the initial task amounts of all devices needs to equal the total amount of the overall production task of the oil field, neither exceeding nor being insufficient. The second is the single-device upper limit constraint, the initial task amount of a single device does not exceed the maximum productivity that the device can complete within the production time limit. The third is the non-negative constraint, the initial task amount of each device cannot be negative to avoid logical contradictions. The optimization objectives are divided into two categories. The first is to minimize the maximum time consumption of a single device to ensure that all devices complete the task as synchronously as possible. The second is to minimize the total energy consumption of all devices.

[0021] Encode the particles, each particle corresponds to a set of initial task allocation scheme, the dimension of the particle is consistent with the number of devices, and each position value of the particle represents the initial task amount of the corresponding device, set the population size, randomly generate the initial particle position that meets the constraint condition, first randomly allocate the task amount of each device within the range of 0 to its maximum capacity, then adjust the task amount of all devices in proportion, ensure the sum equal to the total amount of overall production tasks, and initialize the initial speed, learning factor, inertia weight and maximum iteration number of the particle;

[0022] Initialize the optimization algorithm parameters, encode the particle corresponding to the task allocation scheme, set the population size, randomly generate the initial particle position and speed, and initialize the learning factor, inertia weight and maximum iteration number;

[0023] Calculate the double-objective function value of each particle, i.e. the maximum time consumption of a single device and the total energy consumption, filter non-dominated solutions based on the Pareto dominance relationship, if the two objective values of a particle are better than those of another particle, the former dominates the latter, and the latter is eliminated, and all non-dominated solutions are retained to form a Pareto optimal solution set, secondly, update the individual optimal position and global optimal position, thirdly, update the particle speed and position according to the preset formula, and if the particle position violates the constraint condition after updating, truncate the task amount exceeding the maximum capacity to the maximum capacity, and distribute the excess part to the devices that do not reach the maximum capacity in proportion, then readjust the task amount of all devices to ensure that the total amount constraint is met, finally, repeat the above steps until the iteration is completed;

[0024] When the number of iterations reaches the maximum iteration number, or the objective function values of the Pareto optimal solution set do not fluctuate for ten consecutive generations, stop the iteration, preset the weights of the two types of objectives according to the actual demand of the oilfield, score each solution in the Pareto optimal solution set by linear weighting method, and select the solution with the highest weighted score as the final initial task allocation scheme.

[0025] Further, when the decision control unit performs secondary incremental allocation, the current load state of the non-incremental device includes power redundancy, capacity redundancy coefficient and unit energy consumption coefficient:

[0026] Subtract the current actual operating power of the non-incremental device from the upper limit of its rated power to obtain the space power redundancy reflecting the operating power that can be improved by the device, the larger the value, the greater the potential of the device to undertake incremental tasks;

[0027] First, calculate the maximum capacity that the non-incremental device can generate within the remaining production time according to the upper limit of the rated power, then calculate the initial task amount that the device has not completed, and divide the additional maximum capacity by the initial task amount that has not been completed to obtain the capacity redundancy coefficient, the larger the coefficient, the stronger the ability of the device to undertake incremental tasks;

[0028] The benchmark unit energy consumption of the non-incremental device is obtained by dividing the benchmark energy consumption by the benchmark energy production, and the unit energy consumption coefficient is obtained by dividing the benchmark unit energy consumption by the current operation efficiency of the device, and the smaller the coefficient, the more energy-saving the device is when undertaking the incremental task.

[0029] Further, the decision control unit splits the to-be-allocated increment to the non-incremental device through a dynamic coupling adaptation algorithm, including the following steps:

[0030] The total amount of to-be-allocated increment is obtained by adding the to-be-allocated increment of all incremental devices.

[0031] Based on the power redundancy, the energy production redundancy coefficient and the unit energy consumption coefficient of the non-incremental device, the secondary allocation of the increment is completed through a dynamic coupling adaptation algorithm, the power redundancy and the unit energy consumption coefficient are normalized by dividing the single-device load index value by the maximum value of the load index in all non-incremental devices, the weight coefficients of the three types of load indexes of the power redundancy, the energy production redundancy coefficient and the unit energy consumption coefficient are set according to the oilfield demand through an expert scoring method, and finally the dynamic coupling weight is calculated, the higher the dynamic coupling weight of the device, the more preferentially the device undertakes the incremental task, the allocation ratio is determined according to the proportion of the dynamic coupling weight, and the allocation ratio of a single device is the dynamic coupling weight of the device divided by the sum of the dynamic coupling weights of all non-incremental devices.

[0032] The initial allocation increment of each non-incremental device is calculated according to the allocation ratio, the initial allocation increment is added to the initial task amount of the device, it is judged whether the sum exceeds the maximum energy production of the device within the production time limit, if not, it is determined that the device undertakes the initial allocation increment, if yes, the excess increment is calculated, the excess part is reallocated according to the allocation ratio of other non-incremental devices, and the process is repeated until all to-be-allocated increments are allocated to non-incremental devices with load capacity.

[0033] Further, the abnormal values in the device operation parameters are removed through the Laplace criterion, the collected device operation parameters are divided into three types of actual operation power, cumulative operation time and completed energy production, for each type of parameter, the average value and the standard deviation of the parameter in the recent continuous multiple sampling periods are calculated, the values of the parameter exceeding the average value minus three times the standard deviation to the average value plus three times the standard deviation are determined as abnormal values, and the abnormal values are removed from the parameter sequence.

[0034] Further, the benchmark energy production and the benchmark energy consumption of each production device in the previous sampling period are the actual energy production and the total amount of energy consumed when the production device continuously operates at a designed standard fixed power for a complete sampling period from the parameter collection start time to the collection end time under normal production conditions.

[0035] Compared with the prior art, the present application has the following advantages:

[0036] 1. Implement initial and secondary task allocation through multi-objective optimization, adapt to equipment capabilities, avoid equipment overload and idleness, and improve efficiency and reduce costs;

[0037] 2. Introduce an efficiency decay model to accurately predict production capacity, dynamically adjust power and locate incremental equipment, and avoid misjudgment of production capacity and energy waste. Attached Figure Description

[0038] Figure 1 A flowchart of a remote control and dispatch system applicable to multiple scenarios in oil fields;

[0039] Figure 2 This is a flowchart of the initial task allocation process in this invention;

[0040] Figure 3 This is a flowchart of the efficiency decay model prediction and incremental device positioning process in this invention;

[0041] Figure 4 This is a flowchart of the secondary incremental allocation process in this invention. Detailed Implementation

[0042] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0043] like Figure 1 As shown, this invention discloses a remote control and scheduling system applicable to multiple scenarios in oil fields, including a control terminal and an oil field controlled terminal:

[0044] The control unit includes an instruction generation unit and a decision control unit. The decision control unit is used to execute initial task allocation and secondary incremental allocation. It generates corresponding instructions through the instruction generation unit and transmits them wirelessly to the adjustment control unit. When performing initial task allocation, the decision control unit obtains the production task and production time limit, and simultaneously obtains the baseline capacity and baseline energy consumption of each production device in the previous sampling period. It then calls the particle swarm optimization algorithm to calculate the initial task load for each production device, aiming to maximize global production efficiency and minimize total energy consumption. The instruction generation unit generates initial allocation instructions and transmits them wirelessly to the corresponding adjustment control unit. During secondary incremental allocation, after receiving the device operating parameters, the decision control unit considers the decrease in device operating efficiency with accumulated time. It uses a capacity prediction method based on an efficiency decay model to dynamically calculate the expected capacity that each device can complete within the entire production time limit and compares it with the initial task load to determine whether each device can independently complete its initial task load. If the expected capacity is greater than or equal to the initial task load... If the initial task quantity is less than the production time limit, the task is deemed complete. If the initial task quantity is less than the production time limit, the task is deemed to require adjustment. For equipment requiring adjustment, the theoretical target power to meet the task requirements is calculated based on the initial task quantity and production time limit, and compared with the rated power limit. If the theoretical target power is less than or equal to the rated power limit, a control command is generated and wirelessly transmitted to the corresponding adjustment control unit to adjust the equipment power to the theoretical target power. If the theoretical target power is greater than the rated power limit, the equipment is marked as an incremental equipment. The maximum expected capacity of the incremental equipment when running at the rated power limit is calculated. The maximum expected capacity is then subtracted from the initial task quantity to obtain the incremental to be allocated. The total amount of all incremental to be allocated is calculated. Based on the power redundancy, capacity redundancy coefficient, and unit energy consumption coefficient of non-incremental equipment, the dynamic coupling weight of each non-incremental equipment is calculated through a dynamic coupling adaptation algorithm. Based on the dynamic coupling weight, the incremental to be allocated is optimized, split, and allocated to non-incremental equipment. The theoretical target power required for each non-incremental equipment to undertake the incremental task is calculated. Finally, a secondary allocation command is generated and wirelessly transmitted to the corresponding adjustment control unit.

[0045] The controlled end of the oilfield includes a regulation and control unit and a data acquisition and transmission unit. The regulation and control unit receives initial allocation instructions and secondary allocation instructions in sequence, and controls the production equipment to operate at a specified power. The data acquisition and transmission unit collects equipment operating parameters through sensors during each sampling period, including actual operating power, cumulative operating time, and completed production capacity, and transmits them to the decision control unit via a wireless network.

[0046] like Figure 2 As shown, further, the decision control unit executes the initial task allocation, and combined with the overall production task volume and production time limit within the current cycle of the oilfield, the optimal initial task allocation for each production equipment is solved using the particle swarm optimization algorithm, and instructions are generated and sent to the control equipment, including the following steps:

[0047] Obtain the total production task of the oilfield in the current cycle. and production time limit Record the total number of all production equipment in the oilfield as follows: single device Collect basic parameters of all production equipment, including individual equipment. During the previous sampling period Benchmark capacity within single device During the previous sampling period Reference energy consumption within Maximum rated power of a single device The rated power limit of a single device is used to constrain the maximum workload of the device and avoid overload. The baseline capacity and baseline energy consumption are based on the production equipment operating at a fixed power under normal production conditions during the previous sampling period. Production capacity and energy consumption during operation;

[0048] Before performing multi-objective optimization, first define the decision variables and constraints. To be assigned to the device The initial task load and the task allocation of all devices constitute a decision variable vector. The objective is to optimize the solution to the requirements, while setting constraints to ensure the feasibility of task allocation. These constraints include total quantity constraints, single-device upper limit constraints, and non-negativity constraints. The expression for the total quantity constraint is shown below:

[0049] ,

[0050] in, To be assigned to the device The initial workload. The expression represents the total production task of the oilfield. It means that the sum of the initial tasks of all equipment must equal the overall production task of the oilfield; it cannot exceed or fall short of the target. The upper limit constraint expression for a single piece of equipment is shown below:

[0051] ,

[0052] in, To be assigned to the device The initial workload. For equipment The maximum production capacity that can be produced within the entire production period. This refers to the number of data collection periods included within the production timeframe. For a single device During the previous sampling period The baseline capacity within the range, with nonnegativity constraint expressions as follows:

[0053] ,

[0054] The purpose of nonnegativity constraints is to ensure that the initial allocation amount cannot be negative when allocating initial workload to all devices, thus avoiding logical contradictions.

[0055] An optimization function is constructed with the dual objectives of maximizing global production efficiency and minimizing total energy consumption. Global production efficiency is defined as the total capacity per unit time. Due to total capacity constraints, the total capacity is fixed within a given production time limit. Therefore, in practical optimization, maximizing global production efficiency is transformed into minimizing the maximum time consumption of a single production device. The global production efficiency function is then... The expression is as follows:

[0056] ,

[0057] in, The initial workload allocated to device 1. For device 1 during the previous sampling period The benchmark capacity within, Similarly, the time taken for device 1 to complete the initial workload is also... The time taken for device 2 to complete the initial workload. For equipment The time taken to complete the initial task is expressed as follows: This expression aims to ensure all devices complete tasks as synchronously as possible, avoiding individual devices slowing down the overall progress, thereby maximizing effective productivity per unit time. Total energy consumption is the sum of the energy consumption of all devices; hence, the total energy consumption function is... The expression is as follows:

[0058] ,

[0059] in, For equipment During the previous sampling period The baseline energy consumption within, For equipment The energy consumption per unit of production capacity will be used to power equipment. initial task volume With equipment Multiply the unit capacity energy consumption to obtain the equipment Complete the initial task The total energy consumption is summed to obtain the total energy consumption of the production equipment. The oilfield operating cost is reduced by minimizing the total energy consumption.

[0060] To initialize the particle swarm optimization (PSO) algorithm parameters, the particles are first encoded, with each particle representing a set of task allocation schemes. The particle dimension is... Consistent with the number of devices, at the particle position middle, Indicates the first Each particle corresponds to a device The initial task allocation amount, setting the population size as Randomly generate initial particle positions, that is, randomly generate initial task assignments that satisfy constraints, including randomly assigning each device a position from 0 to 1. The amount of work within the range, and calculate the current sum and... The difference is used to scale the workload of each device proportionally until the sum equals the difference. ,particle speed This indicates the adjustment range of the initial workload for each device; the initial value is set randomly. Within the range, Values Ten percent, and simultaneously initialize the learning factor. and inertia weight and maximum number of iterations ;

[0061] Iterative optimization is performed to find the optimal task allocation scheme by gradually approaching it through swarm intelligence optimization of particle swarms. Calculate its dual objective function value The goal is to minimize both the maximum time consumption and the total energy consumption, and to evaluate the merits of particles based on the Pareto dominance relation. If particle A's two objective values ​​are both superior to those of particle B, then A dominates B, and B is eliminated. All non-dominated solutions are retained, forming a Pareto optimal solution set. Individual and global optimal updates are then performed, with the individual optimal solution being the best. Recording particles The optimal position from the initial stage to the current iteration, i.e., the best-performing task allocation scheme in the history of this particle, is the global optimum. To randomly select a solution from the Pareto optimal set as the global guiding particle and avoid the algorithm getting trapped in local optima, particle velocity and position are updated. The particle velocity update formula is shown below:

[0062] ,

[0063] in, For particles In the In the next iteration, the device The speed of task load adjustment For inertial weights, These are learning factors, used to control the acceleration of particles moving towards their individual optimal and global optimal states, respectively. The random number between 0 and 1 is generated randomly in each iteration, making the adjustment direction and amplitude of the particles uncertain and enhancing the diversity of the search. For particles The corresponding device in the individual optimal position The amount of tasks allocated, For particles In the The corresponding device in the next iteration The current task allocation amount, For the globally optimal position, the corresponding device The task allocation and location update formula are as follows:

[0064] ,

[0065] in, For particles In the In the next iteration, the corresponding device The speed at which the workload is adjusted determines the device's performance in the next iteration. How much the workload will increase or decrease, depending on the device The task allocation is adjusted by adding a certain amount to the original task allocation to obtain the updated task allocation. If the updated location violates the constraints, the task allocation exceeding the upper limit is truncated. The excess workload is proportionally allocated to other devices that have not reached their limits, and the total is recalculated. The workload of all devices is scaled proportionally to ensure that the total limit is met. When the number of iterations reaches [a certain threshold], [the process continues]. The iteration stops when the objective function value of the Pareto optimal solution set no longer fluctuates after ten consecutive iterations, or before the maximum number of iterations has been reached.

[0066] The Pareto optimal solution set contains multiple non-dominated solutions, which are not absolutely superior or inferior to each other. One non-dominated solution exhibits high efficiency but slightly higher energy consumption, while another has low energy consumption but slightly lower efficiency. Based on the actual needs of the oilfield—whether the oilfield prioritizes efficiency or energy consumption—weights are preset. A linear weighting method is used to assign these weights to the dual objective function and perform calculations. The solution with the optimal weighted value is selected as the final task allocation scheme. , the elements Corresponding to the first Task allocation decisions for individual devices;

[0067] Will The data is converted into structured instructions, including device ID, task quantity, and start time, and transmitted to the regulation and control units of each device via the oilfield wireless network to complete the initial task allocation.

[0068] like Figure 3 As shown, furthermore, a capacity prediction algorithm based on an efficiency decay model is used to dynamically calculate the expected capacity that each piece of equipment can complete within the entire production timeframe, and finally locate the incremental equipment, including the following steps:

[0069] The acquisition and transmission unit obtains information from each device. During the current sampling period Operating parameters, including actual operating power Cumulative duration Completed production capacity The data was validated, outliers were removed using the Raida criterion, and the average of the previous three periods was used to fill the gaps after the removal.

[0070] To quantify the mechanical wear accumulated over time and the load fatigue caused by accelerated wear due to high-power operation, an efficiency decay model is constructed by combining time decay terms and load decay terms. The expression of the efficiency decay model is as follows:

[0071] ,

[0072] Simplified and merged into:

[0073] ,

[0074] in, For equipment run The current operating efficiency of the equipment after one hour, that is, the ratio of efficiency to its state when it was brand new. , For equipment initial efficiency, To accumulate runtime, The equipment efficiency degradation coefficient represents the rate of degradation caused by operating time. This is the load attenuation coefficient, reflecting the amplification effect of the load on attenuation. The cumulative average power utilization rate, which is the average load intensity of the equipment from startup to the present, is expressed as follows:

[0075] ,

[0076] in, For the first Average operating power during the data collection period The duration of a single data collection period is represented by the numerator, which is the cumulative power consumed, and the denominator is the cumulative power at the rated power.

[0077] The efficiency degradation model was validated by obtaining historical operating data of the equipment, including the average operating power, cumulative duration, and actual efficiency for each data collection period. The actual efficiency can be obtained by dividing the actual output by the theoretical output, while the average load intensity is calculated and fitted using the least squares method. and This minimizes the sum of squared errors between the predicted efficiency and the actual efficiency, i.e. The efficiency attenuation coefficient is obtained by solving. and At this point, the efficiency degradation model closely matches the actual performance of the equipment;

[0078] Based on the current power and efficiency degradation model, the actual capacity of the equipment is determined by first calculating the ratio of the actual operating power of the current production equipment to the power used by the equipment to produce one benchmark unit of capacity. Power utilization rate The calculation formula is as follows:

[0079] ,

[0080] in, For production equipment Actual operating power Given the operating power of the production equipment at its baseline capacity, and combining this with the power utilization rate, baseline capacity, and efficiency degradation, the operating power of the production equipment can be calculated. Actual capacity during a sampling period under the current conditions The calculation formula is as follows:

[0081] ,

[0082] in, For equipment The baseline capacity is obtained by multiplying the baseline capacity, power utilization rate, and current equipment operating efficiency with reaction decay.

[0083] computing devices Remaining production time and calculate the equipment Cumulative capacity for the remaining period The calculation formula is as follows:

[0084] ,

[0085] in, For production equipment Remaining production time, This is the ratio of the remaining production time of the production equipment to the previous sampling period, therefore the equipment... The total expected production capacity within the entire lifecycle to the production deadline is ;

[0086] Production equipment Total projected capacity throughout the entire production cycle within the production timeframe With the initial workload of the device If a comparison is made, Determine if the device can complete the assigned initial workload. If the equipment is identified as needing adjustment, it will be entered into the theoretical target power calculation process.

[0087] Calculation enables the device The theoretical target power that exactly completes the initial task must satisfy the following equation:

[0088] ,

[0089] The theoretical target power can be obtained by solving and deriving the solution. The calculation formula is as follows:

[0090] ,

[0091] The theoretical target power obtained through this calculation formula is for the equipment. Minimum operating power required to complete the initial workload;

[0092] Based on the calculated theoretical target power, it is compared with the upper limit of the rated power of the equipment. If... The instruction generation unit generates instructions and sends them to the corresponding adjustment and control unit in the oilfield to adjust the power to... ,like The device is marked as an incremental device, and a command is generated and sent to the oilfield regulation and control unit to ensure that all incremental devices operate at their rated power limit, and the incremental devices are calculated. The gap between the production capacity operating at the rated power limit and the initial task is recorded as the incremental capacity to be allocated. .

[0093] like Figure 4 As shown, further, based on the incremental power to be allocated and the equipment operating parameters, and using the dynamic coupling adaptation algorithm, the theoretical target power required for each non-incremental device to undertake the incremental task is calculated, including the following steps:

[0094] The total number of incremental units to be allocated is obtained by summing up the incremental units to be allocated for all incremental units. Its expression is as follows:

[0095] ,

[0096] in, This is the set of incremental devices, i.e., the device numbers marked as incremental devices. For incremental equipment The unassigned increments are filtered, and the non-incremental devices in the device operating parameters are obtained. The set of non-incremental devices not marked as increments is then processed. For each non-incremental device Extract operating parameters, which are consistent with the overall device serial number. Distinguish, Replace with non-incremental device serial number And assign values, including the current operating power. Initial task volume Rated power limit Remaining production time Completed production capacity Current efficiency and the previous sampling period Internal benchmark capacity ;

[0097] For each non-incremental device, calculate the power redundancy, capacity redundancy coefficient, and unit energy consumption coefficient. The power redundancy is the difference between the device's current operating power and its rated power limit. This reflects the potential for increased power output and the capacity redundancy factor. The maximum additional production capacity that non-incremental equipment can undertake during the remaining production time when the equipment is operating at its rated power limit is calculated using the following formula:

[0098] ,

[0099] in, This refers to the power utilization rate of non-incremental equipment operating at its rated power limit. Non-incremental equipment The remaining initial workload, Non-incremental equipment Given a fixed power during the previous sampling period, the numerator represents the additional capacity that can be generated in the remaining time at the rated power. The larger the capacity redundancy coefficient, the stronger the ability to undertake incremental tasks. The above formula can be further simplified as follows:

[0100] ,

[0101] Unit energy consumption coefficient The energy consumption per unit of production capacity when undertaking incremental tasks is used to prioritize the selection of energy-saving equipment. The calculation formula is as follows:

[0102] ,

[0103] in, Non-incremental equipment During the previous sampling period The baseline energy consumption within, Non-incremental equipment The energy consumption per unit of production capacity, divided by the energy consumption of non-incremental equipment. Current efficiency Then, the unit energy consumption reflecting the actual decay state is obtained. The smaller the size, the more energy-efficient the equipment is in handling incremental tasks;

[0104] Based on the power redundancy, capacity redundancy coefficient, and unit energy consumption coefficient of non-incremental equipment, a dynamic coupling adaptation algorithm is used to determine the task allocation ratio of each non-incremental equipment. First, for each non-incremental equipment, the comprehensive power redundancy is calculated. Capacity redundancy coefficient and unit energy consumption coefficient Calculate dynamic coupling weight using three load metrics The calculation formula is as follows:

[0105] ,

[0106] in, The weighting coefficients are determined using an expert scoring method based on the actual needs of the oilfield. , These represent the maximum power redundancy and unit energy consumption coefficient among all non-incremental devices, respectively, used to normalize the indicators to the range of 0 to 1, and dynamically couple the weights. The higher the value, the more priority the device takes on incremental tasks. The proportion of incremental tasks undertaken by each non-incremental device is positively correlated with the dynamic coupling weight. Therefore, determining the task allocation ratio is crucial. The calculation formula is as follows:

[0107] ,

[0108] in, To determine the task allocation ratio, Greater than or equal to 0, therefore, non-incremental devices Incremental workload to be undertaken At the same time, the constraints must be met, namely, the incremental task shall not exceed the maximum capacity of the equipment. If the constraints are exceeded, the excess part shall be distributed to other equipment proportionally and recalculated until all constraints are met.

[0109] Based on the assigned incremental tasks, the required power adjustment value is calculated backwards, which is the theoretical target power for the equipment's operation. The calculation formula is as follows:

[0110] ,

[0111] The above formula has been constrained, so the operating power of the equipment after the power adjustment value will not exceed the upper limit of the rated power of the equipment, and relevant instructions can be generated directly.

[0112] For each non-incremental device, generate a data set including the device ID and the incremental task quantity. Theoretical target power The secondary allocation instructions are transmitted to the corresponding device's adjustment and control unit via wireless network to ensure that the device operates at the target power and undertakes incremental tasks.

[0113] This invention discloses a remote control and scheduling system applicable to multiple scenarios in oilfields. The system includes a control terminal and an oilfield controlled terminal. The control terminal includes an instruction generation unit and a decision control unit, while the oilfield controlled terminal includes an adjustment control unit and an acquisition and transmission unit. After receiving the production task quantity and production time limit, the control terminal constructs a dual-objective optimization function that minimizes the maximum time consumption of a single device and minimizes the total energy consumption using a particle swarm optimization algorithm. Combining the total quantity, the upper limit of a single device, and non-negativity constraints, it completes the initial task allocation and sends the initial instruction. The acquisition and transmission unit periodically collects equipment operating parameters and feeds them back to the control terminal. The decision control unit uses the Laida criterion to eliminate outlier parameters and integrates the time decay term with the negative... The efficiency decay model of load attenuation term predicts the expected production capacity of equipment within the production time limit, compares the initial task volume to locate the equipment to be adjusted, calculates the theoretical target power and marks the incremental equipment whose theoretical target power exceeds the rated upper limit and the incremental equipment to be allocated; then, based on the power redundancy, production capacity redundancy coefficient and unit energy consumption coefficient of non-incremental equipment, the incremental equipment to be allocated is optimized and split into non-incremental equipment through dynamic coupling adaptation algorithm. The instruction generation unit generates secondary allocation instructions and transmits them to the adjustment and control unit. The oilfield controlled end controls the operation of equipment according to the instructions, realizing dynamic and efficient scheduling in multiple scenarios of oilfield, and solving the problems of unscientific task allocation, inaccurate production capacity prediction and disordered incremental allocation in traditional scheduling.

[0114] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A remote control and dispatching system suitable for multiple scenarios in oil fields, characterized in that, Including the control end and the oilfield controlled end: The control unit allocates initial tasks based on the user-input production task quantity and production time limit, generates and sends initial allocation instructions, receives equipment operating parameters, obtains equipment operating efficiency and calculates the expected capacity within the production time limit by using an efficiency decay model that integrates time decay and load decay terms, compares the initial task quantity to locate equipment to be adjusted, filters and marks incremental equipment based on the difference between the theoretical target power and the rated power limit of the equipment to be adjusted, combines the total incremental amount to be allocated that the incremental equipment cannot complete within the production time limit, and completes the secondary allocation of incremental equipment through a dynamic coupling adaptation algorithm based on the power redundancy, capacity redundancy coefficient and unit energy consumption coefficient of non-incremental equipment, generates theoretical target power and sends secondary allocation instructions. The oilfield controlled terminal receives initial allocation instructions and secondary allocation instructions in sequence, controls the corresponding equipment to operate at the corresponding power, collects and processes the equipment operating parameters, and feeds them back to the control terminal.

2. The remote control and dispatching system applicable to multiple scenarios in oil fields as described in claim 1, characterized in that, Using an efficiency decay model, the projected capacity of the equipment within the production timeframe is calculated, including the following steps: Outliers in equipment operating parameters were removed based on the Raida criterion, and the mean values ​​of the parameters from the first three sampling periods were calculated to fill the gaps left after removing outliers. Considering the degradation characteristics of equipment over time, an exponential efficiency degradation model is constructed. Combining historical operating data of the equipment, the least squares method is used to fit the efficiency degradation coefficient, resulting in an efficiency degradation model that fits the actual efficiency degradation law of the equipment. Input the equipment runtime into the efficiency decay model to obtain the current equipment operating efficiency. At the same time, calculate the equipment's power utilization rate. Multiply the current equipment operating efficiency by the power utilization rate and then by the baseline capacity to obtain the actual capacity for a single sampling period. Determine the number of sampling periods included in the remaining production time of the equipment. Combine the actual capacity of a single sampling period with the current operating efficiency of the equipment to calculate the cumulative capacity of the equipment during the remaining production time and add it to the capacity already completed by the equipment to obtain the expected capacity of the equipment over the entire production period.

3. A remote control and dispatching system applicable to multiple scenarios in oil fields as described in claim 2, characterized in that, The efficiency decay coefficient is fitted using the least squares method. By collecting actual efficiency data of the equipment under different cumulative operating times, and substituting it into the efficiency decay model, the sum of squared errors between the actual efficiency and the efficiency calculated by the model is minimized to determine the efficiency decay coefficient that is suitable for the actual operating state of the equipment.

4. A remote control and dispatching system applicable to multiple scenarios in oil fields as described in claim 1, characterized in that, Calculate the theoretical target power of the equipment to be adjusted and locate the incremental equipment, including the following steps: Based on the initial workload, production time limit, current equipment operating efficiency, and baseline capacity of the equipment, and based on the equation that the initial workload equals the sum of the actual capacity of each sampling period within the production time limit, the theoretical target power is derived. The theoretical target power is compared with the upper limit of the rated power of the equipment. If the theoretical target power is less than or equal to the upper limit of the rated power, a control command is generated to adjust the operating power of the equipment to the theoretical target power. If the theoretical target power is greater than the upper limit of the rated power, the equipment is marked as an incremental equipment.

5. A remote control and dispatching system applicable to multiple scenarios in oil fields as described in claim 1, characterized in that, Calculate the incremental capacity to be allocated, and combine the power utilization rate corresponding to the rated power limit, the number of sampling periods included in the production time limit, the baseline capacity and the current efficiency to calculate the maximum expected capacity that the incremental equipment can complete in the entire production time limit when running at the rated power limit. Then, subtract the maximum expected capacity from the initial task of the incremental equipment, and the difference is the incremental capacity to be allocated.

6. A remote control and dispatching system applicable to multiple scenarios in oil fields as described in claim 1, characterized in that, The initial task quantity is calculated based on the particle swarm optimization algorithm, including the following steps: Obtain the total production task, production time limit, and baseline production capacity, energy consumption, and rated power limit of each piece of equipment during the previous sampling period of the current cycle of the oilfield; Define the initial task quantity of each production equipment and form a decision variable vector. Set total quantity constraints, single equipment upper limit constraints and non-negativity constraints, and construct a dual-objective optimization function to maximize production efficiency and minimize energy consumption. The optimization algorithm parameters are initialized, the particles are encoded into the corresponding task allocation scheme, the population size is set, the initial particle positions and velocities are randomly generated, and the learning factor, inertia weight and maximum number of iterations are initialized at the same time. Calculate the dual objective function value of each particle, retain non-dominated solutions to form a Pareto optimal solution set, update the individual and global optima, update the particle velocity and position according to a preset formula, and adjust the task amount when the constraints are violated. Stop when the maximum number of iterations is reached or the objective function value no longer fluctuates. Preset weights according to actual needs, use linear weighting to score the Pareto optimal solution, and select the optimal solution as the initial task quantity.

7. A remote control and dispatching system applicable to multiple scenarios in oil fields as described in claim 1, characterized in that, Power redundancy is the difference between the current operating power and the rated power limit. Capacity redundancy coefficient is the capacity of the equipment operating at the rated power limit for the remaining time divided by the amount of unfinished tasks. Unit energy consumption coefficient is the ratio of the baseline energy consumption of non-incremental equipment to the unit capacity energy consumption, and then the ratio is divided by the current efficiency of the equipment.

8. A remote control and dispatching system applicable to multiple scenarios in oil fields as described in claim 1, characterized in that, The incremental allocation to be distributed to non-incremental devices is achieved through a dynamic coupling adaptation algorithm, including the following steps: The total number of unallocated increments is obtained by summing up the unallocated increments of all devices; based on three load indicators—power redundancy, capacity redundancy coefficient, and unit energy consumption coefficient—for non-incremental devices, a dynamic coupling adaptation algorithm is used to calculate the dynamic coupling weight of the three load indicators, and the task allocation ratio is determined according to the ratio of the dynamic coupling weight to the dynamic coupling weight of all non-incremental devices. Add the incremental task to the initial task of the non-incremental device. If the sum exceeds the maximum capacity of the device, allocate the excess to other non-incremental devices proportionally. Repeat the calculation until the constraint is met.

9. A remote control and dispatching system applicable to multiple scenarios in oil fields as described in claim 2, characterized in that, Outliers in equipment operating parameters are removed based on the Raida criteria. These parameters include actual operating power, cumulative operating time, and completed production capacity. The mean and standard deviation of these parameters are calculated separately. Parameters with standard deviations exceeding the mean by a preset positive or negative multiple are identified as outliers and removed.

10. A remote control and dispatching system applicable to multiple scenarios in oil fields as described in claim 1, characterized in that, The baseline capacity and baseline energy consumption of each production equipment during the previous sampling period are the actual capacity and total energy consumed by the production equipment under normal production conditions, when it operates continuously at the designed standard fixed power.

Images